Geology of the republic of Haiti

MISSING IMAGE

Material Information

Title:
Geology of the republic of Haiti
Physical Description:
Book
Language:
English
Creator:
Woodring, W. P. (Wendell Phillips)
The William L. Bryant Foundation - West Indies Collection ( Contributor )
Manufacturer:
The Lord Baltimore Press
Publication Date:

Subjects

Subjects / Keywords:
Caribbean
Geology -- Haiti   ( lcsh )
Spatial Coverage:
Caribbean

Notes

Summary:
At head of title: Republic of Haiti. Department of public works. Geological survey of the Republic of Haiti.
Acquisition:
.

Record Information

Source Institution:
University of Central Florida Libraries ( SOBEK page | external link )
Holding Location:
University of Central Florida ( SOBEK page | external link )
Rights Management:
All rights to images are held by the respective holding institution. This image is posted publicly for non-profit educational uses, excluding printed publication. For permission to reproduce images and/or for copyright information contact Special Collections & University Archives, University of Central Florida Libraries, Orlando, FL 32816 phone (407) 823-2576, email: speccoll@mail.ucf.edu
Resource Identifier:
QE223 .W65
System ID:
CA01200010:00001


This item is only available as the following downloads:


Full Text

PAGE 2

e 1 tam ryant oun ation _, .. West - South America est n 1es ection I \ -

PAGE 3

( LIBRARY BRYANT FOUNDATION WEST INDIES CENTER -

PAGE 4

, CONTENTS. INTRODUCTION, by w endell p. w oodring ................................ Arrangements for the reconnaissance ................................ Field work Office work ......... Acknowledgments PART I. GEOGRAPHY, by John S. Brown and Wendell P. Woodring ...... Physical geography ...... General relations .... Surf ace features ..... . . . General character ......... . Geographic provinces Drainage General features Drainage slopes and principal streams ...................... Peculiar drainage features .................................. Lakes Drainage of Gonave and Tortue islands ..................... R.elation of drainage to structure ........................... Climate Source of data. Temperature Precipitation Relative humidity Winds Vegetation Types of vegetation Forests Occurrence and general appearance ..................... Kinds of trees Associated v e getation Notes on local areas. Xerophytic vegetation Extent and general features ............................ Kinds of plants ............ Savannas Halophytic vegetation Vegetation of the shore line ................................ Economic geography Population Total population ........ Principal cities and towns .................................. Comparison of urban and rural population .................. Density of population Geographic factors influencing distribution of population ..... Agriculture General features 3 1Aom. 23 23 24 24 26 28 28 28 30 30 31 32 32 33 34 35 35 35 36 36 36 43 55 56 57 57 57 57 57 59 59 62 62 62 63 64 64 65 65 65 66 67 67 68 69 69

PAGE 5

.. I . I < ..

PAGE 6

4 CONTENTS. PAGE. PART I. GEOGRAPHY, by John S. Brown and Wendell P. WoodringContinued. Economic g eography Continued. Agriculture Continue d. Chief export crops of the colony and of the Republic. . . . . 70 Tobacc o and indigo . . . . . . . . . . . . . . . . . . 70 Sug a r . . . . . . . . . . . . . . . . . . . . . . . . 70 Coffee . . . . . . . . . . . . . . . . . . . . . . . . 71 Cotton . . . . . . . . . . . . . . . . . . . . . . . . 73 Cacao 74 Honey .............................. .-. . . . . . . . . 74 Crops grown for domestic consumption.... . . . . . . . . . 74 Live stock and poultry . . . . . . . . . . . . . . . . . . 7 5 Methods of farming and future of agriculture. . . . . . . . 76 Commerce . . . . . . . . . . . . . . . . . . . . . . . . . . 76 General features . . . . . . . . . . . . . . . . . . . . . 76 Exports . . . . . . . . . . . . . . . . . . . . . . . . . 77 Principal articles of colonial export ..................... Trend of changes in exports ............................ 77 78 Imports . . . . . . . . . . . . . . . . . . . . . . . . . 80 Principal countries trading with the Republic of Haiti. . . . 80 Manufactures . . . . . . . . . . . . . . . . . . . . . . . . . 81 Transportation . . . . . . . . . . . . . . . . . . . . . . . . 81 Highways and trails . . . . . . . . . . . . . . . . . . . . 81 Railroads . . . . . . . . . . . . . . . . . . . . . . . . 83 Telegraph and t e lephone systems. . . . . . . . . . . . . . . . 83 PART II. GEOLOGY, by Wende ll P. Woodring, John S. Brown, and Wilbur S. Burbank ................................................... 84 R eco nnaissance geologic map ....................................... 84 Sedimentary rocks, by W ende ll P. Woodring and John S. Brown ...... 84 Pale ozoic ( ?) metamorphic rocks ............................... 84 Cretaceous system ............................................ 85 I Lower Cretaceous ser i es .................................... 86 Description by r egions ................................ 86 Massif du Nord ................................... 86 Plaisance Valley .............................. 86 Les Trois Rivieres between Gros-Morne and Pilate 88 Near Dondon ................................ 88 Near Cerca-la-Source ......................... 89 Morn.e du Cap ............................... 90 Noires ................................ 92 Southern P eninsula ............................... 92 Arrondi ssement of J acmel ..................... 92 Near Petit-Goave ............................ 92 Arrondi ssement of Aquin ...................... 92 Arrondissement of Cay es ...................... 93 Arrondissement of Tiburon .................... 93 Upper Cretaceous series ................................... 93 D escription by regions ............... ................. 93 Massif du Nord ................................... 93 Near the Citadelle of Christophe ............... 93 Near La Tannerie ............................. 94 Morne Grand-Gille ........................... 95

PAGE 7

REPUBLIC OF HAITI DEPARTMENT OF PUBLIC WORKS Luc THEARD CoMMANnEnA. L. PAnsoNs,c.E.c., u.s.N. Secretary of State for Public Works Engineer in Chief GEOLOGICAL SURVEY OF THE REPUBLIC OF HAITI WENDEI.1. P. WoonRING, Geologist in Charge GEOLOGY OF THE REPUBLIC OF HAITI BY WENDELL P. WOODRING, JOHN S. BROWN AND WILBUR S. BURBANK PROPERTY OF D -\ PORT-AU-PRINCE 1924

PAGE 8

CONTENTS. p ART II. GEOLOGY, by Wen dell P. Woodring, John S. Brown, and Wilbur S. Burbank Continued. Sedimentary rocks, by Wend e ll P. Woodring and John S. BrownContinued. Cretaceous system Continue d. Upper Cretace ous series Continue d. Description by regions Continued. Montagne.s de Terre-N euve ....................... Massif de la Selle ................................. Massif de la Hotte ................................ Fossils ............................................... Tertiary system .............................................. Eocene s e ries ............................................. Middle Eocene ....................................... Plaisance limestone ............................... Name .............................. Areal distribution ............................. Stratigraphic r e lations ........................ Lithology .................................... Thickness .................................... Structure ..................................... Fossils ...................................... Upper Eocene ........................................ General features .................................. Ar e al distribution ............................. Stratigraphic r e lations ......................... Lithology .................................... Thickness .................................... Structure ..................................... Description by regions ............................ Mas sif du Nord ............................... M orne du Cap ........................... Near Dondon ............................. Arrondi ss em ent of Borgne ................. Northwest P e ninsula .......................... Bombardopolis Plateau ................... l\iI on tagne s d e Jean Rabel .................. Montagne s de Terre-N euve ................ Vall e y of Riviere d' Ennery .................... Montagnes Noires ............................ Northwestern part ........................ Southe astern part ......................... Chaine d e s Mateux ............................ Monta gn e s du Trou d 'Eau ..................... M ass if d e la S e lle .............................. Massif d e la Hotte ............................ Gonave I s land ................................ Fossils ........................................... 0 ligocene series ........................................... Lower Oligocene ..................................... Middle Oligoc ene ..................................... Description by r e gions .............................. Massif du Nord ............................... 5 PAGE 95 95 96 97 98 99 99 99 99 99 102 102 103 103 103 108 108 108 108 108 109 110 110 110 110 111 112 112 112 113 l 13 122 123 123 126 126 128 129 132 138 139 145 145 146 146 146

PAGE 9

- ee .So-cb ""ttimort {Preas BALTIMORB, KD., U. S. A. /

PAGE 10

6 CONTENTS. PART II. GEOLOGY, by Wendell P. Woodring, John S. Brown, and Wilbur S. Burbank Continued. Sedimentary rocks, by Wendell P. Woodring and John S. BrownContinued Tertiary system Continued. Oligocene series Continued. Middle Oligocene Continued. Description by regions-Continued. Northwest Peninsula .......................... Trois Rivieres Valley ...................... Montagnes de Jean Rabel .................. Montagnes de Terre-N euve ................ Montagnes Noires ............................ Chaine des Mateux ............................. Montagnes du Trou d'Eau ..................... Near J a cm el .................................. Fossils ........................................... Upp er Oligocene ...................................... Description by regions ............................ Tortue Island ................................. Arrondissement of Borgne ..................... Trois Rivieres Valley ......................... Borders of the Central Plain .................... Cbaine des Mateux ............................. Montagnes du Trou d'Eau ..................... Fossils ........................................... M . iocene series ............................................ General features ...................................... Areal distribution ................................. Stratigraphic relations ............................. Lithology ........................................ Thickness ........................................ Structure ......................................... Description by regions .................. .............. J ea.n Rabel Valley ................................ Trois Rivieres Valley .............................. Near Mole St.-N icolas .............................. Arbre Plain ....................................... F os:3ils ............................ Central Plain ..................................... . Artibonite group .............................. Madame Joie formation ................... Fossils ............................... Thomonde formation and Maissade tongue .. Southeastern part ..................... Northwestern part .................... Fossils, Thomonde formation .......... Fossils, Maissade tongue .............. Las Caho bas formation ................... of Las Cahobas formation ...... Fossils of Artibonite group ................. Artibonite Valley and Chaine des Mateux ........... Artibonite Valley ............................. PAGE. 147 147 147 147 148 148 148 148 149 151 151 151 151 151 152 153 153 154 157 157 157 157 158 158 158 158 158 159 160 160 161 161 161 162 163 165 165 168 173 191 196 200 205 206 206

PAGE 11

' CONTENTS. PART II. GEOLOGY, by Wendell P. Woodring, John S. Brown, and Wilbur S. Burbank Continued. Sedimentary rocks, by Wendell P. Woodring and John S. Brown Continued. Tertiary system Continued. Miocene series Continued. Description by regions Continued. Artibonite Valley and Chaine des Mateux Continued. Chaine des Mateux near St.-Marc ............... Momes des Guepes ........................ South of St.-Marc ......................... Valley of St.-Marc ........................ Southwest slope of Chaille des Mateux .......... Fossils ....................................... Southern edge of the Cul-de-Sac Plain .............. General feat urea ............................... Vicinity of Port-au-Prince and Petionville ...... Fossils ....................................... North coast of the Southern Peninsula .............. Morne-a-Bateau .............................. L'Acul ....................................... Grand-Goave ................................. Ta pion du Petit-Goave ........................ .. Near Baraderes ............................... Fossils ....................................... Commune of Jeremie .............................. Grande Riviere and Bras-a-Droit ............... South of Les Roseaux .......................... FossjJs ....................................... Between Port-Salut and Port-a-Piment ............. Asile Valley ...................................... Fossils ....................................... Camp Perrin ..................................... Fossils ....................................... Cayes Plain ...................................... .. B' ossi ls ....................... Gonave Island .................................... FossiJs ....................................... Pliocene series ............................................ Marine deposits ....................................... Valley of Riviere Gauche ......................... Ta pion du Petit-Goave ............................ N onmarine deposits .................................. Central Plain ..................................... Hinche formation ............................. Fossils .............................................. Quaternary system ............................................ General features .......................................... Marine deposits .......................................... Stratigraphic relations ................................ Lithology and local details ............................. Thickness ............................................ Structure ............................................. Fossils ............................................... 7 PAGE. 210 210 212 212 213 214 219 219 220 221 223 223 224 224 224 225 225 226 226 227 227 228 228 231 232 236 237 237 237 238 239 239 239 240 240 241 241 241 243 243 243 244 244 247 247 247

PAGE 12

8 CONTENTS. I PART II. GEOLOGY, by Wendell P. Woodring, John S. Brown, and Wilbur S. Burbank Continued. Sedimentary rocks, by Wendell P. Woodring and John S. BrownContinued. Quaternary system Continued. N onmarine deposits ....................................... Fossils ............................................... Extinct Quaternary mammals and birds .................... Igneous rocks, by Wilbur S. Burbank ............................... General distribution .......................................... Outline of the igneous geology ................................. Northern region .............................................. Extrusive rocks ........................................... General featur es and distribution ....................... Age of eruptions ....................................... Order of eruption ..................................... Earlier basaltic rocks .................................. Distribution and structural relations ................ Petrograpliy ..................................... Olivine-free basalts .............. ............. Hypersthene basalt ........................... Other basaltic rocks ............................. Alteration and metamorphism ..................... Andesites and dacites .................................. Distribution and structural relations ................ Petrography and chemical composition .............. Pyroxene (augite-hypersthene) andesites ........ Hypersthene andesites ........................ Augite-hypersthene-hornblende andesites ...... Hornblende-augite andesites .................. Hornblende andesites ......................... Hornblende-mica andesites or dacites ........... Pyroxene ( augite-hypersthene) dacites ......... Hornblende-augite dacites .................... . Hornblende-mica dacites ...................... Relations of the various types ...................... Alteration and metamorphism ...................... Later basaltic rocks .................................... Distribution and structurai relations ................ P etrogra phy ...................................... General features .............................. Ess.exi te ...................................... Analcite andesites and ana.lcite-olivine andesites .. Amygdaloidal basalts ......................... Zeolitized olivine basalts and diabases .......... Olivine diabas e .............................. Alteration ........................................ Relations of the various types ...................... Age of the lavas ................................... Intrusive rocks ........................................... General featur es and distribution ....................... Earlier basic intrusives ................................ Distribution and structural relations ................ PAGE. 255 256 257 260 260 262 265 265 265 266 267 268 268 270 270 271 271 272 272 272 274 274 277 277 277 278 278 278 279 279 279 280 280 280 281 281 282 283 284 284 285 285 286 286 286 286 287 287

PAGE 13

CONTENTS. PART II. GEOLOGY, by W e ndell P. Woodring, John S. Brown, and Wilbur S. Burbank Continue d. Igneous rocks, by Wilbur S. Burba.Dk Continue d. Northern regionr--Continu e d. Intrusive rocks Continue d. Earlier basic intrusive s Continue d. P etrography ...................................... Metadia base ................................. Augite peridotite ............................. Earlier quartz diorite group ............................ Distributio n and structura l ie lation s ................ Petrography and ch e mical compo sition of the normal quartz diorite ................................. . in or var1a t1 ons ................................. Quartz diorite ................................ Ophitic quartz diorite ......................... Daci te porphyries ............................ Kaolinized porphyry .......................... Qu a r t z diori t e porphyries ...................... Dike roc k s and v e ins .............................. Gene ral f ea tu r e s .............................. Alteration and m etamorphis m ...................... W eathering ...................................... Ag e of intrus i o n .................................. Dacite porphyry of the Montagne s Noire s ............... Distributio n and structural r e l a ti o n s ................ P etrogra phy ...................................... Correlatio n and a ge of the porphyry ................ Late r quartz diorite group of the Montagne s de T e rreN e u ve ....................................... Distribu t i o n and structura l r e lations ................ P etrog1aphy and ch e mi ca l c om p o s i t ion .............. Quartz diorite ................................ Granodiorite ................................. D a cite porphyri es ............................ Dike r o cks ................................... Aplit e s and p egmatites ........................ . R e la t ion s of the v a ri o u s types .......................... Metamorphic rock s e n tire ly or in part of igne ous origin ...... Distribution and g e n e r a l f eature s ....................... Amphibolit es and hornblend e schi sts .................... Talc sc hi s t ........................................... Chloritic sch ists ...................................... Summary of the r e lation s of t h e e x t ru s iv e and intrusive s e ri es. C entra l r e gion ................................................ G e n eral feature s and d ist r i bution of igne o us roc ks ........... Pre-Tertiary or l o w e r Eocen e basa l t ic ro c k s ................. P etrogra phy .......................................... Hype rsth e n e b asa lts .............................. Tuff ............................................. N ep h e lite b asa lts ........................................ . Distribution and struct u r a l r e lations ................... . Near Saut d 'Eau ........................... ....... N ear Tho m a z eau 9 PAGID. 287 2 8 8 2 8 8 289 289 290 293 293 294 294 294 294 295 295 296 297 299 299 299 300 301 301 301 302 302 303 305 305 306 306 306 306 308 309 309 310 312 312 312 312 312 314 314 314 314 3 15

PAGE 14

10 CONTENTS. PAGE. PART II. GEOLOGY, by Wendell P. Woodring, John S. Brown, and Wilbur S. Burbank Continued. Igneous rocks, by Wilbur S. Burbank Continued. Central region Continued. N ephelite basalts-Continue d. Petrography and chemical composition ................. N ephelite basalt ................................... Hai.iynite-nephelite basalt ............................ Zeolitiz e d nephelite basalt ......................... M elili te-nepheli te basalt ............................ Relations and origin of the lavas ....................... Age of the eruptions .................................... Miocene (?) basaltic rocks ................................ General features and structural relations ................ Petrography .......................................... Age and relations of the lavas .......................... Southern region .............................................. General features .......................................... 315 315 317 317 318 318 318 319 319 319 319 320 320 Late Cretaceous basaltic rocks. . . . . . . . . . . . . . . 320 Distribution and structural relations. . . . . . . . . . . 320 Massif de la S e lle. . . . . . . . . . . . . . . . . 320 Etang de Miragoane . . . . . . . . . . . . . . . . 321 Vicinity of A s ile Valley. . . . . . . . . . . . . . . 321 Vicinity of Aquin and St.-Louis du Sud. . . . . . . 323 Western part of peninsula. . . . . . . . . . . . . 323 Sources Chaudes . . . . . . . . . . . . . . . . . 323 Petrography and chemical composition ....... . . . . 323 Basalts . . . . . . . . . . . . . . . . . . . . . . 323 Albitized or spilitic basalts. . . . . . . . . . . . . 326 Diabase porphyries . . . . . . . . . . . . . . . . 326 Basic augite andesi tes. . . . . . . . . . . . . . . . 32 7 Alteration . . . . . . . . . . . . . . . . . . . . . . 327 Origin of the lavas. . . . . . . . . . . . . . . . . . . 328 Age of the lav as........................................ 329 Andesites . . . . . . . . . . . . . . . . . . . . . . . . 329 Distribution, structural r e lations, and age. . . . . . . . 329 Petrography . . . . . . . . . . . . . . . . . . . . . 329 Hypersthene andesites . . . . . . . . . . . . . . 329 Hornblende andesites . . . . . . . . . . . . . . . 330 Post-Eocene ( ?) basaltic rocks. . . . . . . . . . . . . . . 330 T ect onics, by Wendell P. Woodring. . . . . . . . . . . . . . . . . 331 General features . . . . . . . . . . . . . . . . . . . . . . . 331 Tectonic histozy . . . . . . . . . . . . . . . . . . . . . . . 332 Bearing of tectonics on geologic histozy of West Indies . . . . . 337 Earthquakes, by Wend ell P. Woodring. . . . . . . . . . . . . . . 338 Records available . . . . . . . . . . . . . . . . . . . . . . . 338 Disastrous earthquakes from 1551 to 1908. . . . . . . . . . . . 339 Earthquakes from 1909 to 1922. . . . . . . . . . . . . . . . . 343 Conclusions regarding frequ e ncy of shocks in different parts of the Republic . . . . . . . . . . . . . . . . . . . . . . . 346 Preca utions against damage. . . . . . . . . . . . . . . . . . . 348 Summazy of g e ologic histozy, by Wend e ll P. Woodring............... 350

PAGE 15

CONTENTS. PART III. GEOMORPHOLOGY, by Wendell P. Woodring, John S. Brown, and Wilbur S. Burbank ........................................ Tortue Island General relations Land features Upland forms Shore features Sublittoral features North Plain Extent and general features ..................................... Land features Undissected alluvial p1ain Dissected plain Low hills ................................................. Rock platform Shore features Sublittoral features ............................................. Massif du Nord Name and extent General features Land features Eastern part Central part Western part Morne du Cap ............................................ Drainage Shore features Sub littoral features Northwest Peninsula .............................................. Extent and general features .................................... Land features Trois Rivieres Valley .................................. Montagnes de Terre-N euve ............................ Montagnes du N ord-ouest ............................. Bombardopolis Plateau General features Emerged coastal terraces Arbre Plain Jean Rabel Valley ..................................... Shore features Sublittoral features Central Plain Name and extent Surface features N orthwestem part Southeastern part Drainage Montagnes Noires Name and extent General features Surface features Northwestern part 11 PAGE. 354 354 354 354 354 355 355 356 356 356 356 356 357 357 358 358 359 359 359 360 360 361 362 363 363 364 365 366 366 367 367 367 367 368 369 369 371 375 376 376 377 377 377 378 378 379 381 382 382 383 383 383

PAGE 16

12 CONTENTS. PART III. GEOMORPHOLOGY, by Wendell P. Woodring John S. Brown and Wilbur S. Burbank Continued. Montagnes Noires-Continued. Surface features Continu e d. Southeastern part Drainage Artibonite Plain and Artibonite Valley ............................. Name and general features .................................... Land features Artibonite Plain Artibonite Valley Drainage Shore features Suqlittoral features du Trou d'Eau ......................................... Name and extent .............................................. Surface features Eastern part ................ Western part Drainage Chaine des Mateux .. Name and extent. General Land features features Chaille d e s Ma teux proper ................................. Morne s de St.-Marc ....................................... Arcahaie Plain Drainage Shore featur e s Sublittoral featur e s Cul-de-Sac Plain Name and extent .............................................. General features Land features .. Outline and dominant e lem e nts. Shore features ... Sublittoral features Massif de Ia S e lle .................................................. Name and extent. General featur e s Land features Mountains North slope Interior Eastern part, including Montagne de la Selle ........ Western part South slope Plains Leogane Plain .. Drainage Shore and sublittoral f e atures. North coast Shore features Sublittoral features PAGm. 384 384 385 385 385 385 385 388 388 389 3 8 9 389 389 389 390 391 391 391 391 391 391 392 393 394 394 395 395 395 395 395 395 397 397 398 398 399 399 399 399 400 400 401 402 403 403 403 404 404 404 404:

PAGE 17

CONTENTS. 13 PAGE. PART III. GEOMORPHOLOGY, by Wendell P. Woodring, John S. Brown, and Wilbur S. Burbank Continued. Massif de la Selle Continued. Shore and sublittoral featuresN--Continued. South coast . . . . . . . . . . . . . . . . . . . . . . . . 404 Shore features . . . . . . . . . . . . . . . . . . . . 404 Sublittoral features . . . . . . . . . . . . . . . . . . 406 Massif de la Hotte ................................................ 406 406 Name and extent .............................................. General features . . . . . . . . . . . . . . . . . . . . . . . 406 Land features . . . . . . . . . . . . . . . . . . . . . . . . . 407 Eastern part . . . . . . . . . . . . . . . . . . . . . . . 407 North slope . . . . . . . . . . . . . . . . . . . . . 407 Interior 408 Asile Valley . . . . . . . . . . . . . . . . . . . 408 South slope . . . . . . . . . . . . . . . . . . . . . . 409 Western part . . . . . . . . . . . . . . . . . . . . . . . 4 1 O North slope . . . . . . . . . . . . . . . . . . . . . 410 Interior; Montagnes de la Hotte. . . . . . . . . . . . 411 South slope . . . . . . . . . . . . . . . . . . . . . . 412 Cayes Plain . . . . . . . . . . . . . . . . . . . 412 Port-Salut Peninsula . . . . . . . . . . . . . . . 412 Drainage Shore features ............................................ North coast 413 413 413 West coast . . . . . . . . . . . . . . . . . . . . . . 414 South coast 415 Sublittoral features . . . . . . . . . . . . . . . . . . . . . . 416 North coast . . . . . . . . . . . . . . . . . . . . . . . 416 West coast South coast Gonave Island .................................................... General relations Land f ea.tures Southeast half ............................................ Ridge along south coast ................................ Mapoux Plain ........................................ Interior plateau .............................. Dissected plateau along north coast .................... Northwest half ........................................... Drainage . . . . . . . . . . . . . . . . . . . . . . . . Shore features ............................................ Sublittoral features PART IV. MINERAL RESOURCES, by Wilbur S. Burbank, John S. Brown, and Wendell P. Woodring . . . . . .............................. Metals Introduction and summary by metals ........................... Mineral deposits of the Terre-Neuve district, by Wilbur S. Burbank and John S. Brown .................................... Introduction Location Geology and access 416 417 418 418 418 418 418 419 420 420 421 421 421 422 423 423 423 425 425 425 428 General features . . . . . . . . . . . . . . . . . . . 428 Sedimentary rocks . . . . . . . . . . . . . . . . . . 429

PAGE 18

14 CONTENTS. PART IV. MINERAL RESOURCES, by Wilbur S. Burbank, John S. Brown, and Wen dell P. Woodring Continued. Metals-Continued. Mineral deposits of the Terre-N euve district, by Wilbur S. Burbank and John S. Brown Continued. Geology Continued. Igneous rocks ......................................... General features and age ........................... Volcanic rocks .................................... Pyroxene ande sit e s ............................ Hornblende dacites and andesites ............... Bas altic rocks ................................ Mode of occurrence ............................ Intrusive rocks .................................... Quartz diorite and granodiorite ................. Dacite porphyries ............................. Structure ............................................. Folds ............................................ Faults and fissures ................................. Relations of structural f eatures ..................... History of mining developm ent ............................ G eneral charact e r and extent of d e posits .................... Contact meta morphic depo s its ............................. General f e a tu r e s ...................................... Feature s of the m etamorphis m ......................... Mineral composition of the tactite .................. Metamorphism of igneous rocks .................... Alteration of intrusive porpyhries .................. Al t eration of limestone to marble .................. Origin and f eature s of the mineralizing solutions ......... Distribution and structural f eatures ..................... Principal prospects ................................... Possibilities of finding new d e posits .................... Tenor of ore s ......................................... Cost of labor and transportation ........................ Conclusions .......................................... Vein deposits ............................................. General f e a t ures and distribution ....................... Veins n ear Rocher ..................................... Character and distribution ......................... Pros p ects ........................................ Tenor ............................................ Mine r a logy ....................................... Enrichm ent ..................................... Alteration of wall rock s ............................. V e ins near R a vine Jeanty .............................. Characte r and distribu t ion ......................... Prospect-s ......................................... Tenor ............................................ Mineralogy ....................................... Alteratio n of the country rock ...................... Veins at Dolan ........................................ Characte r and distribu tion ........................ Prospects ......................................... Character and origin of mine ralization .............. PAGE. 429 429 430 430 430 430 431 431 431 431 432 432 432 433 433 435 436 436 437 437 439 439 440 440 441 443 446 446 448 448 450 450 450 450 451 452 453 454 455 455 455 456 456 457 457 457 457 458 458

PAGE 19

CONTENTS. 15 PAGE. PART IV. MINERAL RESOURCES, by Wilbur S. Burbank, John S. Brown, and Wendell P. Woodring Continued. Metals Continued. Mineral deposits of the Terre-Neuve district, by Wilbur S. Burbank and John S. Brown Continued. Vein deposits Oontinued. V e ins east of Terre-N euve village. . . . . . . . . . . . 458 General conclusions as to veins. . . . . . . . . . . . . 459 Copper-bearing veins in pre-Tertiary rocks, by Wilbur S. Burbank and John S. Brown. . . . . . . . . . . . . . . . . . . 459 General features and origin. . . . . . . . . . . . . . . . . 460 Enrichment of primary veins. . . . . . . . . . . . . . . . 460 Tenor of veins and general conditions. . . . . . . . . . . . 461 Gen eral distribution . . . . . . . . . . . . . . . . . . . 461 Copper veins near Grande-Riviere du Nord. . . . . . . . . 462 Location and access . . . . . . . . . . . . . . . . . . 462 Geology . . . . . . . . . . . . . . . . . . . . . . . 463 History of development . . . . . . . . . . . . . . . . 463 V eins at habitation Zepiny. . . . . . . . . . . . . . . 464 Veins in Section Cormiers. . . . . . . . . . . . . . . 464 G e ner a l conclusions as to copper-bearing veins. . . . . . 465 Other localities . . . . . . . . . . . . . . . . . . . . . . 466 S ect ion Las Lomas . . . . . . . . . . . . . . . . . . 466 Plaisanc e and vicinity. . . . . . . . . . . . . . . . . 467 Limonade and vicinity . . . . . . . . . . . . . . . . 467 Jean Rabel . . . . . . . . . . . . . . . . . . . . . . 46 7 Iron deposits at Morne Beckly, by Wilbur S. Burbank and John S. Brown ............................................... Geography and geology ................................... Character and occurrence of ore ............................ 0 .. ngin ................................................... Economic value ........................................... D eposits of manganese, by Wilbur S. Bu1bank and John S. Brown .. Manganese deposits in the Commune of Gros-Marne ........ Geography and general geology ........................ Character and extent of faulting ........................ Mineralogy .......................................... Origin of the d eposits ................................. Conclusions as to character of veins .................... Manganese deposits in the Commune of Jacmel near Coteaux .. Residual concentrations of iron and manganese, by Wilbur S. Bur-bank and John S. Brown ............................... General features .......................................... Iron and manganese on the North Plain ..................... Iron and manganese near Paul ............................. Iron in the Southern Peninsula, by John S. Brown ........... Residual deposits on basaltic rocks ..................... Residual deR.osits on limestone ......................... Nonmetals, by John S. Brown and Wendell P. Woodring ............. L t lgill Lignite near Ma!ssade .................................... Lignite near Camp Perrin ................................. Relative heating value ..................................... 468 468 468 469 470 470 470 470 471 471 473 474 475 477 477 477 478 479 479 479 480 480 481 483 485

PAGE 20

;;a16 CONTENTS. PAGE. PART IV. MINERAL RESOURCES, by Wilbllr S. Burbank, John S. Brown, and Wen dell P. Woodring Continued. Nonmetals, by John S. Brown and Wendell P. Woodring Continued. Oil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 488 Possible oil resources of the Central Plain. . . . . . . . . . 488 Stratigraphy of the Miocene rocks. . . . . . . . . . . . 488 Structure of the Central Plain. . . . . . . . . . . . . 488 N 01thwestern part . . . . . . . . . . . . . . . . 488 Southwest limb of syncline. . . . . . . . . . . 488 Fond Bleu dome . . . . . . . . . . . . . 488 Northeast limb of syncline. . . . . . . . . . . 489 Plunging anticline between Maissade and Pignon ... ............................ 489 Southeastern part ................................ . 489 Southwest limb of syncline ..................... 489 West side ................................ 489 Thomonde anticline .................. 490 Chamouscadille anticline .............. 490 Ayaye anticline ....................... 491 South side ................................ 491 Belladere anticline .................. .. 491 Northeast limb of syncline ..................... 492 Structural terrace near Thomassique ........ 492 Possible source of oil ........................... ...... 492 Reservoirs ........................................... 492 Structt1re as affecting accumulation of oil ............... 492 Relations to fields near by ............................. 493 Conclusions as to the pos.sibilities of finding oil .......... 493 Possible oil in other regions ................................ 494 Road material ................................................ 494 Present state of road building .............................. 494 Tests of samples of road material ........................... 495 Summary of available material ............................. 497 Building stone ................................................ 498 Rock for concrete ............................................. 500 Lime A 500 Material for cement ........................................... 501 Clay for bricks ........................................... .... 503 Sand ......................................................... 507 Salt .......................................................... 509 Guano ....................................................... 510 p ART v. w ATER RESOURCES, by John s. Brown ........................... 513 In trod u cti on . . . . . . . . . . . . . . . ....................... 513 Surf ace and ground water supply ................................... 514 Cul-de-Sac Plain . . . . . . . . . . . . . . . . . . . . . ... 514 Importance ............................................... 514 Surface features .......................................... 515 Streams and springs ....................................... 515 I t' rnga I on ..................................... 517 Ground water ............................................ 518 Geologic features affecting ground water ................ 518 Wells and pumping plants ............................. 519 Wells of the Haytian-American Sugar Co ............ 519 Non-flowing well at Peyrard ....................... 520

PAGE 21

CONTENTS 17 PAGE. PART V. WATER RESOURCES, by John S. Brown Continued. Surface and ground water supply Continued. Cul-de-Sac p1ain Continued. Ground water Continued. Wells and pumping plants Continued. Flowing well near La Moriniere .................... 520 Flowing well at La Moriniere ...................... 520 Pumping plant at La Moriniere .................... 521 Pumping plant at Dessources ....................... 521 Flowing wells at Dessources ........................ 523 Pumping plant at Vaudreuil ....................... 524 Pumping plant at Drouillard ....................... 524 Abandoned well at '' Hasco '' mill ................... 524 Abandoned well at '' Hasco '' residences ............. 525 Value of springs ....................................... 525 Conclusions as to ground water in the Cul-de-Sac Plain ... 526 Arcahaie Plain ................................................ 527 Leogan e Plain . . . . . . . . . . . . . ..................... 528 Cayes Plain .................................................. 529 Plain .............................................. 530 Surf ace features .......................................... 530 Irrigation and flood control ................................ 531 Ground water ............................................ 532 Artibonite Valley .......... .................................... 533 Central Plain ................................................. 534 Arbre Plain . . . . . . . . . . . . . . . . . . . . . . . ..... 535 Trois Rivieres Valley and Jean Rabel Valley .................... 536 North Plain .................................................. 537 Surf ace f ea tu res .......................................... 537 Streams .................................................. 537 Utilization of water and flood control ....................... 537 Ground water ............................................ 538 Mountains and highlands ....................................... 539 Gona ve Island ......................... : ..................... 540 Tortu e Island . . . . . . . . . . . . . ...................... 541 Quality of water ................................................... 542 General features .............................................. 542 Streams, fresh lakes, and common springs ....................... 542 Wells . . . . . . . . . . . . . . . . . . . . . . . . . ...... 544 Salt lakes ..................................................... 545 Min er al springs . . . . . . . . . . . . . . . . . . . . . 547 Graphic comparison of analyses ........................... 547 Springs ........................................................... 550 General features and classification ............................. 550 Springs emerging from solution channels in limestone ............ 551 Con tact springs ............................................... 551 Artesian springs .............................................. 552 Springs in fractures in impervious rocks ......................... 553 Unusual types of springs ....................................... 553 Salty springs contaminated by sea. water .................... 553 Sulphur springs (Sources Puantes) ................... : ...... 554: Warm 8})rings ............................................ 557 Sources Chaudes or Eaux Boynes ....................... 558 Sources Chaudes de Los Pozos ......................... 562 2

PAGE 22

18 CONTENTS. PART V. WATER RESOURCES, by John S. Brown Continued. Surfac e and ground water supply Continued. Springs Continued. Unusual types of s prings Continu e d. Warm sp1ings of the Southern P e nin s ul a .... : .......... Sources Chau des de Dame-Marie or de Jeremie ..... Public water supplies .............................................. Port-au 1Prin ce ............................................... Purpose of inve s tigation ................................... Sources and distributing system ............................ Yield of springs ........................................... G e ologic f e atures and their r e lation to water supply ........... Notes on in di vi dual spring s ................................ S D . ource iq u1n1 ............................. > Source Chaudeau ..................................... Source Barron ........................................ Source Le Clerc ....................................... Source Turgeau ...................................... Sourc e Plais a nce ...................................... Sourc e C e risi e r ....................................... Source Carron ........................................ Source Millet ......................................... Source Bon Ami, Source d'Argent, and n earby springs ..... Othe r springs ......................................... Well water ............................................... Qu a lity of water and treatment for hardnes s ................ Surf ace water ............................................ Conclusions as to Port-au-Prince s upply ..................... C H t' a p-a1 ien .................................................. Purpose of investigation ................................... Principal features of pre sent s upply ........................ Geologic f e atures and their relation to wate r supply .......... Rock formations ..................................... Groups distinguished .............................. Igneous rocks .................................... Impure lim est on e and chert probably of Cretaceous ag e Uppe r Eocene limestone ............................ R e c ent alluvium .................................. Circulation of ground water ................................ Note s on present sources of supply ......................... . Source Cinq Carreaux .................................. Source d 'Aubry ....................................... Source Jean .......................................... Sources Ti penne and Source Georges ................... Sources du Buisson and nearby springs .................. Source Bois de Chene ................................. Source Belair ......................................... Source Mansuy ........................................ Additional springs available ............................ Well water .............................................. Wells in the city ....................................... Poss ible yield of wells near the city ..................... W e lls on the North Plain ......................... .... PAGE. 564 565 566 566 566 567 569 570 571 571 571 572 572 572 572 573 573 573 573 574 574 575 576 577 577 577 578 580 580 580 580 581 581 5 8 1 582 582 583 583 583 5 8 3 583 584 5 8 4 584 584 585 585 587 587

PAGE 23

CONTENTS. PART V. WATER RESO U RCES by John S. Brown Continued. Public w a t e r s upplies Continued. Cap-Ha!tie n Continued. Surface wate r ............................................. C o ncl usions as to C ap-Hai'.t i e n supply ....................... Note s on water supply of some othe r towns and villag e s ......... P ort-de-Paix ......................................... Mole St.-Nicolas ...................................... B aie de Henne ....................................... Anse Rouge .......................................... G .. ona1v e s ............................................. St.-Marc ............................................. L eogane ............................................... P eStel ................................................ J , e r em1e .............................................. Les C a y e s ............................................ J acme! ............................................... S a l t r o u ............................................... Grand-Gos i e r ......................................... Notes o n w a t e r p o w e r and d a m s ites ................................ N orlheastern p art ............................................ Pla isan ce Valle y .............................................. Rivi e r e Artibonite and tributaries .............................. Riv i e r e Coupe a l 'lnde ......................................... Grande Riviere du Cul-de-Sac ................................. Grande Riviere de L eogane .................................... G d R .. d J' 'm1e r a n e 1v1ere e e re ................................... BIBLIOGRAPHY, by w endell p. w oodring ................................. APPENDIX I. Some new Eocene Foraminifera of the genus Dictyoconus, by W end e ll P. Woodring ............................................. APPENDIX II. Some new middle Eocene and lower Miocene mollus ks, by Wende ll P Woodring and Wende ll C. Mans field ..................... ILLUSTRATIONS. 19 PA.GE. 589 5 8 9 590 590 590 590 591 591 592 592 592 592 593 593 593 593 593 593 594 594 594 594 595 595 596 608 611 PLATE. PAGE. I. G e ologic s k e t c h map of the Republic of Haiti .............. In pocket. II. G e ologi c m a p of the T erre-N euve region ................. In pocket. III. M a p s howing itinerary. . . . . . . . . . . . . . . . . . 24 IV. A, A mapou tree in M eme Valle y southe a s t of T erre-Neuve; B, Xerophytic vegetation in the Artibonite Plain southeast of Grande -Saline . . . . . . . . . . . . . . . . . . . . . . 58 V. A B a y ahonde thicke t in the Cul-de-Sa c Plain n ear Pont Beud et; B, Thicke t of Cercid ium pra e cox, a palo verde, about 5 kilometers northeast of Gonaives on the trail to Terre-Neuve. 62 VI. A, Savane La Cidra, about 10 kilometers southwest of St.-Michel de l'Atalaye; B, Salt bush, characteristic of soil, in the Artibonite Plain near Grande-Saline ; C, Halophytic vege-, t ation on the nort h coas t of Gona v e I s l and w es t of E troit.... 64 VII. A, Small mill for crushing sugar c ane; B, Method of drying corn. 74

PAGE 24

20 ILLUSTRATIONS. PAGE. VIII .A, Cretaceous ( ?) calcareous argillite on Le Trois Rivieres be tween Gros-Mome and Pilate; B, Pillow lava and limestone of supposed upper Cretaceous age on the north side of the Asile Valley . . . . . . . . . . . . . . . . . . . . . . . 88 IX. Characteristic Foraminifera of the Plaisance limestone (middle Eocene) .................................... . . . . . . 104 X. Some characteristic mollusks of the Plaisance limestone (middle Eocene) . . . . . . . .. . . . . . . . . . . . . . . . . 106 XI. A, Upper Eocene limestone near La Pierre; B, Conglomerate at base of upper Eocene limestone as exposed in Moulin near the trail from Gros-Morne to Terre-N euve.................. 108 -XII. A, Typical exposure of thin-bedded upper Eocene in the western part of the valley of Riviere d'Ennery; B, Thinbedded limestone of supposed upper Eocene age on the Grande Riviere dtt Cul-de-Sac; C, Chalky upper Eocene limestone containing bands and nodules of chert exposed in sea cliff near Saltrou. . . . . . . . . . . . . . . . . . . 122 XIII. Some characteristic upper Eocene Foraminif era. . . . . . . . 140 XIV. A, Thomonde formation on Riviere Bonde on the northeast side of the Central Plain; B, Las Caho bas formation near Las Cahobas on the south side of the Central Plain ; C, Miocene mar 1 near La Chapelle. . . . . . . . . . . . . . . . . . . 168 XV. Some characteristic mollusks of the Thomonde formation...... 176 XVI. Some characteristic mollusks of the Ma!ssade tongue and of the Artibonite group . . . . . . . . . . . . . . . . . . . . 192 XVII. A, Pitted Quaternary limestone (roche .. a-ravet) exposed in fifth emerged sea cliff on trail from Petit Paradis to Baie de Henne; B, Pleistocene conglomerate on the roaq leading up to Fort Nati onale, Port-au-Prince. . . . . . . . . . . . . 246 XVIII. A, Thrust fault in railroad cut south of Morne Deux Mamelles, northeast of Gonaives; B, Bedded volcanic debris on Sa vane Madame Michaud, southwest of Saut d'Eau. . . . . . . . 280 XIX. A, Quartz diorite from Marne Madeleine, south of Les Perches; B, Porphyritic quartz diorite from Mome Madeleine, south of Les Perches . . . . . . . . . . . . . . . . . . . . . 290 XX. A, Dacite porphyry from Savane La Cidra, Montagnes Noires; B, Granodiorite from Meme Valley, of Terre-N euve . . . . . . . . . . . . . . . . . . . . . . . . . 296 XXI. A, Photomicrograph of quartz diorite from Mome Madeleine, south of Les Perches; B, Photomicrograph of dacite porphyry from Sa vane La Cidra, Montagnes Noires.................. 300 XXII. A, Photomicrograph of granodiorite from Meme Valley, Com mune of Terre-N euve; B, Photomicrograph of glassy pyroxene andesite with perlitic texture from south slope of Morne Du-muraille, Commune of Terre-N euve. . . . . . . . . . . . 304 XXIII. A, Photomicrograph of basalt from the Massif de la Selle, north of Riviere GoS5eline; B, Photomicrograph of nephelite basalt from the vicinity of Maneville .................... . . . 316 XXIV. Tectonic trends of the Republic of Haiti. . . . . . . . . . . 332 XXV. A, Supposed fault scarp south-southwest of Cerca-la-Source; B, Fault scarp on the east side of the Gona1ves Plain. . . . . 334 XXVI. A, Asymmet rical anticline of Miocene and limestone at Savanette on Riviere Fer-a-Clreval ; B, Minor b igh-angle thrust fault in limestone of supposed upper Eocene age on the Grande Riviere du Cul-de-Sac.. . . . . . . . . . . . 336

PAGE 25

ILLUSTRATIONS. 21 PLATE. PAGE. LXVII. Geographic provinces of the Republic of Haiti. . . . . . . . 354 XXVIII. Emerged coastal terraces. A, East end of Tortue Island as seen from the channel to the south; B, Northeast end of Mole St.-Nicolas Bay; C, Cap St.-Nicolas. . . . . . . . . . . . 372 XX:IX. A, Accordant crests of strike ridges formed by conglomerates of the Las Caho bas f 01mation on the south side of the Central Plain; B, Stream terraces in the Central Plain, as seen looking southeastward along the trail from Hinche to Thomonde. . . 380 XXX. A, Gravel-covered terrace in the A1i.ibonite Valley near Savane a-Roche; B, Lowland of Miocene marl and ridge of Miooene corallif erous limestone near La Chapelle. . . . . . . . . . 386 XX.XI. Bathymetric map of Port-au-Prince Bay. . . . . . . . . . . 398 XX.XII. A, View of the Citronniers Valley looking northward from a locality a few kilometers north of the crest of the mountains along the trail from Jacmel to Leogane; B, Jagged limestone ridge on the north side of the Mapoux Plain, Gonave Island. 402 XX.XIII. Ore from contact-metamorphic deposits, Roche Glisse, Meme Valley . . . . . . . . . . . . . . . . . . . . . . . . . 43 B XXXIV. A, Ore from contact-metamo1phic deposits, Roche Glisse, Meme Valley; B, Secondary copper ore from enriched vein at Rocher . . . . . . . . . . . . . . . . . . . . . . . 448 XX.XV. A, Secondary copper ore from enriched vein at Rocher; B, Secondary copper ore from partly oxidized vein at Rocher. . 454 XX.XVI. Geologic sketch map of the Central Plain. . . . . . . . . . . 488 XX.XVII. A, Southwestward-dipping beds on the northeast side of the Central Plain near Thomassique; B, Main entrance to Citadelle of Christophe. . . . . . . . . . . . . . . . . . . 492 XX.XVIII. A, Clay pits at brick factory near I' Arcahaie; B, Drying shed and kiln at brick factory near l'Arcahaie; C, Pits for evapo-rating sea water to obtain salt near Grande-Saline. . . . . 506 XX.XIX. Topographic map of the Cul-de-Sac Plain........ . . . . . . 516 XL. A, Flowing well near La Moriniere in the Cul-de-Sac Plain; B, Large salt spring emerging from limestone of supposed upper Eocene age on the beach near Miragoane; C, Outlet of the Sources Puantes at the northwest corner of the Cul-de-Sac Plain . . . . . . . . . . . . . . . . . . . . . . . . . . 520 FIGURE 1. Map showing trend of geographic features of the West Indies.. 29 2. Curves showing monthly mean temperature for different stations 41 3. Curves showing monthly mean rainfall for different stations... 51 4. Generalized section showing fault zone at contact of argillite and volcanic rocks, as exposed in roadside ditch at Plaisance. 87 5. Section across the Chaine des Mateux. . . . . . . . . . . . 128 6. Generalized section of rocks exposed on Riviere Gosseline..... 129 7. Section across the northwest part of the Massif de !_Iotte. . 137 8. Section across the southeast part of Gonave Island........... 138 9. Diagrammatic section of foothills at Marne Madame Joie..... 162 10. Diagram showing lateral change in litl1ology of the Miocene rocks of the Central Plain and the transgressive overlap of the Thomonde formation . . . . . . . . . . . . . . . . . . 167 11. Diagram showing stratigraphic relations of the Ma.lssade tongue of the Thomonde formation. . . . . . . . . . . . . . . . 169 12. Gf-?n1 south side of the Artibonite Valley near La Chapelle, as expdseli Lon Riviere Delean. . . 207 13. Section f Miocene beds in the upper Artibonite Valley....... 208 LIBRARY BRYANT FOUNDATION WEST INDIES CENTER

PAGE 26

22 ILLUSTRATIONS. F1ounE 14. Section the Momes des Guepes. . . . . . . . . . . . 211 15. Composite generalized section of Miocene beds along the south side of the Cul-de-Sac Plain. . . . . . . . . . . . . . . . 219 16. Sketch map of the lignite area n ear Camp P errin. . . . . . . 232 17. Section of Miocene lignite-bearing beds exposed on La Ravine du Sud . . . . . . . . . . . . . . . . . . . . . . . . . 233 18. Generalized sections showing the relations of the igneous and sedimentary rocks in the Massif du Nord.. . . . . . . . . 311 19. Section across the Montagnes du Trou d 'Eau and the Mon-tagnes Noires from Maneville to Las Cahobas.............. 313 20. Generalized sections across the Southern Peninsula showing t .he relations of the basaltic lavas and sedimentary rocks........ 322 21. Section across the Cul-de-Sac trough and adjoining mountains. 335 22. Graph showing the number of earthquakes recorded at different stations from 1909 to 1922. . . . . . . . . . . . . . . . . 343 23. Subaqueous profiles off the north coast. . . . . . . . . . . . 365 24. Subaqueous profile off Pointe La Pierre near Gonaives. . . . 377 25. Subaqueous profiles off the west and south coasts of the Southern Peninsula . . . . . . . . . . . . . . . . . . . 417 26. Sketch map of the Terre-Neuve district showing the roads and principal trails . . . . . . . . . . . . . . . . . . . . . 426 27. Generalized sections showing the relations of the rocks in the Terre-N euve district . . . . . . . . . . . . . . . . . . 442 28. Section across Meme Valley showing the contact metamorphic deposits ......................................... . . . . 444 29. Geologic sketch map of the vicinity of Grande-Riviere du Nord showing the location of the larger copper prospects. . . . . 462 30. Sections showing the manganese deposits in the Commune of Gros-Morne . . . . . . . . . . . . . . . . . . . . . . . 4 71 31. Graph showing the heating value of Ma1ssade and Camp Perrin lignite as compared with other coals and wood.............. 486 32. Graphic representation of analyses of some waters of the Re-public . . . . . . . . . . . . . . . . . . . . . . . . . 548 33. Graphic representation of analyses of some waters of the Re-549 public .................................................. 34. Sketch map of the Sources Chaudes or Eaux Boynes. . . . . 559 35. Sketch map of the Sources Chaudes de los Pozos.............. 563 36. Sketch map showing the water supply of Port-au-Prince and its relation to the geology. . . . . . . . . . . . . . . . . . 568 37. Sketch map showing the water supply of Cap-Ha!tien and its relation to the geology. . . . . . . . . . . . . . . . . . 579

PAGE 27

GEOLOGY OF THE REPUBLIC OF HAITI. By WENDELL P. WooDRING, JOHNS. BROWN, and WILBUR S. BURBANK INTRODUCTION. By WENDELL P. WOODRING. ARRANGEMENTS FOR THE RECONNAISSANCE. On March 24, 1917, Hon. Frank L. Polk, Acting Secretary of State of the United States, addressed a letter to Hon. Franklin K. Lane, Secretary of the Interior, calling attention to Article I of the treaty of September 16, 1915, between the United States and the Republic of Haiti, which provides that the Government of the United States will by its good offices aid the Haitian Government in the proper and efficient development of its agri cultural, mineral, and commercial resources, and inquired whether it would be practicable for the United States Geological Survey to make a geological reconnaissance of the Republic of Haiti. In reply the Secretary of the Interior stated that the Geological Survey would cooperate by furnishing the scientific personnel, but that it had no authority \1nder which any part of the appropriations made by Congress for its maintenance could be expended outside the territorial limits of the United States. In later correspondence estimates of the expenditures necessary for a geological reconnaissance were submitted by the Secretary of the Interior. In the summer of 1919 plans for the reconnaissance submitted by the Director of the United States Geological Survey were approved by the Secretary of State for Public Works oi the Republic of Haiti and by Hon. John A. Mcllhenny, Financial Adviser to the Haitian Government, and Commander E. R. Gayler, C. E. C., U. S. N., Engineer in Chief of the Department of Public Works. In the meantime field work had been begun for a geological reconnais sance of the Domini can Republic under the supervision of the United States Geological Survey. Dr. T. W. Vaughan, of the United States Geo logical Survey, who was in charge of the work in the Dominican Republic, made an inspection trip from Ouanaminthe to Port-au-Prince and return in order to make more effective plans for the field work in the Republic of Haiti. The appropriation for the reconnaissance was made by the Council of State. The field work was delayed until the fall of 1920 in order that all 28

PAGE 28

24 GEOLOGY OF THE REPUBLIC OF HAITI. the information contained in the complete report on the geology of the Dominican Republic might be us e d to the greatest advantage. The arrangements for the work were made with Commander A. L. Parsons, C. E. C., U. S. N., who succeeded Commander Gayler as Engineer in Chief FIELD WORK. Messrs. John S. Brown, Wilbur S. Burbank, Frank G. Evans, jr., and I arrived in Port-au-Prince on Octob e r 1, 1920. We spent the succeeding six and a half months in the field and sailed from Port-au-Prince on April 15, 1921. During most of the time we sp ent in the field we worked in two parties, traveling by automobile, on horseback, a .nd on foot along the routes shown in Plate III and mapping the geology as we t ravel e d, thus making a reconnaissance survey of almost the entire Republic, including Gonave and Tortue islands. General geologic work was done by all the members of the expedition, but Mr. Brown gave particular attention to the under ground-water resources and to the field relations of the mineral deposits and igneous rocks, Mr. Burbank examined the mineral deposits and the igneous rocks, and I studied the stratigraphy and stratigraphic paleon tology and the lignite and oil resources. A detailed reconnaissance of several regions was made in order to ascertain the extent of the mineral deposits and the possibility of utilizing the underground-water resources. We had no adequate b a se map. The triangu lation, which was under the supervision of Mr. Gl enn S. Smith, of the United States Geological Survey, design e d to provide a base for a topo graphic survey of the Republic, bad not progressed far enough to per1nit us to utilize the results. We used the charts of the Hydrographic Office of the United States Navy for mapping regions near the coast, and we used also the maps of M. L. Gentil Tippenhauer, of Port-au-Prince, which cover some parts of the Republic. In other regions we were forced to rely on rough notebook sketches, in whi c h distance s were e stimate d only by the time consumed in traveling. Unless otherwise stated, altitudes are based on aneroid readings. While we were in the field }Ir. Brown prepared and submitte d r e ports on the public water supply of Port-au-Prince and of Cap-Ha! t ien. We also prepared a report suggesting improv e m ents in the water supply of some towns and villag e s, particularly those that now depend on salty water for their principal supply. These preliminary reports have been amplified and included in this volume. OFFICE WORK. As soon as we returned to Washington we b egan to write the text of this report, to examine the collections, and to pre pare m a ps and other illustra tions. The original agre ement provided that the colle c tions should be de posited in the United States National Museum, but that if a national

PAGE 30

INTRODUCTION. 25 muse11m should be established at Port-au-Prince a duplicate set would be sent there. The extensive collections of fossils were examined by the following paleontologists: Corals, Dr. T. W. Vaughan, of the United States Geologi cal Survey; Crinoidea, Dr. Frank Springer, of the United States National Museum; Echini, Dr. R. T. Ja.ckson, of Peterborough, New Hampshire; Bryozoa, Dr. R. S. Bassler, of the United States National Museum; Cre taceous Mollusca, Dr. T. W. Stanton, of the United States Geological Sur vey; Cirripedia, Dr. H. A. Pilsbry, of the Philadelphia Aca demy of Natural Sciences; decapod Crustacea, Dr. Mary J. Rath bun, of the United States National Museum; :fish, Prof. T. D. A. Cockerell, of the University of Colorado; birds, Dr. Alexander Wetmore, of the Biological Survey, United States Department of Agriculture; mammals, Mr. G. S. }liller, j1'., of the United States National Museum; plants, Prof. E. vV. Berry, of Johns Hopkins University. I undertook the examination of the larger Foraminifera, and in collaborat ion with Mr. W. C. Mansfield, of the United States Geological Survey, the examination of the Tertiary and Quaternary Mollusca and Brachiopoda. The smaller Foraminifera and several collections of ostracods and calcareous algae have not been ex amined. Chemical analyses of igneous rocks were made by Dr. H. S. Washington, of the Carnegie Institution of Washington. Samples of ore and of raw material for cement were anal)zed in the chemical laboratory of the United States Geological Survey under the direction of Mr. George Steiger, chief chemist. Otl1er samples of ore were analyzed and assayed b)" Ledoux & Co., of New York. Mr. E. V. Shannon, of the United States National lv[useum, analyzed a sample of kaolinite. Samples of lignite were ana lyzed at the Pittsburgh Laboratory of the United States Bureau of Mines. Samples of rock and other road material, and samples of sand and gravel for making concrete 'vere tested by the Bureau of Public Roads, United States Department of Agriculture. Samples of clay for making brick were tested at the ceramic station of the United States Bureau of Mines at Columbus, Ohio. Analyses of guano were made by the United States Department of Agriculture. Analyses of samples of water were made in the water-resou-rces laboratory of the Geological Survey under the direction of Mr. W. D. Collins. A preliminary account oi the possible oil resources of the Central Plain was published in July, 1922. A brief summary of the results of the recon naissance was published in the Journal of the Washington Academy of Sciences in April, 1923. These and other papers describing collections obtained during the reconnaissance or setting forth some of its results are listed below : BERRY, E.W., Tertiary fossil plants from the Republic of Haiti: U. S. Nat. Mlts. Proc., vol. 62, art. 14, 10 pp., 1 pl 2 text figs., 1922.

PAGE 31

26 GEOLOGY OF THE REPUBLIC OF HAITI. COCKERELL, T. D. A., A fossil cichlid fish from the R epublic of H aiti: U. S. Nat. Mus. Proc., vol. 63, art. 7, 2 pp., 1 pl., 1923. MILLER, G. S., JR., Remains of mammals f1om cave s in the Republic of Haiti: Smithsonian Misc. Coll., vol. 74, No. 3, 8 pp., 1922. PILSBRY, H. A., Miocene Cirripedia from the R epublic of Haiti: U. S. Nat. Mus. Proc. (Awaiting publication. ) RATHBUN, M. J., Fossil crabs from the R epublic of Haiti: U. S. Nat. Mus. Proc., vol. 63, art. 9, 6 pp., 2 pis., 1923. SPRINGER, FRANK, A new Tertiary crinoid in the W es t Indies: U. S. Nat. Mus. Proc. (Awaiting publication.) WETMORE, ALEXANDER, R emains of birds from ca v e s in tl1e R epublic of Haiti: Smithsonian Misc. Coll., vol. 74, No. 4, 4 pp., 2 text figs., 1922. WoooRING, W. P ., Middle Eocene Foram.inif era of the g enus D ic tyoconus from the Republic of Haiti: Washington Acad. Sci. Jour., vol. 12, pp. 244-247, 1922. --Stratigraphy, structure, and possibl e oil r esource s of the Miocene rocks of the Central Plain: Republic of Haiti G e ol. Survey, 19 pp., m ap, 1922. --Tectonic features of the R epublic of Haiti and their b earing on the geologic history of the West Indies (abstract): Washington Acad. Sci. Jour. (Awaiting publication.) --An outline of the results of a geological r e connaissance of the R epublic of Haiti: Washington Acad. Sci. Jour., vol. 13, pp. 117-129, 1923. --Tertiary mollusks of the g enus Orthaulax from the R epublic of Haiti, Porto Rico, and Cuba: U. S. Nat. Mus. Proc.: vol. 64, art. 1, 12 pp., 2 pls., 1923. ACKNOWLEDGMENTS. In the field work and in the preparation of the report we have had the unfailing support and encouragement of Commander Parsons, under whose direction the work was done. The field work and mo s t of the office work was done under the sup e rvision of Dr. T. W. Vaughan while he was chief of the sections of Coastal Plains investigations and of West Indian Geological Surveys of the United State s Geological Survey. Doctor Vaughan has reviewed the entire r eport, and we have thus had the ad vantage of his wide knowledge of the g e ology of the West Indies and regions near by. We wish to record our deep appreciation of the s e rvices of Rev. J. Scherer, Directeur de l'Observatoire Meteorologique du Semi naire-College St.-Martial. In an unassuming way and almost without the knowledge of the general scientific world he has for years been collect ing and publishing invaluable meteorological and seismologi cal data. The matter on climate and earthquakes in this report could not have b ee n written without M. Scherer's records. We have used with great advantag e the published geologic maps of M. L. Gentil Tippenhauer, of Port-au Prince. Only those who have traverse d the rugge d mountains of the Republic can appreciate the labor that the se maps repres ent. Mr. E. L. McN air, in charge of the triangulation party, and the engineers of tl1e Department of Public Works cooperated with us in every way. Mr. R. A. Conard, engineer of the Haytian-American Sugar Co., furnished logs of wells in the Cul-de-Sac Plain and other information concerning them. Mr. Frank G. Evans, jr., one of the members of the party, gave valuable

PAGE 32

INTRODUCTION. 27 help in traversing the country and in making geologic examinations. Dr. W. F. Jones, of the MassaC;husetts Institute of Technology, gave us the advantage of the results of his work in the Republic in 1918. Many oth ers aided us in different ways. We often enjoyed the hospitality of the parish priests and of the officers and men of the United States Marine Corps and of the Gendarmerie d'Haiti. Everywhere we were received with unfailing cordiality, and we will long remember our pleasant experiences Several geologists of the United States Geological Survey have re viewed parts of this report. Mr. E. S .. Larsen, jr., reviewed the descrip tions of the igneous rocks; Mr. H. G. Ferguson and Mr. D. F. Hewett the d escriptions of the mineral deposits; and Mr. N. C. Grover, Mr. 0. E. Meinzer and Mr. W. D. Collins the matter on the wai:er r esources The titles of the illustrations were edited by Mr. F. E. 1Yiatthes, of the United States Geological Survey. The text was translated into French by Mr. M. C. Delporte, and the translation was edited by J. J. de Gryse. The illustrations were drafi:ed by Mr. Lewis B. Pusey, of the United States Geological Survey. ..

PAGE 33

' PART I. GEOGRAPHY. ----By JOHN S. BROWN and W ENDELL P. WooDRING PHYSICAL GEOGRAPHY. GENERAL RELATIONS The most striking thing in the morphology of the West Indies is the arrangement of its geographic f e atures in arcs. Most of the arcs are convex northward, but in Haiti and in islands farther east some of them are convex southward. These arcs, wh i ch are seen in the trend of tl1e islands, of the moun tain range8 on the islands, and of the ridges and troughs in the submerged areas, are shown in Figure 1, a map based on one re ce11tly published by Professor Taber,li to whom we are inde bted for its use. The island of Haiti, the largest of the W est Indian islands ex cept Cuba is between Cuba and Porto Rico. It lie s between parallels 17 39' and 20 north latitude, and meridians 68 20' and 74 30' west of Greenwich. The Atlantic Ocean borders it on the north and the Caribbean S e a on the south. It is separated from Cuba by the Windward Passag e and from Porto Rico by the Mona Passage. The submerged platform in the Mona Passage is relatively shallow, the maxim11m depth of wate r in the central part being but 260 fathoms ( 47 5 meters). The Bartlett Deep, the most remarkable of the deep troughs of the West Indies, extends into the Windward Passage, where the floor of the sea pl11nges southwe stward from a depth of 893 fathoms (1,633 meters) to 1,737 fathoms (3,177 meters). A submerged ridge extends westward :from the southern penin sula of the Republic of Haiti beyond the east ern end of a similar ridge tha. t extends westward from Jamaica. These two ridges are separated by a trough having a maximum depth of 1,573 fathoms (2,817 meters). Between the island of Haiti and the Bahama Banks is an unnamed trough having a maximum depth of 2,388 :fathoms ( 4,367 meters). South of the island lies the deep basin of the Caribbean Sea. The Republic of Haiti occupies approximately the western third of the island and the Dominican Republic the eastern two-thirds. The maximum length of the Republic of Haiti is about 295 kilometers, and its width near the Dominican border is about 183 kilometers. The area of the 1 Taber, Stephen, The great fault troughs of the Antilles: Jour. Geology, vol. 30 pp. 89-114, pl. 1, fig. 6, 1922. 28

PAGE 34

' ' \ ... ' \ \ I '-' t \ \ : \ \ \ \ \ I \ \ I \ GEOGRAPHY. I : 1 \\ lt, l I : : I \ t I \\ \ \ \ 0 .... . ' . \ ,_1 ,,,, ' ,, ft\\\ """ \ ':t'''\ V) \ . .. ... \ ,' : I I \ : I I \ : I I . . : : \ / .. : I I I I I 1' } (/) \ ', ', ,, \ \ .... \ ' \ I \ I l 0 \ t \ ,. .... ,,, I \ ,1\ \\\I 0 '1 ,, \ :1,, \ '' I "''\ f \ \ 'I \' \ .. '--- ell Q,> = - 29 ..

PAGE 35

30 GEOLOGY OF THE REPUBLIC OF HAITI. Republic, including Tortue Island, Gonave Island, Grande Cayemite, and Ile A Vache, is about 27, 700 square kilometers, according to planimeter measurements of a map compiled by the Service des Leves Topographique. The precise area of the Republic is not known, as the boundary between the two Republics has not been definitely fixed. The two prominent westward-extending peninsulas, the northwest peninsula and the southern peninsula:, embrace an extensive body of water in which lies Gonave Island SURFACE FEATURES. GENERAL CHARACTER. Haiti is very mountainous. Towering mountains are visible from tl1e sea in front of any of the open ports of the Republic, and at many ports steep mountain slopes extend down to the coast. The traveler who is a ccustomed to wide plains can scarcely believe that a population so large as that of the Republic can live in a co11ntry so rugged, yet the exports from which the Republic largely derives its revenue consist principally of agricultural products. The ruggedness of the mountains is even more impressive to one who travels along the trails that extend into the heart of all the mountains. The mountain slopes at moderate altitudes yield immense crops of coffee, which forms the leading article of export. The mountain slopes at all altitudes are cultivated in small gardens, which yield the vegetables and fruits that are sold at inn11merable market places, which form the most characteristic feature of the commercial and social life of the ru,ral districts. Large areas of the plains that flank the mountains in some parts of the Republic are semiarid. The rural population is concentrated in the fertile mountain valleys and in the parts of t11e plains that through natural hea, y rainfall or the application of water by irrigation yield sugar cane, cotton, and other crops. Although a large part of the Republic is mountainous, extensive plains flank the mountains at some places along the coast or extend like wedges into the mountainous regions. Many of the mountains consist of beds of limestone, and steep slopes that are scarred by high cliffs form the most characteristic ieature of these mountains. These cliffs appear to be abnormal features; they were probably not formed by the usual processes of subaerial erosion. They are confined almost entirely to regions where the limestone is massive, regardless of its age. The base of the cliffs is determined by the outcrop of strata of thin-bedded limestone or other kinds of rocks. The cliffs have apparently been formed by the l1ndermining and subsequent stoping of blocks of massive limestone by ground water, which has penetrated the rock and along the contact with the 11nderlying beds. The contrast between the surface features produced by the erosion of massive and of thin-hedded limestone is clearly shown on the north and south slopes of Mont Puilboreau, the limestone range between

PAGE 36

GEOGRAPHY. 31 Ennery and Plaisance. The massive limestone on the north slope is scarred by cliffs that attain a height of several hl1ndred meters On the south s lope, where the limestone is thin-bedded, the dissection is more intricate and there are no cliffs GEOGRAPHIC PROVINCES. The following geographic provinces are recognized in this report. The boundaries between these provinces are s hown approximate ly on Plate XXVII, page 354. Tortue Island. North Plain. Massif du Nord. Northwest Peninsu la. Central Plain. Montagnes Noires. Artibonite Plain and Valley. Montag n es du Trou d'Eau. Chaine des Mateux. Cul-d e-Sac Plain. Massif de l a Selle Massif de la Hotte. Gonave Island. As the topographic features of these provinces are deter rnined by t he pl1ysical and structural features of their surface rocks and by thei r geo logic history they are fully described in Part III, which fol lows the text describin g the stratigraphic and structural geo logy and the geolog ic his tory. The surface features are briefly s11mmarized here in order to furnisl1 the geographic setting for the description of the geology Tortue Island is an imperfectly dissected plateau, the margins of wl1ich are modifi ed by late emergence The North Plain extends along the north coast from Acu l Bay eastward to the Dominican border Between Acul Bay and Cap-Ha1tien the plain is shut off from the sea by Morne du Cap, an outlier of the 1viassif du Nord. The Massif du Nord, which is the northwestern prolongation of t he Cordillera Central of the Dominican Republic, extends from the Domini-. can border northwestward to the deep trough of the valleys of Riviere ]a Quinte and Les Trois Rivieres north of Gros-Morne . The Massif du Nord forms the western part of the arc that extends across the island. The entjre massif is mountainous and very rugged The Northwest Peninsula embraces the entire peninsula west of the valleys of Riviere la Quinte and Les Trois Rivieres This region contains a v ari ety of surface features, including mountains, lowlands, and an ex tensiv e plateau, as follo\vs: Montagnes de Terre-N euve, }tlontagnes du Nord-ou est, Arbre Plain, Jean Ra bel Valley, Bombardopolis Plateau. The Monta gnes du Nord-ouest and the Montagnes de Terre-Neuve form a short arc that is convex northward. The most striking features of the peninsula are the magnificent emerged terraces that border t he oute1 margin s of the Bombardopolis Plateau. The Central Plain, which is the only extensive interior p lain in the Republic, extends from the Dominican border northwestward to St.-

PAGE 37

32 GEOLOGY OF THE REPUBLIC OF HAITI. Michel de l'Atalaye as a wedge between the Massif du Nord and the tagnes Noires. It is the northwestward prolongation of the San Juan Valley of the Dominican Republic. The Montagnes Noires form a mountain system that is the northwestern prolongation of the northern part of the Sierra de N eiba of the Dominican Republic. At its northwest end it merges into the Massif du Nord forming a short arc that is convex southward The Artibonite Plain is a wedge extending southeastward between the Montagnes Noires and the ChaSne des Mateux together witl1 tl1e Mon tagnes du Trou d' Eau. The apex of the wedge is near the Dominican border, where the Montagnes Noires join the Montagnes du Trou d'Eau. The southeastern part of the Artibonite Plain, to which the name Arti bonite Valley is applied, has a greater variety of surface features than the northwestern part the plain proper. Morne Grammont is an isolated outlier of Montagnes Noires in the plain southeast of Gona!ves. The Chaine des Mateux and its southeastward prolongation, the Mon tagnes du Trou d'Eau, extend southeastward from St.-Marc to the Domin ican border as an arc that is convex southwa .rd and for10 the prolongation of the southern part of the Sierra de Neiba of the Dominican Republic. The Cul-de-Sac Plain is part of a remarkable depression that extends across tl1e island from Port-au-Prince Bay to N eiba Bay. The Dominican part of the depression is called the Hoya de Enriquillo. The plain is bounded on the north by the Montagnes du Trou d'Eau and on the south by tbe Massif de la Selle. It contains the largest inland body of water in the Republic, the Etang Saumatre, which has no outlet. The Massif de la Selle, which is named from Mont la Selle, the highest peak in the Republic, is the northwestward prolongation of the Sierra de Bahoruco of the Dominican Republic. It extends westward to the gap in the mountains along a line between Jacmel an d Grand-Goave and in cludes a large coasta l plain, the Leogane pla in .. The Massif de la Hotte embraces the entire southern peninsula west of a line between Jacmel and Grand-Goave, where it merges into the Massjf de la Selle. This region includes a large coastal plain, the Cayes Plain, and several interior lowlands, the largest of which is the Asile Valley. The remainder of the region consists of rugged mountains. The de la Selle and the Massif de la Hotte form an arc convex northward. Gonave Island is the largest outlying island belonging to the Republic. The southeastern half of the island is more rugged and has a greater variety of surface features than the northwestern half. DRAINAGE. GENERAL FEATURES. The Republic of Haiti has a great many small streams. Most of them have short courses, and many of them flow directly down steep mountain

PAGE 38

GE OGRAPHY. 33 sides. All streams that contain water during most of the year in any part of their c ourses are called rivieres, but this term does not have the signifi cance of size commonly attached to it in other col1ntries. Considerably more than 10 0 such streams enter the sea along the coast of the Republic. Not more than a third of these, however, are large streams,1 and o f these Riviere Artibonite is the largest. Most of the streams are swift, shallow, and full of rapids, especia lly i n the mountains, and the water is usually clear except during or immediately after rains. Riviere Artibonite, however, contains no rapids in it. s lowe r course, possibly as far up as La Chapelle, and near the sea it becomes rather s luggish. In this region it is also somewhat turbid all the time. Riviere de l'Estere is tidal for several kilometers above its mouth, and a few other streams are tidal for as much as a kilometer or a little more. In colonial days large barges were towed up Riviere Artibonite as far a.s Petite-Riviere de l' Artibonite. The other streams are navigable only for canoes or very small boats and generally not more than a fe w kilometers above their mouths. DRAIN AGE SLOPES AND PRINCIPAL STREAMS. The streams of the Republic may be placed in four groups according to their outlet and general direction of flow. The groups and their principal members are as follows: 1. Streams flowing into the Atlantic Ocean: Riviere Massacre, Grande Riviere du Nord, Riviere du Limbe, Les Trois Rivieres, Riviere de Jean Rabel. 2. Streams flowing westward into the Gona!ves Gulf: Riviere de l'Estere, Riviere Artibonite, Grande Riviere du Cul-de-Sac 3. Streams flowing northward into the Gona1ves Gulf: Grande Riviere de Leogane or Riviere Momance, Grande Riviere de Nippes, Riviere des Baraderes, Grande Riviere de Jeremie or Riviere de la Grande Anse. 4. Streams flowing southward into the c aribbean Sea: Riviere de I' Acul, La Ravine du Sud, and Riviere de l'Ilet on the Cayes Plain, Riviere de Cavaillon, Grande Riviere de J acmel, Riviere de Fesle, and Riviere Pedernales. The fo lio .wing list gives the approximate lengths of the principal streams, all minor sinuosities being omitted. No tributaries of Riviere Artibonite except Riviere Guayamouc are included. The figures are based on measurements made on two or three maps and data published by Tippenhauer and others. Though admittedly inaccurate the relative order o f lengths is probably correct. 1 Fortunat, in his Nouvelle de l'lle d'Haiti, pp. Paris, 1888, gives a list of forty-three principal rivers of the Republic, of which eight are tributaries of Artibonite. 3

PAGE 39

34 GEOLOGY OF THE REPUBLIC OF HAITI. Approximate length of principal streams. Kilometers. Artibonite to headwaters in Dominican Republic 1 236 Artibonite to farthest headwaters on Riviere Tenebres near Cari ce 1 262 Guayamouc, from Bouyaha to headwaters of Riviere Doree near Marmelade . . . . . . . . . . . . . . . . . . . . . . . 108 Le T R .. 96 s roIS 1v1eres ........................................... Grande Riviere du Nord. . . . . . . . . . . . . . . . . . . 82 Estere . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 Grande Riviere des Nippes and Riviere Serpent. . . . . . . . 60 Grande Riviere de Leogane or Momance. . . . . . . . . . . . 56 R .. M 55 1v1ere assacre ........................................... G d R . d J, 50 ran e i vi ere e erem1e .................................. Grande Riviere du Cul-de-Sac. . . . . . . . . . . . . . . . . 45 Riviere Artibonite is by far the largest of these streams and has the only extensive and complex drainage system. In the Republic of Haiti it drains an area of about 7,800 square kilometers, approximately three tenths of the whole country, and in addition about 1,800 square kilometers of Dominican territory.1 Its principal tributary, Riviere Guayamouc, drains about 2,675 square kilometers, all in the Republic of Haiti, and is probably the co11ntry's second largest stream. I ies Trois Rivieres undoubtedly is next in order of size, but the others can not easily be placed without more accurate maps and streamflow data. Grande Riviere de Nippes, Riviere de Cavaillon, Grand Riviere de Jeremie, Grand Riviere du Nord, and Grand Riviere de Leogane are the largest streams except the tributaries of the Artibonite. PECULIAR DRAIN AGE FEATURES. Certain areas have peculiar drainage, due to geologic or climatic conditions. Underground drainage in limestone is one of the commonest of these peculiar features. In areas of 11ndergro11nd drainage the surface water disappears into sink boles or caves in limestone and there are no surface streams. The best example of such an area is on the northern slope of the southern peninsula from Baraderes to Jeremie. This region is full of sink holes that have no outlets and contains few surface streams. Some of the surface streams disappear into limestone caverns; others, such as Riviere Salee, west of Baraderes, appear suddenly from caverns. Areas of sink-hole dra inage are common in regions where limestone is the surf ace rock, but most of them are not very large. Lost streams of the desert type, which disappear by evaporation or by infiltration into allu vium, are rather common jn some areas. Usually they originate on moun-1 Tippenhauer and others have given the length of the Artlbonite as 320 kilometers. This length must include many minor sinuosities, which in the lower course are very numerous. 2 Areas measured by planlmeter on Thomasset map. Although the map is not very accurate, the errors probably very nearly compensate one another.

PAGE 40

GEOGRAPHY. 35 tain slopes, where rainfall is abundant, and disappear at or near the arid lowlands. Riviere du l\1ole, at l'lole St.-Nicolas, which disappears 2 kilo meters from the sea, is a good example. Other examples are Riviere Blanche, which disappears on the Cul-de-Sac Plain, and, to some extent at J13ast, Grande Riviere du Cul-de-Sac. These streams flow through to the sea, however, in time of flood. Some streams of this type are interrupted streams. They :flo w perennially in certain stretches where the structure of the rock b ,rings the water to the surface and disappear in other stretches where the gravel fill is deep. Riviere Colombier at Terre-N euve is a good example. \ LAKES. The Republic of Haiti contains a considerable number of small lakes, most of which occupy sink holes in limestone Some of them are perma .. nent; others are ephemeral. There are also a few small lakes of the desert-playa type, particularly on the Arbre Plain. Perhaps the Etang Bois-Neuf, south of St.-Marc, also is of this type. There are two rather large lakes, the Etang Saumatre, at the eastern end of the Cul-de-Sac Plain, and the Etang de Miragoane. The area of the Etang Sa11matre is about 180 square kilometers and that of the Etang de Miragoane perhaps 25 square kilomete rs.1 The Etang de overflows to the sea, but the Etang Sa11matre has no outlet. West of Etang Saumatre is a shallow lake called Trou cruman, which occasionally overflows to the sea. DRAINAGE OF GoN.AVE AND ToRTUE ISLANDS. Gonave and Tortue islands are characterized generally by ltnderground drainage in limestone. A few subterranean streams break out as springs and flow for short distances, but no perennjal surface streams reach the sea. RELATION OF DRAINAGE TO STRUCTURE. The adjustment of the drainage to the structure of the rocks is dis cussed in Part III separately for eac h of the geographic provinces. Riviere Artibonite and Les Trois Rivieres are the only streams that traverse more than one geographic province, except many smaller streams that rise in mountains and flow into the larger plains. The Guayamouc and Artibonite drainage is closely adjusted to the structure in its upper and lower part.s, but the middle part, between Las Cahobas and Mire .. balais, is unrelate d to the structure. Riviere Guayamouc flows southeast ward in a plunging synclinal trough, and Riviere Artibonite below Mire .. balais flows northwestward in a similar trough. The diversion of Riviere Artibonite across the Montagnes Noires is discussed on pages 381-382. Les .Trois Rivier es flows along the strike of the rocks from its headwaters 1 Areas measure d by planimeter trom Thomasset map. The measurement gives 29 square kilometers for Etang. de Miragoa.ne, but this probably is too great.

PAGE 41

36 GEOLOGY OF THE REPUBLIC OF HAITI. to Pilate and then cuts obliquely across the strike. At Gros-1.ilorne it enters a downfaulted trough and follows this trough northward to the sea. Detailed descriptions of the drainage with regard to water supply and power development are given on pages 513-595. CLIMATE. SOURCE OF DATA. Nearly all the data regarding climate are taken from the bulletins of the Observatoire Meteorologique du Seminaire-College St. Martial,1 which have been published sem.iannually for the period from July, 1909, to the end of 1916, and annually since that time. The data contained in the bulletins were collected under the supervision of J. Scherer, Directeur de l'Observatoire. Records of rainfall and temperature at many stations ex tend back a few years prior to 1909 but are summed up in the Bulletin Semestriel for July to December of that year. Records at Port-au-Prince covering a much longer period of observation also are summarized in dif ferent issues of the bulletin. Data from 1'foreau de Saint-Mery 2 that are definite enough to be of value are used to a small extent to supplement those taken from the bulletins. TEMPERATURE .All parts of the Republic have a warm and notably equable tempera ture. Frost, snow, and ice probably do not form anywhere in the Republic,8 even at the highest altitudes, although the temperatures on the high mountain ranges are appreciably less than those at lower altitudes. Most of the stations at which temperature observations have been made are at low altitudes on plains, in valleys, or near the sea. The records made represent the conditions under which the greater part of the population of the Republic lives, but they do not give an entirely correct idea of the purely physical aspect of the climate. Only one station, Furey, is located at a really high altitude, and the records from this station, though incom plete, are sufficient to show a considerably lower temperature than that prevailing on the lowlands. The following tables give the mean monthly temperatures at a number of well-distributed stations over the period for which records are available. The averages of the monthly means for the years of record available also are given. l Imprimerie Nationale, Port-au-Prince. The bulletin for 1919, published in 1920, was the most recent one used in compiling the tables. The bulletins for 1920 and 1921 have been published in the meantime. 2 Moreau de L. E., Description topograpbique, physique, clvile, et polittque de la partie de l'isle Saint-Domingue, 2 vols., Philadelphia, 1797-98. 8 Moreau de (vol. 2, p. 505) states that in the Canton of Fond Jean-Noi!l, northwest of Sal-Trou, near the crest ot Montagne de la Selle, '' one experiences a sensa tion ot coldness and sometimes one even sees a kind of frost.'' His statement is so qualified as to be rather doubtful, considering the lack ot confirmatory evidence

PAGE 42

Year. Jan. 1908 1910 . 2 0.7 1911 22.6 191! 24.8 1918 23.8 1914 23.2 1915 23.7 1916 Kean. 23.1 GEOGRAPHY. Monthly mean temperature at places in Haiti. [Degrees Centigrade.] Cap-Hai'.tien; altitude, 15 meters. Feb. Mar. Apr. May. June. July. Aug. Sept. 27.1 21.8 21.9 23.3 24.7 25.8 26.3 25.2 22.4 22.7 24.6 25.1 27.0 27.6 27.7 27.8 2 4 .6 25.5 25.8 26.4 27.3 27.2 27.5 27.9 23.8 24.9 23.9 24.4 26.8 26.0 23.6 23.4 25.2 26.1 26.5 26.8 26. 5 27.0 21.5 22.5 22.8 22.9 23.5 24.3 25.3 26.5 26.8 27.0 27.0 Oct. Nov. 25.0 22.5 26.8 24.2 26.4 26.9 27.1 25.8 25.1 26.5 24.8 26.2 24.9 All means are obtained by adding the maximum and minimum and dividing by two. Ba.yeux; altitude, 10 meters. Year. Jan. Feb. Mar. Apr. May. June. July. Aug. Sept. Oct. Nov. 1909 27.6 27.1 26.8 25.0 24.0 1910 21.6 22.2 22.4 23.8 26.2 26.3 26.8 26. 8 26.3 25.8 24.8 1911 22.5 22. 6 22.6 24.2 25.1 26.3 26.8 27. 0 26.3 25.8 1912 1918 23.0 25.6 23.9 23.0 24.1 25.4 26.l 26.3 26.0 25.4 24.2 1914: 23.0 28.2 23.1 25.0 25.9 26.3 1915 27.4 27.4 27.1 26.9 24.9 1916 22.4 22.3 22.3 25.1 25.4 26.8 26.5 27.0 26.6 25.7 24:.5 Mean. 22.5 23.2 22.9 24.2 25.1 26.2 26.9 26.9 26. 6 25.8 24:.7 Port-de-Palx; ltitude 25 meters. Year. Jan. Feb. Mar. Apr. May. June. July. Aug. Sept. Oct. Nov. 1909 26.8 26.2 26.0 26.4 24. 5 1910 20.9 21.8 22.7 23.9 25.1 26.6 26.2 26.S 26.6 26.1 1911 21.6 22.3 22.2 24.3 24.9 26.5 26.9 27.8 27.2 1912 191941 21.6 22.3 21.9 24.2 25.5 26.2 26.8 26.5 26.6 25.7 24.2 Kean 21.4 22.1 22.3 24.1 25.2 26.4 26.7 26.6 26.6 26.1 24.3 Furnished by M. Abbeg, Port-de-Paix. 37 Dec. 21.5 22.5 24.2 24.9 24.4 24.3 23.9 Dec. 22.1 25.2 23.9 23.4 23.6 22.5 22. 4 -Dec. 22.7 23.7 22. 6 23.0

PAGE 43

38 GEOLOGY OF THE REPUBLIC OF HAITI. Monthly mean temperature at places in Haiti Continued. Gonai:ves; altitude, S meters. Year. Jan. Feb. Mar. Apr. May. June. July. Aug. Sept. Oct. Nov. D e c. 1911 25.5 26.8 26.6 28.2 28.7 28.7 28.0 27.4 25.6 1912 25.7 26.0 27.6 27.3 27.6 28.2 27.6 27.4 26.4 1913 26.0 25.4 26.7 25.8 27.1 27.3 27.6 27. 2 26.8 26.9 26.2 25.2 1914 24. 5 26. 1 27.3 28.1 28.4 29.3 29.2 28.8 28.1 26.9 26.9 1915 26.6 25.4 26.9 26.8 28.3 28.5 1916 25.7 25.2 25.6 27.2 28.0 28.0 27.3 26-6 24.4 Mean. 25.7 25.4 26.4 26.9 27.6 28.l 28.5 28.4 27.9 27.6 26.6 26. 7 Gan tier; altitude, 117 meters. Year. Jan. Feb. Mar. Apr. May. June. July. Aug. Sept. Oct. Nov. Dec. 1909 2 7.2 26.7 26.7 26.5 23.6 22.0 1910 22.9 23.7 23.4 2.5.1 26.4 27.5 27 2 27 .4: 26.6 26.5 26.4 23.9 1911 23.7 24.1 24.6 25.5 26.5 26.7 27.8 28.l 27.8 26.9 26.4 24.1 1912 24.4. 25.6 26. 8 27.0 29.0 28.2 28.8 28.8 21.1 26.1 26.9 26.4 1913 24.4 25.6 26.9 25. 8 2 6 .8 27.6 28.1 27. 9 27.5 27.4 26.4: 26.2 1914: 24.0 24.6 25.2 27.4 28.2 28. 0 28.0 28.6 28.2 27.8 25.9 25.9 1915 24.6 25.2 25.5 26. 2 27.1 27.9 28. 5 28.5 28.6 28.6 27.0 25.7 1916 26.9 27.1 27.5 27.6 27.5 27.3 26.9 25.4 23.8 Mean 24.0 24.8 25.4 26.S 27.3 27.6 27.9 27.9 27.5 27.1 26.0 24.6 Port-au-Prince; altitude, 37 meters. Year. Jan. Feb. Mar. Apr. May. June. July. .Aug. Sept. Oct. Nov. Dec \ .. 1909 28.9 28. 0 27.2 27.3 24.8 23.7 1910 24.5 25.2 24.7 26.0 26.6 28.0 28.3 28.2 27.5 26.6 26.1 25.5 1911 25.4 25.3 2.5.5 26.9 26.3 28.4 29.2 28.7 28.2 26.6 26.4 26.o 1912 26.l 26.1 26. 9 26.7 2 7 5 28.4 28.8 28.2 27. 5 26. 9 27.1 2 6 .4 1918 26.2 26.3 27.1 25.6 26.l 27. 9 28.2 28.3 27.4 26. 8 26 0 2 5.4 191j 25.2 25.9 26.4 26.4 27.8 28.1 29.6 28.9 28. 4: 27. 6 26.1 26.2 1915 .. 26.5 26.0 27.0 26.2 28.2 28.8 29.2 28.8 28.2 28.1 2 7 2 26. 6 1916 26.3 26.1 ?J8.5 27.2 27.6 27.9 27.7 28.6 27.9 2 6 .8 26.0 24.7 Mean 25.7 25.8 26.3 26. 4 27.2 28.2 28.8 28.5 27.8 27.1 2 6 .2 26. 6

PAGE 44

GEOGRAPHY. 39 I }.{ onthly mean temp e rature at pkzces in Haiti Continued. P6tionville; altitude, 400 meters . Year. Jan. Feb. Mar. Apr. May. June. July. Aug. Sept. Oct. Nov. Dec. 1900 27.0 26.1 25.6 25.4 22. 4 21.6 1910 21.6 22.5 22.8 24. 0 24.6 26.6 26.0 2 6 .3 2fj. 6 24.8 24 l 22.7 1911 22.3 23.2 2 3. 8 24.7 24.3 25.8 26.6 26.6 25.8 24.9 24 2 23.0 1912 23.8 23.9 2 4.3 24.6 25.4 24.8 26.3 26.2 2 6. 0 25.1 24.6 23. 7 1913 2 6.0 2 6 .1 25. 0 24. 2 23. 8 22.6 191, 22.5 23.3 23. 9 25.2 25.7 2 5 .7 27.3 26.9 26.6 25. 9 25.l 24.3 1915 24.2 23.3 25.1 24.0 2 6 .1 26. 4 27. 2 26.7 26.3 26. 2 25.3 24.9 1916 24.3 24.0 24.0 24.8 25.5 25.8 25.6 25.9 25.9 24.0 23.5 22.4 Kean. 23.1 23.4 24.0 24.5 25.3 25.7 2 6 .5 26.a 26.0 25.1 24.1 23.1 Les Cayes; altitude, 7 meters. Year. Jan. Feb. M ar. Apr. May. June. July. Aug. Sept. Oct. Nov. Dec. 1909 27.4 2 6.7 25.5 26.0 25.5 24-8 1910 24.0 23.8 2 3.7 24.4 25.4 26.7 26.7 26. 9 26.8 26.3 26.2 25.0 1911 24.8 24.0 27.9 28.2 27.2 26.6 25.9 1912 26.0 25.2 2 5 .7 ?,5.5 2 6.5 27.1 27.1 26.8 25.8 1913 24.6 24.5 25.4 24.8 25.8 2 6.8 2 6.3 2 6.7 26.6 26.5 26. 1 25.0 191, 24. 3 24. 7 25.0 25.7 26.7 27.5 27.2 27.1 27.0 26.9 26.0 26.9 1915 24.8 24. 8 25. 5 25.6 26.6 27.3 2 6.9 27.0 26.9 27.1 26.7 26.8 1916 24.8 24.7 24.6 25.3 26.4 26.8 2 6 .9 26.7 26. 7 25.7 25. 6 2,6.l Kean. 24.7 24.7 25. 0 25.2 2 6.2 27.0 26.9 27.0 26. 8 2 6.6 26.2 25.8 Furey; altitude, 1,640 meters. Year. Aug. Days of record. Sept. Day s of reco rd. D e c. -Jan. Days of record. 1906 . 19. 9 20. 3 1-12 190'1 19.6 19.5 1 -15 17. 9 D ec. 1-31. 1908 . 19.7 19.3 1 16 1909 19.0 18.9 1-15 1910 . 19.2 19.8 1-16 1911 . 19.7 19.8 1-14 1912 ... 19.2 3-31 ................... 19.0 1-18 17.4 Dec. 24-Jan 4. 191.3 .. 18.4 18 8 1-17 17.5 Dec. 24-Jan. 4 191 .. 19.4 19.6 1-14 17.3 Dec. 23-Jan. 3. 1915 19.6 1-11, 17-31 ............. 19. 6 1-16 18.0 Dec. 28-Jan. 7. 1916 . 19.7 1-13 16, 81 ....... 19.3 1-13 1917 18.6 18.3 1-15 Mean ... 19.3 19. 8 17.6

PAGE 47

42 GEOLOGY OF THE REPUBLIC OF HAITI. to be about 9 0. The mean daily difference in degrees Centigrade be tween maximum and minimum temperatures at Port-au-Prince by months based on readings taken hourly from 1888 to 1910 is as follows: J anu ary, 10.7; February, 11.3; March, 11.1; April, 10.6; May, 9.8; June, 10.5; July, 11.4; August, 11.3; Septe mber, 10.3; October, 9.6; Novem ber, 9.5; December, 10.1. The variations appear to have some relation to rainfall, as the y are smallest in the months of May, October, and November. May and October have heavy precipitation, but November has only an average amount. The variations are greatest in February, July, and August. July is the driest s11mmer month and February is a dry winter month. At Port-au-Prince the minimum daily temperature, based on monthly observations over a period of 16 years, occurs at 6 a. m. from November to March, inclusive, and at 5 a. m. from April to October. The daily maximum occurs at 1 p. m. from November to April, inclusiv e, and occurs sometimes at 12 and sometimes at 1 p. m. during the rest of the year. For 1909 to 1916, inclusive, the maximum temperature recorded at Port-au-Prince was 37.8 C., on August 2, 1914. The minimum at Po1t au-Prince was 15.2 C., on February 4, 1910. A minim11m of 12.2 was recorded at Petionville on February 3, 1910, evidently an effect of the same cold wave. The lowest recorded temperature elsewhere, except at Furey, is 13 C. at Port-de-Paix on February 20, 1911, and March 9, 1911. At Furey the maximum temperature recorded in the incompl ete records available is 27.2 C. on September 6, 1906, and the minimum is 10.8 C. on January 2, 1913. Moreau de Saint-Mery states that in colonial days maximum temperatures of 40 0. (converted from Reaumur scale) were recorded at Le Trou, Port-au-Prince, and on the Plaine du Port-a-Piment (called Arbre Plain in this report) and that minimum temperatures of 5 C. in the parish of Dondon and 7 C. in the Canton of Nouvelle-Touraine (near Furey) had been observed. These figures prob ably give a fair idea of the range of maximum and minim11m tempera tures throughout the Republic. In the lowlands the high temperatures of s11mmer are rather oppressive, particularly to persons not acclimated to the tropics, and the directness and intensity of the s11n's rays add to the discomfort. It is often dangerous to exercise strenuously in the midday heat. Fortunately the nights generally are cool enough to permit refreshing sleep, although this is by no means always true in summer. At altitudes of 900 meters or more above sea level the winter nights seem chilly to persons accustomed to the lowlands, and at altitudes of 1,500 meters or more even the permanent inhabitants at times suffer considerably from cold. Some extra clothing is necessary for travelers in the mo11ntains, and a camp fire is occasionally very pleasant. A sudden drop of several degrees in temperature often accompa.nies storms and causes 11npleasant coldness even in the low land plains

PAGE 48

GEOGRAPHY. 43 p RECIPITATION. The precipitation in the Republic, which comes chiefly as rain, is notably lacking in the 11niformity and equability that characterize the temperature. The amount of precipitation varies greatly both with the seasons and at different lo calities For the local irregularities the sur face features are mainly responsible. The country is exceedingly moun tainous, and the high ranges cause condensation of the moisture carried by the winds so that the rainfall is great on the windward side and very small on the leeward side, only a short distance away. Within the Repub lic there are all gradations in climate from very hum. id to semiarid and even to arid, and these gradations are characteristic not of large and homogeneous regions but of small and scattered districts. The effect of differ ences in the annual precipitation is most apparent on the vegetation and the agricultural development at different localities. In some places, such as the Arbre Plain, the vegetation is all of arid types, particularly cacti, and agriculture is almost impossible without irrigation. In other places, such as the valley of Dondon, there is a 111xuriant natural for est, and crops flourish without irrigation. },!any mountain ranges, such as the Montagnes la Hotte, are well watered and heavily forested. One featur e of the rainfall, however, is very constant over all the Re public, and that is its concentration into two well-defined wet or rainy seasons, one in the spring, the othe r in the fall. The following tables of rainfall by months at the most important observation stations not only show this feature of the rainfall but give in some detail the available rainfall records for a considerable n11mber of years at stations where observations have been made over a sufficient interval of time to be of considerable value for use as engi neering data. The writer has found it impossible to make some monthly mea ns based on Scherer's yearly records agree with the means published by him. Some of the discrepancies are due to his use of data that had not appeared in the yearly records; others appear to be due to typographical errors in the bulletins. Most of the discrepancies are not large enough to be of much significance. The second table, which contains the stations arranged in alphabetical order, gives only monthly means of rainfall and the number of years of observation at all stations where any records whatever have been kept for more than a year. This table gives a rough indication of the probable rainfall at a considerable number of stations that were maintained for only a short time and places the data for other stations in alphabetical order for more ready reference A final table gives the total yearly rainfall at all stations where the records cover sufficient time to make the means of real value, usually at least seven years. An appendix to this table gives records of annual rain fall in colonial days at a number of places, as published by Moreau de Saint-Mery. These records, although probably not so reliable as those

PAGE 49

\ 44 GEOLOGY OF THE REPUBLIC OF HAITI. published by the Observatoire Meterologique, are valuable because tl1ey give an indication of the approximate amo11nt of rainfall at some important places for which no recent records are availa ble and because they indjcate that the climate has pot. as is sometimes suggested, changed radically in l1istori cal times. The long record at Leogane gives a result no t much different from that of the recent observations. That at Le Trou corresponds closely wi t h Cap-Ha1tien and Limonade, as would be expected. Rain/ all in millimeters by months. Cap-Rai'.tien. Year. Jan. Feb. Mar. Apr. May. June. July. Aug. Sept. Oct. Nov. Dec. 1906128.9 84.9 56.0 27.3 133.7 129.6 45.1 46.2 68.3 216.7 187.7 132.2 19()94 .. 293.8 753.0 159.5 1909 20. 7 209.5 295.2 1910 324.8 40.4 459.6 18.3 7.0 55.9 39.2 46.9 98.6 293.0 146.B 1911 .. 76.8 38.8 15.6 82.0 224.3 8.4 13.8 31.7 97.6 419. 9 67.1 678 8 1912 1g.q. 6 90.1 5.6 86.4 94.5 9.2 17.7 21.4 23.3 370. 3 626.6 14.1 1913 3.9 1.6 38.3 304.8 191.9 10.9 0.6 42.2 80.2 98. 7 223.6 157.7 191, .. 133.9 100.6 152.8 46.3 82.2 64:.6 17.5 0.0 46. 0 301.6 367.4 102. 4 1915 111.9 426.8 237.1 140.9 15.3 13. 0 144. 1 220.l 373.6 133.5 1916 112.S 244.0 42.5 2 0 143.7 24.4 121.5 23.7 87.5 145.6 1.0 1917 10.0 70.0 11.0 149.5 197.0 186.1 o.o 14.0 67.0 127.0 278.4 878.0 1918 117.0 58.0 169.0 55.0 1919 152.5 72.6 117.S 278.9 Mean 123.5 109.7 98.9 77.1 145. 4 79.8 88.4 45.1 92.5 237.6 a. Figures for January to June in this line show mean rainfall, 1906-1909; figures for July to December show mean rainfall, 1906-1908. Port-de-Paix. Year. Jan. Feb. Mar. Apr. May. June. July. Aug. Sept. O ct. Nov D ec . 1907-112.4 115.1 36.9 21.8 64.9 47.0 244.5 100.8 83.9 100.2 129.5 181.5 1909' 1909 . 161.0 186.5 261.0 811.4: 638.0 39.0 1910 582.1 67.0 103.0 32.5 68.0 114.0 66.0 185.0 173.0 297.5 640.0 1911 133.3 35.0 18.5 100.6 151.0 65.0 34.0 57.0 111.0 372.5 1912 124.0 16.0 52.0 36.3 52.0 40.5 102.5 09.0 258.0 666.0 105.0 1913 175.0 27.0 18.0 247.0 86.5 27.0 39.0 66.0 75.0 79.0 182.0 45.0 1914 106.0 67.0 113.0 88.0 30.0 59.0 14.0 15.5 76.6 13.0 111.0 12.0 1916 50.0 104.2 45.0 83.0 46.0 10'2.5 91.5 84.0 2 69.0 46.0 130.0 64. 0 1916 95.0 39.5 86.0 69. 0 120.7 36.5 106.5 48.5 129.5 287.0 23.0 1917 80.0 110.5 28. 0 95.5 107.5 278.0 67.0 131.0 112.0 47.0 180.5 106.5 1918 19.5 70.0 29.5 87.6 42.5 74.0 ]3.0 69.0 79.5 75.5 113.0 97.0 1919 90.5 83.0 42.5 81.0 114.0 37.5 6.0 30.0 112.0 93.0 15.0 62.0 Mean 14:8.6 82.9 43.3 71.6 71.9 90.9 50.2 95.5 146.6 129.8 234.7 140.7 Fii'\ll'ea for January to June in this line show mean rainfall, 1907-1909; figures for July to December show mean rainfall, 1007.

PAGE 50

GEOGRAPHY. 45 Rain/ all in millimeters by months-Continued. MOle St.-Nicolas Year Jan. Feb. Mar. Apr. May. June. July. Aug. Sept. Oct. Nov. Dec. 1905 49.9 27.0 50.3 3.1 65.9 13.7 33.0 13. 7 120.6 100.5 25.8 83.2 > 19094 1909 23.0 175.3 217.6 128.1 196.7 21.0 1910 125.5 17.2 24.5 21.6 o.o 79.9 48.8 33.0 44.6 89.8 22.4 134.4 1911 83.6 o.o 3.4 o.o 103.6 0.0 0.0 5.0 16.5 0.0 6.0 1912 127.1 183. 0 93.7 110.2 o.o 1913 0.1 102.0 o.o 59.0 26.8 20.0 11.0 6.9 51.8 0.6 157.2 8.2 1914 91.2 32.1 32.7 93.0 86.2 36.3 41.6 1915 17.7 77.2 29.5 23.9 44.4 53.3 24.2 41.3 29.6 14.8 13.6 0.4 1916 0.0 8.2 14.3 13.5 23.8 '38.4 19.6 88.9 14.6 8. 76 123.0 0.0 1917 1.9 11.1 0.0 105.9 49.6 24.9 16.9 68.7 40.7 29.7 169.8 85.4 1918 . 11.3 3.3 32.5 23.5 17.0 2'J.7 0 0 51.4 252.2 5.5 13.3 1919 4.8 o.o 1.9 0.0 149.0 8.4 24.7 1.7 5.2 2.2 46.l 58. 4 Hean 46.5 32.4 25.8 26.1 59.7 24.8 2 6. 0 39.2 72.5 79.8 67.7 32.9 Figures for January to June in this line s how mean rainfall, 1905-1909; figures for July to December show mean rainfall, 1905-1908. Gonaives. Year. Jan. Feb. Mar. Apr. May. June. July. Aug. Sept. Oct. Nov Dec. .. 1903-7.8 5.7 9.7 24.4 81.7 35.9 101.2 54. 9 19.9 6.6 84.9 88.3 190911 1909 91.0 202.5 1910 .. 28.0 o.o 50.0 5.0 39.2 182.8 96.5 77.9 64.4 119.4 o.o 66.9 1911 o.o o.o 0.0 21.7 141.2 101.7 33.8 55.3 52.7 32.4 24.0 76.4 1912 .. 6 .0 60.4 44.5 4.1 64.7 83.5 109 4 68.4 31.2 69.3 11.5 o.o 1913 0.0 1.8 13.7 50.1 22. 5 92.3 23.2 66.7 130.5 21.4 10.7 o. (j 1914 .. 18.3 14.8 54.7 72.3 41.0 52.5 30.8 55.0 62.8 4.5 142.2 6.1 1915 18.0 23.3 8.8 10.6 7.1 146.1 18.0 1916 .. 0.0 30.3 0.0 12.2 115.1 148.6 50.0 88.0 175.6 67.0 46.4 o.o 1917 0.0 0.0 3 8 56.0 107.6 132.6 37.6 42.5 17.4 8.5 1.1 4.3 1918 .. 0 .0 27.0 3.7 8.0 70.9 71.1 56.6 7.6 163.6 45.5 0.0 0.0 1919 2.1 0 0 22.4 o.o 95.8 86.5 55.9 o.o 95.2 o.o o.o 2.5 Mean 7.3 11.6 15.8 24.2 76.4 100.9 68.2 58.l 93.8 47.0 23.4 12.7 Figures for January to June and September to D e cember in this line show mean rainfall, 1903-1909; figures for July and August s how mean rainfall, 1903-1908. -St.-Marc. Year. Jan. Feb. Mar. Apr. May. June July. Aug. Sept. Oct. Nov Dec. 1905. 1909" > 7.9 20.4 18.6 60.7 180.4 145.2 124.8 134.8 176.5 149.6 21.3 2.6 1909 145.7 165.7 189.0 91.7 89.6 0 0 1910 .. 54.0 50.2 0 0 o.o 86.2 177.1 119.9 18.5 76.9 0.0 23. 2 1911 o.o 0.0 0.0 28.8 180.9 128.2 96.4 173.8 143.6 72.1 38.6 33.1 1D12 5.5 21.0 18.0 11.3 45.2 180.8 179.0 1919 .. o.o 0.0 7.3 3.4 131.9 82.5 157. 8 69.9 68.2 92.2 12.7 21.9 Mean 11.0 19.3 13.1 38.5 144.0 143.8 133.1 120.8 148.0 120.4 28.8 11.0 F lgUres for January to June in this line show mean rainfall, 1905-1909; figures for July to December show mean rainfall, 1905-1908. I

PAGE 51

46 GEOLOGY OF THE REPUBLIC OF HAITI. Rainfall in millimeters by months Continued. St.-Michel de l' Atalaye. Year. Jan. Feb. Mar. Apr. May June. July. Aug. Sept. Oct. Nov. Dec. .. 190821.6 13.5 46.3 6 8.7 186.8 208.6 212.0 46.7 233.4 1 31.l 101.1 46.0 1909 .. 1910 30.0 0.0 2 6.6 72. 9 65.9 109.9 73.2 176.0 125.0 100.0 75.4 126.0 1911 17.3 0.0 7.0 68. 0 388.0 176.3 49.0 158.0 98. 0 107.6 17.0 65.0 1912 0.0 12.4 78.0 so. o 119.0 182.0 68.9 143. 2 179.4 196.0 48.0 0.6 1913 0.0 0.0 56.0 59.0 114.0 121.0 160.0 110 .0 25.0 8.0 1914 18.6 7.0 41.0 169 .0 62.0 20.0 118 .5 142.4 72,4 29.0 1915 2 2 21.3 4.3 23.9 227.1 592.2 2 10.0 60.1 26.6 70.6 12.0 0.0 1916 . o o 6.0 1 6.0 3 3 .0 163.5 827.6 1 8 0 2 1 3 1 8 140.2 0.0 1917 0.0 2.6 o.o 131.1 176.4 139.l 48.0 105.7 162. 7 90.7 45.0 1918 5.0 2.0 68.0 103.0 115.4 120.0 2 0 8 8 127.2 109.1 J27. 7 12.0 13.3 1919 . 15.1 0.0 21.0 62.2 272.1 175.6 104.4 1 3 7.2 5 5.0 24.5 8.6 Mean 10.9 6.5 34.2 73.3 171.4: 198.4 119.1 1 63.7 143.6 118. 5 75.3 S-.l.3 a Figures for January to June in this line show mean rainfall, 1908-1909 ; figur es f o r July to December sho w rainfall, 1908. Mirebalais. Year. Jan. Feb. Mar. Apr. May. June. July. Aug. S ept. Oct. Nov Dec. -1907.. 3.7 47.2 83.9 169.3 277.4 189.4 22'2.7 220.7 298.2 169.9 35. 0 6.0 1909 1909 121.5 .. . .... 248. 2 173. 0 42.0 1910 20.0 0.0 85.0 32.0 293.7 3 25.0 1 69.0 396.4 348.2 150.0 79.0 207.0 1911 43.5 2 0 3 4.5 105.5 242. 5 192. 5 399.9 403.S 2 81.0 163 .0 118.0 24.0 1912 8.0 16.0 119.0 212.0 304. 0 243.0 3 51.0 289.0 343.0 264.0 1 18.0 6.0 1913 35.0 25.0 81.0 2 0 3.0 277.0 199.0 291.0 6 35.0 525.0 156 .0 17.0 119.0 1914 20. 0 38. 0 125. 0 86.0 472.0 426.0 201 .0 S 8 8.0 554.0 673. 0 405.0 144.0 1915 0.0 2 0 0.0 182.0 509.0 6 27.0 596.0 456.0 434.0 446 .0 100.0 95. o 1916 .. 36.0 287. 0 161.0 333.0 746.0 791.0 436.0 385.0 597.0 7 8 9.0 492. 0 1.7 Mean 17.4 51.2 84.7 166.1 8 67.6 327.2 3 01.0 361.5 397.7 322.9 157.2 65.1 Figures for January to June and August and Septe mber in this line sho w mean rainfall, 1907-1909; figures for July and O c tober t o D ec ember show mea n rain fall, 19071 908. Thomazea.u. Year. Jan. Feb. Mar. Apr. May. June. July. Aug. Sept. Oct. Nov. Dec. 1905.. 16.8 31.9 41.6 107.5 154.8 190941 32.4 42.5 82.5 160.1 150.6 42.7 8.7 -1909 10.0 105.0 119.4 87.6 199.1 o.o 1910 . 6.0 o.o 37.6 87.6 58.5 85.8 88.8 70.1 161.5 62.5 71.0 63.0 1911 0.0 14.5 27.8 78.9 124.0 21.0 27.0 3.0 92.0 136.8 128.0 o.o 1912 o.o 89.1 62.8 114.1 133.5 50.0 83.3 1 24.0 160.7 109.0 4.0 o.o 1913 0.0 0.0 o.o 96. 8 105.8 7.0 96.0 45.0 69.2 155.1 33.3 17.0 1914 .. 14.8 87.8 5 5 45. 0 161. 0 120. 0 29.0 56.5 66.0 180.0 160. 0 20.8 1915 14.0 35.0 10. 5 35.7 151.5 234.5 2 80.8 164.0 109.6 1916 3 6 .4 153.6 143.3 150.0 o.o 1917 0.0 17.3 7.0 148.5 198.1 145 .1 62.4 44.8 75.5 129. 9 78.4 15.6 1918 0.0 17.6 120.2 55.3 281.9 49.7 51.0 58. 1 131.0 144.9 99.0 o.o 1919 0.0 o.o 10.2 147.5 21.2 42.2 35. 2 82.2 140.9 108.4 0.0 27.5 Mean 8.5 22.9 34.9 96.2 144.S 65.5 64.7 73.7 12.5.3 132.8 81.0 12.8 a Figures for January to June in this line show mean rainfall, 1905-1909; figures for July to December show m ean rainfall. 1905-1908.

PAGE 52

47 GEOGRAPHY Rai,nf all in m i llim e t ers by months Continued. Port-au-Prince. Year. j Jan. I Feb. j Mar. I Apr. I May. I June I July, J .Aug. I Sept. j Oct. Nov I Dec 1863 83.0 128.0 257.0 92.0 18.0 1864 1.0 124.0 111.0 212.0 261.0 59.0 109.0 223.0 164 .0 171.0 61.0 4 5 .0 1866 20.0 13.0 78. 0 193.0 451.0 75.0 103.0 129.0 298.0 151.0 158.0 28.0 1866 57. 0 75.0 149.0 862.0 226.0 146.0 150.0 125.0 131.0 110.0 125.0 66.0 1867 51.0 27.0 23. 0 199.0 328.0 177.0 54.0 189.0 53.0 127.0 63.0 ,1.0 1868 1.0 143.0 87.0 10'2.0 817.0 52.0 43.0 129.0 282.0 118.0 118.0 43.0 1869 27.0 141. 0 109.0 123 .0 3 26.0 139.0 97. 0 265.0 151.0 29.0 6.0 1888 . 12.0 36.0 93.0 108.5 34.0 113.0 29.0 62.0 88.0 o.o 1884 59.5 81.9 88. 0 275.5 140. 9 27.2 139.6 129. 4 111.0 44.8 43. 7 1885 .. 42.5 58.0 37.1 23.0 92. 0 58.1 67.5 S9.8 154.6 127.6 19.7 57.6 1886 .. 161.8 7 3 .3 60.6 185.8 204.7 144.9 72.5 223.5 113.5 51.5 1887 . 453.8 50.8 77.5 210.4 191.0 173.0 74.1 0.8 1888 . 5.1 11. 6 37.3 100.4 179.6 113.0 43.7 85.1 87.6 127. 8 165.1 124.7 1889 .. 12.9 21.2 56.5 2'22.3 464. 6 99.2 32.3 108.9 225.9 203.1 4.2 20.1 1890 .. 48. 8 78. 4 127.5 83.5 119. 4 71.2 86.1 106.0 233.5 119.9 94.7 62.0 1891 27.6 116 .3 24.7 169. 7 332.0 148 .6 26.0 198.3 176.3 151.6 124.0 5.4: 1892 . 34.0 26.0 21.9 171.2 169.1 1 0 3.4 106.6 145.7 61.6 161.0 50.9 24.o 1893 80.7 52.4 41.9 111. 4 302.7 6 7.2 146.2 217.5 209.7 181.0 4.5 22.1 1894 5.5 6.6 128.5 220.0 212. 8 7 8 .6 1.8 72. 9 296.9 146. 3 24.6 68.7 1895 .. S.8 62.9 93.4 136.2 209.3 66 6 121.3 173.5 216.3 107.8 114. 1 20.0 1896 . 46.3 26.1 36.2 145.2 170.6 105.5 '4.8 206.8 158. 4 196.8 9.7 80.3 1897 0.4 6.2 109.9 160.7 423.? 37.0 35. 4 105.8 245.2 72.9 44.7 6.8 1898 . 15.1 67. 2 48. 7 82.0 80'2.3 144.4 97. 1 156.3 148.8 112.9 61.6 161. 2 1899 . 5 9.7 13.7 67.2 238.4 2 0 4.2 215.4 88.0 81.4 244.0 870.9 18.6 45.2 1900 . 69.8 8.0 7.6 191.5 209.] 74.1 36.6 99.2 231.6 79.6 121.9 11. 4 1901 45.2 7.6 77.0 90.2 289.0 133.2 81.5 71.3 130.0 262.5 44. 5 17.9 190'2 . 20.6 230.9 3 7.4 222.7 489.6 87.6 3.5 100.2 171. 8 228.0 47. 4 42.0 1903 .. 7.8 o o 106.8 169.0 116 .0 100.6 72.5 110.9 156.5 140.6 209.6 9 .7 1904 .. 44. 2 42. 4 314.7 254.6 195.8 10 8 61.3 71.1 188.6 253. 6 64.7 41.4 1905 . 72.7 79.3 96.1 69.8 2 3 3.3 123. 1 43.3 168.8 211.7 309. 4 126.6 0.1 1906 . 44.5 135.1 124.4 153.6 252.9 96.l 48. 2 194.5 224. 9 2 87.1 60.3 62.0 1907 13.5 85.9 74. 9 92. 8 249.0 22.9 43. 0 108.0 119.1 137. 9 81.7 0.5 1908 8.2 276 .2 160.7 119.8 168.5 4.5. S 63.9 186.8 1 0 4.8 1 6 5 7 23.6 24. 6 1909 83.2 31.8 143.0 249.5 112.9 163.8 82. 8 127. 0 872.2 103.8 246.9 40.8 1910 . 24.9 1.9 198.0 246.6 1 5 9.7 68. 9 55.0 135.8 456.6 115.6 69.8 80.2 1911 1.8 20. S 136.6 166.9 258.8 78.7 74.1 67.1 98.4 353.4 120.4 28.2 1912 1 5 30.5 136.5 267.7 219.0 64.8 85.2 218.S 237.6 161.3 29.0 86.8 1913 20. 4 17.5 67.5 281.0 185.2 94.6 71.2 103.4 182.1 142.6 165.8 9.f 1914 76.7 18.2 1 3 0.8 130.9 138.1 184.7 83.2 248.4 119.2 131.4 189.4 60. 0 1915 89.4 23.9 15.5 71.4 200.7 113.1 104.8 271.5 60.0 199.0 85.8 8.2 1916 12. 1 42.9 127.8 187.0 242.7 346. 6 90.7 61.9 216.0 300.9 SOS.I 1.0 1917 . 1.6 40.2 104.6 188.4 277.2 1 3 9.3 120.S 160.1 241.6 182. 8 51.7 37.2 1918 5.8 30.9 199.4 132.9 317. 8 128.0 11.S 77.3 224.9 280.3 69.7 o.o 1919 24.5 61.9 55.3 99.6 156.3 68.4 95.3 27. 4 133.1 M.8 58.8 18. 2 Kean 8 2.2 57.9 93.8 162.4 249.8 105.5 69.5 139. 4 184.7 169.6 85.7 35. 0 Jacmel. Year. j Jan. I Feb. I Mar. I Apr. I May j June. I July. Aug. j S ept. I Oct. I Nov I Dec. 1906. 85.4 1909 88.4-90.4 118.3 187.4. 136.8 77.7 64.8 1 67.9 98.8 88.8 9. ts -1909 54. 9 165.5 152. 159.6 245.2 7 .2 1910 . 49.4 1.0 215.0 117. 7 123. 9 67.9 46. 9 130. 6 115.3 139.5 16.7 69.0 1911 12.2 41.8 134.1 208.6 249.4 8. 4 46.4 48. 6 00.2 151.6 50.3 7.9 1912 40.7 117.5 19.3 142.3 213 .0 '3.6 127.4 174.6 40.8 170. 1 162.2 16.2 1913 4.4 3 0.6 15. 4 242.5 194.3 Sl.1 75.0 86.7 64.7 79.0 203. 6 45.9 191 90.3 6.3 5 3 .5 1 81.4 2'24.2 34.8 27. 5 143.7 10'2.0 56.5 83.0 17. 6 1916 33.6 57.2 19.9 81.6 372. 2 59. 6 97.4 127.7 91.7 55.6 27.5 112.0 1918 10.8 35.6 99.3 65.6 267. 8 137.0 56.3 327.3 117.6 .o<).5 110. 4 o.o 1917 0.0 39.9 57.4 141.2 629.5 155.1 90.6 174.9 110.5 199.9 37.0 43.8 1918 6.5 16.6 134.7 196. 8 310.8 120 .5 94.6 83.9 72.5 162.1 15. 9 89. 7 1919 . 79.5 1.7 141.3 138.8 139.3 162.5 114.2 111.1 118.0 32. 4 21.2 61. 3 Mean 33.5 50.1 89.4 140. 8 238.9 97.7 76.0 126.0 113.2 135. 9 78. 1 80.6 January to June= 1906; July to December = 1906-1908.

PAGE 53

48 GEOLOGY OF THE REPUBLIC OF HAITI. Rainfall in millimeters by months Continued. Les Cayes. Year. Jan . Feb. Apr. May. June. July. Aug. Sept. Oct. Nov. Dec. 1 006--19094 .. 64.4 135.7 86.4 189.8 297.5 211.7 75.1 132.6 238.0 813.l 199. 3 60.9 .. 1909 46.6 242.3 602.0 4 32.6 300.2 15.3 1910 40.3 122.1 94.9 1-57 .3 314.0 64.6 106.2 208.8 238.4 213. 6 21.5 83.5 1911 25.0 30.0 65.2 136.8 920.0 81.3 18.5 122.2 279.0 423.8 204.1 117.0 1912 . 179.1 68.2 132.9 246.0 320.l 67.7 103.2 342.2 92.0 317.8 80.4 65.5 1918 86.5 65.5 212.0 150.0 104.0 27.0 87.5 62.5 156.2 110.9 60.8 50.0 1914 .. 60.0 37.2 25.4 271.6 144.7 61.1 21.7 154.5 206.0 85.2 1 80 .4 13.5 1915 86. 0 11.2 12.0 140.0 160.8 106.9 1 66.0 843.0 257.6 44.2 33.8 59. 7 1916 .. 33.7 82.0 72.1 255.0 428.0 225.0 195.0 291.9 71.5 794.0 583.8 30.0 1917 16.5 17.8 65.5 142.5 228.3 76.7 132.2 526.8 458.3 1046 .5 315.2 482.4 1918 .. 350.3 162.4 120.0 210.0 464.0 265.5 451.5 276.4 193. 0 578.6 223.7 289.2 Mean 83.5 86.9 87.4 189.9 328.8 140.2 119.4 228.3 251.() 383.6 199.4 106.8 "January t o June= 1906; July to December = 1906-1908 J6r6mie. Year. Jan. Feb. Mar. Apr. May. June. July. Aug. S ept. Oct. Nov D ec 1906-71.l 100.7 70. 6 ... 64.5 113.7 81.9 67.4 147. 1 58.7 69.8 17.8 190911 1909 30.3 257.8 78.6 84.0 214.7 12.0 1910 96.9 1.7 202.1 3j.1 96.5 60.5 36.0 117.7 13.8 70.7 161.3 130.0 1911 .. 91.l 15.0 53.1 48.8 103.8 172. 7 1912 253.4 207.5 26.3 7 6.6 20'2.6 69. 4 89.6 54.6 219.2 144.2 103. 8 181.2 1913 22.3 1.0 94.7 160.1 168. 6 99. 9 105. 0 41.3 78.9 143.4 38.8 71..2 1914 8 1.1 32.7 93.4 1 80. 0 233.5 7 3.8 9 0.9 115.5 114.9 134.0 174.6 5.3 1915 75.0 178.1 63.7 7 3.7 1 31.2 102.3 45.9 120.6 76.6 37.8 218.0 68-4 1916 6.6 221.0 89.9 176.8 214.9 273.2 45.3 147.7 55.8 253.6 120.9 o o 1917 50.8 67.4 a.4 126.6 218.7 319.1 256.8 1918 44.7 140.7 44.4 178.1 96. 1 125.1 126.2 95.9 108.9 232.4 153 .3 l:i 4 1919 312.8 75.0 80.7 152.8 855.9 48.9 239.8 198.6 Mean 93.7 100.1 70. 8 109. 0 166.8 180.2 88. 1 105.1 115.2 110.4 1 24.9 58.8 0January to May= 19061909; July to December= Moron. Year. Jan. Feb. Mar. Apr. May. June. July. Aug. Sept. Oct. Nov. Dec. 1910 205.9 532.5 193.8 539.5 1911 76.7 75.4 18.6 112.0 537.6 81.1 74.2 76.1 75.9 132.5 68.3 291.3 1912 265.2 78.6 84. 9 112.6 233.0 263.4 118.l 1 3 2.5 39.3 1913 22.6 0.1 78.2 276.5 818.8 100.0 1 6 4.1 176.8 843.7 134. 3 369.1 87.4 1914 100.3 20.5 184.4 2'll.6 231.3 150.5 180.0 12,.8 306.7 366. 9 158.5 1915 89.4 183.3 84.0 93.4 123.3 146.3 163.1 260.0 270.0 246.7 205.0 76.0 1916 63. 9 42.8 128.7 171.S 169.8 251.5 163.3 210.0 207.3 840.8 191.8 2.7 1917 23.9 41.8 9.6 270.3 470. 6 306.5 145. 9 184.0 258.0 269.0 441.1 93.8 1918 6.2 54.8 166.S 483.2 824.0 122.5 163.1 S26.5 405.3 255.0 27.0 1919 107.8 24.0 86.7 820.0 125.8 Mean 82.8 57.9 82.4 215.8 30'2.5 178.7 140.5 178.6 225.9 276.2 247.1 146.2

PAGE 54

M eon month.l11 rain/all, i n millimeters, compil.ed /rom all available recMda .,.. Station. Jan. Feb. Mar. Apr. May. June. July. Aug. Sept. Oct. -Abricots (lee) ............ 24.2 22.6 58.9 49.8 105.l 47.4 47.9 21.0 59.0 40.0 An se-A-Vea u ....... 3 6.9 50.1 56.4 1 3 0.0 161.7 130.2 108.6 113.2 114.8 91.3 .Arcahaie (l') 16.8 28.9 34.6 70.1 129.8 64.0 66.6 82.0 109.9 105.3 Bahon ......................... 34.5 14.7 51.9 51.6 144.0 160.4 116.3 141.1 1 6 5.5 214.3 Bain et ..................... ...... 46.7 39.6 96.8 240.6 231.8 110.0 120.3 112.9 167. 0 153.5 Bassin-Bleu 34. 9 33.0 18.9 49.4 123.7 128. 7 83.3 117.2 118.1 111.3 Bayeux ............................ 176. 3 117.5 110.6 193.7 206.1 110.4 46.9 94.2 143.8 2 1 7.4 Borgne (le) ..................... 172.2 112.2 81.8 181.2 115.2 99.8 47.9 50.2 91.7 196.9 Cap-H aitien ....................... 123.5 109.7 98.8 77.1 145.4 79.8 88.5 4 5 .1 92.5 237.7 Cayea (les ) 83.5 8 6.9 87.4 189.9 828.8 140.2 119.4 228.3 2 51.0 383.6 Cbardonnieres ...................... 19.3 66.4 75.4 119.8 76.1 48.6 49.1 94.4 136.1 173.2 Don don 61.7 40.9 64.6 106.7 165.5 116.2 101.5 100.9 1 3 4.3 168.1 Fond V errettes .................... 30.9 38.2 55.2 146. 0 206.1 155.1 74.8 135.1 192.6 180.4 Gan tier 16.3 20.5 41.8 123.7 149.4 51.0 33.2 66.7 116.7 145.2 Gonaives .......................... 7.3 11.6 15.8 24.2 76.4 100.9 63.2 68.1 93.8 47.0 Grande-Riviere du Nord ........... 86.5 64.5 78.9 92.1 204.6 g.q.o 67.9 lll.0 171.4 178.9 Hinche ........................... 24.0 3.0 8.0 88.0 177.0 237.7 149.0 75.0 263.3 68.0 J acmel ...................... 83.5 50.l 89.4 140.8 233.9 97.7 76.0 126.0 113.2 135.9 J e remie ............................ 93.7 ]02.1 70.8 109.0 166.8 130.2 88.1 105.1 115.2 110.4 L e ogane ......................... 20.6 52.1 92.2 137.0 153.6 116.9 84.7 140.2 187.8 91.9 Limonade 31.2 70.l 66.3 136.7 86. 0 54.9 26.8 72.8 104.2 177.9 ....................... 0.0 82.5 157.3 69.5 382. 2 270. 8 98.5 103.2 112.6 254.5 M' A iragoane ..................... SS.5 40.0 64.1 109.8 205.6 107.4 155.2 134.3 203.7 165.6 Mirebalais 17.4 51.2 84.7 166.1 867.6 327.2 801.0 361.5 397.7 322.9 M:Ole St.-Nicolas ............... 46.5 82.4 25.8 26.1 59.7 24.8 26.0 39.2 72.5 79.8 Moron .......................... 82.8 57.9 82.4 215.8 302.5 178.7 140.5 178.6 225.9 276.2 Petionville ...................... 28.9 40.2 92.2 183.2 220.6 111.5 77.7 120.4 19'2.6 162.3 Petit-Goave 15.9 42.6 49.1 129.4 222.9 137. 9 107.5 136.6 139.5 86. 9 Pilate ............................ 58.6 63.7 74.1 168.2 261.7 169.1 123.7 170.7 172.2 171.3 Port-au-Prince (St. -Martial) 28.9 58.l 95.6 168.7 249.4 106.2 74.7 138.7 192.6 177.9 Port-au-Prince (Lalue) 26.2 53.6 122.9 177.2 201.2 119. 2 71.4 137. 5 200.1 181.1 Port-au Prince (Ferme-Ecole Thor). 46.8 38.2 34.9 157.6 99.0 51. 4 107.5 23.2 132.8 91.2 Port-de-Paix .............. 143.6 82.9 43.3 n.5 71.9 90.9 50.2 95.5 146.6 129.3 St.-:Marc 11.0 19.2 13. 1 138.5 144.9 143.8 133.1 120.9 140.8 107.6 St.-Michel d e l' Atalaye 10.9 6.5 83.6 76.1 172.2 215.5 119.0 163.7 144.8 118.5 Thomazeau .................. 8.7 22.9 34.9 96.4 142.8 65.5 64.5 73.9 124.6 132.7 Tiburon ........................ 42.9 17. 1 58. 4 122.7 62.6 7.1 34.1 49.9 114. 0 98.7 Nov. Dec 87.0 55.5 60.9 33.9 64.8 21.1 109.6 70.0 91.2 35.3 106.0 61.6 401.5 2 9 1 8 387.9 185.2 298.3 2 3 0 7 199.4 106.8 89.3 27.8 174.7 139. 0 137.7 24.4 61.2 2 6.7 23.4 12.7 206.4 1 6 7.2 47.5 13.5 78.1 80.6 124.9 5 8.8 63.2 5 6.5 229.6 79.9 69.6 o.o 68.1 49.1 157.2 65.1 67.7 32. 9 247.1 146.2 71.8 26.1 63.4 28.1 235.3 72.8 89.7 35.6 108.4 38.8 109.8 11.7 234. 7 140.6 27.3 9. 9 76.S 31.4 80.9 12.9 100.6 25.1 Years of record. 1and2 9-11 13 8 and 9 13 8-10 17 and 18 8 12 and 13 13 9 9 and 10 12 21 17 8 and 9 2 and S 14 12 and 13 7 and 8 6 1 10 and 11 10 14 and 15 7. g 20 and 21 18 and 1 7 38 13 1 13 8 and 9 12 14 and 15 5-7 0 0 "'d

PAGE 55

50 GEOLOGY OF THE REPUBLIC OF HAITI. Place. Anse-AVeau .................. Arcahaie (l') ............. ........................ Bainet ........................ Bassin-Bleu ................... Bayeux ....................... Borgne (le) ................. Cap-Hartien .................. Cayes (les) ............... Ohardonnieres .............. Dondon ....................... FondVerrettes ............... Gantier ....................... Gonaives ..................... Grande-Riviere du Nord ...... Jacmel ........................ Jeremie ....................... Mean annual rain/ all. Millimeters. 1,061. 7 793.9 1,273.4 1,445.2 986.1 2,109.2 1,728.2 1,584.4 2,107 .6 682.0 1,974.1 1,238.8 852.8 524.0 1 429.4 1,205.6 1,266.8 Place. L ogane ......................... Limon ade ....................... Miragoane ...................... Mirebalais ...................... MOle St.-Nicolas .............. Moron .......................... Petionville ...................... Petit-Goave ..................... Pilate ........................ .. Port-au-Prince (St.-Martial) .... Port-au-Prince (Lalue) ...... Port-de-Paix .................... St. -Marc ........................ St.-Michel de l'Atalaye ......... Thomazeau ...................... Tiburon ......................... Millimeters. 1,153.5 1,136.4 1,329.0 2,659.4 542.4 2,124.8 1,326.8 1,159. 7 1,741.3 1,416.1 1,437 .6 1,282. 7 910.1 1,167.2 859.7 733.2 Colonial records of annual rainfall according to Moreau de Saint-Mery. Place. Millimeters. Years of record. Bombardopolis ......................................... 783 12 Oro ix -de s -Bouquets ..................................... 861 2 Fort-Libem ............................................ 1,510 Several Leogane ................................................ 1,350 20 Le Trou ............................................... 1,478 4 Montagnes de Plymouth (Southeast of Jeremie) .... 4,485 1 Port-au -Prince ......................................... F.87 1 Port-Margot ............................................ S,804. 2 St.-Louis (du Sud) ..................................... 2,295 6 Converted from feet on assumption that 1 foot equals 324 millimeters. The seasonal distribution of rainfall is much more readily apparent in the curves shown in Figure 3 than in the tables. These curves, which cover 13 stations that probably are fairly representative of rainfall at low altitudes over the whole Republic, show that the spring rains gen erally last through April, May, and Ji1ne and reach their maxim11m intensity usually in May. The fall rains come at a more variable date at different stations, ranging from August to De cember, and the average maximum is reached in the months of October and Nove mber. December to February, inclusive, are the d111 winter months, and July is almost always the driest summer month. The fall rainy seaBon is generally longer than the spring rainy season, and the winter dry season is longer than the summer dry season. Unfortunately none of the stations for which records are availab le are at altitudes that represent the conditions on the higher mountain ranges. Fond-Verrettes, which is about 810 meters above sea level, is the highest

PAGE 56

GEOGRAPHY. > z --. -' -. ,,--..... --.-----t 'r-7----1 ,,. ... ''l----+---+----4 .... --#-1 ... ... I .: -ld ":- 11. .. -.. I I 'I. -..- ..... ., I -.. t. -. -' I I ,,.-.... la. J 1-- --.. ' .............. ... ' -. .... o.._ __________ _. ____ ._ ___________ __......_ __________________ ___ Cf) 300 lJ a: I-I2.oo ..I ..I L 100 0 U>300 lJ a: t;l&J l: 200 .J .J I: 100 0 -H<, r .. -. - I L r' ., I "" ., I l .. I - -.Ji/11 I r ,, I .:: .. . ,.. .. . . ,, .... --- .. : I --".. I -- ... I .- '1 -.... -.r' ... .. "''"' -c. --;m. ... .. .-..... .... _jll '. ;li- .. I .. I I .., ., "\ .. ., .. -( -.. ... ti".) 0 0 ,, --I - --I 7 .. I \. .. -zl. " ... .... l I I I. g i(' - .. .... I . 1 .... ... .. :ii ... -. '-. . ,, FI - -.. . -- .,.,. . ....... ......... . .,... FIGURE 3 Curves s ho wi ng mont hl y m ea n r a infall for different stations. 51

PAGE 57

52 GEOLOGY OF THE REPUBLIC OF HAITI. station given in the tables. It has only a moderate yearly precipitation Petionville, at an altitude of 400 meters above sea level, has a slightly lower precipitation than Port-au-Prince, only "I kilometers distant and near sea level The heaviest precipitation, 2,659.4 millimeters, is at Mire balais, a.n inland station only 100 meters above sea level. The lowes t recorded precipitation, 524 millimeters, is at Gona1ves, which is only 3 meters above sea level. Other stations of very heavy annual precipitation, such as Les Cay e s and Bayeux, are on the coast practically at sea leve l. The figures do not seem to signify any clear rel ation between altitude and amount of precipitation. The types of vegetation suggest that the most arid areas are the lowlands, such as the. Artibonite Pla.in, Arbre Plain, and Cul-de-Sac Plain, but their aridity may be due in part to difference in temperature rather than to deficiency in rainfall. The following table gives the available records for Furey, the only sta tion that stands at a really high altitude (1,540 meters), and the records for Port-au-Prince during the same periods. Comparative rainfall in millimeters at Port-au Prince and Furey. Date. July 81 to Sept. 12, 1906 ........................... July 27 to Sept. 15, 1907 .............................. July 28 to Sept. 16, 1908 ............................. Aug. 1 to Sept. 15, 1909 ............................. Aug. 1 to Sept. 16, 1910 ....................... July 27 to Sept. 14, 1911 ................... Aug. 3 to Sept. 18, 1912 .......................... July 29 to Sept. 17, 1913 ................................ July 23 to Sept. 14, 1914 .......................... July 21 to Sept. 16, 1915 .............................. July 28 to Sept. 13, 1916 ............................... July 26 to Sept. 15, 1917 ............................. Total .............................................. Dec. 1-31, 1907' ............................ Dec. 24, 1912 to Jan. 4, 1913 ............ Dec. 24, 1913 to Jan. 4, 1914 ................ Dec. 23, 1914 to Jan. 3, 1915 ................... Dec. 28, 19J5 to Jan. 3, 1916 ................... Total e e e e e e e I e e e e I e e e e e e e e e e e e e I Furey. Port-auPrince. 246.3 250.2 271.2 149.2 303.2 216.9 645.2 282.2 412.0 385.7 298.7 162.5 363. 8 278.8 374.4 184.8 224.4 343.2 532.0 301.0 438.8 100.5 381.4 384.2 4,481.4 8,040. 7 17.3 0.5 0.0 4.6 0.0 0.0 5., 1.7 4.6 1.4 27.3 8.2 These figures seem to indicate that the rainfall at Furey is nearly 50 per cent greater than that at Port-au-Prince, but it is not known whether the rainy seasons at the two places are synchronous and the results there fore can not be regarded as conclusive. Moreau de Saint-Mery states repeatedly that there is greater precipitation in the mountains than i n the plains. Valliere, he says, is much wetter than Fort-Liberte. He esti mates the yearly rainfall in the mountains back of Le Trou at an averag e

PAGE 58

GEOGRAPHY. 53 of 2,320 millimeters, about 60 per cent greater than that at Le Trou. H is figures for Bombardopolis give a particularly valuable comparison with Mole St.-Nicolas, indicating nearly 5 0 per cent greater precipitation on the plateau near Bombardopolis. In general, it seems that the precipitation on the mountains may av erage 5 0 per cent more than that in the lowlands. The controlling feature of the rainfall of the Republic of Haiti is the influence of its many mo 11ntain ranges on the moisture-bearing trade winds in t he paths of which it lies The general direction of these trade winds in the West Indian region is northeasterly, but ove r the Rep ublic of Haiti they appea r to incline somewhat more to the east. These winds, when they strike the high ridge of the Massif du Nord from Ouanaminthe to Port-de-Paix, deposit the greater part of their moisture on the north eastern slope of these mountains and the adjacent North Plain. Thus dried and rarefied they pass down the southwestern slope of these mo11ntains and across the Central Plain and the Gona!ves Plain, and this region becomes increasingly arid as its altitude decreases These winds appear to come so nearly from the east that the whole of the Northwest Peninsula is in their lee and consequently that area is much more arid than its exposed positio n would suggest Only on the northern slopes of its highest ranges is there enough rain to support forest growth and to raise crops without irrigation. The entire bordering low land and much of the western plateau are arid or semiarid. The winds that have crossed the Massif d u Nord o r the contiguous heights of the Dominican Rep ublic and the Central Plain, sweep on across the Montagnes Noires, the Chaille des Mateux, and their connecting ranges. The summits of these mo 11ntains are at some places high enough to cause considerable further condensation and to vroduce rainfall to s up .. port forests and garden crops But the leeward lowlands, the lower Arti bonite Plain and the C ul-de-Sac Plain, are arid or s emiarid, and cultivation is there impossible without irrigation. The upper part of the Arti bonite Valley appears to be so narrow that it shares the precipitation con densed by it.a surrounding mountains, and Mirebalais therefore has the highest rainfall of any station on record. The Massif de la Se lle, although i t contains the highest peak of the Republic, does not appea r to have the most ab11ndant rainfall, beca use the winds have been robbed of so large a share of their moisture before reach ing it. As compared with the Cul-de-Sac Plain and the south coast, however, it receives relatively high precipitation. West of the Massif de la Selle the Southern P eninsula seems to receive more and more rain as it projects farther and farther beyond the shelter of the northern and eastern parts of the island of Haiti, and in the latitude of Jeremie it is one of the best-watered parts of the Republic Here, as elsewhere, the north coast gets more rain than the south c oast, much of

PAGE 59

54 GEOLOGY OF THE REPUBLIC OF HAITI. which is in fact rather arid with the one pe culiar exc e ption of Les Cayes. Perhaps the anomalously high rainfall at L e s Cayes may be explained in part by the fact that although it is on the lee side of the main body of the peninsula it is also on the windward side of the mountainous projection that terminate.s in the Po1t-Salut Peninsula. The precipitation c omes in rainstorms of two very different types. The period from October to March is the season of '' nords '' slow and steady rains of long duration borne by northe ast erly winds. These rains may last from several hours to as muc h as two or three days. Few of these rainsto11ns yield a very large quantity of water. The y are particu. larly characteristic of the north coast from Port-de-Paix eastward and in the western part of the Southern Peninsula. They raise the winter rainfall at Cap-Haitien, Bayeux, Le Borgne, Port-de Paix, Grande Riviere du Nord, Jer emie, and Moron above that of the other stations. The other type of rainstorm prevails in summer and is doubtless due to the northward migra tion of the equatorial storm belt at this season. of the rain in the drier parts of the Republic falls in violent storms of short duration, many of them accompanied by high wind, th11nder, and lightning. Late in the summer many su c h storms, especially those on the south coast, become d e structive hurric anes, causing great damage. Occasionally the country is swept by violent hurricanes, u s ually accom panied by heavy rain, which cause tidal waves that do great damage to shipping and to bui1ldings on the water front. Especially destructive hur ricanes occurred in Augu s t, 1909, and in August, 1915. The long periods of unbrok e n drought at certain stations directly affect the population. At Mole St.-Nicolas, St.-Michel de l' Atalay e and J acmel two consecutive months of no rainfall have been recorded, and at Gonaives, St.-Marc, and Thomaz eau three consecutive months. A very great part of the rain at these stations comes in h e avy storms. During the year 1919 at all the stations of observation 38 p e r cent of the total ra.inf all by months fell in single storms, althou g h s e veral rainy day s usually were recorded each month. The int. e nsity of r a ins is briefly summed up by Scherer 1 as follows: Nos pluies sont relativement courtes. La plupart durent mains d une heure. u ne pluie de deux heures parait deja longue. Rarement ell es depassent quatre heures. Une pluie de 12 heures ou de 24 heures est regardee comme extraordi naire. Elle suppose une p e rturbation atmosph e rique. D'aill e urs ce ne sont pas les longues pluies qui fournissent beaucoup d eau au pluviometre. G e neralement les courtes pluies sont aussi les plus intenses. This characterization of the storm s of the Republic corresponds with a further statement by Scherer in the sam e article: '' Chez nous les pluies sont plutot locales que regionales.'' 1 Scherer, J., Observatoire du St. Martial Bull. Ann., 1919, p. 99, 1920.

PAGE 60

GEOGRAPHY. 55 The hours of rain or the time of day at which rain is most likely to fall has also been determined by Scherer, 1 whose conclusions, based on records tabulated at many stations, are summarized as follows: Avant d'examiner quelques tableaux en particulier, indiquons l es points principa11x sur lesqu els doit porter la l ec ture. Ces points sont le minimum et l e maxi mum de frequ ence et l amplitude, c'est-a-dire l 'ecart entre les valeurs extremes. Au bas de chaque tableau on consultera d'abord les totaux. On pour ra constater vers les 9 h. ou 10 h. du matin un minimum de frequence, commun presque a tout le pays. Les maxima indiques dans l es tableaux surviennent dans l 'apres-mi di, a des heures diff erentes selon ] es stations, l es uns plus repproches de midi, l es autres du coucher du soleil selon la region. D'autres stations, comme les Cayes et Jeremie. ont une phase double: deux minima et deux maxima nettement marques. Si l'on compare l es resultats mensuels d'11ne meme station, on constate d une part 11n leger deplacement pour le minimum du matin et d' autre part pour le maximum une avance vers midi ou un recul vers minuit, suivant les saisons. En certains pays il pleut a peu pres aux memes heures toute l'annee; e n d'autres, des pluies tardives semblent coincider avec l 'affaiblissement du regime pluvieux annuel. Hail, usually accompanying summer thunderstorms, is fairly common in the Republic, and many hailstorms are noted in the bulletins of the Observatoire Meteorologique. They seem to have been very comm on at St.-Micb e l de l' Atalaye. }.1oreau de Saint-11ery speaks of hailstones ''as big as the fist'' which fell at Fort-Liberte on Ji1ne 17, 1785, and as large as pigeons' eggs at Boucassin May 30, 1786. Such hailstorms do much damage to growing crops. RELATIVE HUMIDITY. The relative humidity appears to va .ry within wide limits. It is very different for di 'ff erent localities a t the same time, and for the same lo cality at different times of the day, although the monthly means vary but little from year to year. Tl1e humidity is rarely so great anywhere as to cause great physical discomfort. It is probably less than in many tropi cal regions, so that the warm climate is more endurable. Observations of relative h11midity at Port-au-Prince covering many years give the following monthly means: January, 65; February, 63; March, 64; April, 69; May, 73; J11ne 68; July, 64; August, 68; September, 73; October, 75; November 72; December, 68. These figures show a marked correspondence of high humidity with the rainy seasons and of lower humidity with the dry seasons. The daily fluctuation of relative h11midity at Port-au-Prince usually covers a range of 30, the minim11m being about 50 at midday and the maximum about 80 sometime between midnight and 4 a. m. The follow ing table of hourly means for the year 1910, is characteristic of the daily range of humidity at this station: 1 Idem., p. 100.

PAGE 61

I 56 GEOLOGY OF THE REPUBLIC OF H.AITI. H ourl11 mean humidit11 at Port-au-Prince, Haiti, in 1910. 1 a. m. . . . . . 82.9 2 a. m. . . . . . 83.3 3 a. m. . . . . . 83.6 4 a. m. . . . . . 83.4 5 a. m. . . . . . 832 6 a. m. . . . . . 82.7 7 a. m. . . . . . 79.4 8a.m ............ 71.3 9 a. m . . . : . . 63 .5 10 a. m. . . . . . 56.8 11 a. m. .......... 52.0 12 m. . . . . . . 50.6 1 p. m. . . . . . 52.8 2 p. m. .......... 56.9 3 p. m. . . . . . 60.4 4 p. m. . . . . . 62.9 5 p. m. . . . . . 65.9 6 p. m. . . . . . 71.3 7 p. m ............ 75.7 8 p. m. . . . . . 78.2 9 p. m. . . . . . 79.8 10 p. m. . . . . . 81.5 11 p. m ............ 82.2 12 p. m. . . . . . 82.4 The average for this year, 71.8, is someY\1hat higher than the general yearly average, which is about 68.5 The only other station for which records of observations are published is Furey, and these records like the records of temperature, are incom plete. For the midsummer period, including August and parts of July and September, the mean, based on observations at 7 a. m., 1 p. m., and 9 p. m., is 80, and for the Christmas or New Year period the mean is 78. These figures show that the h11mjdity at Furey is considerably higher than that at Port-au-Prince. Humi'dity above 90 appears to be common at Furey, and the climate there would be very uncomfortable if the temperature were higher. WINDS. As the Republic lies in the belt of northeast trade winds its prevailing winds are from the northeast or east Along the northern coa.st they blow very nearly from the east, probably because they are deflected in that direction by the configuration of the land. The same deflection may occur in Cul-de-Sac Plain. These winds are steadiest and most unvarying in direction in winter. In summer, as the equatorial belt moves northward, they are often replaced by local and variable storm winds. The trade winds are locally much modified and obscured by the daily land and sea breezes of the coast. Few parts of the Republic except, per haps, the Central Plain, are far enough from the sea to be beyond their influence. For this reason very strong and persistent winds from one direction, such as cause the accumulation of wind-bloWh sand and produce marked asymmetry in the growth of vegetation, are uncommon in the Republic. In general the sea breezes blow at right angles to the coast and the land breezes blow directly out to sea, but each shows a strong tendency to veer somewhat to the east, probably in conformity with the prevailing direction of the trade winds. Thus on the south coast the land breezes usually come from the northeast and sea breezes from the southeast The land and sea breezes do much to mitigate the tropical heat. When the sea breeze fails for a day in summer, as it does occasionally at Port au-Prince and elsewhere, the heat is very oppressive. Moreau de SaintMery says that at Croix-des-Bouquets the sea breeze has failed for as much as three weeks consecutive ly. On the coast the breezes generally are very regular and determine the hours of sailing of small vessels engaged in fishing and coastwise traffic.

PAGE 62

GEOGRAPHY. 57 TYPES OF VEGETATION. Although the Republic is a tropical country it includes only a few smal l areas where the vegetation presents the aspect of an impenetrable tropical rain forest, the aspect that is so common ly visualized by one thinking of a tropical region Differences in rainfall, soil, and altitude cause the vegetation to change constantly within remarkably short distances. The indigenous flora is very rich in species and has been little studied. 1"'he observations recorded here are intended only to give an impression of the general features of the vegetatio11, of the relation of the vegetation to the geology and the climate, and of the possib l e value of the fores ts. Most of the area of the Republic is covered by two contrasted types of vegehl. tion the mesophytic type, which is developed where rainfall is normal or abundant and which includes most of the forests, and the xeroph)1tic or desert type which prevails in areas of slight rainfall. There are in addi tion small but significant areas of halophytic and of shoreline vegetation FORESTS. OCCURRENCE AND GENERAL .APPEARANCE The forests of the Rep ublic are generally, though not uniformly, con fined to the mo11ntainous areas the northeast slopes of tl1e mountain ranges receive much more rainfall than the southwest slopes, they are mo1e heavily wooded and bear forests at much lower altitudes Every where, howev e r, the forest growth becomes thinner and more stunted toward the base of the mo11ntains The forests visible from the sea are therefore generally rather scrubby in appearance and do not give a fai r impression of the co11ntry as a whole Probably none of the forest.s any where fairly represents the beauty and the extent of those that existed before the discovery of the island During the days of the French colony large areas were cleared to make room fo r plantations of coffee, cacao, and other crops, and such clearing, frequently assisted by fire, has been co11tinued to some extent eve r since Valuable wood in accessible locali ties has been exploited commercia lly during and since colonia l days Finally, the cutting of firewood and the burning of charcoal, the o n ly natural fuels available, cause a continued and enormous wastage It is therefore not surprising that virgin forests, if they occur at all, are found only in the most remote and inaccessible localities KINDS OF TREES. The writers are familiar with the names of only a few of the more common and conspicuous trees. The native oak (bois de chene), the tavernon, and the sandbox tree ( sablier) are among the larger and more handsom e forest trees The silk cotton tree ( arbre a coton) reaches a

PAGE 63

58 GEOLOGY OF THE REPUBLIC OF H.AITI. large size. Large areas in the Southern Peninsula contain a tree that resembles the American walnut. Most of the forest.s at moderately high altitudes contain large trees of the vV. est Indian cedar ( acajou sen ti). Pines are found on some of the mountain tops and under exception al conditions on the plains.. The royal palm. (palmier royal), with it.s great green head surmou .nting a clean slender trunk from 10 to 18 meters high, is one of the most graceful trees of the Republic. It grows at moderate 01 low altitudes where the soi l retajns moisture fairly well. Fan palms (palmiers a eventail) are confined to low altitudes, usually on calcareous soils. Both the base of the fronds of palms ( tache) and the fronds are used as 1oofing material. The coconut palm ( cocotier), which is very common at low altitudes, is generally a cultivated tree. The calabash tree ( calebassier) bears on it.s trunk great green :fruits resembling gourds, which are much used for water vessels and bowls. The largest tree in the Republic is the mapou, which at some places attains a diameter of nearly 3 meters. (See Pl. IV, A.) It grows singly at moderate or low altitudes w11ere the climate and soil are not excessively dry. At many localities, especially where the rainfall is not excessive, leguminous trees and shrubs are numerous. The coral bean tree (bois immortel) i's used to make living fences. The trumpet tree (bois trompette) is grown as an ornamental tree aro11nd houses, especially in the Southern Peninsula. Many of the native fruit trees are common in the forests. Among them are the alligator pear ( avocatier), sour sop ( corossolier), sapote ( sapotillier), mammee tree (abricotier), sweet sop (pomme cannelle or cachiman cannelle), custard apple ( cachiman creme and cachiman coeur-de boeu:f), star apple ( caimitier), and the native cherry ( cerisier). Several fruit trees from other tropical countries have also spread over the island and many of them grow wild in the forests. Among them are the orange (orange), lime (citron), grape-fruit ( shaddock), bread-fruit ( arbre-A-pain), mulberry ( murier), guava (goyavier), and mango ( manguier). The mango grows singly or in clusters spaced at ample intervals. It.a wide, ro11nded tops bear an jmpenetrable canopy of dark-green leaves, making itt one of the best and most handsome shade trees of the country The Republic contains valuable cabinet woods, especially mahogany ( acajou), but the supply of mahogany has now been so greatly depleted that only a small production from isolated and more or less careful ly gt1arded trees is possible. Some mahogany is used locally and some is exported. Other valuable cabinet woods are the manchineel (mancencillier), satinwood, rose wood (bois rose), cinnamon wood (bois cannelle), yellow acoma ( acoma jaune), a kind of ebony ( ebene), and gris-gris. Lig n1un vitae (gayac), a very hard wood that is in special demand :for making tools and parts of machines, is exported in large quantities when trade conditions are favorable. Dyewoods, especially logwood ( campeche), have been exported on a large scale. Logwood is not indigenous but was intro

PAGE 64

BSPUBLIC OF HAITI GEOLOGICAL SURVEY A. l\fAPOU TREID IN MEME VALLEY NEAR Il E N ElT,.,. E I PLATE IV B. XEROPHYTIC VEGETATION IN THE ARTIBONITE PLAIN SOUTHEAST OF GRANDE-SALINE. Principally cact11 and

PAGE 65



PAGE 66

GEOGRAPHY. 59 duced in colonial days and has spread rapidly over the Republic, where it grows best in dry, calcareous soils. Brazil wood (bresillet), fustic ( fustet), and sassairas are other dyewoods. ASSOCIATED VEGETATION. Associated with the forest trees at high altitudes (above 1,000 meters) there are many ferns ( f ougeres), tree ferns ( f ougeres arborescentes), and a great variety of 11n derbrush. The tree ferns and begonias are especially characteristic of areas of volcanic rocks at high altitudes. At lower altitudes bamboo (bambous) and heliconias (bananier marron) are found in moist ground along streams. Bamboo grows in dense clusters to a height of 20 meters and a diameter of 10 or 15 centimeters. The hollow stalks are cut in lengths of about 2 meters, the ends are plugged up, and the hollow pipe thus m ade is used for carrying water. Bamboo has been introduced from the orient, but there are several similar indigenous plants. The trees at all altitudes contain a large number of epiphytic bromeliads and orchids. A great variety o f lianas (lianes) festoon the trees at all altitudes. Some oI the striking lianas are the barrel liana (liane a barrique)' which is used in making barrel hoops, the water liana (liane-a-l'eau), which yields potable water, the snuffbox sea bean (liane coeur-de-St. Thomas), and ox-eye sea bean (liane a cacone or yeux de bourrique). The seeds of the snuffbox sea bean and ox-eye bean are the most common West Indian seeds that are carried by the Gulf Stream to the shores O f northeastern Europe. Low down on the mountain slop es the forests are replaced by xerophy tic plants and grasses. On the northeastern slope of the Massif du Nord and in the northwestern part of the Southern Peninsula trees and shrubs that are elsewhere found at high altitudes extend down to unusually low altitudes. NOTES ON LOCAL AREAS. The uncultivated land in the Massif du Nord is generally rather heavily forested on the crest and well down on the northeastern slope. The south western slope, facing the Central Plain, and the lowlands of Gona1ves are invaded by xeropbytic vegetation. The crest of the limestone ridge northeast of the Central Plain is covered with thick, scrubby forest, but north of Cerca-la-Source the vegetation changes abruptly to an open pine forest with short brown grass and little underbrush except in ravines. This vegetation gives way to dense forests with much underbrush and no pines along the crest of the divide between Lamielle and Mont Orgap.ise, perhaps owing to a difference in precipitatio n, as there is no apparent difference in soil. In the vicinity of Mont Organise, however, there is open pine forest, and scattered pines extend down into the edge of the North Plain at an altitude of 100 meters above sea level, the lowest alti-

PAGE 67

60 GEOLOGY OF THE REPUBLIC OF HAITI. tude at which they were seen anywhere. Here they are confined to ravines. Pines are common as far to the west as Dondon, but in all this region they are confined to areas of igneous or metamorphic rock. The valleys of Grande Riviere du Nord, Dondon, and those to the west contain magnifi cent sandbox ( sablier) and mapou trees, most of them in waste corners and along fence rows, for the land is culti vated. Only the s11mmits of ranges and some of the mo11ntain valleys in the Northwest Peninsula are much forested. The tavernon, mapou, and mango are conspicuous trees in the valleys. There is a small stand of particularly large trees in the valley west of Terre-N euve. Pines grow only on the crest of a range just north of the Sources Chaudes. The other trees are said to include many valuable species. Mahogany appears to be restricted mainlJT co the lower mountain slopes. There are many isolated trees in the Commune of Gros-Morne but they could not be extensively exploited. The Bombardopolis Plateau contains large areas of very dense forest, apparently new growth, and a great deal of lignum vitae (gayac), which, however, is kept down by constant cutting, for it is the chief article of export at Mole St.-Nicholas. The limestone ridges of the Montagnes Noires are generally thickly timbered, but large and perfect trees are confined to small areas. At places forests extend into the Central Plain. The trough at the foot of the mountains on the southwest and south margins of the plain supports a heavy forest, as does also the depression southeast of Riviere Canot, just northwest of Ma1ssade. A peculiar feature of the Central Plain is the occurrence of pines along certain ridges about its edge and in the interior between Ma1ssade and Pignon. These ridges mark the outcrop of folded beds of coarse sandy rock that appears to furnish a soil espe cially suited to the pines. The sides of the Chaine des Mateux bear only scrubby brush, and most of its crest is a grassy or bushy savanna. Guavas are thick and attain almost the dignity of trees. Northeast of Couyau and elsewhere there are considerable areas that have a thin stand of pines, which grow in tall guinea grass. Farther southeast the conditions are much the same, and there is little forest. Grands-Bois, or Cornillon, gets its name from the forests which once existed there. Moreau de Saint-Mery states that the forests were cut down in colonial days to make room for co:ffee.1 Several years ago pines and other trees from this region were cut into lumber at a sawn1ill at Glore, on the northern shore of Etang Saumatre, but the mill has burned down. There is now only a thin stand of pines in this region. The pines of the ChaSne des Mateux and Grands-Bois grow on limestone, and are virtually confined to altitudes of 1,000 meters or more above sea level. 1 Op. clt., vol. 2, pp. 294 et seq.

PAGE 68

GEOGRAPHY. 61 The higher uncultivated parts of the Massif de la Selle are generally forested. The lower slopes contain merely brush. Along the Dominican border, south of Fond-Verrettes (also called Mission), scattered pines grow on a lim estone terrane from an altitude of about 1,000 meters up to 1,400 meters above sea level, where they give way to a dense tropical rain forest. Above 1,600 meters, however, an open pine forest in a thick stand is fo11nd again. This pine forest extends up to the very peak of }.font La Selle. On the south slope, from 1,500 meters down to 1,000 meters, the belt of the tropical rain forest is strongly developed and grades into xerophytic vegetation lower down, near the south coast. Similar conditions are found at corresponding altitudes farther west. Near Furey there is a large area of depleted pine for est, some of it on limestone but most of it on basaltic rock. The s11mmit of the divide between Jacmel and Port-au-Prince, at an altitude of about 1,300 meters above sea level, supports a dense tropical rain forest. The limestone lo wer down on the north side of the mountains contains many trees but no r eal forest. The mo11ntains of the Massif de la Hotte, particularly the Montagnes de la Hotte and their northern slopes, in some places even down to the coast, bear heavy forests that probably contain valuable species. 'l,he vici nity of Baraderes Bay was one of the most famous lumbering regions of colonial days, but the virgin forests of the present day are confined to the interio r, and roads would have to be made to exploit them. There are practicable routes up the valleys of the Grande Rivi ere de Jeremie, the Rivier e de Tiburon, the Riviere des Baraderes and the Grande Riviere de Nippes. The route up the Grande Riviere de Nippes would tap the rich agri cultura l land of the Asile Valley. This area probably offers the best opportunity for the development of the lumber resources of the Re public, and it is at least worthy of a survey by a qualified forester. Gonave Island contains no extensive forests. Large isolated mahogany trees are fo11nd in the interior of the island. All the upland is covered with thick shrubbery except areas that were cleared and have grown up in guinea grass. The inclosed lowland called the Plaine des Mapo11x gets its name from some large mapou trees. The forests of Tortue Island were famous in colonial days and were then extensively exploited }.foreau de Saint-Mery states that nearly all the valuable species of the colony grew there, among them mahogany, gayac, and bois-marie or dame-marie.1 At present the eastern end of the island is partly cleared, and large areas in it are abandoned to guinea grass. The western third and most of the northern slope are forested. During a brief visit to the region near Vallee our party saw some gris-gris but no mahogany. A few years ago a Belgian company attempted to exploit the timber on this island by building a narrow-ga ,ge railway north ward from Vallee. The grade was established, a little track was laid, and Op. cit., vol. 1, p. 7 40.

PAGE 69

62 GEOLOGY OF THE REPUBLIC OF HAITI. two locomotives were installed, but they are now idle, and no timber appears ever to have been cut by the company. Logwood has been exploited with more success in recent years than any other forest product. At Grande-Riviere du Nord a plant for the extrac tion of dye liquids was operated for several years. It burned down but has now been replaced. This plant is supplied mainly from groves of logwood on the north side of the Central Plain near St.-Raphael and Pignon. At Port-de-Paix a remarkable aerial tramway has been constructed to convey logwood a distance of about 18 kilometers from Bassin-Bleu to the coast. The supply is drawn from the slopes of the val ley of Les Trois Rivieres but is said to be very much depleted. Good groves of logwood were seen in the valley of the Artibonite southeast of La Chapelle. There is much logwood along the south coast, and it is exported in considerable quantity from Miragoane. Figures showing the export value of different forest products are given under the heading '' Commerce '' (p. 79). XEROPHYTIC VEGETATION EXTENT AND GENERAL FEATURES. The xerophytic vegetation of the R e public is typical of its plains, par ticularly the arid plains in the lee of the mountain ranges, such as the Cul-de-Sac, Artibonite, and Arbre plains. The vegetation at some places on these plains is of an extremely arid type The xerophytes usually ex tend some distance up the mountain slopes, especially the southwestward facing slopes, where they even approach the crests of high ranges, as at Ennery. They include innumerable spiny plants, among which cacti are prominent and at some places dominant. The genera l aspct of these arid plains is sterile and forbidding, and their great extent, particularly over the more parts of the Republic, is 11npleasantly surprising to travelers who see the co11ntry for the first time. KINDS OF PLANTS. The cacti are the dominant plants in extremely arid regions. There are n11m. erous kinds of cacti, many of which seem to be similar to species fo11nd on the mainland of North America. Arborescent fonns are won derfully developed and some of them attain a height of 6 or 10 meters and have trunks more than 30 centimeters in diameter. Cacti are espe cially typical of the northwestern part of the Cul-de-Sac Plain, the lower Artibonite Plain, and the vicinity of the Arbre Plain (Pl. IV, B). Cer tain kinds of cacti, particularly opuntias ( raquettes), are widespread and are not confined to the typically arid regions. Coll1mnar cereus is used for hedges at many localities outside the zones of xerophytic vegetation. Op11ntias, associated with pine forests, are reported from Mont la Selle, the highest peak in the Republic.

PAGE 70

RllPUBLIC O F HAITI GEOLOGI CAL SUR\'EY PL.ATE V .tl. IX TlIE CUL -DE-SAC rr, .. NEAR BEUDET n. THICKET OF O ERCIDIUM PRAEC O Y ( A PAT.10 "ERDE) ABOU T ;:> N ORTITEAST OF GO:\' .<\ OX TIIE l TJ 'l' ERI{

PAGE 71



PAGE 72

GEOGRAPHY. 63 Cacti are not by any means the only plants of the xerophytic regions. Where the soil is fertile and retains moisture well large areas of the plain are forested with a thick tangle of bayahonde, a scrubby leguminous tree r esembling the American mesquite (Pl. V, A). This wood is h .ard and t akes a good polish. The trunks of some of these trees reach a diameter o f 0 .6 meter, but they are knotty and gnarled. The bayahonde is used a great deal for fuel, some of it as charcoal. It is interesting to note that a small palo verde, a tree that has not been previously reported from the R epublic, was fol1nd on the desert plain north of Gonfilves, on the trail t o Terre--Neuve. }tlr. Paul C. Standley, of the United States National Museum, reports that it is probably Oercidium praecox, the Mexican species. (See Pl. V, B.) Acacias, yuccas, and agaves, of different kinds, as well as palmetros, grow on the mountain slopes. Many palmettos rise s omewhat abo ve the other trees and are spaced at intervals so regular t hat they give the landscape the aspect of a weird orchard. Palmettos are c h aracteristic of transition zones between the xerophytic and mesophytic vegetation. Certain species furnish material for baskets, and the aloes f u rnish fibers for rope, two articles that are manufactured and used l ocally in great quantities. Many of the trees and shrubs in the xerophytic z ones are covered with masses of Tilloodsia and other bromeliads, as well as with other epiphytic plants. Agaves are at many places used as h edges. SA v .ANN .AS. Grassy savannas are numerous and at places extensive. Some of them, particularly those on the plains, a.re natural prairies; others, especially those in the mo11nta.ins, occupy land that was cleared for agriculture and abandoned. Most of the cleared areas are covered with guinea grass, but the natural savannas are covered with native grasses. The grass in the more hi1mid areas is often 0.5 to 1 meter tall, but throughout most of the year it is rather dry and brown, especially on the natural savannas. Where sto c k raising is f e asible it is often eaten down. The savannas of the plains are most extensive on the North Plain and the Central Plain. On the North Plain there is a belt of savanna 4 or 5 kilometers wide at the base of the mountains from Les Perches to Ouana minthe. The savanna is very flat except where it is dissected by narrow, steep-sided ravines. In the ravines there is a healthy forest growth, includ ing pines, but else where there is scarcely anything but grass. At its north ern border the savanna grades into a lowland covered with xerophytic shrubs, probably because rainfall is less abundant at this distance the mountains. The northwest part of t he Central Plain is a flat, grassy savanna, in which there are a few ravines that contain a growth of shrubs and small trees which causes them to stand out as sinuous green lines in the general expanse of brown gra"8s. Southeast of this large savanna the plain is more

PAGE 73

64 GEOLOGY OF THE REPUBLIC OF HAITI. dissected and consequently more wooded. The interstream areas are grassy, but are invaded by a shrub resembling greasewood. In this regio n there are a few areas of pine and other for est trees. (See p. 60.) Smaller savannas of this general type are seen in the Artibonite Valle y near Mirebalais and farther down the valley, as well as in the uper pa .rt of the Cayes Plain. Savannas of a second type, which occupy areas once forested, are found in the Massif du Nord, particularly in the Chaine des Mate11x and its connected ranges to the east They are found also in the Massif de la Selle, and on Tortue and Gonave islands. They generally occupy fairly flat or rolling uplands, not the steep and rugged mountain sides. The grassy savannas are associated with a thin stand of pines. (See pp. 59-60.) A third type of savanna is fo11nd in areas of volcanic rock, in the southern two-thirds of the Republic. These areas, especial ly the areas of basalt, where the rainfall is not very great, are covered with short brown grass (Pl. VI, A) and present a strong contrast to the adjacent areas of lim estone, which are densely wooded, although all the forests do not contain large trees Contrasra of this kind are very common througl1 -out the Southern Peninsula and near Saut d'Eau, on the north slope of the Chaine des Matel1x. A similar contrast may be observed between areas of the rather impervi ous chalky limestone and areas of the more massive and porous limestone The areas of chalky limestone support only low shrubs and flowering plants and a growth of short grass resembling that of the areas of vol canic rock. The areas of massive limestone are forested or are covered with tall guinea grass. This contrast was noted at places on the mo11ntains north of Ennery, on Gonave Island, and on the so,uth slope of the South ern Peninsula, in the region between Jacmel and Cotes-de-Fer, where the chalky limestones are extensively exposed. HALOPHYTIC VEGETATION. On the lower Artibonite Plain, where the soil is poorly drained and very alkaline, there is an extensive area of halophytic vegetation, in whi c h a variety of salt bush is particularly ab11ndant. (See Pl. VI, B.) There are regions of similar vegetation near the sea at the west end of the Cul-de-Sac Plain and at the outer edge of many other smaller coastal plains. Plate VI, 0, sho ,ws the halophytic vegetation on a mucl flat b ehind a lagoon and mangrove thicket on the north coast of Gonave Island west of Etroit. In all these areas the halophytes are replaced by xerophytes farther back from the shore. VEGETATION OF THE SHORE LINE. Extensive mangrove thickets are conspicuous along the shore line of the low coastal plains, especially in the North, Artibonite, and Cul-de-Sac

PAGE 74

REPUBLIC OF JTATTI GEOLOGfCAL St..:1 1 \.WY PLATE VI A. SA i;T A :\TF, r ... \. CIDRA, ABOLT 10 KIT,,O:\IE'l'ERS SO UTII" rEST OF I CII I> L'1 \ f ALA 111 Hc)n1e 1):1rts of t l1r I?e pt1bl i c g rassco\<'recl tl\T annas a r c c h a r t 1cteristi c <>f tloor ec 1 , itl1 \ O l cnnic B. SAI"' l J { l T SII JN l 'IIE ARTIBONl' l E J >LAIN N J ijAlt C h a racteris ti c of alkalin e so il 0. HALOPIIYTIC "\7EGETATION ON THE NORTH COAST OF GONAVE ISLAN D W EST OF ETROIT. In the back g1ou11
PAGE 75



PAGE 76

; GEOGRAPHY. 65 plains. The thickets are composed of several kinds of mangroves (mangliers or paletuviers), which are called red mangrove ( manglier rouge o r manglier chandelle), gray mangrove ( manglier gris), and black mangrove (manglier noir). These trees supply firewood and tan bark. Among the plants and trees that are conspicuous on beaches where there are no mangrove thickets are seaside grape ( raisinier bord-de-mer), seaside potato (patate bord-de -mer), and manchinee l tree ( mancenillier), which has a poisonous fruit ECONOMIC GEOGRAPHY. PO PULATION. TOTAL POPULATION. The favorable cljmate and the productive soil of t .he island have brought to it a dense population throughout most of its known history, which begins with its discovery by Columbus in 1492. At tha, t time the numb e r o f aborigina l Indi 'ans in the whole island was estimated at from 1,000,000 to 2,000,000, the lower estimate probably being more nearly correct. Under Spanish rule, within a few decades after the island was discov ered, this population was practically a .nnihilated by war and s lav ery and was replaced gradually by African negroes, who were imported in great numbers in the eighteentlt century, particularly by the French, who had taken possession of the western part of the island the part that now constitutes the Republic of Haiti. Moreau de Saint-Mery,1 the most trustworthy historian of the colonial period, estimates the total pop, u l ation of the French colony in 1788, near the close of the colonial era, at 520,000, of whom 40,000 were white, 28,000 were ''people of color'' (''gens de couleur ''), and tl1e remaining 452,000 were black. This popu lation inhabited an area smaller than that of the present Republic, as the Central Plain and adjacent mountain slopes were then Spanish territory. No reliable census of the Republic has ever been taken. War and famine and the exodus of the whites reduced the population considerab ly in the years during and immediately following the Revolution. Tippenhauer 2 regards Humboldt's estimate of 375,0 00 people in 1802 BB reliable and considers conservatively some later estimates. Even the best estimates, which are probably those made by the clergy for church use, he regards as generally rather large. The clergy estimated the population in 187 6 at 960,000 and in 1887 at 1,017,000, which Tippenhauer would reduce to 800,000 in 1880 and 90 0 ,00 0 in 1888. The latest ecc l esiastica l estimate in 1905 was 1,425,00 0 and more r ecent estimates range from 1,500,000 to 2,500,000. The former figure probably is more nearly correct. 1 Op. ctt., vol 1, p. 'Tippenhauer, L. G., Die Insel Haiti, pp. 420 et seq., Leipzig, 1893. 5

PAGE 77

66 GEOLOGY OF THE REPUBLIC OF HAITI. PRINCIPAL CITIES .AND TOWNS. Port-au-Prince, the capital, which is by far the largest city of the Republic, had a few years ago a population that was conservative ly est i mated at 100,000 Very recently the population has grown so rapidly that it is now estimated by some oi its inhabitants as h:igh as 200,000, but perhaps 125,000 is a safer figure. Fortunat 1 placed the population in 1888 at 60,000, probably including all the commune, which, however, would not greatly increase the total for the city The capita l has not always been the largest city of the Republic, for in colonial days it was greatly overshadowed by Cap-Frangais ( Cap Ha1tien) The population of Port-au-Prince in 1789, according to Moreau de Saint-Mery, wa.s abou t 6,200, to which he adds a transient population of 3,200 sailors and sol diers. Cap-Frangais was about twice as large. Cap-Ha1tien, called Cap-Frangais in colonial days and Cap-Henri during the reign of Christophe, had 12,151 inhabitants according to the official census of 1788. Moreau de Saint-Mery thought this figure too low and estimated the resident population at 15,000 and the transients (sol diers and sailors) at 3,550. At all eventiS the Cap was the largest and busiest city of the colony. At present it is not a great deal larger than :lt was then, for a census taken in December, 1918, is said to have shown 1 4, 000 people. Conard,2 however, thought the population wa.s neare r 20,000. Fort11nat gives 29,000 for the whole comm11ne in 1888, but this appears to be rather high for the city proper. I,e s Ca yes, frequently called Aux Ca yes in English literature, was the third largest city of the colony and probably stil l r etains that position In 1789, according to Moreau de Saint-Mery, it had 4,550 permanent and 1,100 transient inhabitants. The figure given by Fortunat for 1888 and still published in recent encyc lop edias is 25,000, but this represents the population of the whole commune and includes many residents of the Cayes Plain. The population of the city proper probably is not over 15,000. Conservative estimates of the population of other larger cities of the Republic are : Gona1ves, 12,000; St. Marc, 10 ,000; J acmel, 10,000; Portde-Paix, 7,500, and Jeremie, 7,500. All these cities are busy open ports. Among the other ports that have a population ranging from 500 up to a few thousand are Petit-Goave, Miragoane, Anse-a-Veau, Dame-1Iarie, Aquin, St.-Louis du Sud, l' Arcaha ie, Mole St.-Nicolas, and Fort-Liberte Among the inland towns whose population ranges probably from 5,000 down to about 1,000 are Mirebalais, Croix-des-Bouquets, Petite-Riviere de l'Artibonite, Grande-Riviere du Nord, Leogane, Gros-Morne, Hinche, St.-Michel de l' Atala ye, Las Cahobas, and J ean Rabel. 1 Fortunat, Nouvelle de l'tle d'Haitl, p. 349, Port-au-Prince and Paris, 1 888. 2 Conard, R. A., RePort on water supply of Cap-Haitlen to Engineer in Chief of Republic 01. Haiti, April 2, 1919.

PAGE 78

GEOGRAPHY 6 7 COMP.A.RISON OF URBAN AND RURAL POPULATION. The estimated population of the larger cities as given above is s11m marized as follows : Port-au-Prince ... Cap-Haitien ... Les Cayes ...... Gonaives ..... ....... 125,000 20,000 15,000 12, 000 10,000 J acmel .................... Port-de-Paix .............. Jeremie 10,000 7,500 7,500 207,00 0 Probably the people who live in other towns having a population of more than 1,000 would increase the total urban population to about 225,000, possibly even to 250,000. If the total population is 1,500,000 the percentage of city dwellers is about 15 to 17, figures which show that the population of the Republic is essentially rural and that agriculture is the chief industry. Although the population is so largely rural and depends for its living mainly on the soil, much of it is concentrated in small villages and rural comm11niti es of a hundred people or l ess rather than in detached houses occupied by one family. Whenever possible the Haitians live in little communiti es, some of them consisting of only three or four families, and as e ach of these communities has a distinctive name, an almost infinite number of place names appear on the maps and in the literature. DENSITY OF POPULATION. The area of the R epublic is about 27, 700 square kilometers. This figure includes the islands of Tortue, Gonave, and Vache, which belong to the Republic and are clos ely relate d to the main island geographically. If the total population is estimated as 1,500,000 or a little more the average density of population i s about 55 to the sq1iare kilometer, but if the city population is omitted the average density for the rural regions is about 45 to the square kilometer. The rural populatiion is, however, very 11 nevenly distributed, and larg e areas that have unfavorable climate, soil, or surface features are virtually unsettled. The whole of Tortue Island, for example, is spars ely inhabited. Gonave Island has a small population, probably considerably l e ss than 10,000, which is concentrated in little coastal villages and O n bodies of good agricultural land in the eastern half of the island. Ile-a-Vache and Grande Cayemite Island also have a relatively thin agricultural population. The density of population in fertile, thickly settled parts of the Republic is undoubtedly double the figure given, or at least 100 to the square kilom eter. The average density of population per square kilometer in France in 1910 was 78, in Germany 124, in Great Britain 154, and in the United States 12.

PAGE 79

68 GEOLOGY OF THE REPUBLIC OF HAITI. GEOGRAPHIC FEATURES INFLUENCING DISTRIBUTION OF POPULATION. The re is a v e ry natural concentrat i on of the population on the long, crooked coast. All the larger c ities are ports, which depend in great m e asure upon for e i g n c omm e r ce. Many small e r towns on the c oast are not open to foreign trade, but the y have exten s ive c oastwise commerce, par ticularly as f e eders for the large r ports. The almost impassable moun .. tains and the general lac k of roads greatly incre ase the volume of ocean and transportation. Mor e over, many of the alluvial plains that c on stitute the bulk of the agricultural land border the shore and favor the concentration of population nea-1' the sea. Fishing also is an industry that provides both o c cupation and su s tenance for a large number of the inhabi tant.s of the c oast. Som e lon g st1 .. e t c hes of coast, however, are 11napproach able even by small boats and are unfit for cultivation, and are therefore very thinly inha. bi ted. Over the stric tly agri c ultural r eg ions the density of population is closely proportion e d to the produ c tivity of the land in articles of food and in the f e w staple export crop s c off ee, cotton, sugar, and cacao. ""t\.lthoug h the forests at time s s upply a large part of the e xports they are work e d for the most part by labor drawn temporarily from near-by town s or agri cultural r eg ions. The produ c tivity of any part of the land depends on the surface fea tures, the climate, and the soil Closely r e lated to the surface feature s is accessibility to markets. Some small areas that have good soil an d favorable climate are i s olated by s o many nearly impas s abl e mountai n ranges that they are not thic kly s ettle d or carefully tilled. Such area s are found on the h e i g hts of the Massif de la S e lle and the 1Yiassif de la Hotte. l\fany other areas, e.specially pla c e s on the mountain sides, are so ste ep, rocky, and unapproa c hable tha. t the y are left in forest and senr e only as pasture for cattle, goats, and ho g s. Good examples are the rugged slop e s of the Morne du Cap and of the :Mont agnes Noir e s. In both mountain and p1ain the climat e especially the rainfall, greatl y influ e nces the produ c tiv ity of the land. Larg e are as that stand in the lee of mountain ranges and that are thus cut off from the life-giving east or northeast rains are arid and ste rile. Without irrigation they can suppor t only a s canty population. Among the se barre n areas are the Arbre Plain, the lower part of the Artibonite Plain, much of the Central Plain, and many small plains of the south coast. The upper part of the Artibonite Plain, the Arcahaie Plain, the Cul-de-Sac Plain, and other areas are saved only by irrigation. The lower southwest slopes of the high mountain ranges generally are not susceptibl e of r e clamation. Not the least es s ential r e quirement for successful agriculture is a good soil, for only where the soil is deep enough and fertile enough to produce crops year after year can a dense population be developed and maintained. The composition and texture of the soil is closely related to the area l

PAGE 80

GE OGRAPHY. 69 geology and is more fully d iscussed in Part II. The soils of the Republic may be classified briefly as alluvial soi l s, soils residual from limestone, and soils residual from igneous rock. The alluvial soils cover the surface of most of the larger plains and many similar but smaller valleys and plains. Although they constit11te probably less than 3 0 per cent of the total area they support perhaps more than 50 per cent of its population. The greater productivity of these plains is due not entirely to the natu1e of their soil but in part to their smooth surface and their superio r accessibility. The limestone that covers so large a part of the Republic yields on weathering a red clayey soil which is very fertile and which for some crops is superior to the alluvial soils Areas in which the surface is smooth enough to permit the retention and cultivation of this soil are well cultivat ed and thickly settl ed. Many small bodies of sucl1 land are scattered through the mountains, especially on the summits of tl1e ranges. Examples are the Bombardopolis Plateau, the crest of the Chaine des Mateux, the mountains of Dame-Marie and Jeremie, and the interior plateaus of Gonave Island. The soils derived from igneous rocks are generally thin and poor and support only a scanty population. Some of the granjtoid rocks of the north eastern part of the Republic, such as those at Valliere and the vo l canic rocks of Plaisance and Terre-N 0 uve are an exception to the rule. These rocks yield a fairly deep and fertile soil, which with sufficient rainfall is highly productive, and the areas so favored are wel l settled AGR ICULTURE GENERAL FEATURES. Agriculture has been the basic and dominant industry under both the colony and the Republic About 80 per cent of the people (see p 67) live in the rural districts and get their living from the soil. In this description the leading agricultural and live-stock product s are divided into two distinct classes those grown for export and those grown for local consumption The first class includes coffee, cotton, cacao, sugar, honey, and hides and skins The second c lass includes live stock and a great number of grains, vegetab l es, and fruits. As any discussion of the present state of agriculture inevitably l eads to comparisons with that i n the French colony, which was in its day the most productive and pros perous region in the New World, a brief review of the state of agriculture in colonial times is given with the description of its present state. Mos t of the information regarding colonial agriculture other than statistics of exports is taken from the work of Moreau de Saint-Mery, to wl1ich refer ence will be made by v olume and page wherever direct credit seems to b e necessary.

PAGE 81

70 GEOLOGY OF THE REPUBLIC OF HAITI. CHIEF EXPORT CROPS OF THE COLONY AND OF THE REPUBLIC. TOBACCO AND INDIGO. Tobacco and indigo, two crops that are no longer of great value, occu pi ed a prominent place among the colonial exports. Tobacco, the c rop .first grown, was cultivated by the buccaneers on Tortue Island.1 It was the leading crop 11n til the cultivation of indigo was begun. According to Moreau de Saint-Mery,2 the colony owed its first real prosperity to indigo Although wild indj go (indigo marron) grows in the Republic, the introduced variety (indigo franc) seems to have been preferred wherever it would grow. In certain soils, however, only the wild indigo would thrive In the ear ly part of the ei 'ghteenth century indigo was the leading export crop, but later it yielded its rank in turn to sugar, coffee, and cotton, and at the end of the colonial period it occupied the fourth p1ace. It was still the leading crop in a few areas, such as Jean l{abel and parts of the Artibonite Plain, where the lancl was too low, hot, and dry for coffee anu could not be irrigated to raise sugar c ane. The quantity of indigo and the leadin g ports from whi ch it was exported in 1791 are shown in the table on page 78. More than a third of the total was produced in St.-Marc, although Cap-Ha.ltien, Portau-Prince, and Les Cayes exported consider able quantities. Probably this supp ly was drawn from a much larger area than that immediately around the smaller ports, which owing to their commanding situations, were the principal ports of major divisions of the colony. One of the chief disadvantages of the culture of indig o seems to have been that it impoverished the soil rapidly and ceased to flourish Its c ultivation has been abandoned since colonial days, and the general use of synthetic indigo will probably prevent any considerable revival. Tobacco culture was dead at the end of the colonial days, and although the plant seems to succeed well, all later attempts to revive its cultivation have been short-lived, and the R e public now imports nearly all the tobacco it cons11mes. SUGAR. Sugar cane was introduced into Spanish Santo Domingo soon after the island was discovered, but it appears to have been first cultivated in the French colony on the Leogane Plain in 1680.8 It proved to be the most lucrative crop of the colony, and its cultivation spread rapidly to all areas where the conditions were at all favorable for it. The most famous plantations of the colony were in the North Plain, especially near Limonade and Quartier Morin and in the near-by valley of Limbe. Water 1 Labat, Voyage du Labat aux Isles de l' vol. 5, pp. 63-64, La Haye, 1724. 1 Idem, vol. 1, p. 24. 1 Idem, vol. 2, p. 450.

PAGE 82

GEOGRAPHY 7 1 power was us ed extensive ly to turn the cane mills, but irrigation seems t o have been re lativ ely little practic ed, as the natural rainfall genera lly suf ficed t o produce good crops In the west and south, however, on the Artibonite Plain, Cul-de-Sac P lain, and even the Cayes Plain, irrigation generally was imperative It was for the extension of sugar culture in these areas that the French built the remark abl e systems of irrigation works that stil l form the basis of the irrigation that is now practiced in the Republic. These works are more f ully described i n Part V The necessity for thorough tilling and preparation of the land and gen erally for irrigation also, as well as for a high temperature, has always restrict ed the raising of sugar cane to the all uvial plains and valleys, where the land is fairly smooth and f r ee from stones. It was on the four larger plains North Plain, Cul-d e-Sac Plain, Leogane Plain and Artiboni te Plain that the great bulk of colonial sugar was raised, and it is on the se and the Central Plain that any large modern sugar indust1y must be developed. Nevertheless the industry i n co lonial days extended to nearly every small a lluvial plain or narrow val ley where the conditions were favorabl e or could be adapted to the growth of suga r cane On the Leogane Plain long-continued cropping is said to have dep l eted the soil so much that fertilization was necessary .1 Sugar refining was highly developed in colonial days, and nearly half the quantity exported in 1 79 1 was refined The table on page 78 shows that the val ue of the sugar exported \Vas more than 50 per cent of the value of total exports in 1791 Political as well as geographical conditions made Port-au-Prince, Cap-Ha1tien, and Les Cayes the leading ports N orma lly Cap-Ha1tien probably surpassed Port-au-Prince, but its exports in 1791 were greatly curtaile d by revolutionary disturbances Sugar cane has always b een grown to supply the domestic demand Un refined brown sugar (rapadou), is an artic le of food, and raw sugar cane is highly prized A considerab l e quantity of sugar cane is used in making liquors, particularly rum. Plate VII, A shows a small cane mill of the type that is common throughout the Since 1915 an effort has been made to deve lop the sugar industry on a large scale, and an American com pany has estab lish ed a big plant at Port-au-Prince and laid out extensive plantations on the Cul -de-Sac Plain. The sugar exported in 1921 and 1922 (see table, p 79) was principally the output of this company COFFEE. Coffee was introduced from Martinique by the Jesuits 2 and was first grown successfully at Dondon, w here the first establishment for its prepa ration was installed in 1738. Its culture sp read with amazing rapidity, 1 Moreau de op. cit., vol. 2, p. 450. 1 Idem, vol 1, p. 164.

PAGE 83

72 GEOLOGY OF THE REPUBLIC OF HAITI for it was found to be especially adapted to the mountainous regions, where sugar cane could not be raised and where indigo had given only desultory success. Forests were cut down everywhere and coffee plantations took their place. They were still being extended rapidly at the time of the Revolution. It was estimated in 1908 that the total area of coffee plantations in the Republic was more than 500 square kilometers.1 Coffee is grown successfully between la.titude s 25 N. and 25 S. It thrives best between 300 and 1,500 meters above sea level, although under exceptional conditi0ns it is grown commercially at lower altitudes. In the neighboring island of Porto Rico the best coffee is said to grow on soils derived from igneous rocks,2 but although soils derived from igneous rocks produce excellent coffee in the Republic of Haiti they are probably no better ada.pted to its cultivation than some of the ocherous soils derived from limestone. Deep soil is necessary, for the plant has a long vertical tap root, but loose stones are not objectionable if the so i l does not dry out too rapidly. Excellent plantations are fo11nd on steep and stony slopes A fairly hca vy rainfall is nec essary, but some dry weather also is required. Deep, well-drained soils that sl1pport a good gro,vth of natural forest are generally suitable for coff Pe. Some shade is required, and several kinds of forest trees are used to obtain it.3 Coffee generally grows better on north slopes, mainly, no doubt, because they receive more rai11 but partly, perhaps, be ca use they are sheltered from tl1e excessive heat of the sun. Nearly all the north slope of the }fassif du Nord produced excellent coffee in colonial days, that of Le Horgne being especially famous. Coffee was the chief crop of the Montagnes Noires, of and of practically all tl1e mountainous area of the Southern Peninsula. The coffee of Grands-Bois was perhaps the most famous in the colon3r Coffee is cultivated now in much the same way that it was forn1erly. The plant grows virtually wild at many places. The chief producing areas are about the same as tl1ose of the colony. The coffee of is particularly esteemed in the French market. Plaisance and Petit-Goave are well-known producing areas The value of the coffee exported in 1791 constituted about one-fourth of the total value of the exports. In recent years the value of coffee has constituted about two-thirds of the total value of exports The quantity exported annually varies greatly, as is shown by the following table: 1 Internat. Bur. Am. Republics Bull., Dec., 1908, p. 1095. 2 McClelland, T. B., Indlcaciones acerca de la slembra de en Puerto Illco, Estac16n Experimental Agrfcola de Puerto Rico ci1c. No. 15, Mayagiiez, Oct., 1914. a A., Culture pratique du cafeier et p1epa.ration du cafe, p. 28, Paris, 1908. ,. Moreau de op. cit., vol. 2, p. 293.

PAGE 84

GEOGRAPHY. 7 3 P ounds of coffee exported annually in 1901-1905, 1909, and 1910.1 1901 . . . . . . . . . . 56,372,192 1 905 . . . . . . . . . . 38,306, 0 55 1902 ..................... 64,950,274 1909 ..................... 4 1 6 34,47 0 1903 ..................... 51,446,193 1910 ...................... 79,023, 1 68 1904 81,617,568 The Report of the Receiver of Customs for 1920 truly says that'' coffee is the principal crop of Haiti, the staple on which her people in the main dep end for exi stence, and the export tax thereon is the largest single source of revenue to the state.'' COTTON. C otton is indigenous to the West Indies, and in the island of Haiti i t grow s as a perennial shrub, attaining a height of 3 to 5 meters It grows wild but when desired for market is generally cultivated in a crude w ay I t thrives best if replanted every few years. The native plant seemB to have supplied practically all the cotton exported by both the colony and the R epublic. Moreau de Saint-Mery 2 records an attempt to introduce, at St.-L ouis du Sud, an improved variety, which however, does not appear to 11ave become extensively cultivated. The cotton plant requires some rain in the early p eriod of its gro\vth but will endure much drought later. It appears to grow best on calcareous soils and at low altitudes. I t is the l e a ding crop of certain low land areas, such as the Arbre Plain, parts of the Artibonite Plain, and Cul-de-Sac Plain, where irrigation is impractica ble and the climate is too hot and dry for coffee. In colonial days it was exported chiefly from and Port-au-Prince. (See table, p. 78. ) Cotton is now generally cultivated in about the same way and to the sa me extent as in colonial days. The quantity exported varies greatly, as the price fluctuates, depending upon the quantity picked rather than upon the quantity grown, which usually exceeds that marketed. When prices ar e 11igh cotton is marketed in great quantities, but when prices are low it m ay nearly disappear from the market. An American company recentl y attempted to introduce cotton-growing on a large scale in the Central Plain. An impro'\ r ed American cotton was introd-qced, but it failed to m ature properly during the first season, and further experiments have b een checked temporarily by a great drop in price. Cotton raising, howeve r appears to offer much promise of success when its details have been th oroughly worked out. Similar difficulties were encountered and surmounted in introducing the cultivation of sea-island cotton into the British West Indies.a 1 Figures from Rapports de la Chambre des Comptes, in Le Moniteur. For 1901-2 see No. 27, p. 205, 1905 ; for 1903 see No. 42, p. 327, 1905 ; t.or 1904-5 see combined No. 1 6 and 17, p. 101, 1910 ; and for 1909-10 see No. 34, p. 243, 1912. 1 Op. cit., vol. 2, p. 648. 1 See publications of Imperial Dept. Agriculture for the West Indies.

PAGE 85

74 GEOLOGY OF THE REPUBLIC OF HAITI. CACAO. The cacao tree is another indigenous plant. Its v alue was recognize d early in colonial days, but it never attained the importance of coffee It ranked fifth, however, in the order of value of colonial exports in or sixth if syrup, a by-product of the sugar industry, is considered. The principal area of production in colonial days, as now, was the west en d of the Southern Peninsula, in the comm11nes of Dame-Marie and Jeremie. Here cacao in great measure takes the place of coffee, growing in large forests or orchards on the gentler slopes of the limestone mountains Early in the colonial period St. -Michel de Fonds-des-Negres was famous for its cacao. Cacao grows between latitudes of 20 N. and 20 S., and thrives best at altitudes less than 800 meters above sea level, in deep, well-watered, but well-drained soil It requires close shade and thorough protection from strong winds Banana trees, rubber trees, bois irnmortelle, and other trees are us ed for shade and windbreaks. More rain is required for cacao than for coffee. These exacting conditions therefore restrict the range of the cacao tree considerably more than that of the coffee tree.1 HONEY. Although honey is only a minor export, it appears regularly among the list of exports of the Republic. Bee culture was introduced in colonia l days but does not seem to have attained much success. Now, however, it is a recognized industry, probably capable of considerable extension Haitian honey is of excellent quality and probably could with little effor t obtain a wider market. CROPS GROWN FOR DOMESTIC CONSUMPTION. Many grains, vegetables, and fruits, both cultivated and wild, form a part of the dietary of the people but seldom find their way outside the country. The principal grains are corn ( ma1s), petit-mil, and rice ( riz). Corn is cultivated successfully on the gentler mountain slopes up to alti tudes of 900 to 1,200 meters where there is enough moisture. Also, although the climate is rather too bot, much of it is raised in some of the low lands, especially the Artibonite Pla, in. The ears are small but good, the commonest corn being yellow. Most of the crop is ground into meal and used for food, although a little is fed to stock. The high prices offered in 1917, 1918, and 1919 caused the export of a considerable quantity, most of which went to Cuba,1 but its export is very unusual. Ears of corn in the 1 See article ''Cacao'' in International Bur. Am. Republics Bull., Sept., 1908, English section, pp. 471-482. 2 See report of Consul John B. Terres of Port-au-Prince, in Supplement to Commerce Reports, Bureau of Foreign and Domestic Commerce, Washington, November 9, 1918

PAGE 86

REI J.BLIC OF HAITl S UR\" E Y A. SMALL MILL FOR CRUSHING SUGAR CANE. B. OF DRYING CORX. PLA'rE VII

PAGE 87



PAGE 88

GEOGRAPHY. 75 husk tied together and h11ng up in great bunches on trees or poles for protection from animals form a conspicuous feature of the corn-raising areas. (See Pl. VII, B.) Petit-mil js one of the nonsaccharine sorghums, probably a variety of durra. It is raised extensively on the valleys and plains. The grain is used for food and the fodder is fed to stock. Considerable rice is raised locally, and more is imported to supply the great demand for the national dish rice and beans. Beans and peas of several varieties are raised for domestic cons11mption. The c ongo pea (pois congo) from Africa is one of the most esteemed varieties. A very important food plant is manioc, commonly called cassava, from the roots of which cassava bread is made. In some localities it is the staple food. Yams and sweet potatoes (patate) are among the most widely cultivated and used vegetables. Of the fruits, perhaps the commonest are the plantain (banan), banana (figue banan), orange (orange), alligator pear ( avocat), mango (mango), pineapple ( anana), and coconut (coco). The plantain, picked green and cooked, forms a staple like bread at many places. Many less common fruits are highly prized in season, such as the grapefruit ( shaddock), breadfruit ( arbre-a-pain), grape (raisin), mulberry ( mure), cherry (cerise), apricot ( abricot), corrosol, cayemjte, and sapotille. Fruits of temperate climates can be raised on some of the high mo11ntains. Blackberries ( mures sauvagoo) and straw berries ( fraises) grow at Furey. Nearly all these fruits and vegetables are raised in small gardens on little individual plots of land. They are characteristic of the mountains rather than of the plains, which are more often devoted to cotton, sugar cane, grains, and grass. Some localities are known throughout the Repub lic for the excellence of certain fruits or vegetables that they produce, and the local trade in these commodities is a rather large industry. No attempt has been made to standardize products, such as fruits, and raise a marketable article of uniformly superior quality. This neglect appears to be most 11nfort11nate, for the country can produce oranges, bananas, pineapples, and other frui'ts that could compete successfully in any market with fruit raised elsewhere, and the geographical location of Haiti is more favorable than that of several fruit-growing countries of tropical America. LrvE S.rocK AND PouLTRY. The live stock of the Republic includes mainly horses, mules, donkeys, cattle, goats, sheep, and hogs, all originally imported from Europe. They are raised entirely for local use, excep t that their hides and skins are exported. Stock raising is the principal industry in areas where the soil and climate are unsuited for intensive agriculture. The savannas of the North Plain, Central Plain, Artibonite Valley and Cayes Plain, and of certain mountain lands are devoted mainly to raising horses and cattle.

PAGE 89

76 GEOLOGY OF THE REPUBLIC OF HAITI ... The horses, which are small but hardy, are used for riding and to a small extent for driving. Mules are used for riding and for packing heavy loads, as they are stronger and tougher than the horses. Donkeys, however, are the real burden bearers; nearly all the produce that is carried over the mountain trails is borne b y them. Cattle are raised chiefly for meat and hides, ra. rely for milk. Oxen are used for drawing loads in the plains. Short-haired goats of many colors are perha ps the commonest domestic animals. They are valued chiefly for their meat and skins, although they also yield most of the milk produced in the country. They are especially adapted to browsing rugged mo11ntain sides that are 11nfit for cultivation and inaccessible to other pasture animals. Sheep are raised in only a few places. The wool is not of very good quality. Hogs are nearly as common as goats. They for age everywhere, living on seeds of trees and on fruits, vegetation, and waste. They are of a lean, rangy type, but their meat is highly esteemed. Chickens, turkeys, and guineas are the commonest fowls and are raised nearly everywhere, both for eggs and meat. Wild guineas a .nd wild pigeons are common at some places, particularly in the Central Plain. All the domestic animals and the poultry could be greatly improved by careful breeding and selection. METHODS OF F .ARMING AND FUTURE OF AGRICULTURE. As has already been intimated, the methods of farming are rather crude, nearly all the work being done by hand. Doubtless much of it must always be perfor1ned in this wa.y, for much of the land is rough and stony and is divided into small plots. On the plains, however, traction plowing and cultivation has been introduced successfully in recent years and can be greatly extended. Any great increase in the production of export crops is inextricably bound up with the problem of conserving and distributing water for irrigation. (See Part V.) Much can be done also, however, by improving and standardizing farm products and fruits and by perfecting methods of marketing. These are greatly to be desired, for agriculture is and undoubtedly must continue to be the country's mainstay and support. COMMERCE. GENERAL FEATURES. Foreign commerce has remained much the same under the Republic as it was in colonial days. The raw products of the plantation and forest constitute the great bulk of the manufactured goods, particularly textiles and prepared foodstuffs, comprise the greater part of the imports. The value of the exports has generally exceeded that of the imports. These features are brought out in the following table, which gives tl1e

PAGE 90

GEOGRAPHY. 77 official records of foreign trad e for several years at different periods d ur ing and since colonial days. In recent years, howeve r, the ratio of imports to exports has b e en unus ually hi g h. Value in United State s curr ency of imports, exports, and total for eign trade of the colony and of the R e public of Haiti at difj erent periods. Year Imports Exports. Total foreign trade. 1788. $20 ,590 5 72 1791. $40,060,'627 1889 $26,188,569 1890 10, 060,979 14,1 66 ,789 24,226,758 1891 8,8 2 3 ,776 14,340,234 23,164,01 0 1903. 8 966,388 8 92 6 ,939 12,893,327 1918. 9,876,555 17,285,485 28,499,075 1917 f 8 ,606 086 7,22 0 ,290 15,826,376 1918, 10,180,6 9 3 6,27 6 ,321 16,457,014 1919 f 17 117, 608 21,460,045 38,577,653 1920 f 27, 3 98,411 1 8 ,990,0 32 46,388,443 f 11, 9 57,205 4,953,570 16,910 775 1922 f 12,350, 271 10, 712,21011 23,062,487 The fiscal year u n der b oth the Repub l i c and the colony has extend e d fro m October 1 to September 30, and this i s the p erio d c ov e r e d. The figure s f o r 192 0 f o r i nstance c over the period October I, 1919, to S e ptember 30 1 9 20. From Edward s Bry a n Historical surve y of the Frenc h c o lony in the i sland of Santo Domingo, pp. 199, 206, L ondo n, 1797. Edward s c om piled his data from colonial re c ords during a protracted visit to Oap -Har tien. The livre tourn ois is v alued at $ 0.20 From Handbo ok o f Haiti, Bureau of the Ameri ca n Republi c s, Bull. 62, pp. 76, 77, Washington, 1892. Le Moniteur, No. 42, p. S27, 1905. From Internationa l E nc y c l opedia, Artic l e H aiti. f Ba s ed on annual r e p orts o f Admi n i stration of Custo ms, Haitian Cu stoms Receiver ship, Wash ington, D C 'See note at bottom of p. 79. E X PORTS. PRINCIPAL ARTICLES OF COLONIAL EXPORT At the end of the colo nial period the f our great staple exports of the colony, nam e d in the ord e r of th eir importanc e w e re su g ar, coffee, cotton, and indi go. Forest produ c ts and hides and skins occupied a leading place among the minor artic l e s of e xport. Tobacco was once a staple, and indig o was formerly the l e adin g produ c t, but these were forced to yield to sugar and coffee the product i on of whi c h had grown to overwh e lming propo r tions. In colonial d ays su gar manufacture had been developed to a high degree, and almo s t half the quantity exported was refined. The followin g tab le shows the e xport value of the four great staples during the y ear 1 791 and the relative rank of the ports of origin The figur e s for 1791 are sa i d to be 25 to 30 per cent less than those for the few y e ars imme diat e ly precedin g on acco11nt of disturbances growing out of the R e volution. Ordinarily c onsiderably more sugar was exported from Cap-Ha1tien than from Port-au-Prince, and the poor showing of Cap Haitien is due to the ear ly outbreak there of revolutionary movements

PAGE 91

78 GEOLOGY OF THE REPUBLIC OF HAITI. Quantity in French pounds a and value in United Stat e s currency of exports o/ jour leading products in 1791.b Sugar. Coffee. Port-au-Princ e ................... 61,441,142 Cap-Haitien . . . . . . . . . 45,482,041 Les Cayes . . . . . . . . . . 23,360,052 Fort-Liberte ....................... 10,249,158 St. -Marc . . . . . . . . . . 10 238,639 Leogane ............................ Port-de-Paix ........................ Petit-Goave ........................ J . e r e m1e ....................... Tiburon ........................... MOle St.-Nieolas .................. Jacmel ............................ St. -Louis du Sud ................ 9,181,520 1,29 8 ,300 1,074,103 496,249 841,650 128,180 116,176 11,600 To'ta.l ...................... 163,405,220 Approximate value in United States currenc y 0 $23,432,470 Cotton. I Cap-Haitie n ...................... 29,367,382 Port-au-Prince ...................... 14,584,023 St.-Marc . . . . . . . . . . . . . 5,521,237 J e r emie . . . . . . . . . . . . . . . 4,453,331. J a cme I . . . . . . . . . 4,357 ,'no Fort-Liberte . . . . . . . . . . . 2,321,610 Les Cayes . . . . . . . . . . . . . 1,843,403 Port-de-Paix . . . . . . . . . . . . . 1,829,75' L eogane . . . . . . . . . . . . . . . . 1,786,48' Petit-Goa ve . . . . . . . . . . . . . . 1,395,690 Tiburo n . . . . . . . . . . . . . . . 305,740 MOle St.-Nicolas . . . . . . . . . . 294,55 0 St.-Louis du Sud . . . . . . . . . 90,706 To"ta.l . . . . . . . . . . 63,151,180 Approximate -alue in United States currency o $10,378,150 Indigo. St. -Marc . . . . . . . . . . . . . . . 8,008,163 St. Marc ............................. 357,630 195,099 176,918 105,456 Port-au-Prince . . . . . . . . . 1,370,021 Cap-Haitie n ...................... Les C a y es . . . . . . . . . 720 770 Port-au-Princ e .................. Jacmel . . . . . . . . . . . . . . 613,019 Les Cayes ............................. J e r e mie . . . . . . . . . . . . . 189,194 Port-de-Paix ......................... L e ogane . . . . . . . 154,<>84. L eogane ...................... Otl1er ports . . . . . . . . . . . 230,875 Other pom ........................... 61,472 12,520 21,021 Total . . . . . . . . . . . . . . Total . . . . . . . . . . . . 930,016 Approximate value in Unite d States Approximate value in United States currency 0 $3,514,450 curre n c y 0 $2,175,024 a To obtain equivalent values in pounds avoirdupo i s add about 8 per c ent. 1> Data rec a s t from Edwards Bryan, His t orical survey of the Frenc h colony in the island of Santo Domingo, pp. 198, 199, London, 1797. c Computed on basis $0.20 for the value of the liv r e turnois. The a ctual value was a trifle less. TREND OF CHANGES IN EXPORTS. For some time during a .nd after the Revolution the exports decreased greatly, but they have gradually grown up again, though with some modi fications, to something like their form e r volume. Coffee has replaced sugar as the leading staple exported. Sugar almost disappeared from the list of exports until very recent years, when an attempt was made to revive the industry. Naturally indigo has completely disappeared as an export. Cotton, though it has had many vi c is situdes, has n e arly retained its former standing. Wh e n prices are high it appears in great quantities; when they are low, its output shrinks to small proportions. Cacao has retained an important place; s o also have dyewoods and lignum vitae, hides and skins, and generally honey. Other articles, like castor beans and corn, appear and disappear sporadically. In general coffee, cotton, forest

PAGE 92

GEOGRAPHY. 79 products, and hides and skins are stap les. Their volume an d relative value can be judge d from the fol lowing tab}es : Principal products exported from Haiti in 192 1 and 1922.a Material. (about 8 per cent i s triage) .......... (Jct'ton ............... (Je:Jit"t;on seed (Jc)ttonseed oil ............ .................. c::Jlic!&O I.ogwood ........... ................ Bides ........... .. .. .... Gm.t skins ....................................... S'ugar .............. Boney ..... . ... Llgrlum vitae ................................. beans ..................................... O>rn ..................................... Orange peel ...................................... l'ertiliz'er ..................................... Oct. 1, 1920, to Sept. SO, 1921. French pounds. b 45,689,687 3,987,430 8,014,774 611,012 2,355,319 74,893,411 40,729 207,340 10,51 1,004 890,68 4 6,537,006 146,4 6 1 226,060 1,907,888 O ct. 1, 1921, to Sept. 30, 1922. Frene h pounds. b 58,425,61 4 8,567,058 16,568,649 212 714 4,396,620 59,960,87 6 1,280,596 26,951 184,937 20,597,925 1,445,580 4,141,500 25,044: 372,632 568,24 9 From report of Administration of Haitian Customs, Haitian Customs Receiverships, fiscal period ending Sept. 30, 1922. To convert into Englis h pounds add 8 per cent. Value in United States currency of products exported from Haiti, 1917 to 1922.a Material. O c t.1,1916, to Sept. 30, 1917 O c t.1,1917, t o S ept. 30, 1918. Oct.1,1918, to Sept. 30, 1919. Oct.1,1919, to Sept. 30 1920. Oct.1,1920, to Sept. 30, 1921. Oct. 1 1921, to Sept. 30, 1922.l> .. Coffee . . . . . $4,620,554 $3,147,282 $16,407,234 $2,912,575 $7,487,312 Cotton (including cotton seed and cotto nseed o il ) Logwood (including extrac t and roots) ............. . . . . . . . ...... Sugar (raw) ............. Bides and skins ........ Boney ....... Castor beans ......... Lignum vitae ... Sisal ............... Shells (tortoise) ........ Fertil' izer ................ Beeswax .............. Orange peel ....... Mahogany .............. Co!-'n ....... Old copper .......... Iron ore .................. All other artic les ......... Total ................ Perce11tag e shippe d to U. S. 485,1 9 9 780,7 6 8 489,695 2 291,621 191,096 49,541 107,113 14, 322 1,562 955 23,4. 3 6 13,800 53,517 3,248 93, 861 54.39 850,274 319,364 4 3 0,348 69 162,144 174,089 3 3 6,707 36,386 15,552 657 6,308 10,942 11,274 681,390 2,33?. 92,202 $6,276,32 1 80.83 2 005,374 640,149 648,396 506,960 519,952 260,566 231,454 70,825 36 ,406 20,326 13,222 17,673 3, 414 4,436 12,417 4,627 56,714 $21,460,045 44.49 2,463,631 2,868,411 606,801 897,197 354,448 131,2 3 5 100,512 1 14,923 25,490 28,809 19,477 5,614 25,019 15,289 12,789 1,800 3 1 1,891 $18,990,0 3 2 62.15 475,425 744,311 94 ,487 457,166 76,738 28,883 S,724 62,013 116 7,621 7,949 2,814 7,411 7,120 859 66,858 $4,953,570 82 38 1,647,205 506,953 274,12 4 537,475 65,125 62,076 674 26,25 7 535 13,310 97 7 20,130 586 5,660 2,20'2 161,809 $10,712,210 13.44 Annual r eports of Administration of Customs Haitian Customs Receivership, Washington, D. C b Value of exports for 1922 include the export tax. Figures for previous years do not include the export tax. In 1922 the export tax was $2,015,598, ot which $1,752,768 was on coffee

PAGE 93

80 GEOLOGY OF THE REPUBLIC OF HAITI. IMPORTS The imports at the present time probably do not differ greatly in gen eral features from those of earlier or even colonial days. By far the larg single items are cotton cloth, flour, and soap. Recent years have seen a growing demand for machinery. The values of the materials impo rted during the last six fiscal years are shown in the following table : Value in United States currency of products imported into Haiti, 1917 to 192!. Material. Cloth and fiber products (about 90 per cent cotton Oct.1,1916, to Sept. 30, 1917. Oct.1,1917, to Sept. 30, 1918. Oct.1,1918, to Sept. 30, 1919. Oct.1,1919, to Sept. 30, 1920. Oct.1,1920, to Sept. 30, 1921. Oct. 1, 1921., to Sept. 3 0, 1922 cloth) . . $2,876,617 $3,627,186 $5,252,192 $9,852,533 $2,272,493 $3,670,189 Foodstuffs (about 40 per cent flour) .......... Soap .................... Iron and steel manufactures Tobacco ................. Petroleum products ..... Liquor, beer, and other beverages .............. Machinery (agricultural, automobiles, vehicles) .. Lumber and manufactures (wood products) .... Leather goods ....... Chemicals and drugs ..... Cement ................. All products ........ 2,540,958 494,263 121,165 229,189 175,410 167,614 65,216 151,841 108,924 88,960 46,765 1,539,164 1,694,332 926,836 844,812 255,825 282,960 142,384 99,759 205,960 159,267 53,562 71,267 1,816,544 5,964,608 839,780 731,228 381,884 397,915 129,213 197,345 186,062 94,976 107,063 82,546 2,752,798 8,607,240 972,924 957,505 626,251 454,576 505,107 277,083 341,657 290,176 236,766 149,833 4,126,760 ---3,654 480 615,530 805,365 698,725 327,477 285,402 90,837 169 ,231 126,772 157,969 103,345 2,649,579 3,849,59 1 639,01 4 427,895 454,71 5 315,675 251,0 16 106,158 181,873 120,310 155,022 56,701 2,122,112 Total . . . . . . $8,606,086 $10,180,693 $17,117,608 $27,398,411 $11,957 ,205 $12,350,27 1 Percentage of imports from u. s. .................. 86.39 9'2. 48 93.12 83.12 79.82 83.87 0 Annual reports of Administration of Customs, Haitian Customs Receivership, Washington, D. C. PRINCIPAL COUNTRIES TRADING WITH THE REPUBLIC OF HAITI France has always consumed the greater pa .rt of the output of Haitian coffee. This feature and the inherited kinship of language and politica l thought have caused the Republic of Haiti to depend upon France for certain special classes of products, such as books and the more fashionable c l othing materials and works of art. There is a small amount of trade with the surrounding West Indian Islands, dependent on transient loca l conditions. For instance, in 1917 and 1918 a la rge quantity of corn ( ma!ze), which is seldom exported from the Republic, was shipped to the neighboring Republic of Cuba. During the last century the commerce of the Republic with the United States grew gradually to large proportions, and during the World War, i n 1917 and 1918, it was almost exclusively with the United States. Since then, however, there has been a gradual readjustment. The preceding

PAGE 94

GEOGRAPHY. 81 tables show that about 84 per cent of the imports in 1922 came from the United States and that about 13 per cent of the exports were sent to the United States. MANUFACTURES. The manufacturing industries are small and not complexly organized Basketry, rope, and simple household articles, such as chairs and straw mats, are made by hand in large quantities for domestic use. Small woodtuming shops that produce chiefly curios for tourist trade are common Small cabinet shops produce cabinet work of real excellence, especially in mahogany. A plant for the extraction of dye liquor from logwood was snccessfully operated at Grande-Riviere du Nord for a few years It burned down but was rebuilt and is now in operation. One company manufactur es cigarettes at Port-au-Pri nce. The capital also has the only considerable ice factory in the Republic, but ice is made at small plants in other cities Plants for cleaning and preparing coffee for market, known as'' usines a cafe,'' are common, and some of them are rather large. Con siderable cotton is ginned in plants in several of the larger cities, and some c ottonseed oil is made. The largest of these plants is in St.-Marc. Crude plants for making unrefined brown sugar (see Pl. VII, A) are com mon but are of small capacity. The'' Rasco'' sugar mill, owned and oper ated by the Haytian-American Sugar Co. at Port-au-Prince, is a large and modem plant, which is said to be capable of grinding 250,000 tons of cane annually, but it has not yet been operated to its full capacity. Several small brick plants supply virtually all the domestic demand for bricks. (See pp. 503-507.) 1fost of the domestic supply of salt comes from several evaporating works. (See pp. 509-510.) TRANSPORTATION. As most of the area of the R e public is exceedingly mo11ntainous inland transportation is very difficult. All the larger cities are on the coast, where the bulk of the freight can be handled by water. Throughout the periods of both the colony and the Republic the coastwise traffic and pas seng er service in small sailing vessels has been great. These small vesseI s bring in large quantities of goods to the larger cities for sale or export. HIGHWAYS AND TRAILS. In colonial days oxcarts and carriages were used on all the la.rger plains, where roads were easily made, but intercommunication was very diffi cult. After much trouble and labor a vehicle road from Cap-Ha1tien to Port-au-Prince was opened in 1787, and vehicular traffic between Portau-Prince and Les Cayes was virtually unknown.1 All parts of the colony were reached by horseback, however. As early as 1700, according to Labat ,2 it was possible to travel from Cap-Frangais t.o Leogane by way of the Central Plain. 1 Moreau de op. cit., vol. 1, pp. 104, 662, et seq. 2 Labat. op. cit., vol. 5, pp. 134-135. 6

PAGE 95

82 GEOLOGY OF THE REPUBLIC OF HAITI. Since 1915 the roads have been improved as far as funds would perm it, and at present (1923) there are 960 kilometers of highway under con tinuous maintenance and passable for vehicles, including au ton \obiles at nearly all seasons. During the height of the rainy season there are two regions, one in the Plain of St.-Michel and the ot her between Miragoa ne and Aquin, where traffic is difficult because the soil is alluvial The improvements most needed on the highways are bridges. There a re not more than six steel highway bridges i'n the co11ntry, and many of t he small rivers become dangerous or impassable after rains. Ferries are used to cross some of the largest streams, but fords are generally used and traffi c must wait until floods subside. Work has been started (1923) on bridges for the larger streams and construction has been begun on a road fro m Las C1ahobas to Hinche and a road from Las Cahobas to Belladere, on the Dominican border. This road to the border will make it possible to trave l from Port-au-Prince to Santo Domingo City by automobile in twelve hours, whereas it now takes three days by W ay of Cap-Ha1tien, Ouana minthe, and Santiago. Inn11merable horse trails and footpaths lead from the mai n roads to almost every nook and cranny of the mountains Over these rough an d difficult trails and paths an enormous amount of coffee, vegetables, and foodstuffs are carried to the markets on the backs of burros or on the heads of market women. Although more and better trunkline highways are needed on the plains to reach areas that are not now accessible to wheeled traffic, large quantities of the country's products will necessarily continue to be tra. nsported by primitive methods over unimproved trails. The following table shows the distances between the principal towns along the main highways. Distances between towns in Haiti, in kilometers. Port-au-Prince St.-Marc . . . . . . . . . . . . . . . . . . 98 St.-Marc Gonaives . . . . . . . . . . . . . . . . . . . . . 74 Gonalves Ennezy . . . . . . . . . . . . . . . . . . . . . . 29 Ennery Plaisance . . . . . . . . . . . . . . . . . . . . . 27 Plaisance Limbe ........................................... 21 L b C H .. t. 26 1m e ap-. a1 I en ........................................ Port-au-Prince Cap-Haitien ................................. Cap-Ha.ltien Ouanaminthe .................................. Port-au-Prince Leogane .................................... Leogane Petit-Goave ....................................... Petit-Goave Miragoane ..................................... M A A rragoane qu1n .......................................... Aquin Aux Cayes .......................................... 275 71 32 39 22 47 61 Port-au-Prince Aux Cayes . . . . . . . . . . . . . . . . . 201 Port-au-Prince Pont-Beudet . . . . . . . . . . . . . . . . 18 Pont-Beudet--Mirebalais . . . . . . . . . . . . . . . . . . 35 Mirebalais Las Cahobas . . . . . . . . . . . . . . . . . . 29 Port-au-Prince Las Cahobas . . . . . . . . . . . . . . . . 82

PAGE 96

GEOGRAPHY. 83 RAILROADS. The Republic has 259.5 kilometers of narrow-gage railroads, which are operated in several disconnected l1ni ts by two companies. The following table sho"rs the distance between the ter1ni nal stations of the di:fieren t sectors: Distances in kilometers between terminal stations of railroads. Compagnie N ationale des Chemins de Fer d'Ha1ti: Port-au-Prince St.-Marc . . . . . . . . . . . . . . . . 105 Gonaives Ennery . . . . . . . . . . . . . . . . . . . 33 Cap-Haiti en Bah on . . . . . . . . . . . . . . . . . . 39 177 Compagnie des Chemins de Fer de Ia Plaine du Cul-de-Sac: Port-au-Prince Manevill e . . . . . . . . . . . . . . . 42.7 Port-au-Prince Leogane . . . . . . . . . . . . . . . . 34.5 Street railways in Port-au-Prince . . . . . . . . . . . . 5.3 82.5 The Compagnie N atronale des Chemins de Fer d'Ha1ti hopes to com bine its 11nits into one line connecting Port-au-Prince and Cap-Ha1tien by way o f Mirebalais, Las Cahobas, Hinche, and Bahon. The railroads of the Republic are greatly hampered by the lack of large business or agri c ultural enterprises to furnish patronage and by the lack of any large vol11me of passenger traffic. Labor is so cheap that transportation by animal s competes with transportation by rail. TELEGRAPH AND TELEPHONE SYSTEMS. The telegraph and telephone systems of the Republic are operated by t he Government. There are 39 telegraph offices which handle foreign as well as local messages. The city of Port-au-Prince has a modern auto matic telephone exchange with a capacity for 300 subscribers, which is now being increased to 500 Small exchanges are maintained in several other cities. The volume of the telegraph and telephone business in the Republ ic has doubled during the last two years.

PAGE 97

PART I I. GEOLOGY By WENDELL P. WOODRING, JOHNS. BROWN, and WILBUR S. BURBANK. RECONNAISSANCE GEOLOGIC MAP. The reconnaissance geologic map (Pl. I, in pocket) is accurate to the published scale only along the routes traversed during the reconnaissanc e shown on Pla. te III (p. 24). In compiling it the maps published by Mr L. Gentil Tippenhauer (see bibliography, p. 603) have been free l y used. Regions distant from the routes traversed were mapped principal ly by long-range observations supplemented by Tippenhauer's maps. The base map was compiled by the Service des Leves Topographique s under the supervision of Mr. Glenn S. Smith, chief of the Division of West Indian Surveys of the Topographic Branch of the United State s Geologic Survey. The delineation of almost the entire shore line, of the features near the shore line, and of the course of Riviere Artibonite fro m its mouth to Mirebalais is bas e d on aerial photographs controlled by tri angulation, work done under the direction of Mr. E. L. McNair, of the United States Geological Survey. The location of towns and other fea tures in the interior is based on all available published and 11npublishe d information supplemented by personal observations. The topograph ic surveys were suspended in the winter of 1921 22 from lack of funds SEDIMENTARY ROCKS. By WENDELL P. Woona1No and JoHN S. BaowN. PALEOZOIC (?) METAMORPffiC ROCKS. Sedimenta ry rocks are exposed at the surface over fully four-fifths of the Republic. Their age ranges from probably Paleozoic to Rec ent. Metamorphic rocks, probably of Paleozoic age, occur as float on the North Plain and on the Leogane Plain and are fo11nd in place on Tortue Island. Quartz schist and mica schist were found as float on the North Plain east of Limonade. Although no bedrock exposures of the schist were seen the fragments fo11nd presumably represent a metamorphic basement on whiclt the old basaltic rocks lie These schists are much more metamorphosed than any of the old volcanic rocks or Cretaceous sediments and are therefore considered Paleozoic or early :1Iesozoic. A sp e cimen of garnetiferous quartz-mica schist that was found as floa,t on the plain a little east of Le Trou shows the highest degree of metamo r phism seen in any rock collected in the Republic This rock is greenish gray to dark gray, fine grained, and distinctly schistose. It is spotted 84

PAGE 98

SEDIMENTARY ROCKS 85 with pinkish to reddish-brown patches of garnet. In thin section the rock is seen to be composed essentially of quartz, chlorite, biotite, magnetite, and garnet. The quartz comprises about 70 to 75 per cent of the volume and has a characteristic mortar structure, produced by granulation of the borders of the grains. The chlorite and brown biotite comprise 15 to 2 0 per cent of the rock and together with the finely crushed quartz occupy the spaces between the larger grains of quartz Considerable magnetite and some ro11nded patches of garnet are scattered through the rock The garnet has a slight reddish tinge and apparently has replaced the chlorite, mica, and quartz between the larger grains of quartz. The rock may have been formed by the metamorphism of an jmpure sandstone It shows evidence of intense dynamic and thermal metamorphism of a character not observed as the result of deformation caused by intrusions of igneous rock in Cretaceous and later times Exposures of these older rocks were fo11nd on Tortue Island beneath limestone that is probably of upper Oligocene age There the metamorphic rocks consist principally of much sheared limestones that have been com pletely recrystallized and in some places partly replaced by chlorite, epi dote, and quartz. The limestone is exposed in a sea cliff a short distance east of the landing at La Vallee, on the south coast. It is a hard bluish gray rock and contains a network of seams of calcite. About 150 meters east of the landing greenish limestone containing chlorite and stringers of quartz crops out in a sea cliff. Bluish schistose limestone under lies this rock. The beds are contorted and wrinkled The schistosity planes in general strike N 80 W. and dip 20 NE. Similar rocks crop out in the huge amphitheater inland from La Vallee, where the cover of upper Oligo cene limestone has been stripped by erosion, but along trails their exposures are weathered. Float of a granitic rock, which presumably intrudes the schistose limestone, was seen on the trail to Source Lavier. Schi stose rocks were also seen northwest of Pointe des Oisea11x, where the trail from La Vallee descends to the coast. The schistose limestone re sembles similar rock on the south slope of Samana Peninsula, Domini can Repu blic .1 No exposures of schists consjdered to be older than Cretaceous were found in the southern part of the Republic, but float consisting of mica and quartz schists like those from the northern part of the Republic was seen on the western part of the Leogane Pla.in. This material probably comes from an unexplored part of the mountains south or southwest of t he plain. CRETACEOUS SYSTEM. Rocks that can be positively identified from fossil evidence as Creta ceous are confined to the Upper Cretaceous series. They consist princi pally of limestone and were found in only a few small patches in the 1 A geological reconnaissance of the Dominican Republic: Dominican Rep. Geol. Sur vey Mem., vol. 1, pp. 53, 83, 182-183, 1921

PAGE 99

I 86 GEOLOGY OF THE REPUBLIC OF HAITI. arrondissements of Cap-Ha1tien and Grande-Riviere du Nord Detai l e d examination of the Republic doubtless would disclose other similar patches, but the area covered by such rocks is probably very small Rocks that are tentatively referred to the Lower Cretaceous system on stratigraphic, structural, or lithologic grounds 11nderlie rather large are as, chiefly in the Massif du Nord. Lithologically most of these rocks in the northern part of the Republic are grouped under the term argilli te, although at places they show wide variation from this type Rocks in the southern part of the Republic that are considered of the same age o n indefinite grounds generally consist of metamorphosed limestone. It is not at all certain that all these rocks called Lower Cretaceous are of the same age. All of them are older than Tertiary, but some may be even olde r than Cretaceous The Upper Cretaceous rocks probably rest lm comformably on the rock s of supposed Lower Cretaceous age, but the actual contact was not ex amined. They contain small pebbles of the older igneous rocks. The Lower Cretaceous rocks in the northern part of the Republic appear t o have been deposited on a basement composed mainly of volcanic rocks of which they contain recognizable pebbles at some places. The relations of the argillite and volcanic rocks are considered elsewhere. (Seep. 273.) The fossils show that the Upper Cretaceous rocks are marine. (See list, p 98.) The lithology and the absence of marine fossils in much of the Lower Cretaceous series of the northern part of the Republic indicate that they are marginal deposits, laid down in part on flood plains. These rocks are nearly everywhere thinly and evenly bedded, and in many places they show mud cracks and contain fragments of lignite. Some impure limestones, calcareous sandstones, and argillites in this series contain small Foraminifera, indicating that they were deposit e d in shal low marine waters. Most if not all of the ljmestone of supposed Lower Cretaceous age in the southern part of the Republic is marine. Both series of Cretaceous rocks were once much more exten sive, but erosion prior to upper Eocene time removed the greater part of them. The solubility of the limeston e s facilitated their rapid removal The argillite, even though now in some places much indurated, is not ordinarily very hard and generally remains only where it has been pre served in the troughs of faulted or folded areas. Detailed descriptions of the Cretaceous rocks as observed at different localities are given below. LOWER CRETACEOUS SERIES. DESCRIPTION BY REGIONS. MASSIF DU NORD. Plaisance Valley. That part of the valley of Les Trois Rivieres above Pilate is known as Plaisance Valley. All this rolling valley between t he bordering mountains seems to be underlain by argillite. On the north bank

PAGE 100

SEDIMENTARY ROCKS 87 of Les Trois Ri' ieres at the bridge near Plaisance an excellent exposure shows a thickness of about 40 meters of beds that strike north and dip 400 E. The beds are 2 to 5 centimeters in thickness and very regular. They consist of indurated dark-brown fine-grained sandstone or sandy slaty shal e Many of the weathered blocks contain peculiar ridges that suggest fossils, but that are apparently fillings of mud cracks About 300 meters southwest of the bridge on the road to Ennery brown to gray b eds resembling those just described but more clayey and con taining no visible sand g:ains are exposed in a roadside cut. An analysis of a sampl e of material from this exposure is given on page 502, under the discussion of possible raw mat.erials for cement Silica in the moisture free sa mple forms about 56 per cent and calci11m and magnesi11m car bonates only about 17 per cent of the total. A num ber of other exposures of this series are found along the road farther southeast. The rocks differ chiefly in color, which varies from ashy gray to deep purple. At all the exposures the beds are steeply tilted S.O NE Argilit,e creta.cee Jv k< et cal.ca.ire meta:rno iqa.e .. e re e s volcEU?iques ( ?) FIGURE 4. Generalized section showing fault zone at contact of argillite and volcanic rocks as exposed in roadside ditch at Plaisance. Rather different rocks, probably of the same series, are exposed at the foot of Mont Puilboreau in a roadside cut about 100 meters long The thi ckness of beds exposed is apparently 60 to 75 meters. The beds dip ste eply to the southwest, into the mountain. They consist in part of dark gray metamorphosed cherty limestone, which is partly crystalline Interbedded with the limestone is at least an equal amo11nt of shale, indurated and at places slaty and showing ridges due to hardened fillings of mud cracks The contact of these rocks with the igneous rocks to the south is probably marked by a fault. At the north side of Plaisance Valley, just at the north edge of the town of Plaisance, the contact of the rocks of the valley with the volcanic rocks of Plaisance Mountain is well exposed in a ditch beside the road. The ro cks exposed are purple argillite, metamorphic cherty limestone (which is partly changed to marble and which ranges in color from brown to white), and reddish volcanic rock. The three rocks are intermixed along a zone of step faults as shown in Figure 4

PAGE 101

88 GEOLOGY OF THE REPUBLIC OF HAITI. Similar argilli te and oth e r elastic rocks are exposed in the Plais ance Valley between Plaisanc e and Pilate. They consist principally of dark chocolate-brown thinb e dded argillit e and thick e r beds of harder sa nd stone. The argillite bre aks into small blocks that have conchoidal sur fac e s. The general strik e of these bed s is northwest and they dip to the northeast and southwe s t at various angles. Les Trois Rivie r e s b e t ween Gros-Morn e and Pilate. Farther w est the same series of ro c ks is expos e d on Les Trois Rivieres along the trail fro m Gros-1\forn e to Pilate. They crop out on both banks of the river abov e the s e cond cros sing toward Pilate. Their outcrop is a narrow northwes t wa. rd-trending band betwe e n volcanic ro c ks on the west and middle Eocen e limest one (Plaisanc e limeston e ) on the east. At this place the ro c ks con sist of dark thin-bedd e d c al care ous argillite. About 200 meters above the second cros si ng, on the right bank, a bed of limestone changed to whit e marble containing greenish streaks is interbedded with the argillite. At most of the exposur e s the b e ds stril { e northwestwa.rd and dip steep ly southwestward An expo s ure on tlie l eft bank about 1 kilometer abov e the second cro s sing s how s that the b e ds are crumpled. (See Pl. VIII, A .) The contacts with both the volcanic ro c ks and the limestone are probably along faults. N e ar Dondon. Argill i te and other rocks of the same series probabl y unde rlie n e arly all the mountain-wall e d vall e y at Dondon, but ove r mos t of its are a they are cover e d by alluvium. Good exposures are found in the trail b e tween Dond on and Grande-R i viere du Nord at the north end of the valley, about 3 kilomet e rs north of Dondon. The rock consists partly of brown or purple cal c areous argillite and partly of similar but wl1iter rock, in fairly thin b e ds. Inte rbedded in the fine-gra ined argillite are one or two conglom eratic b e ds, about h alf a m e t e r thick, which contain cobbles of igneous roc k, some as muc h as 25 centimeters in diameter. The matrix of these beds con s ists in part of angular fragments of dacitic lava, quartz, and p1agio c lase in a calcareous c ement containing a few small undetermined Foraminife ra. Smaller l e ns e s of sandy material not more than a f e w c entime t ers in thic kness contain angular fragments of plagio clase, quartz, a n d vol c ani c ro c ks in an impure calcareous matrix containing numerous s mall undetermine d Foramin i fera. B e tween tl1is place and Carre four Menard, the nearest station on the railway betwe e n Cap-Ha!tien and Grande-Riviere du Nord, there are numerous e xposures of the argillite along the trail that follows a deep ravine l e ading down to the Grande Riviere du Nord. Most of the rock is thin-b e dd e d fine-grained brown sandstone, which contains ab11ndant fragments of lignitized wood. Some of the rock grades into impure lime stone, which is brown and partly recry s talliz e d. At some places it is a coarse conglomerate, compos e d of fragments of the underlying igneous rocks that are expos e d at s ome places in the ravine. The rock everywhere is complexly folded and has been considerably metamorphosed.

PAGE 102

\ Rl!lJ,URLIC OF HAITI "SUR\'EY PLATE VIII (?) CALCAREOUS ARGILLITE LES TROIS Rl,"Il!iRES AND llJLAr.rE. R PJr,r,ow T JA\T A AND OF RUPPOSED UPPER C'RETAC'EOrs .\Gl
PAGE 104

ROCKS. 89 N ortbwest of this locality, on the trail leading from Milot tkl Chris tophe's Citadelle, on the northeru;t slope of the ridge called Bonnet-a l'Eveque, exposur es of purple argillite are common between altitudes of 300 and 650 meters above sea le,Tel. The rock is intricately folded and at places has a slaty cleavage. Near Cerca-larSource. The largest area of Lower Cretaceous argillite is in the ea.stern part of the Massif du Nord, embracing the southern part of the arrondissement of Valliere This area is probably continuous under the cover of Tertiary limestones with the area near Dondon and with that in Plaisance Valley. Near the Dominican border the area covered by these rocks is about 11 kilometers wide, but the width is less towa ,rd the northwest. I ts length is about GO kilometers These rocks are well exposed in bluffs on the right bank of R1 viere l'Ocean 100 meters below the crossing of the trail from Thomassique to Cerca -la-Source. At thjs place they consist of thin-bedde d variegated argillite and thicker beds of greenish sandy material. The colors of the argillite are somber shades of chocolate-brown, olive-green, purple and brick-r ed The argillite weathers into small blocks that have conchoidal surfaces In the bluffs the beds have slumped, but ledges in the stream s how that they strike northwestward and dip 40-80 NE. or SW. or even stand vertically. The greenish sandy beds consist of angular to subangular grains and crystals of plagioclase, grains of quartz, and flakes of brown mica. Grains of magnetite, a few prisms of apatite, and a very few grains of epidote and zir con are present. The brown mica is partly altered to greenish mica ceous and chloritic material, which has partly replaced the plagioclase and quartz, giving the sandstone its greenish color and acting as a cement. The plagio clase is also pa1tly replaced by calcite, sericite, or brownish products of the alteration of the mica. There is abundant calcite in the ro ck, but owing to al tera ti on and recrystalli zatio n it is impossible to determine whetl1er this mineral was an original constituent oi the sandstone The brown shaly rock carries small fragments of plagioclase and quartz in a matrix of finer material containing an iron-stained calcareous and chloritic cement. The shale contains thin seams of coarser gritty material. T he argillite lies at the surface northeastward from Cerca-la-Source to the foot of the north slope of the ridge south of Lamielle, where the llnder lying pyroxene andesite crops out. The soil formed by the argillite is very thin, and the rock is well exposed at many places along the trail. On weathered surfaces it is c1ark brown or dark gray, and contains angular to subangular fragments of quartz and a few fragments of altered pla gioclase, as well as flakes of white mica and grains of magnetite. The cement is largely calcareous, but there is considerable secondary chloritic material. Beds of gray calcareous sandstone, which weather brownish, contain angular to subangular fragments of quartz, a few fragments of

PAGE 105

90 GEOLOGY OF THE REPUBLIC OF HAITI. plagioclase largely altered to sericite or kaolinite, and a few grains o f m agnetite. The grains are embedded in a calcareous matrix, which con .. stitutes a large part of the rock. Some of the rock has an incipient cleavage almost at right angles to the bedding. The beds undoubtedly are intricately crumpled, BB in the sha l low exposures along the trail they dip northeastward and soutl1westwar d at steep angles, the direction of dip changing within short distances. On the south slope of the first ridge northeast of Cerca-la-Source a networ k of branching quartz veins about 10 meters wide cuts the argillite. The white quartz debris from the veins makes a conspicuous band that i s visible on the next two spurs to the northwest. In this dista nce the vein s are offset twice by faults. Morne du Cap. In the Morne du Cap near Cap-Ha1tien there are n11merous small patches of a formation whi ch, although rather differen t from anything else in the Republic, apparently should oo classed with the Lower Cretaceous rocks. This formation consists mainly of very dark chert and of brown or yellow indurated mudstone and siltstone, consider ably more metamorphosed than the Eocene rocks. Moreover, it is over lain unconformably by upper Eocene limestone. It probably rests uncon formably on a basement of igneous rocks that are mainly of volcanic origin. This formation in its relation to the water supply of Cap-Ha1tien is described on page 581. The formation is typically develop e d at the shore and in bluffy slopes just above the shore near Carenage, the most northerly section of the city, and on Morne Calvaire, perhaps 200 meters to the southwest. I t consists partly of brown or yellow siltstone, claystone, or fine sandstone, partly of dense black or blue chert, all in generally rather thin beds that average perhaps 8 to 15 centimeters in thickn03s. The chert at some places forms solid beds or constitutes a whole series of beds several meters in thickness, but at other places it forms nodules or concretionary masses. The claystones at places appear to grade into very impure limestone. The whole formation is much shattered and breaks into minute polygonal blocks, which at many places form a talus over weather e d slopes. Such material is successfully used as top dressing for roads in the city. (See p. 497.) Thin sections show that some of the chert probably is in part of radiolarian origin, as Radiolaria can be re cogn ized. It also conta ins some Fora minifera, but the rock is so greatly metamorphosed that the proportion of the two kinds of organisms is not e asily detennjnable. In many fea tures the rock resembles some of the Franciscan chert of California. Beds of coarser, distinctly d etrital material and even of coarse con glomerate appear at places, the conglomerate probably being near the base of the formation. At Carenage the bluffs of blue chert above the shore are underlain by sandstone and conglomerate, which crop out on

PAGE 106

SEDIMENTARY ROCKS 91 \ the beach. This conglomeratic material contains much-weathered frag ments of dark igneous rocks and also rounded masses of chert, apparently concretionary. Farther northeast the conglomerate appears to be under lain by much-weath ered greenish volcanic rock. The following section was measured in a ravine about 200 meters north of the end of the street in Carenage. The beds strike N. 10 E and dip 30 NW. Section of Lower Cretaceous beds near C ap-H a?tien Meters. Thin beds of yellowish sandstone and interbedde d chert. . . . . 41 Same as above, but thicker beds, cherty nodules. . . . . . . . . 3 6 T hick beds of blue chert. . . . . . . . . . . . . . . . . . . . 8 Soft gray sandstone contain ing a bed of blue chert about 25 centimeters thick . . . . . . . . . . . . . . . . . . . . . 5 Greeni sh weathered igneous rock. . . . . . . . . . . . . . . . 2 Beach. Thickness of Cretaceous beds exposed. . . . . . . . . . 92 On either side of the east gate to Morne Calvaire there is a bluff, 5 to 10 meters in height and nearly 100 meters in total length, in which the formation is well exposed. The rocks are chiefly cherty beds and brown silt stone. The siltstone has blue-black or brown iron stains on fractured surfaces. In this exposure there is one peculiar bed about 2.5 meters thick, apparently a fine conglomerate composed mainly of fragments of greenish igneous rock but partly of small pebbles of chert resembling the llnder lying cherts. However, the conglomerate appears to be perfectly interbedded in the series, and whether it denotes an 11nconformity is doubtfu l. Other exposures of a similar bed of conglomerate were noted at several places in the Ra vine de la Belle Hotesse. Many other small exposures of the formation were noted, particularly around the shore of the cape north and west of O ap-Ha1tien. Those near the town are shown in some detail in Figure 3"/ (p. 5"/9) in connection with the description of the water supply. The structure of the formation is complex. Everywhere it is sharply folded, and in good exposures many faults are visible. In the bluff east of Morne Calvaire the beds dip intA> the hill, but at the north end of the bluff there is part of a small anticlinal fold, much resembling a fan, in which minor are seen. ,Jus t south of the steps leading up to the gate there is a normal fault, which shows an offset of about a meter in the thick bed of conglomerate described above. A nearly vertical normal fault is exposed near the gate south of the city on the road to Port-au Prin ce, where road metal has been dug off the hillside. The contact of tl1e c herty formation with the porphyritic igneous rock is marked by a fault, which has an apparent displacement of at least 20 or 30 meters

PAGE 107

92 GEOLOGY OF THE REPUBLIC OF HAITI. MoNTAGNES N ornES. About 4 kilometers southwest of St. -l\1ichel de l' Atalaye, on the trail through Section Paul to Dessalines, stretches a broad interior v alley cal l ed Savane la Cidra, which lies back of the first range of the Mon .. tagnes Noires at the border of the Central Plain. This valley con tains r1umerous exposures of a thin-bedded slaty argillite. The rock is g r a y or brown tinged at places with tints of blue or green Part of the series c on sists of a black slaty limestone, which contains small undetermined F ora minifera. This rocl{ weathers yellowish white. The beds are conside rably folded In contact with tl1is argillite appears a gray or green altered hornbl e nde andesite Both the argillite and the andesite have been metamorphos ed by an intrusion of dacite porphyry The argillite is partly replace d by epidote, chlorite, and quartz and contains cubes of pyrite. SOUTHERN PENINSULA. Rocks of doubtful age that are known to be older than the oldest Tertiary rocks were examined at several localities in the Southern Peninsula. They are tentatively classed with the rocks of supposed Lower Cretace ous age in t h e northern part of the Republic A rrondissement of J acmel. In the area of basalt north of J acmel the re are several exposures of schistose limestones apparently have bee n engulfed by the basaltic eruptions. An exposure of this limestone w as fo11nd north of Coteaux on the trail from J acme to Leogane, about 500 meters north of the southern boundary of the basalt. The limestone is b r own and dense and is much sheared but contains eyes of uncrushe d l imestone I n tllln section the limestone is seen to be only partly recrystall ized and contains small undetermined Foraminifera. The schistosity strikes N. 60 W and dips 80 S Another thin zone of metamorphic limestone was seen just south of Corai lBrache on the same trail. North of the divide along Ravine la Rououne, the western branch of the Riviere des Citronniers, there are exposures of light-brownish m u c h sheared metamorphic limestone The beds strike N 8 0 W and dip 50 S. These metamorphic limestones appear to be overlain by the basa lts, a l though the contacts generally were concealed Near Petit-Goave About 4 kil ometers southeast of Petit-Goave, o n a trail l eading up the mo 11n tain slope, there are large exposures of a gray to grayish-brown impure metamorphic limestone, but the extent of the beds was not deter1nined The rock is in regular and rather thin beds and some beds have a conchoidal fracture. The grayish-brown limeston e contains Globigerina, Textularia, and other small Foraminifera. I t prob ab l y is o verlain by upper Eocene limestone, but the contact was not see n. Arrondissement of Aquin. About 2 kilometers west of the little v il l age and chape l of Cha.ngier, at the highest point on the trail betwee n

PAGE 108

SED IMENTARY ROCKS 93 I' Asile and Ca vail l on, some small exposures of a reddish schistose l i m e stone were n oted The area covered is small, and the relations of tl1i s rock to the s u r r ounding rocks are not known, but it is 11ndoubtedly olde r than the prevailing upper Eocene limestone and is tentatively referred to the Cre t aceous Such highly metamorphosed sediments are not com mon in the Cretaceous but probably form local masses in less metamo1 phosed r ock, as at Camp Perrin. Arrondi ssement of Cayes. Ro c ks which on rather indefinite grounds are tentatively assign e d to the Lower Cre taceous were found at a n 1 1mber of oth e r p laces in the Southern Peninsula. The principal area is north of the lowlan d near Camp Perrin, which is underlain by the lignite bearing Miocene b eds described on pages 236, 483 Here the rock is a thin-bedded, very ch erty, generally chocolate-brown limestone, but some beds are whiter. Some bed s are recrystallized and contain many closely spaced fractures, which are filled with coar s ely crystalline calcite. About 2 kilo met e rs n orth of the diversion dam of Canal d' A.vezac the limestone appears to contain l arge zones of schist, which is probably derived from impure limeston e by metamorphism The beds are crumpled into close folds, tl1e strike changing greatly withi n a f e w meters, but the prevailing st;rike is appro ximately east. No fossils were found in this limestone, but the de gree o f metamorphism indicat e s that it is at l east as old as Cretaceo u s It was seen in place only along La Ravine du Sud, beginning at a locality abou t a kilometer north of the diversion dam Float of a similar rock is v e r y abundant in the lowland, indicating that it is extensive in the mo11ntains to the north. Float or similar brown metamorphosed lime ston e which was noted at Port-a-Piment and elsewhere along the south coast possib l y indicates an extensive area of such rocks in the unexp l ored int erior of the llf ontagnes de la Hotte. Arr ondissement of Tibu ron. Somewhat similar brown limestone was foun d along the trail near the north coast of Tiburon Bay Basalt seems to o verlie the limestone, but this relation was not definitely estab l ished. If the l imestone really underlies the basalt the l i mestone probably is Cre taceous; if not, it probably is a pl1ase of the upper Eocene limestone, whi c h appears to cap the high ranges north of Tiburon. UPPER CRETACEOUS SERIES. DESCRIPTION BY REGIONS. MASSIF nu Noan. N e
PAGE 109

94 GEOLOGY OF THE REPUBLIC OF HAITI. The extent of the exposure is undetermined, but probably it is not more than 100 to 300 meters in diameter. The rock consists of white limestone, apparently massive and partly broken down by solution into a mass of large boulders. In the calcareous matrix there are a few small fragments, probably of volcanic rock. The whole limestone mass is richly fossiliferous. It contains some undetermined Foraminifera, but the conspicuous fossils are large, mostly fragmentary rudistid mollusks. (See list, p. 98, stations 9880 and 97.46.) 1 Some single specimens on weathered exposures are more than a meter in length, but it is difficult to extract them from the tough rock. The limestone is virtually a reef rock formed by these mollusks. The relations of this limestone to the underlying rocks were not deter mined. Two exposures of a thick-bedded tilted conglomerate found near by probably should be referred to the upper Eocene basal conglomerate. (See p. 111.) Near La Tannerie. Upper Cretaceous limestone was found near the railway station called La a few kilometers north of Grande Riviere du Nord, at the northern border of the mo11ntains. The limestone at this place forms the first low mountain ridge bordering the North Plain. It is exposed just north of La Tannerie, in a ra. ilway cut about 75 meters long and 7 meters in maximum height. The rock at the south end of the exposure consists mainly of dark-blue metamorphosed limestone containing prominent veins of crystalline calcite. At the north end the limestone is mixed with a considerable amo11nt of purple argillite and dark-brown sandstone, evidently belonging to the Lower Cretaceous series as developed near Dondon. (See p. 88.) The whole mass is intricately faulted and badly crushed, and the intermixture of argillite and limestone is probably due to the faulting. On the ridge west of this cut the limestone is exposed for several hundred meters. It is rather thick-bedded, the beds ranging from 20 to 40 centimeters in thickness. The strike is somewhat south of west and the dip vertical or very steep to the north. The rock is brown on weathered surfaces and is etched and furrowed by peculiar ridges. The broken in terior is dense and dark blue . The rock contains small scattered fragments of igneous rock. No fossils show on the interior, but weathered surfaces contain undetermined Foraminifera and numerous fragments of rudistids (see list, p. 98, station 9881), which appear to be indigenous to the cal ca r eous matrix. To the south the limestone is in contact with quartz porphyry. Near La Tannerie the contact is along a ravine and is covered by detrital material, but westward it rises up the northern slope leading down to the ravine and curves gradually southward. Although the relations were not definitely 1 The numbers designate stations at which collections were made by members of the U. S. Geological Survey. The collections are deposited in the U. S. National Museum.

PAGE 110

SEDIMENTARY R OCKS. 95 determined the contact is believed to be marked by a fault. Along the contact there is a zone, 2 or 3 meters wide, of altered green igneo us rock Directly east of La Tannerie, on the mo11ntain s lopes east of the Gran de Riviere du Nord, there are sm. all outcrops that appea r from a distance t o be limestone, probably the same as that at La Tannerie. M orne Grand-Gille. Morne Grand-Gil le is an isolated hill in the North Plain about 3 kilometers northeast of La Tannerie It stands just north of the road connecting l\Iilot and Limonad e and just east of the Grande Rivi ere du Nord. This hill is capped by dark limestone. Float found alon g the road at the foot of the hill indicates that the rock is the same as that at La Tannerie. A collection of float fossi ls made here was lost. MoNTAGNES DE TERRE-NEUVE. At a n altitude of about 800 meters above sea leve l, on the west slope of Morne Guimbi, float limestone containing small indetermjnab le gastro pods and an Anomia-lik e bivalve was collected This rock may be of Upper Cretaceous age, a l though at the same locality float containing upper Eocene Foraminifera (see list opposite p 144, station 9816) was obtained MASSIF DE LA SELLE. In the southern part of the Republic impure limestone and tufface ous rocks, apparently interbedded in basalt, are considered Upper Cretaceo us, althou gh the evidence is rather meager Some of these rocks contain fresh tuffaceous material and also marine fossils The volcanic material must either have been blown into the sea during eruptions or was quick ly washed from the land surf ace during floods Other tl1inly laminated rocks that contain no marine fossils probably were laid down on flood p lains. Such rocks were seen at many places in the basalt areas of the }.lassif de la Selle In the western part of the Commune of Marigot, along t he trail leading from the coast to Etang Bossier, a brownish-gray sha ly limeston e was found in the area of basalt at a lo cality about a kilometer south of the lalce. It contains smal l pieces of thick shell that have the fine fibrous texture characteristic of Inoc eramus A dark impure limestone containing finely laminated bands crops out on the same trail about half a kilom eter closer to the lake This rock contains fish vertebrae and c ar boniz ed plant fragments Rocks of the same age were seen at several places in the large basalt area around Furey. They consist of black laminated calcareous tu.ff, sandy tuff, and banded chert. The black laminated tuff contains scattered fresh angular fragments of augite and fragments and crystals of plagioc lase. Thin seams in the tuff consist largely of crystals of p lagioclase and f rag m ents of augite The matrix is calcareous and contains Globigerina, Textularia and other small Foramjnifera, some of which have been re placed by hematite The matrix is stained brown by iron oxides and othe r impurities

PAGE 111

96 GEOLOGY OF THE REPUBLIC OF IIAI'rI. Similar rocks crop out on the south slope of the mountains along th trail from Jacmel to Carrefour. Dark banded calcareous tuff and impur limestone are exposed on Riviere Gosseline, at the south end of the bas al t gorge between the locality marked by the contact 'vith upper Eocene lime stone and Carrefour Andral, where the trail leaves Riviere Gosseline and ascends Riviere Mabial. These rocl{s are crumpled at the place wh ere the trail crosses from the right to the left side of the river. Along the san1e trail near the crest of the mountains, at an altitude of 1,050 meters above sea level, on the south slope, a bed of black tuffaceous limesto ne stands vertically in a cut along the trail. Numerous exposures of dark shaly limestone or tuffs were seen in t he area occupied by basalt along the trail from Corail-Brache to tl1e he ad of the valley of Riviere des Citronniers, on the easte1'n route from J acm el to the Leogane Plain. Most of the rock exposed a ppeared to be in thin beds interbedded in the basalts. As these exposures were near the cen t e r of the basalt band it is probable that they are in the lower part of t he basalt series. At all these localities the exposures of these roclrs are very narrow acro s s the strike, and the beds seem to lie between basalt They have a re markably uniform northwesterly strike, indicating that they belong to t he same series as the basalt They apparently lie near the base of this series At all the exposures examined the beds dip steeply, usually to the southwest, indicating that the rocks are folded, although the folding is not apparent from exposures of the basalt. MASSIF DE LA HoTTE. Limestone that is probably of Upper Cretaceous age was found near the Grande Riviere de Nippes north o.f the Asile \ T alley, along the trail be tween l Asile and Anse-a-Veau. In this region the Grande Riviere d e Nippes occupies a narrow gorge cut mainly in basalt. About 3 kilometers from l' Asile the river bends sharply eastward, and the trail also turns eastward along the north bank of the river. At this turn a ledge of brow n limestone, apparently interbedded with the basalt, is exposed. It strike s about N 75 W. and dips 35 NE. Perhaps 300 meters east of the turn in the trail and river the trail parts from the river, ascends a steep mo11ntain side, and crosses a divide i n basalt at an altitude of about 300 meters above sea level. At an elevatio n of 5 0 meters south of the crest of this divide there is an exposure of brow n limestone, which strikes ea.st and dips 55 N beneath basalt. The trail for several meters follows a sharp V -shaped ditch along the contact. A little l ower on the hillside an exposure shows basalt containing fragments of limestone of irregular sjze and shape. The limestone is dark gray and partly recrystallized. Plate VIII, B, a view of this exposure, sho ws tl1at ellipsoidal masses of basalt fit into cusp-shaped edges of pieces o f

PAGE 112

SEDIMENTARY ROCKS. 97 r.&llestone. The head of the hammer shown in the view rests on one of 8 ellipsoidal masses of basalt. These features suggest pillow structure and that the basalt flowe d under water where calcareous muds were being c)eposited, the calcareous mud filling the open spaces between the pillows of basalt. At all these exposures the limestone contains 11ndetermined calcareous algae, and at the exposure near the crest of the hill it contains small indeterminable gastropods. It is of marine origin and apparently was Jaid down when the basalt was being poure d out. The masses of lime stone that are included in the basalt contain small fragments of igneous rock clouded with iron oxides. In the level savanna south of the first range of mountains south of Anse8-Veau, alon g the same trail, there is a small ledge of brownish-yellow limestone which protrudes through the basalt that floors the valley. The rock is entirely recrystallized. It may or may not be the same as the rocks described in the preceding paragraphs. Another exposure of brownish-yellow limestone was found 5 or 6 kilo meters east of Baraderes on the trail to Anse-a-Veau. It is just east of the crest of the third ridge east of Baraderes. Basalt appears to overlie the limestone, which is poorly exposed. The limestone contains the coral Actinacis ? (station 9639). In the northern part of the arrondissement of Nippes there is prob ably much lime stone overlain by or interbedded with basalt, but some of it may b e older than Upper Cretaceous. FOSSILS. The only fossils of undoubted Upper Cretaceous age come from the limestone in the arrondissements of Cap-Ha1tien and of Grande-Riviere du Nord. Rudistid mollusks form huge reefs in this ro c k at some locali ties, particularly on the mountain side east of the Citadelle of Christophe. They are similar to some of the peculiar rudistids described by Whitfield 1 from Jamaica Similar mollusks have b een found in the Dominican Re public, Cuba, and St. Croix, indicating that this fa11na is characteristic of the Uppe r Cretaceo us of the West Indies. Although the evidence is rather meager, Dr. T. W. Stanton, of the United States Geological Survey, who examined the collections from the Republic of Haiti, believes that the fauna indicates a rather late Upper Cretaceous age. Stations Upper Cretaceous. 9881 (B 300 F). Arrondissement of Cap-Haitien, mountain slope just west of La Tann erie, a station on railro a d from Cap-Hai:tien to Grande-Riviere du Nord, about 10 kilometers northwest of Grande-Riviere du Nord. J. S. Brown, collector. March 11, 1921. 1 Whi tfield. R. P., Descriptions of species of Rudlstae from the Cretaceous rocks ot Jamaica, W. I., collected and presented by Mr. F. C. Nicholas: Am. Mus. Nat. Hist. Bull., Yol. 9, pp. 185-196, pls. 6-22, 1897. '1

PAGE 113

98 GEOLOGY OF THE REPUBLIC OF HAITI. 9880 (B 296 F). Arrondissement of Grande-Riviere du Nord, trail from Dond to Citade ll e of Christophe, about halfway up mountain, about 3 kilometers of Citadelle and 10 kilometers north of Dondon; altitude, 620 meters above level. J. S. Brown, collector. March 8, 1921. 9746 (B 317 F). Arrondissement of Grande-Riviere du Nord, same locality 9880. W. P. Woodring and J. S. Brown, collectors. March 21, 1921. Upper Cretaceous jossils from limestone of Massif du Nord.a Ma ssif du Nord. Species. Cap-Haitien. Grande-Riviere. 9881 0880 9746 Undetermined coral ................................. x Mollusca: Pelecypoda : Radiolites nicholasi Whitfield ................ x Radiolites sp. cf. R. cancellatus Whitfield .... x x Radiolites sp. ............................. x Genus ? Caprinidae ......................... x x In the lists of fossils the names in the first enclo s ed beading (for example, Massif du Nord) indi cate geographic divisions; those in the second beading (for example, Cap-Haitien) indicate 1ner1ts. The numbers at the beads of the columns are those of the stations in the preceding list. Stations probably Upper Cretaceous 9663 (W 67 F). Arrondissement of Jacmel, trail from Cayes de Jacmel to Etang Bossier, about a kilometer south of Etang Bossier; altitude, 200 meters above sea level. W. P. Woodring, collector November 2, 1920. 9639 (B 113 F). Arrondissement of Nippes, trail from Baraderes to Anse-a-Veau, third ridge east of Baraderes. J. S. Brown, collector. November 22, 1920. Upper Cretaceous (f) fossils from arrondissements of Jacmel and Nippes. Coral: Station. Actinacis ? sp. . . . . . . . . . . . . . . . . . . . . . 9639 Mollusca: Pelecypoda: Inoceramus sp. indet. . . . . . . . . . . . . . . . . 9963 TERTIARY SYSTEM. Sedimentary rocks of Tertiary age are the most extensive surface rocks in the Republic, covering probably three-q1iarters of area. Middle and upper Eocene, middle and upper Oligocene, lower and middle Miocene, and Pliocene deposits were seen during the reconnaissance. Rocks of lower Eocene age have not yet been recognized anywhere in the West Indies proper. The deposits that are considered middle Eocene in this report are the first deposits of that age recognized in the West Indies proper

PAGE 114

SEDIMENTARY R OCKS. 99 out-side of Jamaica. Rocks of lowe r Oligocene age that is, of the age of the lower part of the Vicksburg group of the southeastern Unite d States were not definit e ly recognized in the Republic, but some o f t he li mestones that are tentatively considered uppe r Eocene may be lower Oligocene. None of the Miocene deposits appear to be as young as the upper M iocene Cerros de Sal formation of the Dominican Republic. The Tertiary de posits a nd their e quiva l ents in the Dominican Repub l ic and othe r regions near by a re given in the table on pages 100 and 10 1. The ro cks of Eocene and Oligocene age, corresponding to the N um mulitic series of the Mediterranean region, consist almost exc lusively of limestone, but the Miocene and Pliocene deposits consist principally of detrital rocks The Eocene and Oligocene limestones form many of the mountains of the Republic and give rise to rugged surface features in which cliffs, caverns, and sink holes are conspicuous. The Miocene and Pliocene detrital rocks are almost entirely confined to the larger valleys and p lain s and underlie intricate ly dissected lowlands The extensive distribut ion of the limestones and of the younger detri tal rocks to which they have so larg e ly contributed accounts for the high percentage of ca l cium carbonate in the rocks, soils, and waters of the Republic EocENB SERIES. The Eocene is the most extensive series of rocks in the Republic, both in area covered and in thickness. By far the greater part of the series is of upper Eocene age, but ro cks of middle Eocene age attain a remarkab le thi ckness in the arrondissement of Plaisance. Both the midd l e and upper Eocen e rocks are characteristically limestones MIDDLE EOCENE. PLAISANCE LIMESTONE. Name The name Plaisance limestone was used by Vaughan 1 for the lime stone on the northeast slope of 1Iont Puilboreau When the Plaisance limes tone was named it was supposed to be upper Eocene, but Woodr ing 1 has recently shown that it is middle Eocene. Areal distribution. The Plaisance limestone crops out in a northwest ward-tr ending strip in the southern part of the arrondissement of Plai sance and adjoining parts of the arrondissements of Gona!ves and Marme lad e This area comprises the crest and the upper part of the northeast s lop e of Mont Puilboreau and its prolongation northwestward and south eastward. The limestone was recogn i zed as far to the northwest as La Pi erre, on Les Trois Rivieres, in the western part of the arrondissemen t o f 1 A geological reconnaissance of the Dominican Republic : Dominican Rep. Geo l. Sur vey Mem., vol. 1, pp. 58, 94, 1921. 2 Woodring, W. P., Washington Acad. Sci. Jour., vol. 12, pp. 244-247, 1922.

PAGE 115

c-:> l r I 0 ::0 s c: en I I ( s: 3: .,, ::0 0 -0 ..._, m time ...,._, subdivision. Tertiary s edimenta ry deposits of the Republic of Haiti. Republic of Haiti. Dominican Republic. Localities of some other American equivalents. I European time subdivision fll"'V. a .,, I I I I Hinche formation (non-marine) I Las Matas formation Conglomerates and marl near marine.) Pliocene. Jacmel. (nonI Panama, Rica. Jamaica, Cuba, and Costa I Astian. P laisancian. I Upper. Middle. Miocene. Lower. C erros d e Sal formation. Yorktown and Duplin formations, I T rt V . d N h C 1 o oruan. Conglomerate, sandstone, and limestone along south edge of the Cul-de -Sac Plain. p. :::s 0 Las Caho bas formation. p. 0 s... bO Q) Q Q) ..., Thomonde formation Maissade tongue. i andl >4 s 0 ,.Q Madame Joie formation irg1rua an ort aro ina. La Cruz marl, Cuba; Bowden marl, Mao clay. Mao Adentro limeI Jamaica; Gatun formation, Panama stone. Gurabo formation. Cercado formation. Baitoa formation.a Bulla conglomerate.a and Costa Rica; Porto Rico; Cal-I Helvetian. vert, Choptank, and St. Marys formations, Maryland and Virginia. I I----------Porto Rico; Shoal River marl and Oak Grove sand members of Alum Bluff formation, Florida.. 1 Burdigalian. Uscar1 formation, Panama; eastern Porto Rico ; Chipola marl member of Alum Bluff formation, Florida. 0 0 0 t,:tj 0 0 Q t-
PAGE 116

, Oligocene. Upper. Middle. Cevicos limestone. LimeAnguilla; St. Croix ; Porto Rico ; \ Aquitanian. Limestone around border of t hel stone eas t of Bani and Cuba; Panama; Tampa, Florida. Central Plain and elsew here. I elsew here. Byram ca lcareous marl,o Mississippi. \Chattian. Limestone in Mon tagnes N oires, IT b a era Cha1ne des Matheux, and else -1 t s one where. formation.b near Tubano. L. I Antigua; St. Croix; Porto Rico; imeCuba; Panama; eastern Mexico; I Rupelia n. Glendon formation, Gulf States.0 -------1 I I I------------Lowe r. (?)Limestone near Dondon and in the Chaine des Mateux. Red B luff cla.v, Alabama and Missis -sippi; Marianna limestone, Gulf I Lattorfian. States. ----------1 I I I 1---------- I I Upper. Limestone at St. Bartholomew; Cuba; Trinidad; Limestone in all the mountain! in the Sierra de Neiba, Ocala limestone, and Jackson forI Priabo nian. ranges. I Sierra de Bahoruco and mation, Gulf States. Eocene. elsewhere. Middle. Plaisance limestone. Lower. Largely contemporaneous. b May include lower Oligoc e ne '' Y e ll ow limestone,,, Jamaica; borne group, Gulf States C lai1 A uversian. Lutetian. Ypresian. Trinidad; Midway and Wilcox groups, Spamacian. Gulf States. Thanetian. Montian. o See Cooke, C. W., The correlation of the Vicksburg group ; U S. Geol. Survey Prof. Paper 133, 1923. 00 t;lj s 8 0 (') 00 ...... 0 1-L

PAGE 117

102 GEOLOGY OF THE REPUBLIC OF HAITI. Gona1ves, and as far to the southeast as a locality in the arrondissement of Marmelade 3 to 5 kilometers northwest of St.-Michel de I' Atalaye. It probably extends farther northwestward in the a rrondissement o f Gona1ves It is the limestone called '' gelbe Kalke '' by Tippenhauer .1 In the eastern part of the lvlontagnes de Terre-N euve there is a con glomerate several hundred meters thick between the old lavas and the upper Eocene limestone Several collections of fossils (see lists, p. 1 07, stations 9892, 9893, and 9894) indicate that the lower part of this con glomerate may be of middle Eocene age, but the fossils were collected fro m float, and the stratigraphic relations of the beds from which they were derived are not known The exposures on the northeast s lope of Mont Puilboreau, along the road between Plaisance and Ennery, constitute the type locality Stratigraphic relations At its type locality the Plaisance limestone rests 11nconformably on volcanic rocks, probably of early or middle Meso zoic age, and has a thin but conspicuous conglomerate at its base This basa l conglomerate contains pebbles of the volcanic rocks and also of the argil lites of supposed Lower Cretaceous age in the Plaisance Valley, although the contact between lime s tone and argillite was not seen Lithology At its type locality the Plaisance limestone consists almos t entirely of hard limestone, usually gray, yellow, or white, and remarkabl y pure, as indicated by the analysis on page 501 of a sample that contained more than 99 per cent calcium carbonate (Ca C03 ) Only the beds near the base are materially different The basal beds consist of conglomerate, no t particu larly coarse, composed of ro11nded fragments of igneous rocks and argillite, interbedded with shaly material and with thin lenses of impure brownish-yellow limestorte, which also contains detrital fragments The proportion of limestone increases steadily above the base and there is a transition zone of considerable thickness between the conglomerate and the typical white or yellow limestone. In this transition zone the rock is chiefly yellow or buff limestone with some shale The beds are 5 to 1 0 centimeters in thickness and the rock has peculiar knots or lumps, which give outcropping ledges in roadcuts a rough surface The main mass of the limestone, above the transition beds, occurs in beds 10 to 25 centi meters in thickness, and on weathered surfaces it commonly app e ars mas sive It is readily disso lv ed and generally pitted with large solution cavi ties On weathering it yields a red clay soil The fossi ls listed on page 107, stations 977 1, 9857, 9888, 9889, 9858, 9441, 9859, 9860, 9442, 9861, and 9991, were collected on the north slope and crest of Mont Puilboreau At La Pierre, on Les Trois Ri vieres, the Plaisance limestone is brownish yellow, partic ularly on weathered surfaces, but some of the unweathered rock is dark gray The fossi ls listed on page 107, stations 9753 and 9754, were co llected at this locality 1 Tlppenhauer, L. G., Petermanns Mitt., Band 45, pp. 153-155, pl. 10, 1899.

PAGE 118

SEDIMENTARY ROCKS 103 The c onglom erate a t the base of the Plaisance limestone is exposed in 8 roadcut on the east slope of Crete Salee, along the road from Ennery to de l' Atal aye. At this locality the conglomerate consists principally of pebb les of purple andesite and other igneous rocks embedded in 8 purplish calcareous matrix. Shaly beds near the base of the limestone are expos ed along the west slope of Crete Salee along the same road Limestone crops out higher on the west slope and on the crest. Weathered surfaces of this limestone have the peculiar lumpy appearance of the lowe r part of the Plaisance limestone at its type locality The limestone seems to cons ist of calcareous lumps embedded in buff marl. Most of the corals, echinoids, and mollusks listed on page 107, stations 9792, 9952, 995 0 and 9951, were found weathered out and lying on the surface in roadcuts or on d umps from roadcut5 Thickness. Only a rough estimate of the thickness of the Plaisance limeston e is possible, as there are no accurate maps to show the width o:f the outcrop. The thickness seems to be at least 1,000 meters at the type locality, where there is no evidence of duplication by faulting. Stru .ctu1e. On the northeast slope of Mont Puilborea u the Plaisance lime stone dips southwestward from 20 to 30, and its strike parallels the trend of the crest of the mountain. The contact with the upper Eocene limestone just south of the crest appears to be marked by a steeply in clined norma l fault that strikes N 75 W and dips southwestward At other localities the Plaisance limestone is folded and crumpled like tl1e upper Eocene limestone Fossils. Some of the chara cteristic fossils of the Plaisance limestone are shown on Plates IX and X The most common fossils are two species of Foraminifera, Dictyoconus puilboreauen .sis Woodring (Pl. IX, Figs. 3-5, 7-8) and Dictyoconus codon Woodring (Pl. IX, Figs. 1-2, 5-6; description s on pp. 608-609) At the type locality they are found in the basal conglom erate and at different horizons up to the highest beds exposed D. codon is very similar to D egyptiensis (Chapman), a characteristic fossil of the lower Mokattam group of middle Eocene (Lutetian) age of Egypt. No orbitoidal Foraminifera were found in the Plaisance lime stone, so that its foraminiferal fauna is in striking contrast to the rich orbi toidal fauna of the upper Eocene limestone Although the species of Dictyocorz us are similar to Mediterranean species the Plaisance limestone contains none of tl1e large n un1m uli tes that reach t heir culmination in t he middle Eocene of the Mediterranean region This absence of large nummuli tes is the most striking difference between the early Tertiary Foraminif era of the Mediterranean and West Indian regions. The Plaisance limestone yielded the largest collection of echinoids ob. tained in the Republic Though poorly preserved, some of them seem to be simila r to lower Mokattam species Some of the Plaisance mollusks are similar to species from the Mediterranean region Pseudomiltha. 1iaitensis Woodring and Mansfield (Pl. X, Fig. 1), and Chama engonia

PAGE 119

PLATE IX. CHARAc-I'ERISTrc FoRAMINIFERA OF THE PLAISANCE LIMESTONE (MIDDLE EoCENE). FIGURES 1, 2. Dictyoconus codon Woodring (p. 608). Side (1) and basal (2) views of type, X 3. U.S. G. S. station 9857. U.S. N. M. catalogue No. 350586. FIGURES 3, 4. Dictyoconus puilboreauensis Woodring (p. 609). Side (3) and basal (4) views of type, X 3. U. S. G. S. station 9857. U. S. N. M. catalogu e No. 350587. FIGURE 5. Dictyoconus codon Woodring (large) and Dictyoconus puilboreauensis Woodring (small), X 3. U.S. G. S. station 9441. FIGURE 6. Dictyoconus codon Woodring. Vertical section, X 20. U. S. G. S. station 9441. FrGuBEB 7, 8. Dictyoconm puilboreauensis Woodring. Verti ca l sections, X 20. U. S. G. S. station 9441. Fig. 8 shows early spiral stage. 104

PAGE 120

11PUBLl C OF :f I A I T I :EOLOO TCAL S UR \ .EY )( 3 l x 3 x 20 x 20 X 3 7 PLATE IX X 3 ') 6 x 208 CIIARAC' ERISTIC OF P I.iAISANCE LI:\1ESTO ...

PAGE 121



PAGE 122

SEDIMENT.ARY ROCKS. 105 Woodring and Mansfield (Pl. X, Figs. 2, 3; descriptions on pp. 612-613) were collect ed at several localities. Coll ections of fossils recently obtained in Jamaica by Dr. C. A. Ma tley, Government Geologist of Jamaica, show that the ''Yellow limestone'' of Jamaica is of the same age as the Plaisance l imestone The '' Yellow lime stone'' fauna is very similar to the Plaisance fauna, but is more varied Gigantic Cerithia, which are conspicuous in the'' Yellow limestone,'' have not been fo11nd in the Plaisance limestone. Almost a century ago De la B eebe 1 considered the'' White limestone'' 2 to be of the age of the Calcaire g ro ssier of the Paris Basin that is, middle Eocene (Lutetian) .1 Stations Plaisance limestone (middle Eocene). 9771 (B 260 F). Arrondissement of Plaisance, road from Plaisance to Ennery, at north foot of Mont Puilboreau, about 5 kilometers from Plaisanc e ; altitude, 515 meters above sea lev el. J. S. Brown, collector. Februa1-y 12, 1921. 9857 (B 260 Fa). Arrondissement of Pla isance, same locality as 9771. W. P. Woodring and J. S. Brown, collectors. February 25, 1921. 9888 (B 325 F). Arrondissement of Plaisance, same locality as preceding stations but from higher beds. J. S. Brown, collector. March 24, 1921. 9889 (B 326 F). Arrondissement of Plaisance, same locality as preceding stations but from higher beds. W. P. Woodring and J. S. Brown, collectors. March 24, 1921. 9858 (B 262 F). Arrondissement of Plaisance, road from Plaisance to Ennery, north slope of Mont Puilboreau, about 7 kilometers from Plaisance. J. S. Bro\vn, coll ector February 7, 1921. 9441 (H 6 V). Arrondissement of Plaisance, road from Plaisance to Ennery, north slope of Mont Puilboreau; altitude, 475 meters above sea lev e l. T. W. Vaugh an, collector. April 18, 1919. 9&59 (B 263 F). Arrondissement of Plaisance, road from Plaisance to Ennery, north s lo pe of Mont Puilboreau, about 8 kilometers from Plaisance; altitude, 620 met ers above sea l eve l. J. S. Brown, collector. February 24, 1921. 9860 (B 266 F). Arrondissement of Plaisance, road from Plaisance to Ennery, north slope of Mont Puilboreau, about 8.5 kilometers from Plaisance; altitude, 685 meters above sea level. W. P. Woodring and J. S. Brown, collectors. February 25, 1921. 9442 (H 7 V). Arrondissement of Plaisance, road from Plaisance to Ennery, north s lope of Mont Puilboreau; altitude, 700 meters above sea level. T. W. Vaughan collector. April 18, 1919. 9861 (B 267 F). Arrondissement of Plaisance, road from Plaisance to Ennery, north slope of Mont Puilboreau, about 10 kilometers from Plaisance; altitude, 905 met ers above sea level. J. S. Brown and W. P. Woodring, collectors. February 25, 1921. 9991 (B 236 F). Arrondissement of Plaisance, crest of Mont Puilboreau. J. S Brown collector. February 16, 1921. 1 De la Beebe, II. T .. Remu1ks on the geology of Jamaica: Geoi. Soc. I4ondon 2d ser., vol. 2, pp. 143-194, pls. 18, 19, 21, 22, 1829. (See p. 171.) 2 The lower part of the ''White limestone'' ot De la Beche, which contained most of the f ossils listed, was called the ''Yellow limestone '' in the report of the Jamaica n Star vey of 1869. a Since this report was written Dr. C. T. Trechmann bas published a d(\scrlptlon or the Yellow limestone and its mollusks: The Yellow limestone of Jamaica and Its Mollusca: Geol. Mag., vol. 60, no. 710, pp. 337-367, pls. 14-18, 1923.

PAGE 123

106 GEOLOGY OF THE REPUBLIC OF HAITI. 9871 (B 284 F). Arrondissement of Marmelade, large cave about 3 kilometers northeast of St.-Michel de l'Atalaye. J. S. Brown, collector. March 4, 1921. 9872 (B 285 F). of Marmelade, float on hill about 100 meters south of large cave, 3 kilometers northeast of St.-Michel de l Atalaye. J. S. Brown, collector. March 4, 1921. 9873 (B 287 F). Arrondissement of Marmelade, south slope of mountains near plantation of United West Indies Corporation, 4 to 5 kilomet e rs northeast of St.-Michel de l' Atalaye. J. S. Brown, collector. March 5, 1921. 9734 (W 268 F .) Arrondissement of Marmelade, trail leading northward into mountains from plantation of United West Indies Corporation, east of St.-Michel de l'Atalaye, about 1 kilometer from plantation. W. P. Woodring, collector. February 8, 1921. 9792 (W 316 F). Arrondiss ement of Gonaives, road from Ennery to St.-Michel de l' Atalaye, west slope of Crete Sal e e, about 1 kilometer from crest. T. W. Vaughan and W. P. Woodring, collectors. March 4, 1921. 9952 (W 345 F). of Gona.lves, road from Ennery to St.-Michel de I' Atalaye, west slope of Crete Salee, altitude 580 m e ters above sea level. W. P. Woodring and J. S. Brown, collectors. March 24, 1921. 9951 (W 344 F). Arrondissement of Gonaives, road from Ennery to St.-Michel de l' Atalaye, west slope of Crete Salee, altitude, 565 meters above sea level. W. P. Woodring and J. S. Brown, collectors. March 24, 1921. 9950 (W 343 F). Arrondissement of Gonalves, road from Ennery to St.-Michel de l'Atalaye, west slope of Crete Salee; altitude, 535 meters above sea level. W. P. Woodring and J. S. Brown, coll e ctors. March 24, 1921. 9753 (W 287 F). Arrondissement of Gona.lves, trail from Gros-Morne to Pilate, right bank of Les Trois Rivieres, 2 kilometers above second crossing. W. P. Wood ring, collector. February 16, 1921. 9754 (W 288 F). Arrondissement of Gonaives, trail from Gros-Mome to Pilate, right bank of Les Trois Rivieres, 50 meters above third crossing. W. P. Woodring, collector. February 16, 1921. 9892 (B 338 F). Arrondissement of Gona1ves, trail from Gros-Morne to TerreN euve, float at ba"8e of mountains near Savanne Moulin in S e ction Moulin (300 meters west of station 9891). J. S. Brown, collector. March 27, 1921. 9893 (B 339 F). Arrondissement of Gonalves, fl.oat on trail from Gros-Marne to Terre-N euve, 1 kilometer west of station 9892. J. S. Brown, collector. March 27, 1921. 9894 (B 340 F). Arrondissement of Gonalves, float along trail about 1 kilometer north of little chapel in Section Moulin, north of Terre-N euve. J. S. Brown, collector. March 27, 1921. 9&59 (B 342 F). Arrondissement of Gona1ves, Meme Valley, float at Des Hammes prospects. J. S. Brown, collector. March 28, 1921. -

PAGE 124

REPUBLIC OF HAITI GEOLOGIC.AL SURVEY 2 I r r ,,,,., PLATE X 'II .\Il .. \C'l'ERIS' l I C Oli" I>LAISANCE ( l\IIDDLE EOCENE). Figure 1. Pseudomiltha haite118i8 Woodring and Mansfield (p. 612). Type, X 1. U. S. G. S. station 9792. U. S. N. catalogue No. 35057 2 Figures 2, 3. Oha111a engonia '\"\'oodring and (p. 612). Two views of type, X 1. U. S. G. S. station 9792. U. S. N. No. 350573.

PAGE 126

Fossils from Plaisance limestone (Middle Eocene). Species. Foraminifera: Nummulites (small sp.) ..... Dictyoconus codon Woodring. Woodring ..... I I I I I I I I I I I I I I I I I I I I I I I I Corals: Stylocoenia sp. . . . . . . . . . I I I I t I I I I t I I I I I I I I I ' I I I Stylophora n. sp. (massive form cf. S. ponderosa Vaughan) ...... Genus ? fungid coral, cf. Haloseris ............................ Oladocora jamaicensis Vaughan ............................... Astreopora n. (very small calices) .......................... Goniopora sp. indet .................. Echinii: Linthia sp. ct. L. cavernosa de Loriol ............................ Schizaster ? sp .................................................. E. Eupatagus n. sp. cf. Plagiobrissus sp. cf. P. mortoni (Conrad) loveni (Ootteau) .. Mollusca: Gastropoda : Volutospina sp. Latrunculus sp. cf. L. stromeri Oppenheim ................. Cassis ? sp ................................................... Amauropsis Pelecypoda : Ostrea sp. ? sp. Spondylus sp. cf. S. dumosus Morton ........................ SpondylUB ? sp ............................................... Modiolus sp. Cardita sp. cf. C. chmeietensis Oppenheim ................. Obama sp. Chama engonia Woodring and Mansfield .................. Pseudomiltba baitensis Woodring and Mansfield ........ Veneridae, genus ? . Decapod Crustacea : Zanthopsie ? sp. Massif du Nord. Plaisance. Mar1uelade. f'o. co CJ) C'l ,...... f""'4 C'l CQ I r-4 lO g Mont. d e T erre N e u ve. Gonaives. lO e-l Ol cl') I I I . I I 5 0) &3 0) S) Ol f'o. C'l lO CJ) o:i 0) 0) fg s s s O> x x x x x . . . . . . . . x x x x x x x x x I I I x x x x x x x x I I I x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x r:n t.;j s t:9 z 8 k1 c 00 0 --=l

PAGE 127

108 GEOLOGY OF THE REPUBLIC OF HAITI. UPPEB EOOENE. GENERAL FEATURES .A. real distribution. Upper Eocene rocks cover large areas on most of the mountains in the R e public and on Gonave Island. (See Pl. I.) Stratigraphic relations. The only observed contact of upper Eocene limestone with the Plaisance limestone appears to be marked by a fault (see p. 103). Upper Eoc e ne limestone overlaps the Plaisance limestone and rests unconformably on the floor of igneous and other Mesozoic rocks. This overlap is shown in the relatively short distance between the north east and southwest slopes of Jtiont Puilbor e au. Lithology. The upper Eocene rocks consist almost entirely of lime stone, but at some localities beds of shale, soft sandstone, and even coarse conglomerate are interbedded with the limestone, particularly at the base or in the lower part. A conglomerate generally lies at the base. It may be either thic k and prominent or thin and inconspicuous. The type of lime stone differs considerably at different lo c alities and also at different hori zons at the same loc al i ty. For this reason it seems inadvisable to name the upper Eoc ene limestones. The following are the most common types: One type is white to yellow and has a porous chalky or sandy appear ance, although unweathered specimens are rather hard. It breaks with a conchoidal frac ture. The beds are generally rather regular and even and most of them range from 5 to 15 centimeters in thickness. Shaly partings are common, and some beds contain sandy detrital material and ev e n coarse con g lomerate. Chert lenses and concretions a re common. F e w of the chalky b e ds contain fossils, but certain beds of more granular lime stone that show trace s of detrital material contain abundant Forami nifera. Some of the cherts also contain Foraminifera. In weathering this rock l e aves a gray or chalky soil and many reddish or brownish pieces of silic e ous limestone. This type of limestone is well developed around the borders of the Montagnes de Terre-N e uve, on Morne Bienac north of Gonaives, and along the railroad between Gonaives and Ennery. In these reg i ons it appears to form the lower or basal part of the upper Eocene, but part of it may be middle Eocene. It is probably equivalent to part of tl1e coarse basal conglomerate that contains a few beds of limestone which occurs in great thickness at places in the Montagnes de Terre-Neuve. Another type that is found in most of the larger areas where upper Eocene ro c ks crop out is a hard limestone that is gray on weathered and white on fresh surfaces. It occurs in reasonably regular beds ranging in thickne s s from 10 to 30 centimeters, but on weathered slop e s or bluffs it usually appears massive. Some of the beds are composed of elongate nodules such as characterize the lower part of the Plaisance limestone. The beds contain little or no shale or other detrital material and generally very little chert. The surface is roughly pitted and fretted. Such rock is called roche-a-ravet. The rock yields a red clay soil. This type of lime-

PAGE 128

REPUBLIC OF GEO LOG I CAL s l,, R \ L:..: A. UPPER l Jil\1ES1'0NE I,A PIERRE. PLATE :XI T he lime stone h
PAGE 130

SEDIMENTARY R OCKS. 109 atone commonly contains Foraminif era, some mollusks, and a few c orals. It is well developed south of Terre-Neuve on the road to Gonaives, along the crest of the Montagnes Noires between Dessalines and Paul, near Montagnac east of Dame -Marie, and at other places in most of the moun tains. In the Gonaives region and elsewhere it overlies limestone of the first type and forms the upper part of the upper Eocene deposits A third type consists of very evenly bedded cherty limestone, which at many places contains beds and lenses of nearly pure chert The beds are 10 to 15 centimeters in thickness The rock is white to bluish, its color depending probably on the proportion of siliceous material it contains. It is very hard and breaks with a splintery conchoidal fracture. Such beds contain little or no shale or other detrital material, except near tl1e base. At some localities b eds of white porous granular limestone are interbedd e d with cherty limestone. Chert fragments are conspicuous in the r esiduum from this limestone. The chert may contain Foraminifera, and well-pr eserved Foraminifera are fo11nd at some places in the granular limestone. This type of limestone is common in the Port-Salut Peninsula near Cavaillon and along the south coast from Marigot eastward to the Dominican border. Sim .ilar rock in the mo11ntains just south of Anse-a Veau contains many beds of dark-blue or black chert. Oth er types of upper Eocene limestone might be distinguished or different features might be chosen as criteria for thei r separation, as there are many variations from the types described The so lubility of most of the upper Eocene limestones accounts for their great changes in physical appearance, which are especially conspicuous near the surface but at places probably extend to great depth. The cover ing of soil, which is red or white, chalky or cherty, has been mentioned Oth er features include pitted or honeycombed su rfaces, sharp pinnacled projections, and surfaces coated by a crust of disso lv ed and redeposited calcareous material, which may contain excellent specimens of Recent shells or impressions of plants. Cobbles of limestone embedded in a cal careous crust along trails may simulate a conglomerate Cavities dissolved in the limestone are commonly refilled with soil and sediment that is r ecemented with calcareous materia l producing a mass of rock that may aptly be ter1ned solution breccia and that is i llustrated in Plate XI, A. Such solution breccias are common in all areas underlain by the second type of limestone described on page 108. '!hickness. The structure of the upper Eocene limestone is so complex and the areas in which it lies at the surface are so difficult to explore fully on account of their ruggedness and the thick growth of vegetation that no accurate estimate of its thickness was obtained In a ll the princi pal a reas where it is found, however it has a thickness of at l eas t several h11ndred meters an d at a few places of 1,0 00 to 2,00 0 meters The section on the B ras-A-Gauche south of Jeremie is believed to represent at least 1, 300 meters of limestone. Limestone at le&Bt 5 00 meters thic k is

PAGE 131

110 GEOLOGY OF THE REPUBLIC OF HAITI. exposed on the north slope of the mountains immediately south of Rendez-vous, on the trail from Petionville to Furey. (Seep. 131.) Structure. The upper Eocene limestone is nearly everywhere sharply folded. In general, a s has already be e n state d, the mountain ranges con sist of huge compo11nd anticlines (such as are commonly but incorrectly called a nticl i noria) or are r e mnants of such anticlin e s modified by erosion. At some places the limestone is intensely crumpled and small thrust faults are common, and at a f e w places there are probably high-angle thrust faults of consid e rable mag nitude. The folding has here and there caused shearing and produced a sheet e d structure in the rock that makes the determination of the bedding difficult or impos s ible. DESCRIPTION BY REGIONS. Massif du Nord. M orne du Cap. The crest of the Morne du Cap is form e d by a ridge that trends east and west and that is several kilom eters in length and from 100 to more than 200 meters in height. The crest and most of the north slope of the mountain are covered by lim e stone of upper Eoc e n e age, which lies unconformably on vol c anic rocks and on an impure ch erty limestone of supposed Cr e tac e ous age. The basal contact of the lime stone along the short, steep escarpment south of the crest was not e xamined but is plainly visible from a distan c e and apparently is 50 to 100 meters below the summit of the ridge. On the north slope the limestone in most places extends down to the sea. It therefore seems that the limestone dips seaward and over lies the older rocks. A basal conglo merate marks the contact of the upper Eocene and older ro clcs. This conglomerate was found in place at only one lo c al i ty, near Source Cinq Carreaux and Source d' Aubry of the Cap-Ha1ti e n water supply. The exposure is not very large but indicates a thickness of prob ably 50 meters or more, as the beds apparently are steeply title d. The rock is hard and dark and consists mainly of chert, though in part of fragm ents of volcanic ro ck. B e ds half a mete r thick were noted. At an other locality, about 1 kilometer west of Source Belair, float of a similar conglomerate was found near the base of the limestone. It cont.a.ins a greater proportion of fragments of volcanic rock, generally subangular and very firmly cemented. The main mass of the limestone a ppears to resemble the second type described on page 108. It is hard and is gray on weathered and whit e on fresh surfaces. Some of it appears to be crystalline, and it contains coarse crystals of calcite. The ro c k is gen e rally massive on weathered exposures. Its surface is pitted with s olution cavities, and at almost every locality ob serv e d it consists of huge blocks left by solution, making the land surface very jagged and uneven. Such features are conspicuous west of the light house at Fort Picolet. No reliable estimate of its thickness is possible, I

PAGE 132

I SEDIMENTARY ROCKS. 111 the upper Eocene limestone is at least a few hundred meters thick. Upper Eocene Foraminifera were collected at four localities. (See list at p. 144, stations 9854, 9853, 9847, 9798.) Near Don don. Upper Eocene limestone forms a high mo11ntain just west of Dondon and extends northward beyond the Citadelle of Chris, tophe, capping a sharp, narrow ridge known as the Bonnet-a-l'Eveque. The limestone overlies volcanic rocks, old argillites and upper Cretaceous limestone with angular unconformity. On the trail to the C 'itadelle of Christophe and the Bonnet-a-l'Eveque the contact is at an altitude of about 650 meters above sea level, but farther south the upper Eocene l imestone descends to the border of the valley of Dondon, which is only about 400 meters above sea level. At the base of the upper Eocene limestone is a conglomerate, which is exposed at several places on the trail leading down from the Citadelle to Dondon in the first half a kilometer south of the place where the trail branches from that leading down to Milot. Near the fork in the trail are dark sandy beds that may possibly belong in the basal upper Eocene. A little farther south hard conglomerate occurs in beds ranging from 15 to 40 centimeters in thickness. ri,hese beds are exposed again about half a kilometer from the fork of the trail in contact with upper Cretaceous lime stone, which they apparently overlie. A sample of conglomerate from the last exposure contains llndetermined species of Orthophragmina The hard conglomerate in both exposures is composed mainly of a gray or bluish calcareous matrix which cements a considerable proportion of small pebbles of the older igneous rocks, Cretaceous ( ?) argillites, and upper Cretaceous limestone containing rudistid mollusks. Float from the basal upper Eocene conglomera .te was found on the trail to Milot. The limestone above the basal beds in this region is very similar to that on the Morne du Cap and contains the same Foraminif era. It most closely resembles that of the second type described on page 108. It is gray and appears massive on all weathered surfaces. The surfaces are pitted by solution cavities, and the whole mass is commonly honeycombed by caverns. Several such caverns were seen along the trail to the Citadelle. The thickness of the limestone, at least on the mountain just west of Dondon, must aggregate several hundred meters. The only fossils found were Foraminifera. Good specimens were obtained from float just southwest of Milot (see list opposite p. 144, station 9744) and from a block of the masonry in the Citadelle (station 9797), which is in pa.rt built of limestone from the mountain to the south. To the east and south of Dondon there is supposed upper Eocene lime stone of a type considerably different from that to the west and north and more similar to the first type, described on page 108. This rock occurs in very even beds, which are generally 10 to 20 centimeters thick, though at some places they reach 30 centimeters. Some beds are white and have

PAGE 133

112 GEOLOGY OF THE REPUBLIC OF HAITI. a chalky granular appearance, but most of them are dense and full of blu .. ish chert lenses. The beds are sharply folded and at places show a sheeting transverse to the b edding Good exposures are fo11nd on the mountain slopes east of Dondon and south of Dondon on the trail to St.-Rafael in bluffs along Riviere Bouyaha. This rock is apparently overlain by middle Oligocene limestone, and it may therefore be lower Oli g ocene instead of upper Eo ce ne. Arrondissement of Borgne The eastern part of the coastal ridge be tween Ans e -a-Fol eur and the plain at the mouth of Riviere de Port llargot, in the arrondissement of Borgne, is composed of upper Eocene limestone which rests on basaltic volcanic rocks. A conglomerate at the base of the limestone conta ins pebbles of the volcanic ro ck and also of lime sto ne, p1ob ably of Cretaceous age. The low e r part of the limestone contains beds of shale and sandy detrital material, and tl1e upper part consists of hard gra y limestone and thinner b e ds of d ense porcellanous limestone Upper Eocene Foraminifera were collected at a locality on the west side of Baie de la Riviere Salee. (See list opposite p. 144, station 9769.) The limestone along the coast between Anse-a-liarigot and Acul Bay probabl y is upper Eocene, but was not examined. Northwest Peninsula. Bombardopolis Plateau Rocks of probably upper Eocene age were found on the Bombardopolis Plateau along the trail from Baie de Henne to Bombardopolis, where the y are w ell exposed on three or four north south ridges, described on page 370. Thes e rocks appear to overlie the reworked fragmental volcanic rock de scr ibed on page 266, but their exact relations are obscure. They are overlain unconformably by Quaternary coralliferous limestone. The suppos e d Eocene rocks consist mainly of dark to bluish very cherty limestone, in thin and even beds, but at some places they contain thick layer s of fine, dark-brown sandy material. The sandy materia l appears to contain fragments of igne ous rocks, whi c h are probably de1ived from the same source as the fragments of the underlying tuffaceous ro c k. The beds are sharply folded and generally strike a little west of north, but they dip either to the east or west at angl e s of about 30. According to Liitgens 1 the same series of rocks occurs on 11orne Chien, west of Baie de Henne. His descript i on 2 sl1ows that apparently similar rocks crop out on the edge of the plateau along the trail leading down to La Plateforme. He b e lieved, however, that the basaltic rocks unde rlying the lime stone and the fragmental volcanic beds at or near the base of the limestone are of late Tertiary or early Quaternary age. a LUtgens, R., Geog. Gesell. 1n Hamburg Mitt., Rand 32, pp. 68-69, tlg. 5, 1919. 1 Idem. p. 73, fig. 7, 1919.

PAGE 134

SEDIMENT.A.RY ROCKS 113 Farther north, near Bombardopolis, a white chalky limestone is poorly exposed at several localities. This limestone probably underlies the Qua ternary coralliferous limestone and may also be upper Eocene. Montagnes de Jean Rabel Upper Eocene limestone forms a thick cap on the top of the Montagnes de Jean Rabel, south of the town of Jean Rabel, along the crest of the Northwest Peninsula. I t rests unconformably on quartz diorite and vo l canic rocks and is overlain by Oligocene and Miocene beds At the base of the upper Eocene limestone lies a conglomerate composed of fragments of the underlying igneous rocks The only good exposure of this conglomerate that was seen is about 5 kilometers southwest of Jean Rabel, on Riviere Cadet. The conglomerate pebbles genera lly are less than 1 0 centimeters in diameter and are well rounded. The rocl{ is firmly ce ment ed, well bedded, and has a thickness of about 35 meters. It grades upward into sandy limestone that conta .ins much detrital material. This zon e is about 50 meters in thickness and grades upward into pure white lime stone like that of the second type described on page 108 This rock appar ently constitutes the great mass of the limestone in tl1is region The limestone and conglomerate on the Riviere Cadet are tilted north eastward at angles of 15 to 30, and the limestone on the Jean RabelAnse Rouge trail dips northward. The major structural feature of the mountain ra. nge appears to be an anticlinal arch Foraminifera, particularly Cuban species of Orthophragmina, are abun dant in the conglomera ,te and limestone at most of the places examined (See list at p 144, stations 9795, 9842, 9843, 9845, 9796 and 9964.) M ontagnes de T erre-N euve Upper Eocene deposits cover most of the Montagnes de Terre-N euve, although manjr patches of the underlying igneous rocks are exposed, and in the Terre-N euve mineral district the ign eous rocks are predominant at the surface. The upper Eocene deposits consist mainly of limestone, which rests as a thick cap unconformably on a basement of basaltic and andesitic rocks The lower part of the formation comprises a thickness of several hundred meters of impure beds which grade from coarse conglomerate, composed almost entirely of igneous debris, into thin-bedded limestone containing beds of sand and shale Some of the conglomerate and thin-bedded limestone may be middle Eocene. The upper part of the series, also some hundreds of meters in thickness, consists of purer limestone, generally in thicker beds and more commonly fossiliferous The thickness of the entire series was not meas ured, even roughly, but it is probably not less than 1,000 meters. Suc cessively higher beds overlap the lower beds, as the upper pure lime stone at many localities in the central part of the region rests almost directly on the old igneous landmass. The lower beds are most widely exposed and are thickest around the borders of the mo11ntain mass, whera as the purer limestone is predominant in the interior. This relation is 8

PAGE 135

114 GEOLOGY OF THE REPUBLIC O F HAITI. significant not only stratigraphically and structurally but also beca u se it seems to afford some indication that the old landmass had its axis in part in the same area that constitutes the present axis of the mountains How ever, the existence of a trough in the central part of the region must be postulated to account for the anomalous .development of thin-bedded im pure limestones on 1Iorne Miguinda Exposures of these rocks as they were found at a number of different localities are described below Some of the best exposures of the basal conglomerate are fo11nd in Section Moulin, Comm11ne of Gros-Morne, northeast of Terre-N e u v e, in the deep valley drained by Riviere Lhormand. About 400 meters north west of the Chapel, which is in the western part of this valley (for l oc a tion see Pl. II, in pocket), the conglomerate is exposed on low hills that border a trail. It contains many cobbles, 20 to 30 centimeters in diamete r, composed of the volcanic rocks common in the region (Pl. XI, B). Inter bedded with this coarse material are local beds of shaly and impure lime stone, some of them 1 to 2 meters thick. On the Terre-N euve trail 5 00 to 1, 0 00 meters southeast of the Chapel there are many thin beds of impu re bluish limestone and sandy detrita. l material, which overlie patches of weathered igneous rock About 4 kilometers east of this same Chapel and about 1 kilometer wes t of a village called Savane Moulin the Riviere Lhormand passes thro u g h a gorge in the easternmost ridge of the Montagnes de Terre-Neuve, at the border of the Trois Rivieres valley, along the main trail between Gros Morne and Terre-Neu ve. Near the west end of the gorge a reddish po r phyritic andesite is exposed in the stream bed and is overlain by a con glomerate composed of igneous materjal The conglomerate extends down stream for perhaps 200 or 300 meters and apparently dips eastward It grades upward through sandy and conglomeratic limestone into limestone of the massive type, which crops out to the east in the mai n part of the gorge The exposures of conglomerate in the stream bluffs are masked by talus. Much better exposures are found directly to the south on the southwest slope of the ridge through which the gorge has bee n eroded Here the conglomerate contains coarse and angular fragments as well as some well-rounded pebbles, and appears to be several hundre d meters thick Similar basal conglomerate of comparable thickness is found several ki l ometers to the southeast, probably along a part of the same ridge The conglomerate is associated with the manganese deposits in the vicinity o f Mor ne Macat (See p. 470 ) It crops out on the west slope and partly o n the crest of this somewhat broken ridge for an undetermined distance The observed extent is more than 3 ki l ometers. The conglomerate contai ns many large, rough cobbles of dark andesitic rocks, and also a considerab l e amount of fine sand, the whole mass being rather poorly sorted The u n d erlying vol canic rock appears to be exposed on the south en d of

PAGE 136

SEDIMENTARY ROCKS. 115 Morne Macat, but near the north end, on the west slope, the conglomerate seems to re..st on a bluish metamorphosed shaly limestone, fragments of which it contains. The beds of conglomerate generally dip eastward or northeastward, and on the crest of the ridge they are overlain in places by white limestone, probably of the pure massive type. Between the two there i s a transition zone of sandy and conglomeratic ljmestone, from which the upper Eoc ene Foraminifera listed at page 144 (stations 9868, 9869, and 9870) were collected on the crest of Marne Macat. The upper Eocene rocks in this region are cut off to the east by a fault, and at places the basal conglomerate is in cont.ract with Oligocene rocks, as described on page 121. No othe r exposures of the basal conglomerate were found in the region. Traces of it were noted about 1.5 kilometers southeast of Terre-N euve, on a trail that skirts the southwest side of Marne Miguinda, near the northwest end of that mountain. The material is composed of fragments of igneous rocks interbedded with limestone, but the rock has been brecciated by faulting, so that the relations are not clear. The fossils listed opposite page 144 (station 9800) are from this exposure. Traces of conglomerate beneath the upper Eocene limestone were also found about 3 l\:ilometers west of Terre-N euve, in a ravine south of Morne Avocat. In the interior of the de Terre-N euve the conglomerate either is thin and masked by talus or is composed entirely of fragments of igneous rock and can in f ew places be distinguished from weathered vol canic rock. As suggested on page 102, some of the conglomerate in the eastern part of the Montagnes de Terre-N euve may be of middle Eocene age. The best section of the thin-bedded limestone containing detrital mate rial is exposed on the south slope of Morne Bienac, a prominent isolated mountain directly north of Gona1ves. This face of the mountain forms a cuesta slope, on which the rocks strike approximately east and west and dip 15-30 N. On the southwest slope the beds are exposed almost con tinuously from an altitude of about 90 meters above sea level to the top of the mountains, at about 350 meters above sea level. The lowest exposed beds are thin-bedded yellowish chalky or sandy limestone with a little interbedded clay. At an altitude of about 160 meters above sea level the thickness of the beds increases rather abruptly to perhaps 35 centimeters, and at this horizon there is a conspicuous bed of conglomeratic limestone about a meter thick, which was traced for several hundred meters along the mountain side. It is composed of small pebbles, most of them less than 3 centimeters in diameter, of white or yellowish limestone that resembles the underlying beds and of dark-blue chert and smaller, badly weathered fragments of igneous rock. The pebbles are subangular to well rounded and are firmly cemented in a calcareous matrix. Foraminifera apparently

PAGE 137

116 GEOLOGY OF THE REPUBLIC OF HAITI. embedded in the matrix were collected at two neighboring localities from this bed (see list opposite p. 144, stations 9848 and 9851) The beds b elow the conglomerate may be of middle Eocene age The conglomerate is overlain by a m eter or more of crumbly gray s and stone, followed by thin beds of limestone and by shaly or sandy beds much like those on the lower slope of the mountain and containing no fossils. These beds extend to an altitude of about 300 meters above sea level In the upper part they became whiter and more chalky and con tain l ess shale Capping the mountains and overlying the thin-bedded limestone lies a massive gray limestone, 40 to 50 meters thick, which has been bro k e n down by solution and redeposition of material into a typical solut ion breccia. It is pitted and pinnacled in odd shapes, and has been so fil le d with r e d clay and stained by iron oxides that it has a peculiar reddish ti11t, noticeab l e even from the plain to the south Much crystalline ca l cite b as also been deposited in the cavities. This rock extends an undetermin e d distance northward down the dip slope of the mountain, suggesting that its dip conforms to that of the llnderlying beds I t is believed to be t he same as the pure massive type of limestone described on page 108 The only fossils obtain e d were specimens of a poritid coral of massive g rowth form (Station 9850 ) The thickness of the thin-bedd e d limestone containing detrital mate r i al exposed is fully 400 and probably more nearly 500 meters About 1 or 2 kilom e ters east of the section just described, on the lowe r slope of Morne Bienac, there is a very prominent verti c al cliff, about 5 0 meters high, which is especially conspicuous from the harbor The bas e of this cliff was examined hastily. It appears that the cliff consists of massive grayish limestone, partly conglomeratic, which probab l y dips northward. This bed probably is lower in the section than any of tho s e farther west, but its exact relations are not clear The eastern face of 1\1orne la Pierre, although not examined close l y bears a striking resemblance to the south slope of Morne B i enac I t is a long and regular escarpment that rises gradually from the sea on the south to a height of about 600 meters on the north, west of Morne So l ei l. T he lower slopes appear to be composed mainly of the thin-bedded l ime stone About 100 or 200 mete rs below the top of the escarpment stands a conspicuous, very reddish bluff, appar en tly over lain by gray or massive limestone The reddish color of the b l uff doubtless is due to the formatio n at the contact of the t1vo types of limestone of a sol ution breccia simi l a r t o that on the top of Morne Bienac Thin-bedd e d limestone simi lar to that on Morne Bienac occu pies the upper part of the south slope and all the north slope of Morne So l ei l t l 1 e first ridge crossed by the trail from Gona1ves to T erre-N e uve, north of Gona1ves. O n the south slope and near the s ummit it contains mo re shale and sandy material than is common on Morne B ienac, a t some p laces

PAGE 138

SEDIMENT.ARY ROCKS 117 enou g l 1 to constitute fully half of tl1e section About halfway up the soutl1 slope there is an exposure of badly 'vcatl1ered amygdaloidal basalt. The lim estone beds overlie the basalt, probably unconformably. Only talus is e xp osed on the lower slope. Near the foot of the north slope of Morne Sol ei l along the trail, the beds become much t11icker and contain less detrit al material. One bed of semicrystalline limestone in this section cont ains many small undetermined Foraminifera. Tl1e summit of Morne Sol ei l which rises considerably above the trail, probably has a cap of the mor e massive limestone The rocks along the t .rail are complexly folded, and t he total thiclcness of the beds can not be estimated. Just north of Soleil, on the trail to TerreN euve, lies a deep vall ey, chiefly soil-covered, known as On the slope between !fare a-Colas and the community called Figuier t .here are a number of e x posures of soft bro\vn sandstone or fine conglomerate containing pebbles o f b asalt and limestone, generally not more than 2 or 3 centimeters in di ameter. The rocks appear to be interbedded with chalky limestones. N e a r the top of tl1e main ascent, about 2 kilometers southeast of Figuier, there are some more massive limestone beds from which the Foraminifera li sted under station 9827 at page 144 \Vere obtained. All these heels appear to dip from 15 to 40 southeastward, down the slope of the mountain. Morne Bla. nc, which is scarcely more than a hill, on the nort. h shore of G ona!ves harbor, is composed almost entirely of the thin-bedded and e l astic type of limestone, apparently identical with that 011 l\Iorne Bienac T he surface of a bed on the north slope of the hill is pitte d with numerou s regular depressions that were probably made by raindrops. Near t he summit of the north slope, at an altitude of 90 or 100 meters above sea level, a bed of crystalline limestone about a meter thick, containing a few small pebbles of igneous rock, yielded Foraminifera similar to those f r om the bed of conglomerate on Morne Bie11ac. (See p. 144, station 9 735.) Near the seashore on the southwest point of this mountain the rock i s well exposed at a number of abandoned quarries In the quarries there is one unusually massive bed of limestone about 2 meters thick. At this l ocality the thin-bedded yellowish limestone containing sandy or shaly beds grades upward conformably into very white dense chalky limestone, also thin-bedded, which at many places shows a peculiar sheeted structure transverse to the bedding. There is no shaly material in this rock. On Morne Blanc the beds are generally tilted 15 to 20 in different direc tions, generally southeastward. There are a few very short and sharp flexures that have almost developed into thrust faults. Nearly all the sout h end of 1'1 orne la Pierre, along the trail from Go na1ves to La Pierre, is composed of massive gray limestone, very much altered by weathering, which at many places forms the typical solution breccia described on page 109 and shown in Plate XI, and which closely resembles the rock capping Morne Bienac. The bedding js determinable at only a few places The only fossil obtained was a coral. (See p 144,

PAGE 139

118 GEOLOGY OF THE REPUBLIC OF HAITI. station 9737.) Beautifully preserved Recent shells of the genera Urocopti& and Ohondropoma are included in the fillings and coatings of the breccia. At La Pierre a small valley cuts a deep nokh in the steep slope down to the coast. The lower part of this valley, apparently up to an altitude of 50 to 100 meters above sea level, is underlain by a pure white cha lky sheeted limestone, which is believed to be equivalent of that in the upper part of the section on Morne Blanc. This rock yields a very pure whit,e clay soil. It is definitely overlain by the more massive type of limestone form ing the solution breccia, and the massive limestone appears to constit ute the mountain heights back from the coast. La Pierre Spring, the sou rce of the water supply of Gona1ves, issues at the contact of the two types of limestone in the bed of the Ravine la Pierre and owes the position of its outlet to the presence of the underlying less pervious chalky limestone. (Seep. 591.) Higher in the section the overlying limestone, which can be traced for some distance up the ravine, appears to be less brecciated and probably grades into the pure massive type of limestone common in the interior of t he Montagnes de Terre-N euve Limestone of the thin-bedded type, togetl1er with much sandy and shaly material apparently derived from the erosion of basalt, is widely expose d in the valley of Riviere Bras-a-Droit northwest of the de Terre-Neuve and aro11nd the borders of the Arbre Plain from the Sources Chaudes eastward and southeastward for several kilometers. Near Phi lippe, on Riviere Bras-a-Droit, half or more of the section appears to con sist of detrital material. At a locality about 2 kilometers southeast of .Philippe the limestone appears to be interbedd e d with basalt, but this relation may be due to a strike fault. Upper Eocene Foraminifera were collected at this locality (see list at p. 144, station 9824) from a bed that probably belongs in the thin-bedded and elastic type of rock. Good exposures of the thin-bedd e d limestone are fo11nd on the south bank of Riviere Colombier about 1 kilometer west of Source Marianne and beyond. The rock is rather cherty. It appears to be in a fault contact with the pure massive type of limestone to the east. The fault trace occupies a deep ravine that trends N. 15 E. The thin-bedded limestones dip northwestward about 45 from an escarpment along this ravine. The presence of the massive limestone is indi cated by huge weathered blocks lying on the ground. It seems that a thrust fault would be necessary to explain this contact, and it is possible that the thin-bedded limestone is overturned along the contact On a trail leading to the Sources Chaudes, about 2 kilometers southwest of Source Marianne and just north of a settlement called Petite Place, there is an isolated series of low ridges of the massive limestone that contains many Foraminifera. (See list at p. 144, station 9828.) The mas sive limestone appears to be surrounded by the thin-bedded type of lime stone that contains much detrital material. -

PAGE 140

SEDIMENTARY ROCKS. 1 1 9 With the exception of that on Morne Miguinda (see below) the lime stone i n the high interior ranges of the Montagnes de Terre-N euve appears to b e l ong exclusively to the purer massive or thick-bedded type that carries littl e or no shaly or sandy materja l. At some loca lities, apparently near the base of the limestone, there are thinner beds that contain nodules or lum ps These beds were noted in the Section Bois-N euf northwest of Terr e Neuve and on Riviere Bassin between Meme and Bassin The more typical limestone is exposed in b l uffs, in places 100 to 300 meters high, alon g Riviere Colom bier west of Terre-N euve from Bois-N euf to Source Mar i a nne Some beds contain n11merous Foraminifera and others contain no f ossils Stations 9825 and 9826 are in this region. (For lists see p. 1 44 ) The rocks in this region are sharply folded, and strikes appear to vary a great deal Other collections of fossils from the Terre N euve valley are listed opposite page 144 (stations 9813 and 9814) E qually good exposures are common along the trail between Terre-N e uve and Gona}, r es, in the region between Dolan and Darane, where th e r e are many high bluffs. The Foramini!era listed at page 144, station 9821, were collected from a cliff near Dolan Around Darane, which is n ear the crest of the divide on this trail, there are many large, flat sink holes surrounded by cliffy slopes, about 200 meters high, of tl1e pure thi ck-bedded limestone, the bedding being approximately horizontal This locality is not far from Figuier, }1orne Solei l and the nort11 end of Marne la P ierre, where the massive limestone appears to overlie the thin-bedded l i m estone, and it seems probable that at this locality the sinks have been for med on the soluble massive limestone, which overlies less soluble and impure limestone The less soluble limestone doubtless underlies the flat-bottomed sinks and prevents their enlargement downward, so that they have begun to expa .nd laterally. Foraminifera (see list, p 144, station 9816) we1e obtained from massive li mestone on the mountain slope northeast of Meme In a gorge along R iviere Bassin, about midway between Bassin and Meme, the pure massive or thick-bedded limestone forms high cliffs. Other exposures in Section Moulin and on Morne Macat are described on pages 114-115 The limestone on lt1orne }.1iguinda, a comparatively low mountain spur in the central part of the Montagnes de Terre-N euve, a short distance southeast of the town of Terre-N euve, does not conform exactly to any of the other types but probably should be correlated with the thin-bedded and detrital limestone of Marne Bienac and elsewhere The mountain ex tends northwestward as a spur from Morne Dumuraille and is flanked on either side by deep, narrow valleys, so that it forms a rugged r idge. I t appears to be composed almost entirely of rather cherty and siliceo u s lime stone in even beds that range in thickness from 1 to 20 centimeters and are separated by shale partings hardly as thick as the limestone beds T he rock i s yellow or bluish It has been greatly fractured and the j oints

PAGE 141

120 GEOLOGY OF THE REPUBLIC OF HAITI. are usual l y cemented with calcite In weathering many polygonal b locks are broken off or etched along the bedding planes, giving the surface an unusually rough, jagged appearance. No fossils could be found in the limestone, but possib l y tl1ose in the conglomerate near the west end of the mountain may come from the base of these beds The rocks over mo s t of the mountain appear to stand nearly vertica l. Still another rather divergent type of limestone is found on Morne Bouvard, an isolated ridge about 300 meters high along the coast betwee n L a Pierre and Coridon This mountain is composed of a limestone that is black outside and commonly very dark inside. From a distance the ro c k l ooks like a lava I t is weathered into peculiarly pitted, pinnacled, and jagged forms and is very hard and dense This rock contains an unus ual species of Orthopliragmina, which is found also on Morne Grammont (see list, p. 144, stations 9865, 9866, 9867), where the lithologic type is t he same It is inferred that this limestone is mere l y a phase of the lighte r col ored pure mass ive limestone found in the upper part of the norm al section, especia l ly as there is some indication on Morne la Pierre of a gradation phase between the two types The relation of t h e structure in the central part of the Montagnes de Terre N euve to the ore deposits of the Terre-N euve district is conside r ed on pages 432-433 and is illustrated by figures 27 and 28, pages 442 and 444 As a whole the Montagnes de T erre-N euve constitute a broad, gentle anti c l inal arch trending northwestward Numerous minor folds are superi m posed on this arch, which is modified near its crest by a deep sync linal trough corresponding approximately to the depression along TerreN euve and Meme valleys On both sides of this syncline the base of the upper Eo cene limestone lies from 600 to 1,000 meters above sea level, but the dip o n the flanl(s carries the pure massive limestone down from an altitude o f 200 to 300 meters above sea level on the northeast, at the borde r of the trough of the T r ois Rivieres Valley, to sea level at many places on the southwest The crest of the major arch does not seem to pl1 1nge much in eithe r direction, and deep erosion has cut down to the lower part of the series if not to the underlying bedrock both on the southeast, at Morne Solei l and Morne Bienac, and on the northwest, in the valley of Riviere Bras a -Droit. The antic linal flank is broken on the northeast by one of the mos t prominent faults yet recognized in the Republic This fau l t is expose d on the trail between Gros-Morne and Terre-N euve, about half a kilomete r west of Savane Moulin, near the entrance to the gorge cut by Rivie r e L hormand in the low front range of the Montagnes de Terre-N euve ( Se e P l. II, in pocket ) A l ong the trail the fault r11ns along a low ridge It trends about N 55 W and seems to dip steeply northeastward, but this o bser vatio n has not b een verified On the northeast is pure white t1nf o s -

PAGE 142

SEDIMENTARY R OCKS. 121 siliferous lim estone, of chalky appearance but dense, appa rently thinbedded and prominently sheeted It r e sembles the chalky limestone found at La Pierre and elsewhere, supposedly the middle of the upper Eocene beds, but as apparent ly identica l rock occurs in the Oligocene, and as upper Oligo cene rocks are found in the lowland to the east, this limestone js tentatively correlated with the Oli gocene Southwest of the fault is massive gray limestone of upper Eocene type containing abundant Foraminifera (see list, p 144, station 9891). For 2 or 3 meters along the fault this lime stone is somewhat si li cified and is large ly replaced by b rown o r yellow limonite It a lso contains black stains, probably of manganese minerals. The pure white lim estone strikes about N 75 W., and dips 25 northeastward. The bedding of the massive limestone is not apparent, but the structural and topographic features to the west indicate that i t dips northeastward toward the fault, as the basa l cong lom erate of the upp e r Eocene is exposed not far away at higher altitudes. A few kilometers south by east, along what is probably the contin uation of the same ridge, which strikes out into the Trois Rivieres Valley, the fault was found on Morne Ma c at, where it is intimate ly related to t he deposit of manganese described on pages 47 0-475. Near this place the white sheeted unfossiliferous limestone, of supposed Oligocene age, was found east of the fau lt, which strikes about N 2 0 W West of the fault the c rest of the ridge is formed of upper Eocene basal conglomerate, which is overlain by smal l patches of grayish lim estone containing uppe r Eocene Foraminif era The sheeted limestone to the east stands nearly vertica l. The gray limestone and conglomerate to the west dip 20 or 3 0 north eastward toward the fault The fault p l ane is very steep but probab ly dips eastward. The fault trace in this lo cality is marked by a reef of altered lime stone, much of which is r e p l aced by jasper and by oxides of iron and man ganese It was followed for a distance of about 3 ki lom eters Further indications of this fault were fo11nd stil l farther south, north of the road that branches from the Gona1ves-Gros-Morne road and goes north west to Bassin North of a dry rav i ne that runs eastward past the north end of 1Iorne d' Anneau is a bluff, 20 meters high, formed of the white thin-b edded unfos silifero u s limestone, in b e ds that strike northward and stand vertical A short distance to the west are andesites and other vol canic rocks. No search was made for the fault, but it doubtless comes in somewhere in the intervening space Its trend in this locality appears to have changed to nea1ly north. This sout h ernmost locality at which the fault was noted is about 14 kilom e ters f rom its exposure n ea r Sava ne Moulin, and from this poin t it appears to extend a considerab l e distance both northward and southward. The throw of the fault is apparently eq ual to at least the total thic kness of the upper Eocene se1ies, certain ly several hundred meters It is prob ably a normal fault, a lthou gh norma l faults so great appear to be unusual in Haiti, the thrust fault being the more common type

PAGE 143

122 GEOLOGY OF THE REPUBLIC OF HAITI. Some not able faulting has undoubt e dly occurred along the centra l s yn cline in the Terre-Neuve district, but its details were not traced out, a nd indeed the work of tracing them would be v e ry diffi c ult. Faults are s u s pected to exist at other places, particularly at the c ontact of massive with thin-bedded limestone west of Source Mariann e Valley of d'Ennery. -The vall e y of Riviere d'Ennery, as the t erm is here used, includes tl1e mountain slopes north and south of that stre am, the north slope belonging to the provin c e of the Massif du Nord and the south slope to the Mon tagnes Noires. Most of the surface ro c ks in this area are of Upp e r Eocene age. In their lithology and fauna they r e semble the upp e r Eocene rocks i n the Montagnes de Terre-N euve. At the border of the plain near Gona1ves, on a small hill just north o f the road to Enne ry, thin-b e dded sandy limestone similar to that on Morne Bienac rests on dark basalt. F .arther east, where the road crosses a low ridge at the edge of the mountains, there are crumpl e d thin-bedd e d lime stones and shaly beds, probably also the same as those on Morne Bienac The same beds of limestone crop out in numerous cuts along the rail road to Ennery and at many loc alities farther east, toward Ennery A railroad cut west of Passe Re i ne along the south side of Riviere d'En nery exposes white, dense limestone in thin and very even bed s separated by beds of shaly limestone. (See Pl. XII, A..) The Foraminifera listed at p. 144 (station 9923) wer e collected at this locality. Some of the beds found still farther east are chalky or sandy and contain bluish black chert, in nodules as well as in bands that lie parallel to the bedding. Such beds contain little or no detrital ma terial At Passe Garde, about 4 kilometers east of Passe Reine, a massiv e bed of conglomerate about a mete r in thickness is interbedded in the limestone. The conglomerate is composed mainly of small rounded p e bbles of white limestone, a ppar ently similar to tha. t in the beds below and above it, emb edde d in a cal c areous matrix containing the Foraminifera li s ted a t page 144 ( s tation 9804). White, dense unfossiliferous thin-bedded limestone interbedded with brownish gray granular limestone containing pebbles of weathered igneous rock is exposed in a railroad cut on the s outh side of Riviere d'Enne ry, 3 kilo meters southeast of Passe Garde. The Foraminifera li s t e d at page 144 (station 9989) were collected from a b e d of the granular limestone. Rather similar limestone c overs most of the south slope of Mont Puilboreau, on the road from Ennery to Plaisance. At places the limestone is thin-bedded and contains shaly or sandy beds; at other places it is more massive and contains less detrital material. Thick beds of coarse crystal line limestone rich in upper Eoc e ne Foraminifera (see list, p. 144, stations 9862, 9863, and 9864) were noted at several places on this slope. Some thick but regular beds of coarse conglomerate containing large cobbles of dark igneous rock are also exposed on this slope. About 2 kilometers

PAGE 144

PLATE XII A. TYPICAL EXPOSl:.TRE OF L1)1ESTONE IN 'VESTER:N PAR" OF VALLEY OF RIVIERE D'ENNERY B. TIIIX-BEDDED Lll\IID S l 'OXE OF SUPPOSED urPER EOCENE AGE ON GRANDE RIVIERE DU CUL-DE-SAC 0 CHALKY UPPER EOCENE LIMESTONE CONTAINING BANDS AND OF EXPOSJiJD IN Sfi J A CTJTFF NEAR SALTROU.

PAGE 145



PAGE 146

I SEDIMENT.ARY ROCKS 123 from Ennery, on the lower part of the slope, there is an exposure of solid black basalt, probably part of the underlying basement of igneous rock and doubtless the source from which the cobb l es in the conglomerate were derived. Limestone belonging to the same series crops out southeast of Ennery, along the road to St.-Michel de l' Atalaye, between Ennery and the foot of Crete Salee. Foraminifera were collected from gray granular limestone about a kilometer southeast of Ennery. (See list, p 144, station 9924.) r1,l1e basalt underlying the limestone crops out at the foot of a cliff on the south side of Riviere d'Enne ry, about 3 kilometers southeast of Ennery. A bed of limestone, 50 centimeters in thickness, above the basalt contains small weathered pebbles of the basalt. The Foraminifera listed at page 144 (station 9925) were collected from this bed. This locality is one of the few at which free te sts of upper Eocene Foraminifera were obtained A mas. sive conglomerate containing cobbles of basalt as large as a man's head em bedded in a calcareous matrix overlies the lower bed of limestone The h igh est beds exposed consist of thinbedded white limestone. The beds in this cliff strike N. 45 E. and dip 15 NW. and are broken by a normal fault of small throw that dips steeply southeastward. 'fhe rocks in this region that is, along Riviere d'Ennery and on the north slope of its valley eastward and westward from Ennery appear to belong mainly to the thin-bedded type that carries a large amount of detrital material, which is so prominently developed on Morne Bienac and elsewhere in the Montagnes de Terre-N In all this region the pre vailing dip is southward, even somewhat south of the course of Riviere d'Ennery. This dip, alt. hough it is doubtless reversed in the region not far beyond, appears to carry the thin-bedded limestone beneath the great mass of limestone that caps the Montagnes Noires to the south and that forms conspicuously black and cliffy slopes However, this steep front of the Montagnes Noires was not examined, and its exact relations are 11nknown. There appears to be some question whether the limestones near Ennery are identica l with those of Morne Bienac and others of the same type farther west. The Foraminif era obtained in this region generally more closely resemble those found in the massive and pure type of limestone of the Montagnes de Terre-N euve, and possibly the beds to the east, near Ennery and beyond, sl1ould be correlated with those in the higher part of the series in the Montagnes de Tef!e-N euve, although lithologically they rather resemble the beds in the lower part of the series. It can ha.rdly be doubted that the beds bordering the plain near Gona1ves are almost if not identically the same as those of Morne Bjenac Montagnes Noires. Northwestern part Southeast of the valley of Riviere d'Ennery, in the region between the Artibonite Plain and St.-Michel de l' Atala ye, the

PAGE 147

124 GEOLOGY OF THE REPUBLIC OF HAITI. Montagnes Noires consist of ranges composed almost entirely 0 upper Eocene limestone, generally very pure and usually either thick-bedded or massive, closely resembling the limestone that forms the interior ranges of the Montagnes de Terre-N euve. The thin-bedded limestones containjng detrital beds, which are typically exposed about the borders of the Montagnes de Terre-N euve and in tl1e valley of Riviere d'Ennery, are almos t absent, so far as shown by the exa .mination of the mountain front between Gona1ves and Dessalines and by the traverse of the ranges from Dessaline s through Paul to St. -1Iichel de l' Atalaye In this region the limesto n e r ests on andesitic volcanic rocks and old argillites, probably of Cretace ous age The older rocks are exposed about the borders of the mountain range s and underlie extensive valleys in the interior of the mountains. The con tact undoubtedly is unconformable, although no basal conglomerate was found on the weathered and talus-covered contacts Descriptions of the upper Eocene rocks at different localities follow Morne Grammont, although separated by an alluvial plain fro1n the Montagnes Noires, is geologically a part of that province. This mountain appears to be composed mainly if not entirely of upper Eocene l i mestone of a peculiar type, semmingly identical with that on Morne Bouvard (see p 120) both lithologically and faunally This rock is very dark, both on the weathered exterior and on broken surfaces It is unusually pitted and r oughened by solution In general the fossils are rather sparse, although at certain horizons they are abundant (see list, p 144, stations 9770 an d 9742) and include a peculiar Ortliop}i1.,agmina, which is found also on Morne Bouvard Structurally the mountain appears to form a tilted bl ock that dips northeastward, although no positive dips were obtainab le, as the rock appears massive Similar limestone is exposed on some small hills in the plain east of the mountain and on one of them the limestone rests on dark volcanic rock Igneous rock probably underlies the lime stone of Morne Grammont a t the southwest base of the mountain, which was not examined It seems lil\:ely that here, as at Morne Bouvard, this limestone of unusual facies is the equivalent of the lighter colored rocks to the northeast, which contain a somewhat different fauna, particula1ly as the front range of the Montagnes Noires toward Dessalines appears to contain much dark limestone that indicates a transition between the two types The front range of the Montagnes N oircs between Dessalines and Go na1ves has a steep slope, which overlooks the Artibonite Plain. The northern half of this slope is peculiarly straight and regular and apparently is a fault scarp (See p 336 ) It wa. s examined at a locality about 1 2 kilo meters southeast of Gona!ves, where its crest is about 350 meters above sea level and 300 meters above the bordering plain. From the border of the p lain up to an altitude of about 200 meters above sea level the slope is composed of volcanic rock, and from that altitude to the summit of gray -

PAGE 148

SEDIMENTARY R OCKS. 125 ish limestone, at most places rather white in t he interior and apparently massive. It is much pitted and pinnacled by solution. The configuration of the slopes on opposite sides of the range unmistakably indicates that the limestone is tilted to the northeast, fo r the northeastern s lop e is long and gentle compared to that facing the plain. Very high mountains of dark li me stone rise in a much higher range to the northeast. Farther southeast, where the road from Gona1ves to Port-au-Prince skirts the base of this range, its front is more irregular and not so steep The ro ck exposed in the foothil l s is limestone, commonly rather dark on broken surfaces Similar limestone forms the main front range in the gorge of Riviere Coupe-a-l' Inde east of Dessalines, on the trail to St. Michel de l' Atala ye In front of this range is a ridge of limestone, perhaps 10 0 meters high, which was not examined. Behind the gorge of Riviere Ooupe-a-l'Inde lies a valley that is underlain by andesi tic rock, and beyond this valley stands a second much higher range, which is composed of limestone that apparently rests on volcanic rock. This range, probably 1,000 meters in height, seems to be composed entirely of pure, wl1ite, thick-bedded or massive limestone, from which the fossi l s from stations 9885, 9886, and 9887 were obtained. (For list seep. 144.) The beds in this range appear to be crumpled into a n11mbe r o f folds of northwesterly trend. The third and highest range, whose peaks probably rise 1,0 00 to 1,400 meter s above sea level, lies betweE:n the valley of Riviere Paul and a similar fiat ' alley to the northeast (Sa vane L e Cidra), both of which are about 400 meters a bo''e sea level 1 tl1ough the structure in this range and the one just considered is not very well understood, the thickness of limestone in each must be at least 500 to 1,000 meter s and perhaps conside1 abl) more. In fact, the relations observed at the border of the third range suggest a synclinal structure, but this doubtless is modified by further folding in the interior of the range. The rock probably is mainly pure massive lim estone, although outcrops of a thin-bedded white limestone withou t fossil s and suggestive of tl1e upper Eocene on l\iorne Bienac and in the vall ey of Riviere d'Ennery were found in the central part of the gap that trav erses this range east of Paul. Fosils collected from more massive, browni sh beds to the \vest are listed at page 144 under station 9883 The fourt. h and last range, which borders the Central Plain, is lower, risin g only about 7 00 o r 800 meters above sea level, and it is also narrowe1 than the other ranges. The rock in the western part of this range appears to be white massive, rather cherty limestone It yielded the Foraminifera listed opposite page 144 (station 9890) In this range the rocks appea r to be tilted steep ly, if not actually overturned, as is suggested by appa .rently s out hwestward dips at the edge of the plain about 3 kilometers southwest of de l' Atalaye The lower part of the southwestern slope of the 1'.Iontagnes Noires was examined northeast of Savanne-a-Roche on the trail to Perodin. Here

PAGE 149

126 GEOLOGY OF THE REPUBLIC OF HAITI. the rock is a white massive or very thick-bedded limestone, presumab ly of late Eocene age, although no determinable fossils w ere found in it. The top of what corresponds to the front range in this local ity is a wide bench that stands at an altitude of about 700 meters above sea level. B ehind it rises a much higher range, the rocks on which, all the way to its c rest, appea r to be lim estone At this loc ality the anticlinal arch of t he mo11ntains seems to rise much higher than in the area between Dessalines and St.-Michel de l' Atala ye. Southeastern1 part. Along the trail from Mirebalais through Dufailly to Thomonde the main part of the Montagnes Noires forms a single anti clinal range. Thin-bedded limestone, which weathers chalky and conta ins bands of chert, crops out on the crest of the range and on the slopes nea r the crest. Foraminif era of upper Eocene age were collected from the lime stone a .nd chert. (See list, p. 144, station s 9789 and 9919.) The single anticlinal ridge known as Morne Michel, extending from the gorge of Riviere Artibonite southeastward to the abandoned gorge followed by the road from Mirebalais to Las C1ahobas, was exa mined only along the southwestern slope near Mirebalais Dense, brittle white lime stone that breaks with a conchoidal fracture crops out on this slope. Float of this lim estone, which is unfossiliferous at the outcrops examined, con tained the upper Eocene Foraminifera listed at page 144 (station 9456) The basaltic rocks that underlie the upper Eocene limestone are expose d in the deep gap across the mountains southwest of Las Cahobas. Out crops of the upper Eocene lim estone were not examined at this locality, but Foraminifera of upper Eocene age were collected from float of white lime stone that probably comes from the same beds as the large block of massive limestone that stands along the road. (See li st, p 144, station 9903.) Limestone of upper Eocene age forms the crest of the range at the south eastern end of the Montagnes Noires, but the rocks are not well expose d along the trail from Belladere to Savanette, the only route along whicl1 the range was crossed. Most of the rock seems to be thin bedded, and its weathered surfaces have a chalky appearance. Foraminifera were obtained from float at two localities. (See list, p. 144, stations 9914 and 9915.) Chatne des Mateux. The Chaine des Mateux was examined principally along a trail leading from I' Arcahaie north by east across the central part of the range to Marche Desarmes. In this region the surface rocks on all the higher part of the range, at an altitude of more than 800 or 900 meters above sea level, are of upper Eocene age. These rocks probably rest unconformab ly on a basement composed mainly of volcanic rocks, but this basement is not exposed along the route traversed. Upon the upper Eocene beds on the flanks of the range, and to all appearances conformably, rests limestone of supposed Oligocene age

PAGE 150

SEDIMENTARY ROCKS. 127 The succession of the upper Eocene rocks was not ascertained very clearly. B eds that probably lie in the lower part of the series are exposed east of the habitation called Couyau on Morne Couyau, a high ridge along t .he crest of the range. These beds consist partly of rather pure white lime. stone, gray on weathered surfaces and generally decomposed and soft, like chalk. They contain numerous poor casts of fossils, including corals. The beds are from 15 to 30 centimeters in thickness. Interbedded with this limeston e is a considerable amount of gray or brown sandy material that is composed chiefly of very weathered grains of basaltic rock. On soil covered slopes this material is easily mistaken for weathered basalt. It yielded a foraminifer, probably of upper Eocene age. (See list, p. 144, sta tion 9993.) The presence of these detrital beds, which are absent elsewhere in the upper Eocene of this region, and the anticlinal structure of the range sugg est that these are the lowest beds of the series exposed on the route described. The first ridge northeast of Morne Couyau is but slightly lower and is more rugged. It seems to be composed entirely of upper Eocene limestone that is pure and massive or thick-bedded, white on fresh and gray on weather ed surfa. ces It is considerably pitted by solution but at places break s into smooth-weathered blocks. This limestone rather c lo sely re sembles the massive upper Eocene limestone of the Montagnes Noires and the Montagnes de Terre-Neuve. At some places it contains many speci men s of Nummulites parvula Cushman (see list, p. 144, stations 9487 and 9502), as well as indeterminable corals. On the south slope of Morne Oouyau there is white limestone like that just des c ribed. Farther southwest, in the vicinity of the habitation Caill e-Mare, the l imestone is thick-bedded but very cherty, full of large diskshaped nodules of chert t .hat contain many small undetermined Foraminif era. South of Caille-Mare the limestone is more thin-bedded but contains here and there thick beds of crystalline limestone full of larger Foraminifera (see list, p. 144, station 9486) and suggestive of the fossil iferous beds in the upper Eocene on the south slopes of Mont Puilboreau, near Ennery. Structurally the Chaine des Mateux is an anticlinal range, probably the most perfect example found in the Republic. The anticline is modi fied by secondary folds, and the resulting structure of the upper Eocene bed s is a series of broad anticlirtes and synclines. (See Fig. 5.) On the lower s lopes of the major anticline the upper Eocene beds plunge beneath the younger Oligocene and Miocene beds. The observations made afforded n o means of estima .ting the of the upper Eocene beds in this area, but it is certainly hundreds of meters if not perhaps 1,000 or 2,000 m eters. The upper Eocene limestone of the Chame des Mateux was seen at a lo cality about 3 kilometers southeast of Mont Rouis, where the bluffs of

PAGE 151

128 GEOLOGY OF THE REPUBLIC OF HAITI. (}] .If ,.._ l'--J I CJ : LS \ \ \ ' ' the mountajn range border the s ea. From a bluff of pure, white badly brecciated lime ston e b e side the railroad the fos sils li sted a t pag e 144 ( station 9548) were col lected. The 0 ligoc e ne lime s tone that borders the range farther to the southeast is mis sing at thi s locality, and this r e lation prob ably holds around all the north west t ermination of the range, for the cliffs and dark slopes of the Hauts de St.-Marc look from a distance lik e typical upper Eocene limestone g Along the trail from Saut d'Eau (Ville Bonh eur) to Port rn. au-Prince, at the southeast end of the Chaine des Mateux, thin
PAGE 152

SEDIMENTARY R OCKS. Massif de la. Selle. Limestones of known or sup posed upper Eocene age are the most extensive surface sedi mentary rocks in the Massif de la Selle The limestones rest on an eroaed surface of basalt, but the lower part of the series was seen at only a few locali ties. The contact between the basalt and the overlying lime stone is exposed on the south coast just west of the small stream at the habitation called Adieu-au-Monde (or Guillau mone), in the eastern part of the arrondissement of J acmel Here massive limestone rests on an irregular eroded surface of basalt, although no basal conglomerate was noted Beds at the base of the series of up per Eocene limestones are well exposed on Riviere Gosseline, along the trail from J acme l across the mountains to Car1e four They consist a lmo st en tirely of detrital material and are made up of shale, sand stone, conglomerate, and sandy limestone. A layer of sandy limestone half a meter thick contained carbonized plant fragments and an undeter mined species of Cornuspira. Basalt crops out a short dis tance farther upstream, but the contact was not seen. (See Fig. 6.) Similar detrital rocks and dark impure limestones occur in the same stratigraphic position on the Grande Riviere de J acmel, on the trail from Jacmel to Leogane They are 9 129 0 fl.l.

PAGE 153

130 GEOLOGY OF THE REPUBLIC OF HAI'rl. more fully described on page 476, in the matter on deposits of manganese minerals. Above these basal b e ds the upper Eocene limestone resembles rock of the same age in other parts of the Republic and consists of two principal types thin-bedded limestone, dense or chalky on weathered surfaces and in most places containing nodules and bands of chert parallel to the bed ding, and massive limestone that at many localities forms a solution breccia. The highest beds on the north slope were examined on the Grande Riviere du Cul-de-Sac. Both above and below the narrow gorge above Bassin General the limestone occurs in thin even beds from 3 to 10 centi meters in thickness. (See Pl. XII, B.) The narrow gorge is due to steep slopes in lower massive lime stone on the north flank of an anticlinal arch. The thin-bedded limestone is crumpled and ruptured along small high angle thrust faults. Similar limestones crop out on Morne Hopital, the conspicuous ridge sout h of Port-au-Prince. The rock is typically gray on weathered surfaces and white on freshly broken surfaces and is considerably pitted and rough ened by solution. Much of it is massive or rather thick-bedded, but there is considerable variation, and in some localities the beds are regular and not more than 10 or 15 centimeters thick. The rock of a few beds is soft and chalky rather than dense and fine-grained. The limestone contains some chert, but it is nowhere abundant. The relation of this limestone to the water supply of Port-au-Prince is discussed on page 57 0. These upper Eocene limestones dip northward rather steeply, but the folding is complex in detail and is modified by ma .ny small thrust faults that result from the breaking of overturned folds. Such structure is well exhibited in the first deep ravine east of Petionville, near Source Millet. The folds here are overturned to the north. Thin-bedded limestone is exposed on Riviere Froide, west of Port-au Prince. Farther west, along the main road that connects Port-au-Prince and Leogane, in the region between the villages of Carrefour and Gressier, there are a number of road cuts in which supposed upper Eocene limestone is exposed. At all these exposures the rock is notably thin-bedded, the beds ranging generally from 3 to 10 centimeters in thickness. The rock is a white limestone, usually rather soft and chalky where weathered. The beds are nearly everywhere steeply tilted in different directions and thrust faults are associated with the cr11mpling. Similar dense thin-bedded limestone forms the surface rock along vir tually the entire distance between Petionville and Kenscoff, as seen along the trail. In the southern part of this region .the beds apparently are arched into broad, rather open anticlines, which are separated by similar synclines. (See Fig. 21, page 335.) Farther north, near the edge of the mountain, the beds are more sharply folded and the folds are overturned to the north, some of them being broken along thrust A thickness

PAGE 154

SEDIMENTARY R OOKS. 131 of at least 5 00 meters of these thin-bedded limestones is exposed on t he steep mountain slope south of Rendez -vous, where the beds di p s outhward with an average dip of 30, but the tota l thickness is probably severa l hundred meters more Northeast of Furey more massive lower beds are silicified along the fault between the limestone and basalt The only significant fossils found in this whole series of limestones south and southeast of Port-au-Prince are the Foraminifera that were collected near Kenscoff and on the crest of Morne Hopital (see list p 144, stations 9587 and 9553, respectively), from beds near the middle of the section. The Foraminifera that were obtained from the highe r beds (see list, p 144, stations 9608, 9443, 9583, 9585) have no stratigraphic signifi cance, and these beds may be of lower O ligoc ene age Limesto nes, all presumably of upper Eocene age, are the only r ocks seen in crossing the mo11ntains from the Cul-de-Sac Plain t o the south coast, a lon g the trail from Fond-Parisien through Fond-Verrettes to Grand-Go sier, near the Dominican border In the first exposures, a fe w kilom eters south of Fond-Parisien, the rock is gray and massive on the weath ered ledges and yielded an indeterminate species of Stylophor a (station 9505) The same kind of massive limestone crops out where t he trail reaches a stream northwest of Fond-Verrettes. Near the spring that feeds the right branch of this stream the limestone for1ns a solu tion breccia. Near F ond-V errettes crumpled thin-bedded limestone is interbedd ed with massive limestone. Thin-bedded limestone crops out at several places along the north slope of the main central range. On the crest of the central range, which bas an altitude of 1,700 meters above sea level, the rocks are obscured by a thick mantle of red clay An Operculina similar to 0. ocalana Cushman was collected from a pinnac le of massive lim estone that projects above the clay on the south s lop e not far from the crest (station 9594). Typical upper Eocene Foraminifera, similar to those in the massive limestone of the northern part of the Republic, were c oll ected farther down the south slope from a ledge of jagged solution pitted limestone (See list, p 144, station 9596.) All along the south coast from the Dominican border westward to J ac m e the most common type of upper Eocene rock is white limestone in thin, even beds from 2 to 20 centimeters thick This l imestone contains c hert in nodules and in bands that lie parallel to the bedding planes The thickness of the bands ranges from a few centimeters to almost a mete r. Some of the beds of limestone are hard and dense ; other beds are dazz ling white and chalky on the weathered surface; and stil l others consis t o f granular limestone Good exposures of this t hin-bedd ed limestone may be seen at many lo calities, particu larly along Riviere Pederna les on the trail from Grand Gosier to Anse -a-Pitre. The firs t high sea c liff no rthwest of Grand-Gosier, which is 200 meters high, is composed from top to b ottom of thin-bedded hard white limestone, which weathers chalky and which contains a few lenses o f chert. The beds in this c liff are broken by faults.

PAGE 155

132 GEOLOGY OF THE REPUBLIO OF HAITI. In Plate XII, 0, is given a view of the first sea cliff west of Saltrou, show ing the thin-bedded limestone that weathers chalky and that contains nodules and bands of chert Upper Eocene Foraminifera were collected at several localities but only from beds of granular limestone. (See list, p. 144, stations 9510, 9511, 9444, 9513, 9600.) Similar thin-bedded limestone that weathers chalky forms the upper part of the upper Eocene series of rocks on Riviere Gosseline. Numerous specimens of a small stellate Orthophragmina were collected at Passe la V oute, the fourth crossing of the river as one goes from J acmel to Carre four (station 9605). Farther upstream the lower beds described on page 129 crop out. (See Fig. 6.) A conspicuous massive limestone, which usually forms a typical soluti on breccia, crops out in the desolate rock-floored plain at Anse-a-Pitre at the Dominican border and in the ridge west of the plain. The rock is sta ined light red or even blood-red by recemented clay Its surface is deeply pitted and it makes curiously shaped pinnacles. This rock forms the cliffs marked'' red cliffs'' on chart No. 2653 of the Hydrographic Office of the United States Navy. From a distance it could easily be mj staken for a scoriaceous lava. The stratigraphic relations of this limestone to the thin bedded limestone are not precisely known, but it probably lies higher in the section. The two kinds of limestone seem to be in fault contact in a bluff on the right bank of Riviere Pedernales, about 3 kilometers north of Anse-a-Pitre, but the thin-bedded limestone may dip under the mas sive limestone, as the lower slope of the bluff is covered with talus. In the :first high sea cliff southeast of Grand-Gosier, about half a kilo meter from the village, the thin-bedded limestone is crumpled, and toward the east end of the cliff a breccia of massive limestone containing angular pieces of the thin-bedded limestone and chert rests on an irregular surface that truncates the bedding. It is not known whether this breccia bas any structural significance. Massif de Ia llotte. The most widespread series of rocks in the Massif de la Hotte is of known or assumed upper Eocene age and here as elsewhere it consists mainly of limestone. This series rests unconformably on a basement com posed mainly of dark basaltic rocks and in minor part of older limestones and metamorphic rocks. Rocks of Oligocene age were not recognized, although some of the limestones here grouped in the upper Eocene may really be of 0 ligocene age. At the base of the upper Eocene there is commonly a considerab l e thick ness of conglomeratic or sandy beds composed mainly of basaltic debris, in many places difficult to distinguish from weathered igneous rock but generally interstratified with impure, dark, shaly limestone beds and so far as observed almost 11nfossilif erous No good sections of these beds were obtained, but they probably form a series that reaches a thickness of a few hundred meters at some localities and at other localities is very

PAGE 156

SEDIMENTARY ROCKS 133 thin. A number of exposures of beds believed to be basa l upper Eocene are described below. North of the road on the east side of Pont de :U1iragoane bMaltic sand stones and conglomerate are interbedded with very dark or black rock that is probably a very impure limestone These beds dip northward toward the base of Tapion de Miragoane, where higher and purer limestone crops out. At the east end of the Asile Valley, about 4 kilometers from Asile, a detrital basalti c rock underlies a high bluff of supposed upper Eocene lim estone. Other exposures of beds at the base of the limestone were seen along the trail between Aquin and the Asile Valley. The beds consist of dark-brown impure lim estone containing detrital material that is probably derived from basalt The basal conglomerate of the limestone is exposed on the trail from Cavaillon to St. -Louis du Sud, about 5 kilometers east of Cavaillon. It contains small pebbles of weathered basalt and dark limestone in a calcareous matrix. The outcrop of the conglomerate was traced to a point within a distance of a few meters of an exposure of weathered basalt. Beds of shale and shaly dark limestone overlie the conglomerate. A roadcut 8 or 9 kilometers west of Aquin exposes basalt overlain by brownish and greenish shale and impure shaly limestone. Similar beds, which are believed to li e near the base of the upper Eocene limestone, are exposed for about a kilometer along the trail between Les Cayes and Port-Salut on the eastern slope of the Port-Salut P eninsu la. The beds consist partly of rather hard blue or gray marl and partly o.f interbedded shal e and impure limestone. They are sharply folded and their strike and dip vary within short distances. These beds are overlain by white limestone of the thin-bedded and rather cherty type, of supposed upper Eocene age. The volcanic rocks that presumably lmderlie the basal beds are not exposed along the trail. Other beds very much like those just described were found about 4 or 5 kilometers north of the Sources Chaudes de Dame-Marie or de Jeremie, where unfossiliferous gray sandy beds overlie dark limestone. In the same lo cality, just at the bed of the Bras-a-Gauche, is a bluff of dark sandy and shaly limestone, very hard, partly thin-bedded and partly thick-bedded. Rounded nodules break out of certain beds Carbonized plant remains are rather abundant in the rock. This whole series is probably basal upper Eocene, for the underlying basalt crops out a short distance to the south in the river bed. At the cemetery just south of the town of Anse d'Hainault beds that probably lie near the base of the upper Eocene form a prominent ridge. They are thin-bedded shaly or fine, sandy, dark-gray calcareous sediments, which break with conchoida l fracture The thickness of the exposed beds must be from 100 to 200 meters if not more. The beds appear to overlie basalt and dip steeply northward. I

PAGE 157

134 GEOLOGY OF THE REPUBLIO OF HAITI. Certain beds of marl and conglomerate east of Baraderes, describe d as of qu estionable Miocene age on page 225, are possibly basal upper Eocene If so, the limestone in the conglomerate must be derived from older, probably Cretac e ous, limestone that underlies the basalt. The remaining and principal part of the upper Eocene section in the Massif de la Hotte may be classified into two types of limestone: ( 1) white thin-bedded limestone, generally dense but here and there open textured and chalky, usually breaking with conchoidal fracture, most of it more or less cherty and some of it extremely so, in few places containing determinable fossils but rarely rich in orbitoid Foraminifera; (2) massive or thick-bedded white limestone, in some places crystalline weathering gray and to pitted for1ns, at many places full of small 11ndeter mined Foraminifera but at a few containing large Foramjnifera. The second type closely resembles the massive upper Eocene limestone in the northern part of the Republic. It yields a characteristic red ocherous soil and in many places forms a solution breccia. Cherty limestone was fo11nd north of Pont de Miragoane in association with impure basal beds, and it also occurs abundantly as float farther north, at the base of the escarpment of Tapion de Miragoane, which ap pears to be composed mainly of the massive type of limestone. Most of the limestone on the road between Miragoane and Aquin, in the interior of the peninsula, is of the thin-bedded type containing chert, float of which furnished the upper Eocene Foraminifera listed at page 144 (station 9518). At the south border of the first mountain range south of Anse-a-Veau chert makes up virtually the entire section of supposed uppe r Eoc e ne beds. It is so black that it has been mistaken for coal. The chert beds here are 11ncommonly thick, in some places 30 centimeters in thickness. On the north side of the range white limestone, probably'" thick-bedded, consti tutes the main part of the section. In the second range between Anse-a Veau and l'Asile much similar float of dark chert is associated with white limestone, probably of upper Eocene age, but no good exposures were noted. Cherty white limestone was found on the first slope southeast of Roseaux, but its relations were not determined. Southwest of Jeremie, at and near Moron, thin-bedded cherty limestone is very ab11ndant and appears to dip northward beneath more massive purer limestone. East of Dame-Marie, on the trail to Montagnac, white limestone with chert is very abundant and yield e d upper Eocene Foraminifera. (See list, p. 144, station 9629.) The beds here are from 10 to 20 centimeters thicksomewbat thicker than elsewhere. Very similar cherty limestone is ab11ndant south of Dame-Marie on the ridge just north of Bariadele and contains abundant upper Eocene Foraminifera (se e list, p. 144; stations 9631, 9632.) This cherty limestone is suspected to be in the lower part of the upper Eocene series. The chert in this region is lighter colored than that near Anse-a-Veau and generally is slightly bluish.

PAGE 158

, SEDIMENT.ARY ROCKS. 135 Most of the limestone exposed along the south coast from Port-Salut to Jacmel is white and thin-bedded, commonly dense and in places porous or chalky, has a splintery conchoidal fracture, and contains more or less chert. It is quite devoid of large fossils, although at some localities it contains abundant Globigerina and other minute Foraminifera. '!,his rock forins the plateau of the Port-Salut Peninsula, where it lies nearly hori zontal and rests on the impure, probably basal beds described on page 133. On the east slope of the peninsula the beds pl11nge steeply beneath the Cayes Plain, if indeed they are not broken by a fault along that side. Beds almost identical in appearance and also nearly horizontal compose the ridge between the Cayes Plain and Cavaillon and were recognized between Cavaillon and St.-Louis du Sud, where they overlie the impure transition beds described on page 133, which rest on the basalt. On the east side of Aquin Bay the trail from Aquin to Cotes-de-Fer crosses low bills composed of limestone in beds 20 to 30 centimeters thick. This limestone is gray and chalky on weathered surfaces and contains bands of chert. Along the coast the upper Eocene rocks are generally con cealed by Quaternary deposits. Chalky limestone of assumed upper Eocene age crops out in a ravine north of Cap Raymond, on the trail between Mayette and Bainet. Farther east, between Bainet and Jacmel, white granular limestone with chalky weathered surfaces is exposed on l'Eau Genee, west of Morne Fontai. Foraminifera similar to those in the granular limestone farther east were collected from similar limestone on tl1e west slope of Morne Fontai. (See list, p. 144, station 9649.) It therefore seems that thin-bedded cherty and chiefly unfossiliferous limestone is characteristic of the lower part of the upper Eocene section pver virtually all the Massif de la Hotte and lies in nor1nal succession above the more impure basal beds. It is quite possible, however, that cherty beds occur in the higher part of the section at some places, and this criterion should be used with caution. The thickness of the upper part of the upper Eocene section is not known but probably aggregates some hundreds of meters. On the Port Salut Peninsula and between Les Cayes and Cavaillon the thickness can not much exceed 300 meters, as the beds lie nearly flat and the relief would not allow for a greater thickness above the basal beds. The massive or thick-bedded purer type of limestone covers large areas in and around the section called Plymouth, southeast of Jeremie, and is exposed along the trail to Jeremie most of the way from Petit-'l'rou de Nippes to Roseaux. A small species of Nummulites, similar to one found in upper Eocene rocks of other parts of the Republic, was collected from this limestone between Petit-Trou de Nippes and Baraderes (station 9547). This limestone is also found in part of the first range south of Jeremie and at places in the succeeding ranges to the south and is well developed near Montagnac east of Dame-Marie. Similar limestone ap pears to constitute the mass of the first range north of Port-a-Piment, -

PAGE 159

-136 GEOLOGY OF THE REPUBLIO OF HAITI. the range south of Camp Perrin (station 9 509, see list, p. 144), and the heights of the range south of the Asile Valley. Massiv:e limestone is found near Miragoane (see Pl. XL, B) and prob ably forms most of Tapion de Miragoane. On the road leading eastward from Miragoane, about half a kilometer from the town, the small Dicty oconu.s, which is very abundant in uper Eocene rocks of the northern part of the Republic, was collected from this massive solution-pitted limestone (station 9517). West of the locality where the Aquin road branches off the limestone is faulted against basalt. This limestone appeal's generally to overlie the thinner bedded and cherty limestones, but at some lo calities it probably rests directly on the dark impure basal beds, as for instance along the trail between Aquin and l' Asile. Thus either the upper beds are transgressive with overlap or at places the thin-bedded limestone passes by lateral gradation into more massive and less cherty rock. Along the south coast between Mayette and Jacmel massive limestone overlies the thin-bedded chalky limestone. At some localities it seems to dip more gently than the thin-bedded limestone, but it probably is the same massive limeston e that elsewhere is known to be of upper Eocene age, although it may be Oligocene or Miocene. No fossils were found in it except poorly preserved indeterminate corals. This massive limestone yields a heavy residu e of red clay that is much more fertile than the lightcolored soil formed by the thin-bedded limestone. The structural features of the upper Eocene rocks of the Massif de la Hotte were not very clearly ascertained, but in its western part they appear to be generally folded into sharp, open folds that trend more or less nearly east and west. The general features are apparent in Figure 7, a section across the north side of the massif from Jeremje to the Sources Chaudes. Although not accurate in detail, this section gives the general features of the structure. An exception to the genera l folded structure is found on the south coast, where the thin-bedded cherty limestones of the Port-Salut P eninsula lie nearly horizontal. Faulting also has taken place on a considerable scale, although gener ally it is not easily recognized. Topographic and stratigraphic evidence indicate that thrust faults of considerable magnitude exist at the north side of the Miocene area southwest of Jeremie (see Fig. 7) and at the south side of the lowland near Camp Perrin. (See pp. 235-236 and Fig. 17'.) A fault probably borders the south side of the Asile Valley. All these faults are of post-Miocene age. A normal fault is exposed about 1 kilometer southeast of Miragoane, on the Port-au-Prince road. Massive upper Eocene limestone has been dropped down against the older volcanic rocks. A weathered brecciated zone 3 or 4 meters wide in the volcanic rock marks the fault plane.

PAGE 160

Calce:ire corellien pleistocene Kilometres Tm .......... Te---<' 0 10 15 -----------"T-'=----Hauteurs double es et gt:e5 Cal I caJre eocene m1ocenes I super1eur rl <.., "''k "' -:-....,.. ,:5:: '?..j,u<1-: Ba.salte cretace FIGURE 7. S ect ion across the northw est p art of the Massif de la .Hotte Niveau dek.M"er rn t:j ..... 0 a t;q rn C.:>

PAGE 161

138 GEOLOGY OF THE REPUBLIC OF HAITI. Gona ve Island. Limestones of upper Eocene age cover a large part of the southeas tern half of Gonave Island. They are made up of a lower part consistin g of thin-bedded limestone having a chalky appearance on w e athered surfaces and an upper part consisting of massive limestone. (See Fig. 8.) The greater solubility of the massive limestone has a marked effect on the sur face features and on the supply of water. The maximum exposed thick ness of the lower chalky limestone is about 100 meters, but its base was not seen. The upper massive limestone is probably several hundred meters thi c k. Chalky limestone crops out in the ravine southwest of Anse-a-Ga lets at and below the spring that supplies this part of the island with water. Here the limestone is rather thick bedded, but it inc ludes thinner beds of harder, denser limestone, less than 30 centimet ers thick The beds .. ... . ..... .... Q ar: ... =:. . f: .: : . .. . ... .Alluvions l Te Tm I Q,} Kil 2. :re Calcajre Cales.ire eos:ene superieur Calcaire miocene FIGURE 8.-Section across the southeast part of Gonave Island. probably crop out on the crest of a low arch, but at the spring they dip gently north e astward and farther downstream they dip 11nder massive lime s tone. Simjlar chalky limestone is poorly exposed in the ravine northwest of Picmi. It covers a large area west and northwest of the Mapoux Plain and probably underlies the covering of soil in the 1Iapoux Plain. The overlying massive limestone forms the ridge along the south coas t and apparently covers most of the southeastern end of the island south east of Fond-I' Aurore. Foraminifera are very abundant in it at some localities (see list, p. 144, stations 9682, 9669, 9670, 9668, 9665, 9681, 9671, 9672, and 9674), and a numb e r of mollusks were collected at the west end of the Mapoux Plain. (See list at p. 144, station 9673.)

PAGE 162

SEDIMENTARY ROCKS 139 Fossils. Foraminifera are the most abundant fossils in the upper Eocene lime stone, as is shown by the list at page 144, which records Foraminifera definitely of upper Eocene age from 75 localities Faunules from 16 additional localiti e s a re added, either because. they come from beds that contain significant uppe r Eocene fossils elsewhere or because the beds are supposed to be upper Eocene Although the maximum thickness of uppe r Eocene rocks is v ery great (between 1,000 and 2,000 m eters ) it has been impossible to dis criminate faunal zones because of the lack of detai led work. Early workers in othe r West Indian islands have ass11med that similar limestones are de e p-water d e posits, but the living Foraminifera that are most similar to the fossils obtained are tropica l shoal-water species. The almost compl ete ab s ence of othe r groups of organisms at many localities is difficult to explain. Some of the characteristic upper Eoc ene Foraminifera are shown on Plate XIII. Orbitoida l Foraminifera of the genus Orthophra : gmina are very common in these d e po s i t s So far as known this genus is restricted to ro c ks of Eocen e age in the \iVest Indies and regions near by, although it is found in upper C retaceous deposits (Danian). In the northern part of the R e public many of the sp e cies are identica l with or very simila r to the following species des c ribed by Cushman 1 from ro c ks of the same age in Cuba: Orthophragmina cub e nsis, 0 crassa, 0. soulpturata, and 0 pustulata. Another species, which has been obtained from upper Eocene deposits of the southeastern Unite d States, 0. fiintensis Cushman, is com mon in nearly all tl1e r e gions where collections w e re made Species of Lepidocyclina are not so wid espread nor so abundant as Orthophragmi'Tl)(L. The assocjation of Orthophragmina and L e pidocyclina is a chara c t e ristic faunal f eature of uppe r Eocene rocks of the W est Indies and regions near by, and als o of those of Italy and other areas in the Mediterran e an region. Dictyoconus puilbor e auensis nannoid e s Woodring (se e p 609 and Pl. XIII, B, 0) is the most common foraminifer in the northern part of the Republic. It is very similar to the Plaisance s ubspecies D. puilbo reauensis puilboreauensis Woodring Spe c ies of Nummulites and Operculina, though abundant, are difficult to det e rmine b e caus e most of the m are known only from natural rock sec tions or thin s e ctions Poorly pres erv e d cora ls, which have virtually no stratigraphic signifi c ance, were col lected at only three localities Mollu s ks are not very abundant in the upper Eocene ro c ks The largest collection, 9 or 1 0 species, represented by casts and impressions, was col lected on Gonave Isla.nd, at the west end of the inclosed va lley cal led 1 Cushman, J. A., Carnegie Inst. Washington Pub. 29 1, pp. 52-55, pls. 9, 10, 15, 1919.

PAGE 163

140 GEOLOGY OF THE REPUBLIC OF HAITI. Plaine des Mapoux. No Foraminifera were obtain e d at this lo c ality The Terebellum is very similar to an unnam e d sp e ci e s in the coll e ctions of the United States National Museum from a locality near Port Antonio, Jamaica. At the pres ent time this genus is confined to the Indo-Pacific region, but during Tertiary time it lived in all parts of the equatorial geosyncline. Oth e r mollusks from this locality are similar to species from the Ocala limestone of Florida. Limestones of upp e r Eocene age whi c h contain the sa m e or a simila r fauna are widely distributed in the W est Indian region and in adjoining parts of C entral and South Ameri ca.1 The St. Bartholom e w limestone of the island of St. Bartholomew is c on s ide red the type of this horizon. Stations Upp e r Eoce n e 9854 (B 239 F). Arrondissement of C a p-Ha.ltien tra il l e ading up Marne Lorry, west of Cap-Haltien, about halfway up mountain. J. S. Brown, coll e ctor. February 18, 1921 . 9853 (B 237 F). of C ap-Ha 1ti e n, float on hill slope just north of Carenge at north end of Cap-Hai:tien. J. S. Brown, collector. F ebruary 18, 1921. 9847 (B 242 F). Arrondissement of Cap-Haitien, Source Cinq Carre aux, CapHaitien water supply. J. S. Brown, coll e ctor. February 19, 1921. 9798 (B 243 F). Arrondissement of Cap-Haitie n, Fort Picol e t lighthouse, about 3 kilometers northeas t of Cap-Haltien. J. S. Brown coll e ctor. February 20, 1921. 9744 (B 314 F). Arrondiss ement of Cap-Haitie n, float on trail from Milot to Citadelle of Christophe, northeast slope of Bonnet-a-l 'Eveque, about 5 kilometers south of Milot. J. S. Brown and W. P. Woodring, collectors. March 20, 1921. 9797 (B 294 F). Arrondissement of C ap-Hait ien, from masonry of Citadelle of Christophe. J. S. Brown, coll e ctor. March 8, 1921. 9769 (W 306 F). Arrondissement of Borgne, trail from Le Borgne to Port Margot, about 3 kilom eters southeast of Le Borgne, west side of B a i e d e la Riviere Salee. W. P. Woodring, collector. February 23, 1921. 9862 (B 268 F). Arrondissement of Gonaives, road from Pla isance to Ennery, just south of crest of Mont Puilboreau .. J. S. Brown colle ctor. February 25, 1921. 9863 (B 269 F). Arrondiss ement of Gonaives, road from Plais ance to Ennery, south slope of Mont Puilboreau, altitu<;le 690 meters above sea l e v e l. J. S. BrQwn, coll e ctor. F ebruary 25, 1921. 9864 (B 270 F). Arrondi ssement of Gonaive s, road from Pla isanc e to Ennery, south slope of Mont Puilboreau, altitude 585 meters above s e a l e v e l. J. S. Brown, collector. February 25, 1921. 9924 (W 228 F). Arrondiss ement of Gonaives, road from Ennery to St.-Michel d e l Atalaye, 1 kilom eter southeast of Ennery, altitude 380 m e t e rs above se a l e vel. W. P. Woodring, collector. J anuary 24, 1921. 9925 (W 229 F). Arrondiss ement of Gonaive s, road from Ennery to St.-Michel de l'" Atalaye, cliff on left bank of Riviere d 'Ennery about 3 kilomete rs southeast of Ennery. W. P. Woodring, collector. January 24, 1921. 9989 (W 227 F). Arrondissem ent of Gonaives, railroad from Gonaive s to Ennery, cut 3 kilometers east of Passe Garde. W. P. Woodring coll e ctor. January 24, 1921. 1 For a recent summary of the Tertiary deposits of the W est Indies and adjoining regions s e e Vaughan, T. W., Correlation ot the Tertiary formations of Central America and the West Indies: Bernice P. Bishop Mus. Special Pub. 7, pp. 819-844, 1921

PAGE 164

REPUBLIC OF HAITI GEOLOGICAL SURVEY .. B. PLATE XIII A C. SO:\IE CHARACTERISTIC UPI>Eil EOCENE FORA)1INIFERA .tl. Orthophragmi1ia ctassa Cushman. Vertical sections, X 20. U. S. G. S. station 98rl4. B, 0. Dictyoc: ori1ts pitilbo1eaiten sis '11an1ioides Woodring ( p. B. Horizontnl section ne;1r base. X 20. U. S. G. S. station 9821. 0 section, tJ pe, X 20. U. S. (j, station 9821. U. S. N. M. catnlogtte No. 3=>05 8.

PAGE 165

'

PAGE 166

SEDIM:ENT.AB.Y ROCKS. 141 99'13 (W 226 F). Arrondissement of Gonaives, railroad from Gonaives to Ennery, cut about 4 kilometers northeast of Les Poteaux. W. P. Woodring, col lector. January 24, 1921. 9804 (B 146 F). Arrondissement of Gona1ves, railroad from Gonaives to Ennery, cut 27 kilometers from Gona1ves and 4 kilometers east of Passe Reine. J. S. Brown, collector. January 6, 1921. 9868 (B 276 F). Arrondissement of Gona1ves, Morne Macat, west of road from Gona1ves to Gros-Morne, about halfway between the two t owns. Float at summit of bill. J. S. Brown, collector. March 2, 1921. 9869 (B 277 F). Arrondissement of Gona1ves, Mame Macat, a short distance northeast of 9868. Float at summit of hill. J. S. Brown, collector. March 2, 1921. 9870 (B 278 F). Arrondissement of Gona1v es, Mome Macat, a short distance northeast of 9869. Float at summit of hill. J. S. Brown, collector. Ma.rch 2, 1921. 9827 (B 192 F). Arrondissement of Gona1ves, trail from Gonaives to TerreN euve, about a kilomete r southeast of Figuier. J. S. Brown, collector. January 21, 1921. 9735 (B 221 F). Arrondissement of Gonaives, north slope of Marne Blanc near crest, small mountain on coast west of Gona1ves. J. S. Brown, collector. Febru ary 11, 1921. 9848 (B 216 F). Arrondissement of Gonaives, southwest slope of Mome Bienac, about 2 kilometers north of Gonai:ves. J. S. Brown, collector. February 8, 1921. 9850 (B 219 F). Arrondissement of Gonai:ves, float on crest of Mame Bienac. J. S. Brown, collector. February 8, 1921. 9851 (B 220 F). Arrondissement of Gona1ves, about halfway up southwest slope of Marne Bienac. J. S. Brown, collector. F ebruary 8 1921. 9737 (B 227 F). Arrondissement of Gona1ves, trail from Gonai:ves to La Pierre, float at east base of Mome la Pierre. J. S. Brown, collector. F ebruary 11, 1921. 9865 (B 273 F). Arrondissement of Gona!ves, near coast between Gonaives and Anse Rouge, float on crest of Morne Bouvard at Pointe Coridon, altitude 180 meters above sea level. J. S. Bro\vn. collector. F ebruary 27, 1921. 9866 (B 274 F). Arrondissement of Gona1ves, float on cr es t of Mome Bouvard, a short distance east of 9865, altitude 260 meters above sea level. J. S. Brown, collector. February 27, 1921. 9867 (B 275 F). Arrondissement of Gona1ves, float on crest of Mame Bouvard, a short distance east of 9866, altitude 300 meters above sea level. J. S. Brown, collector. February 27, 1921. 9813 (B 163 F). Arrondissement of Gonaives, Terre-N euve Valley, float east of Germinie J. S. Brown, collector. January 12, 1921. 9814 (B 167 F). Arrondissement of Gonaives, Terre-Neuve Valley, float on trail to Raboteau, about a kilometer north of habitation Gaillarde. J. S. Brown, collector. January 14, 1921. 9824 (B 184 F). Arrondissem ent of Gonalves, trail f ram Terre-N euve to Philippe, north slope of Marne Songe, about 5 kilometers southeast of Philippe and 7 or 8 kilometers northwest of Terre-Neuve. J. S. Brown, collector. January, 18, 1921. 9800 (K 204 F). Arrondissement of Gonalves, Terre-N euve region, southeast of Rocher toward foot of Marne Miguinda. W. S. Burbank, collector. March 28, 1921. 9891 (B 337 F.) Arrondissement of Gonalves, trail from Gros-Mome to TerreN euve, foot of mountains near Savanne Moulin, Section Moulin. J. S. Brown, collector. March 27, 1921. 9821 (B 180 F). Arrondissement of Gona!ves, cliff about a kilometer south of Dolan, near trail from Terre-N euve to Gonaives. J. S. Brown, collector. January 17, 1921.

PAGE 167

142 GEOLOGY OF THE REPUBLIC OF HAITI. 9816 (B 170 F). Arrondis;ement of Gonaives, float on mountain slope northeast of Meme, just above large prospect excavation on east side of valley. J. S. Brown, collector. January 15, 1921. 9825 (B 190 F). Arrondissement of Gonaives, trail from Terre-N euve to Anse Rouge, gorge about 7 kilometers northwest of Terre-Neuve and 4 kilometers east of Source Marianne. J. S. Brown, collector. January 19, 1921. 9826 (B 191 F). of Gona1ves, trail from Terre-N euve to Anse Rouge, about a kilometer east of 9825. J. S. Brown, collector. January 19, 1921. 9828 (B 194 F). Arrondissement of Gonaives, trail from Terre-N euve to Anse Rouge, about 2 kilometers southwest of Source Marianne, just north of Troupaux. J. S. Brown, collector. January 25, 1921. 9845 (B 212 F). Arrondissement of Mole St.-Nicolas, trail from Jean Rabel to Anse Rouge, Riviere Cadet, about 5 kilometers southwest of Jean Rabel. J. S. Brown, collector. February 2, 1921. 9842 (B 210a F). Arrondissem ent of Mole St.-Nicolas, trail from Jean Rabel to Anse Rouge, Riviere Cadet, about 5.5 kilomete rs southwest of Jean Rabel. J. S. Brown, collector. February 2, 1921. 9843 (B 211 F). Arrondissement of Mole St.-Nicolas, trail from Jean Rabel to Anse Rouge, Riviere Cadet, about 6 kilometers southwest of Jean Rabel, near contact of limestone and volcanic rocks. J. S. Brown, collector. February 2, 1921. 9795 (K 120 F). Arrondissement of Mole St.-Nicolas, trail from Jean Rabel to Anse Rouge, Riviere Cadet, about 6.5 kilometers southwest of Jean Rabe l, from limestone overlying volcanic rocks. W. S. Burbank, collector. February 2, 1921. 9796 (K 121 F). of Mole St-Nicolas, trail from Jean Rabel to Anse Rouge, about 7 kilometers southwest of Jean Rabel. W. S. Burbank, col l ector. February 3, 1921. 9964 (K 123 F). Arrondissem ent of Mole St.-Nicolas, trail from Jean Rabel to Anse Rouge, float about 12 kilometers south of Jean Rabel. W. S. Burbank, col lector. February 3, 1921. 9742 (B 231 F). Arrondissement of Gonaives, north end of Morne Grammont, about 5 kilometers southeast of Gonaives. J. S. Brown, collector. February 12, 1921. 9770 (B 229 F). Arrondissement of Gona 1ves, east slope of Mome Grammont, about halfway to crest. J. S. Brown, collector. February 12 1921. 9890 (B 327 F). Arrondissem ent of Marmelade, trail from St.-Michel de l'Atalaye to Dessalines, about 3 kilometers southwest of St-Michel de l'Atalaye. J. S Brown, collector. March 25, 1921. 9883 (B 331 F). Arrondissement of Marmelade, trail from St.-Michel de l'Atalaye to Dessalines, near Paul. J. S. Brown, collector. March 25, 1921. 9885 (B 332 F).. Arrondissement of Dessalines, trail from St.-Michel de l' Atalaye to D essa lines, float at entrance to gap in mountains about 3 kilometers west of Paul. J. S. Brown, collector. March 25, 1921. 9886 (B 333 F). Arrondissement of Dessalines, about halfway down mountain slope between Paul and Dessalines, probably 7 kilometers northeast of Dessalines, altitude 390 meters a.hove sea level. J. S. Brown, collector. March 25, 1921. 9887 (B 334 F.) of Dessalines, float a. short distance southwest of 9886, altitude 300 meters above sea. level. F. G. Evans, jr., collector. March 25, 1921. 9789 (W 220 F). Arrondissement of Mirebalais. trail from Thomonde to Mirebalais, about 0.5 kilometer south\\est of crest of mountains at Bois-Joli, altitude 660 meters above sea leve l. W. P. Woodring, collector. January 18, 1921. 9919 (W 221 F). Arrondissement of Mire balais, trail from Thomonde to Mirebalais, float about 3 kilometers southwest of crest of mountains at Bois-Joli, altitude 535 meters above sea. level. W. P. Woodring, coll e ctor. January 18, 1921

PAGE 168

SEDIMENT.ARY ROOKS. 143 9456 (W 133 F). Arrondissement of Mirebalais, float on southwest s lope of Morne Mich el, about 8 kilometers northeast of Mirebalais, altitude 300 meters above sea level. W. P. Woodring, collector. December 6, 1920. 9903 (W 184 F). Arrondissement of Mirebalais, float from gap in mountains along road from Mirebalais to Las Caho bas. W P. Woodring, collector. January 8, 1921. 9915 (W 202 F). Arrondissement of Mirebalais, trail from Belladere to Sava nette, float on south slope of mountains about 4 kilometers northeast of Savanette, altitude 825 meters above sea level. W. P. Woodring, collector. January 13, 1921. 9914 (W 201 F). Arrondissement of Mirebalais, trail from Belladere to Sava nette, float on south slope of mountains about 4.5 kilometers northeast of Sava nette, altitude 1,100 meters above sea level. W. P. Woodring, collector. January 13, 1921. 9548 (B 119 F). Arrondissement of St.-Marc, railroad cut about 4 kilometers southeast of Mont Rouis. J. S. Brown, collector. December 7, 1920. 9993 (K 56 F). Arrondissement of St.-Marc, trail from l'Arcahaie to Marche Desarm es, 5 kilometers northeast of Couyau halfway down ridge, altitude 1,230 meters above sea level. W. S. Burbank, collector. December 15, 1920. 9502 (K 58 F). Arrondissement of St.-Marc, trail from l'Arcahaie to Marche Desarm es, northeast of Couyau, on crest of ridge overlooking Artibonite Valley. W. S. Burbank, collector. December 15, 1920. 9487 (B 130 F). Arrondissement of St.-Marc, trail from I' Arcahaie to Marche Desarmes, float on slope of Mome Archeveque, about 11 ki l ometers southwest of Marche Desarmes. J. S. Brown, collector. December 16, 1920. 9486 (B 129 F). Arrondissement of Port-au-Prince, trail from l' Arcahaie to March e Desarmes, 1 kilometer southwest of habitation Caille-Mare. J. S. Brown, collector. December 14, 1920. 9459 (W 144 F). Arrondissement of Port-au-Prince, trail from Fond-des Orang ers to the Cul-de-Sac Plain, float on north slope of ridge south of Fond-des Orangers, altitude 610 meters above sea level. W. P. Woodring, collector. D ecember 8, 1920. 9790 (W 313 F). Arrondissement of Port-au-Prince, road from Port-au-Prince to Mirebalais, float on north slope of Morne Terre-Rouge, altitude 590 meters above sea level. W. P. Woodring, collector. February 28, 1921. 9608 (W 79 F). Arrondissement of Port-au-Prince, north slope of ridge between Riviere Froide and Riviere Momance, along trai l leading southward from Carre four, altitude 360 meters above sea level. W. P. Woodring, collector, November 5, 1920. 9553 (W 11 F). Arrondissement of Port-au-Prince, float on crest of Morne Hopital, east of triangulation station, altitude 950 meters above sea W. P. Woodring and J. S. Brown, collectors. October 9, 1920. 9443 (W 9 F). Arrondissement of Port-au-Prince, north slope of Morne Hopi tal, altitude 760 meters above sea level. W. P. Woodring and J. S. Brown, col lectors. October 9, 1920. 9584 (W 13 F). Arrondissement of Port-au-Prince, trail from Petionville to Furey about 2 kilometers southwest of Petionville. W. P. Woodring, and F. G. Evans, jr., collectors. October 12, 1920. 9583 (W 14 F). Arrondissement of Port-au-Prince, trail from Petionville to Furey, about 3.5 kilometers southwest of Petionville. W. P. Woodring, collector. October 12, 1920. 9585 (W 15 F). Arrondissement of Port-au-Prince, trail from Petionville to Furey, about 2 kilometers south of Rendez-vous. W. P. Woodring and F. G. Evans, Jr., collectors. October 12, 1920.

PAGE 169

144 GEOLOGY OF THE REPUBLIC OF HAITI. 9587 (W 25 F). Arrondissement of Port-au-Prince trail from Petionville to Furey, upper part of village of K e n s koff, altitude 1,51 5 meters above sea l evel W. P. Woodring, and F. G. Eva ns, jr., coll ec tors. O ctober 14, 1920. 9505 (W 33 F). Arrondiss e m ent of Port-a u-Prince, trail from Fond-Parisien to Fond-Verrettes, about 3 k i l o m e t e r s sout h-south east of FondParisi e n, altitude 275 m e t e rs above s e a l e v el. W. P. Woodring, coll e ctor. October 21, 1920. 9594 (W 37 F). Arrondi ssem ent of Saltrou, trail from Fond-Verrettes to Grand Gosi e r, south slop e of mountains n ear crest, altitude 1,570 mete rs above sea l e vel. W. P. Woodring, coll e ctor. Oc t ob e r 22, 1920. 9596 (W 39 F). Arrondiss e m ent of S a ltrou, trail from Fond-Verrette s to Grand Gosier, south slop e of mountains about 13 kilometers northeas t of Grand-Gosier, altitude 940 mete rs above s e a l e v e l. W. P. Woodring, collector. Octob e r 23, 1 9 20. 9510 (W 50 F). Arrondisse m ent of Sal t rou trail from Grand-Gosier to Saltrou, 6 kilome t e rs northwest of Grand-Gos i e r, altitude 210 meters above s e a leve l W. P. Woodring, collector. Octob e r 28, 1920. 9511 (W 51 F). Arrondi ssem ent of S a l t rou, trail from Saltrou to Marigot, 4.5 kilom e t e rs w es t of S a l t r o u a l titude 50 m e t e rs above s e a l e v e l. W. P. Woodri ng, coll e ctor. October 30, 1920. 9444 (W 52 F). Arrondi ssement of Saltrou, trail from Saltrou to Marigot, about 7 .5 kilome t e rs west of S a ltrou, altitude 45 m eters above sea level. W. P. Woodri ng, collector. October 30, 1920. 9513 (W 54 F). of S a ltrou, trail from Saltrou to Marigot, about 12.5 kilome t e rs w est of S a ltrou, altitude 350 m eters above sea lev e l. W. P. Wood ring coll e ctor. Octob e r 30, 1920. 9600 (W 59 F). Arrondi ssem ent of J a cm e !, trail from Marigot to J acmel, floa t 6 kilom e t e rs west of Marigot. W. P. Woodring, coll e ctor. Octob e r 31, 1920. 9605 (W 69 F). of J a cm e l, right bank of Riviere Gos s e line a t Pas s e la Voute fourth cro s sing abov e Jac m e l. W. P. Woodring and F. G. Evans, jr., coll e c t ors. Novembe r 3, 1920. 9649 (K 28 F). of J a cm e l, trail from B ainet to Jacmel, about halfw a y u p south slop e of Marne Fontai. W. S. Burbank, coll e ctor. November 14 1920. 9517 (W 81 F). Arrondissem ent of Nippe s, road from Miragoane to Petit-Goave abou t 0 5 kilometer eas t of Miragoa n e altitude 45 meters above sea level. W. P. Woodring collector. Novembe r 14 1920. 9518 (W 88 F). Arrondissem ent of Nippe s, road from Miragoane to Aquin, floa t about a kilomete r south of Cha p e ll e d e Virgile, altitude 275 meters above se a level. W. P. Woodring, collector. November 15, 1920. 9547 (B 114 F). Arrondissem ent of Nippe s, float on trail from Baraderes t o Petit-Trou de Nippes, about halfway b etwe en the towns. J. S. Brown, collect or. November 22 1920. 9509 (B 56 F). Arrondissement of Cayes, bluff a kilometer west of Camp Perrin. J. S. Brown, collector. Novembe r 5, 1920. 9631 (B 82 F). Arrondi s s ement of Tiburon, trail from Anse d'Hainault to Dame-Marie, float on small hill just north of Bariadele. J. S. B1own, collecto r. November 15, 1920. 9632 (B 83 F). Arrondiss e m ent of Tiburon, float on trail from Anse d' Hainault to Dame-Marie, about 2 kilometers north of 9631. J. S. Brown, collector. November 15, 1920. 9629 (B 84 F). Arrondis sement of Tiburon, trail from Dame-Marie to Source s Chaudes, about 6 kilome t e rs east of Dame-Marie. J. S. Brown, collector. Novem ber 16, 1920. 9682 (W 177 F). Gonave Island, about 200 m e t e rs S. 60 E. of triangulation station at Fond-Plaisir in southeastern part of island. W. P. Woodring, co l lect or. December 26, 19'20.

PAGE 171

SEDIMENTARY ROCKS. 145 9669 (W 163 F). Gonave Island, trail from Anse-a-Galets to Picmi, about 2 .kilometers north of Picmi, altitude 180 meters above sea level. W. P. Woodring, collector. December 19, 1920. 9670 (W 164 F). Gonave Island, trail from Anse-A-Galets to Picmi, float about 1.5 kilometers north of Picmi, altitude 70 meters above sea level. W. P. Woodring, collector. December 19, 1920. 9668 (W 162 F). Gonave Island, trail from Anse-a-Galets to Picmi, bottom of ravine at spring north of Picmi. W. P. Woodring, collector. December 19, 1920. 9665 (W 158 F). Gonave Island, trail leading southwestward from Anse-8. Galets, about 4 kilometers from Anse-a-Galets, altitude 310 meters above sea level. W. P. Woodring, collector. December 18, 1920. 9681 (W 176 F). Gonave Island, trail from Anse-a-Galets to Fond-I' Aurore, about 6 kilometers southeast of Anse-a-Galets. Altitude 380 meters above sea level. W. P. Woodring, collector. December 26, 1920. 9671 (W 165 F). Gonave Island, trail from Anse-a-Galets to Plaine des Mapoux, about 2 kilometers west-northwest of triangulation station on Morne Chien Con tent. W. P. Woodring, collector. December 20, 1920. 9672 (W 166 F). Gonave Island, trail from Anse-a-Galets to Plaine des Mapoux, about a kilometer east of east end of Plaine des Mapoux, altitude 525 meters above sea level. W. P. Woodring, collector. December 20, 1920. 9673 (W 167 F). Gonave Island, northwest corner of Plaine des Mapoux, altitude 540 meters above sea level. W. P. Woodring, collector. December 21, 1920. 9674 (W 168 F). Gonave Island, north slope of Morne la Pierre, about half a kilometer northeast of triangulation station, altitude 700 meters above sea level. W. P. Woodring, collector. December 21, 1920. 0LIGOCENB SERIES. Oligocene deposits in narrow bands usually border upper Eocene lime stone along the lower slopes of the mountain ranges. These Oligocene bands consist principally of limestone that can scarcely be distinguished from the upper Eocene limestone except by the fossils it contains. LOWER OLIGOCENE. None of the deposits seen during the reconnaissance are considered of undoubted lower Oligocene (Lattor:fian) age, as none of them contain Lepidocyclinas of the type of L. mantelli (Morton), a species which, in the southeastern United States, is confined to the lower part of the Vicks burg group (Marianna limestone) Some of the limestone in the Cha.lne des Mateux, on the north slope of Morne Hopital (pp. 130-131), east and south of Dondon (pp. 111-112) and elsewhere, described as upper Eocene, may be lower Oligocene. These deposits are unfossiliferous or contain fossils that have no stratigraphic value. Deposits of lower Oligocene age might be expected in such regions as the Cha1ne des Mate11x, where the series from upper Eocene to Miocene seems to be conformable. 10

PAGE 172

146 GEOLOGY OF THE REPUBLIC OF HAITI. MIDDLE OLIGOCENE. Middle Oligocene deposits were seen principally in isolated exposures or as :float, and at most places their stratigraphic and structural relations are obscure. As their lithology differs at different places no general de scription of them can be given. DESCRIPTION BY REGIONS. Kassif du Nord. A typical middle Oligocene fa11na was obtained from a float boulder of yellowish white lime stone in a ravine a few h11ndred meters northwest of the town of Bahon (see list, p. 150, station 9884). The country rock at this plac e is quartz diorite, and t he fl.oat in the ravine consists principally of quartz diorite and of slaty argillite of supposed lower or middle Cre taceous age, but limestone boulders, which have probably been transported from an l1ne xplored lo cality some kilometers to the southwest, are fairly common. Rocks of known or supposed middle Oligocene age are exposed in a large area along the northern border of the Central Plain near St.-Raphael. They were examined along the trail from St.-Raphael to Dondon for a distance of about 3 kilometers, where they overlie limestone of supposed upper Eoc e ne age (s e e pp. 111-112) and dip southward or southeastward beneath the yo11nger rocks of the plain. The lowest beds, which are exposed farthest north, consist mainly of brown or gray marl and fine-grained sandstone in very even, rather soft beds, from 2 to 5 centimeters thick, which occupy a small lowland along the narrow valley of Riviere Bouyaha. Lithologically they are similar to some Miocene rocks. A few beds of coarser material containing small pebbles of white limestone, presumably upper Eocene, suggest an erosional 11nconformity. Inte rbedded with the sand and marl are thic k e r beds of brown crystalline limestone containing the Foraminifera lis t e d on page 150, stations 9877 and 9878. Overlying these elastic b e ds to the south, at the border of the plain, is pure limestone, white and partly crystalline, thick-bedded or massive, and hard, forming steep walls along the gorge of Riviere Bonyaha. The Foraminifera listed on p. 150, station 9876, were collected from this limestone. As the dips along the trail appear to be invariably to the south or south east, and as the width of outcrop is at least 3 kilometers, the total thick ness of middle Oligocene at this locality is several hundred meters. Massive limestone 11nderlies the Pie de Pignon, a conspicuous conical hill north of Pignon. The middle Oligocene Foraminifera listed on page 150, station 9940, were collected from loose pieces of limestone at the foot of the hill. This limestone, which is apparently the same as the massive

PAGE 173

SEDIMENT.ARY ROCKS 147 limestone along Riviere Bouyaha below St.-Raphael, forms the ridge ex tending southea stward from Pignon.1 Northwest Peninsula. Trois Riv ieres Valley. Rocks of middle Oligocene age are apparently wid e ly distributed in the Trois Rivieres Valley north of Gros-Morne. All the rocks in this troug h are fold e d in narrow anticlines and synclines which are parallel to the length of the trough. The relations of the middle Oligo c ene to the upper Oligocene south of Gros-Morne and to the Miocene in the low e r part of the vall e y were not deterrnined. The middle Oligocene rocks were seen on the east side of Les Trois Rivieres along an l1nfre quented trail from Gros-Morne to Port-de-Paix. The corals listed on pag e 150, station 9756, wer e collect e d from pieces of hard gray limestone w eathe red from an outcropping ledge about 9 kilometers north of Gros Morne. Float containing similar corals was collected a short distance farthe r north (see list, p. 150, station 9757). Beds of marl, shale, sandstone, and a conglom erate consisting entirely of large cobbles of basaltic rock crop out along the trail north and south of this locality. These beds w e re seen only in isolated exposures, and as their folding is complex the question whether all the rocks involved are of middle Oligocene age could not b e determined. M on. tagnes de Jean Rabel. A seri e s of brown or gray marls containing many thic k bed s of dense yellowish limestone is exposed along the trail f1om J ean Rabel to Anse Rouge just south of the crest of the Montagnes de Jean Rabel. These beds overlie upper Eocene limestone and probably overlap it in the area to the north, wher e they rest on quartz diorite. To the south the y dip b eneath marly b e ds of supposed Mio c ene age underly ing the Arbre Plain. An Ope rculina, apparently a common undescribed middle Oli g o c en e spe c ies, was collected from one of the beds of limestone. (See list, p. 150, station 9962). The boundary between Oligocene and Miocene in this lo c ality was not determined. M ontagn e s de Terre-N euve. About 5 kilometers southeast of TcrreN euve, at a place known as Hilaire, just east of the mineral prospects known as Casseus and Germinie, the mountain side is covered with large float blo c ks of limestone, gray on weathered surfaces and white in the inte rior, similar to the prevailing upper Eocene limestone. Some of the boulders contain numerous spe c imens of Ope rculina, apparently of middle Oligocen e age, etch e d out on weath ered surfaces. (Se e list, p. 150, stations 9810 and 9812.) The collection listed at page 144, station 9813, probably of upper Eoc ene age, was obtained in the same locality. The middle Oli gocene float probably r e presents rock in place somewhere on the high mountain ridge to the east, at the locality just north of Habitation Dumuraille. 1 On the map ot the Central Plain, Pl. XXXVI, the limestone near Plinon and St. Raphael is lnco1r ectly shown as upper Oligocene.

PAGE 174

148 GEOLOGY OF THE REPUBLIC OF HAITI :Montagnes Noires .. Thin-bedded chalky limestone crops out on the northeast slope of the Montagnes Noires, in the section called Bois-Joli, along the trail from l\firebalais through Dufailly to Thomonde. The beds strike N. 40" W. and dip 20 NE. Pieces of chert lying on the surface of the limestone contain numerous specimens of the middle Oligocene Foraminifera listed on page 150, station 9918. The chert probably came from bands or nodules in the chalky limestone, although none were seen in the exposure. Litho logically this limestone is indistinguishable from thin-bedded chalky upper Eocene limestone containing bands of chert. Chatne des Mateux. Some of the limestone on the north flank of the Cha.lne des Mateux is of middle Oligocene age. Thin-bedded white brittle limestone containing chert nodules crops out at the foot of the mountains south of Savane Madame Michel, on the trail from Saut d'Eau to Fond-des-Orangers. A piece of chert lying loose on the trail contained the middle Oligocene Foraminifera listed on page 150, station 9658. The nephelite basalt described on pages 314-318 overlies this limestone. Farther northwest, on both flanks of the Cha!ne des Mateux, limestone of supposed Oligocene age crops out. It is described on page 153. Kontagnes du Trou d'Eau Middle Oligocene limestone probably extends eastward from the Chaine des Mateux, as a Lepidocyclina similar to L. canellei yurnagunensis Cushman was obtained from float on the south slope of the Montagnes du Trou d'Eau along the road from Port-au-Prince to Mirebalais (see list, p. 150, station 9900). Near Jacmel. Granular white limestone containing nodules of chert is exposed in tl1e northernmost sea cliff on the west side of J acmel Bay, just south of the mouth of a small stream. The bedding in this exposure is not apparent, but the rock is probably crumpled, as pieces from almost any part of the exposure are slickensided. A network of thin seams in the rock contains dark-brown clay. This limestone contains the largest foraminifera l fauna discovered in any of the Tertiary rocks of Haiti. In addition to the species listed on page 150, station 9601, the limestone contains some small undetermined Foraminifera. This is one of the few localities where it is possi ble to obtain entire tests from the rock. The tests stand out in relief on weathered surfaces and can easily be dug out, as the rock is soft. This is the only locality on the south coast where fossils of middle Oligocene age were collected.

PAGE 175

SEDIMENT ARY ROCKS. 149 FOSSILS. Foraminifera are the most abundant middle Oligocene fossils. They include species of Lepidocyclina described by Cushman from Antigua and Cuba and an unde.scribed species of Operculina found in Antigua and other regions. Foraminifera are particularly abundant at station 9601, on the west side of J acme l Bay, where hand specimens containing h11ndreds of tests of Lepidocyclina can be obtained. The few corals collected are similar to species from Antigua. The same fauna has been found at many localities in the Dominican Republic, in other parts of the West Indies, and in Georgia. The Antigua formation of the island of Antigua, which contains an unusually rich coral fauna, is the type formation of this horizon. On account of its stratigraphic relations in Georgia and of its cora ls it is correlated with the Rupelian of Italy.1 Stations Middle Oligocene. 9884 (B 298 F). Arrondissement of Grande-Riviere du Nord, float in ravine about a kilometer west of Bahon. J. S. Brown, collector March 10, 1921. 9876 (B 290 F). Arrondissement of Grande-Riviere du Nord, trail from St.-Raphael to Dondon, about a kilometer north of St.-Raphael. J. S. Brown, March 7 1921. 9877 (B 291 F). Arrondis.5ement of Grande-Riviere du Nord, trail from St.-Raphael to Dondon, about 3 kilometers north of St.-Raphael. J. S. Brown, collector. March 7, 1921. 9878 (B 292 F). Arrondissement of Grande-Riviere du Nord, trail from St. Raphael to Dondon, about 200 meters north of 9877. J. S. Brown, collector. March 9, 1921. 9940 (W 264 F). Arrondissement of Grande-Riviere du Nord, float at foot of Pie de Pignon, 0.5 kilometer east of Pignon, altitude 365 meters above sea level. W. P. Woodring, collector. February 71 1921. 9756 (W 291 F). Arrondis.5ement of Port-de-Paix, trail from Gros-Mome to Port-de-Paix on right side of Les Trois Rivieres, about 2 kilometers north of crossing of Riviere l'Aqui. W. P. Woodring, collector. February 17, 1921. 9757 (W 292 F). Arrondissement of Port-de-Paix, trail from Gros-Morne to Port-de-Paix on right side of Les Trois Rivieres, float about 4 kilometers north of crossing of Riviere l'Aqui. W. P. Woodring, collector. February 17, 1921. 9962 (K 125 F). Arrondissement of Mole St.-Nicolas, trail from Jean Rabel to Anse Rouge, south slope of mot1ntains in valley of Riviere de Port-a-Piment, about 20 kilometers south of Jean Rabel. W. S. Burbank, collector. February 3, 1921. 9810 (B 160 F). Arrondissement of Gona!ves, float at Germinie, in Terre-N euve region, south of Casseus prospect. J. S. Brown, collector. January 12, 1921. 9812 (B 162 F). Arrondissement of Gonaives, float south of Germinie in Terre N euve region. J. S. Brown, collector. January 12, 1921. 9918 (W 219 F). Arrondissement of Las Cahobas, trail from Thomonde to Mirebalais, float on northeast slope of mountains at Bois-tToli, near crest, altitude 610 meters above sea level. W. P. Woodring, collector. January 17, 1921. 1 Vaughan, T. W., Fossil corals from Central America, Cuba., and Porto Rico, with an account ot the American Tertiary, Pleistocene, and Recent coral reefs : U. S. National Museum Bull. 103, pp. 199 and 202, 1919.

PAGE 176

150 GEOLOGY OF THE R EPUBLIC OF HAITI. 9658 (W 140 F). Arrondissement of Mire b a lais, trail from Saut d 'Eau (Ville Bonheur) to Fond-des-Orang e rs, float at f oat of mountains about 6 kilometers eouthwest of Saut d 'Eau, altitude 500 mete rs above eea l e v e l. W. P. Woodring and F. G. Evans, jr., coll e ctor8. Decembe r 8 1920. 9900 (W 179 F). Arrondi ss e m ent of Port-au-Princ e, road from Port-au-Prince to Mire balais, float on south s lop e of m o untains a l t i t ude 360 meter s a bove sea level. W. P. Woodring coll e ctor. J anuary 7, 1921. 9601 (W 60 F). Arrondi ss em ent of J acmel, northe rnmost s e a cliff on we s t side of Jacmel Bay. W. P. Woodring, coll e ctor. Novembe r 1, 1920. Middle Oligoc ene fossils. Statio ns. = 0 ... ca +a QJ 0 = -::s 0 Nor t bweat z "'O et Massif du Nord. m CD Penins ula. co QJ "O C1> C> i::: Q "'O bl Q) be ::s -... Species. !1 Q cd t1S .... .... r= t'IS 0 t3 0 "'O )1 )1 d.i 41 0 0 r= t'IS ...... z .... Gra n de -Riv i ere ..c llQ "4 I 0 .... co du Nord. 41 Q) .c -I 00 CIS ::s "O > cd ""' 0 .c t'IS Q) clS I a Q) Q) ... Q ... ... 0 <> 0 .... 0 <:) 0 at F oraminifera: Oper culina sp. c f. unde sc r i bed spec i es fro m An tigua x x x x x Heteroste gina antillea Cu s hman ? ..... x Heterostegina sp. x L epidocyclina canell ei yurnagun e nsi s Cushman x ? ? ? Lepid oc y clina g1gas Cu s hman ............. x Le pidoc y clin a undosa Ct.Ish man ........... x x ? x Lepidocy clina u n dul ata Cushma n ........ x x x Lepido c y c l ina fav o s a Cu shman ....... x Lepido c y clina sp. c f L. praemarginata R. D o uvill e x x x Lepidoc y clina sp. c f. L. sumatren s ie (Bra dy ) x Lepidocy clina s p. cf. L. morgan1 Lem oi n e and R. Douvill e .................. x x L e pidocyclina sp. ? x Corals: Orbicella sp. x Antiguastre a cellulosa (Duncan) ? ...... x Favites po]ygonalis (Dunc .an) ? ..... x Cyatbo m orpha sp. cf. c. belli V augha n .. x Oyathomorpba n. IP ...................... x Mollu sca: Pelecypoda : Phacoidea (Parvilucina) RP I x

PAGE 177

SEDIMENTARY ROOKS. 151 UPPER OLIGOCENE. Rocks of upper Oligocene age seem to be confined to the northern and central parts of the Republic. They consist principally of limestone, gen erally massive, resembling the massive limestone of Eocene and middle Oligocene age. The upper Oligocene deposits probably are conforinable to the middle Oligocene, although the actual contact was not seen. The lithology and thickness differ in the regions where rocks of this age were examined, as is shown below. DESCRIPTION BY REGIONS. Tortue Island, The limestone that covers the interior of Tortue Island is at least 100 meters thick. It is massive, and most of the weathered outcrops on the surface of the plateau are stained red by recemented residual clay. It rests unconformably on a basement of schistose limestone. The structure of the limestone was not determined, but the surface features indicate that it is arched in a broad anticline. The only fossils obtained are the Foraminifera and mollusks listed on page 156 (stations 9761 and 9762). The evidence with regard to age furnished by these fossils is not con clusive, and the limestone may be of Miocene age. The relation of this limestone to the water supply of the island is dis cussed on page 541. Arrondissement of Borgne. Massive white or grayish limestone crops out in the coastal ridge east of Anse-a-Foleur, in the western part of the Arrondissement of Borgne. The Foraminifera and corals listed on page 156 (stations 9766, 9767, and 9768) were obtained from this limestone. Its stratigraphic and structural relations to the upper Eocene limestone in the same ridge farther east are not known. Troia Valley. The surface rocks in virtually all the southern part of the trough known as the Trois Rivieres Valley, which extends northward from Gonaives to Port-de-Paix, are of upper Oligocene age. These deposits consist of lime stone, gray or brown marl, :fine-grained sandstone, and shale. Brownish y e llow limestone is interbedded with the marl. The limestone, which is in thick or thin beds, is harder than the marl and crops out in conspicuous ledges. Partly crystalline massive limestone and sandy limestone in thin"' ner beds are exposed in low, bare hills along the west side of the road from Gona.lves to Gros-Morne south of the divide. The Forarninifera listed on page 156 (station 9751) were collected from the massive }jme stone. The marl is well expoged in cuts along the same road just south of the divide. It has a conchoidal fracture and closely resembl e s Miocene marl. A bed of limestone interbedded with the marl at a locality about a

PAGE 178

152 GEOLOGY OF THE REPUBLIC OF HAITI. kilometer north of the divide contained the Foraminifera listed on page 156 (station 9943). Along the east and west borders of the trough there are a n11mber of exposures of a white thin-bedded unfossiliferous limestone that commonly is sheeted transverse to the bedding. It seems to be in fault contact with upper Eocene limestone or older volcanic rocks. As the trough has prob ably been formed by normal faulting (see p. 121), this limestone appears to be younger than Eocene. It is tentatively referred to the upper Oligocene. The upper Oligocene rocks, like other rocks in the trough, are folded in narrow anticlines and synclines, and the dip and strike change within short distances. Borders of the Central Plain. Limestone of upper Oligocene age crops out in the lower slopes of the mo11ntains bordering the southeastern half of the Central Plain. At the northwestern extremity of the plain this limestone seems to be concealed by flood-plain deposits, which are not shown on the map (Pl. I). The limestone in the mountains bordering the plain from St.-Raphael southeastward an unknown distance beyond Pignon is middle Oligocene. At Bassin Zinn, northeast of Hinche, where Riviere Samana cascades down the dip slope of the limestone into the plain, the limestone is of upper Oligocene age. The surface of the limestone along the stream is covered with travertine. A piece of limestone at the foot of the cascade contained the Foraminifera listed on page 156 (station 9936). At this locality the limestone has the same strike and dip as the overlying Miocene conglomerate. The same limestone was examined along the trail from Thomassique to Cerca-la-Source, where it crops out in the ridge bordering the plain. 'l,he Miocene beds in the plain and the upper Oligocene limestone in the ridge dip about 20 SW. The southwest slope of the ridge, which faces the plain, is a dip slope. The northeast slope, which over looks the valley of Riviere !'Ocean, is a steep scarp, apparently a fault scarp The limestone appears massive in outcropping ledges. The coral listed on page 156 (station 9949) was collected from a loose piece of limestone on the northeast slope. Massive limestone of supposed upper Oligocene age flanks upper Eocene limestone on the north slope of the Montagnes Noires, along the south side of the plain. It also crops out on the south side of the mountains facing the valley of Rivi ere Fer-a-Cheva l. On the southwest side of the plain the upper Oligocene limestone crops out in the lower slopes of the Montagnes Noires. It forms the conspicuous hill northwest of Thomonde, on the crest of the Thomonde anticline. The corals listed on page 156 (stations 9934 and 9741) were collected at the foot of this hill. Massive gray limestone of upper Oligocene age was seen on the northeast slope of the mountains on the trail from Thomonde to

PAGE 179

SEDIMENTARY ROCKS 153 Mirebalais through Bois -Joli. The fossils listed on page 15 6 (stations 9786, 9787, 9917, and 9788) were collected from float derived from this limestone Low, isolated hills, which rise above the cover of flood-plain deposits in the northwestern extremity of the plain, between St.-Michel de l' Atala ye and Pignon, are composed of limestone, probably of upper Oligo cene age, although it may be middle Oligocene. An indeterminate species of Goniopora was collected from this lim estone (station 98 7 5). Chatne des Mateux. Along the trail from l' Arcahaie through Couyau to Marche Desarmes a band of limestone of supposed Oligocene age, from 2 to 4 kilometers in width, flanks upper Eocene limestone on the limbs of the anticlinal arch of the Chaille des Mateux. The band of Oligocene rocks is narrower on the northeast side than on the southwest side because of the steeper dips The structural relations are shown in Figure 5 (p. 128). On the south west slope of the mountains the Oligocene rocks are thrust southwestward over Miocene rocks. No fossils were found in the beds of supposed Oli gocene age and their age is not definitely known The supposed Oligocene limestone rests without apparent discordance on upper Eocene limestone. It is white and p artly chalky and contains no chert In many exposures it is sheeted transverse to the beddings, appar ently as a result of fractures produced in folding The sheeting may ob scure the bedding and at places resembles bedding. Most of the sheeting p lanes a re only 2 to 5 centimeters apart and the bedding planes are far ther apart. The thickness of this limestone, as can be seen from Figure 5, is at lea .st several hundred meters if not more than 1,000 meters. Montagnes du Trou d'Eau. The surface rock in virtually the entire eastern part of the l\iontagnes du Trou d'Eau east of Morne Trou d'Eau, is limestone of upper Oligocene age The thickness of this limestone is probably several hundred meters Its structural relations are obscure because at almost all the localities examined it appears massive. The limestone is gray or yellowish on weathered surfaces and white on unw eathered surfaces. Weathered surfaces in most places are deeply pitted, and at some places the rock forms a solution breccia. The lime stone rests on the nephelite basalt described on page 315. A conglomerate at the base of the limestone contains large cobbles of basalt and smaller pebbles of limestone, presumably of upper Eocene age. This conglomerate is well exposed on the trail from Thomazeau to Cornillon, about 1.5 kilo meters from Thomazeau. Numerous collections of fossils were obtained from this limestone along the north shore of Etang Sa11matre and along the trails from Thomazeau

PAGE 180

154 GEOLOGY OF THE REPUBLIC OF HAITI. to Cornillon, from Cornillon to St.-Pierre, and from Cornillon to Marche Canard. (See list, p. 156, stations 9898, 9521, 9657, 9556, 9557, 9558, 9451, 9449, 9450, 9559, 9452, 9453, 9560, 9454, and 9561.) Corals are particularly abundant, and at some places for example, on the crest of the :first ridge northwest of Cornillon on the trail to Marche Canard, where large heads of Orbicella and branches of Stylophora and Porites are strewn along the trail the lim estone is a reef rock. FOSSILS. Foraminifera are less abundant in the upper Oligocene limestones than in the older Tertiary rocks. Sorites americana (Cushman), described from the Emperador limestone of Panama, and Miogypsina antillea (Cushman), described :from the Anguilla formation of the island of Anguilla, are the most common species. The deposits south of Gros Morne contain Lepidocyclina g iraudi) described by R. Douville from beds on the island of Martinique that are considered to be of Aquitanian age by Giraud and of Burdigalian age by Douville.1 This species duplicates in a remarkable fashion the surface sculpture o:f the Cretaceous species Orbir toides media ( d' Archiac). At some localities, particularly in the Montagnes du Trou d'Eau, corals are ab11ndant in the upper Oligocene deposits. Some of the species, such as Orbicella imperatoris Vaughan, Orbicella canalis Vaughan, and Sider astrea silecen.sis Vaughan, have been described from upper Oligocene rocks in adjacent regions. At least two species, Stephanocoenia sp. cf. S. intersepta (Esper) and Goniopora jarobiana Vaughan, are of Miocene aspect. The upper Oligocene deposits of the Haitian Republic seem to be of the same age as the upper part of the Culebra formation and the Emperador limestone of Panama and the Anguilla formation o:f Anguilla. This hori zon probably is the equivalent of the Aquitanian, which is called Miocene by most European geologists. Stations Upper Oligoc e ne. 9761 (W 298 F). Tortue Island, trail from La Vallee to Pointe des Oiseaux, about 8 kilom eters east of La Vall ee, altitude 265 meters above sea level. W. P. Woodring, colle cto r. February 20, 1921. 9762 (W 299 F). Tortue Island, trail from La Vallee to Pointe des Oiseaux, steep slope l eadi ng down to coast about 3 kilometers northwest of Pointe d es Oiseaux, altitude 235 meters above sea level. W. P. Woodring, collector. February 20, 1921. 9766 (W 302 F). Arrondissement of Borgne, trail from Anse-a-Fol eur to Le Borgne, float in small stream about 2 kilometers east of Anse-8.-Fol e ur. W. P. Woodring, collector. February 22, 1921. i R., Sur des nouvelles : Soc. France Bull., 4th ser., vol. 7 pp. 307-311, pl. 10, figs. 9, 10, 15, 16, text figs. 1, 2, 1907

PAGE 181

SEDIMENTARY ROOKS. 155 9767 (W 303 F). Arrondissement of Borgne, trail from Anse-a-Foleur to Le B orgne, first ridge along coast on east side of An se -a-Fol e ur. W. P. Woodring, c ollector. F e bruary 22, 1921. 9768 (W 304 F). ArrondiSBem ent of Borgne, float in mountains about 7 kilo m eters west-northw est of L e Borgne, altitude 195 m e ters above sea l e v e l. W. P. Woodring, collector. F ebruary 22, 1921. 9751 (W 276 F). Arrondi ssement of Gona.lves, road from Gonaives to Gros M orne, low hill on northwest side of road about 4 kilomet e rs from Gonaiv e s. W. P Woodring and J. S. Brown, collectors. February 15, 1921. 9943 (W 280 F). Arrondissement of Gana.Ives, road from Gona1ves to Gros M ome, about 7 kilometers south of Gros-Morne. W. P. Woodring and J. S. Brown, c ollectors. F ebruary 15, 1921. 9936 (W 244 F). Arrondissement of Hinche, float at foot of cascade on Riviere S amana at Bassin Zinn. W. P. Woodring, collec t or. January 31, 1921. 9949 (W 324 F). of Valli e re, trail from Thomassique to Cerca laSource, float on north slope of ridge overlooking valley of Riviere I' Ocean, al titude 435 meters above sea level. W. P. Woodring, collector. March 13, 1921. 9741 (W 240 F). Arrondissem ent of Las Cahobas, crest of Thomonde anticline, f oot of Mame Marnont, about 4 kilomet e rs west-northwest of Thomonde, altitude 360 meters above sea level. W. P. Woodring, collector. January 29, 1921. 9934 (W 241 F). Arrondissement of Las Cahobas, float at same locality as 9741. W. P. Woodring coll e ctor. January 29, 1921. 9786 (W 215 F). Arrondissem ent of Las C a hobas, trail from Thomonde to Mirebalais, float about a kilometer southw est of cem etery at Caille-Pain, altitude 480 meters above s e a l e vel. W. P. Woodring, collector. January 17, 1921. 9787 (W 216 F). Arrondiss ement of Las Cahobas, trail from Thomonde to Mirebalais, about 1.5 kilomet e rs southwest of cemetery at Caille-Pain, altitude 4&5 mete rs above sea l e v el. W. P. Woodring collector. January 17, 1921. 9917 (W 217 F). Arrondissem ent of La.s Cahobas, trail from Thomonde to Mire balais, float about 3.5 kilom e t e rs southwest of cemetery at Caille-Pain, altitude 570 meters above sea level. W. P. Woodring, coll ecto1 .. J anuary 17, 1921. 9788 (W 218 F). Arrondi s s ement of Las Cahobas, trail from Thomonde to M irebalais, float on northeast slope of mountains at Bois-Joli, n ear crest, altitude 610 meters above s e a lev el; same locality as 9918. W. P. Woodring, collector. J anuary 17, 1921. 9875 (B 289 F). Arrondissement of Grande-Riviere du Nord, float on road f rom St.-Michel de l Atalaye to St.-R a phael, about 8 kilom e ters southwest of St.-Raphael. J. S Brown, coll e ctor. March 6, 1921. 9898 (B 350 F). Arrondiss ement of Port-au-Prince, float 50 meters west of Source Maneville. J. S. Brown, coll e ctor. March 30, 1921. 9657 (W 105 F). Arrondissem ent of Port-au-Prince, float on small island along north shore of Etang Saumatre, about 6 kilometers southeast of Maneville. W. P. Woodring, collector. November 23, 1920. 9521 (W 104 F). of Port-au-Prince, float short distance east of 9657. W. P. Woodring, coll e ctor. November 23, 1921. 9556 (W 114 F). Arrondissement of Port-au-Prince trail from Thomazeau to Cornillon, eouth slope of mountains about 10 kilometer!! northeast of Thomazeau, altitude 640 meters above !ea level. W. P. Woodring, collector. December 3, 1920. 9557 (W 116 F). Arrondissement of Mirebalais, trail from Thomazeau to Cornillon, about 2 kilometers northwest of Cornillon, altitude 905 mete rs above eea level. W. P. Woodring, collector. Decemb e r 3, 1920. 9558 (W 118 F). Arrondissement of Mirebalais, entrance to cave on north eide of trail from Thomazeau to Cornillon, about 3 kilometers northwest of Comillon, altitude 915 meters above sea level. W. P. Woodring collector. December 3, 1920.

PAGE 182

Species. -Foraminifera : (Cushman) Sorites americana Orbicul ina sp. Operculina sp. Operculina ? sp .................................. Miogypsina antillea (Cushman) ................ Lepidocyclina giraudi R. Douville ................ Corals: Stylophora sp. Stylophora sp. a. b .. Stephanocoenia sp. cf. S. intersepta (Esper) ..... Orbicella imperatoris Vaughan ............... Orbicella canalis Vaughan ...................... Solenastrea sp. Lamellastrea sp. silecensis sp. n. Siderastrea Vaughan. Psammocora n. Psammocora s p. a. Porites sp. aff. P. furcata Lamarck .............. Goniopora jacobiana Vaughan .................... Goniopora sp. indet .......... Mollusca: Gastropoda : Cerithium Pelecypoda : sp. Arca sp. Ohlamys lum Ch la mys cf. A. umbonata Lamarck ... (Aequipecten) ep. cf. C. (A.) flabel (Cooke) ( Aequipecten) sp. Tortue Island. Port dePaix.: l"'"4 s x &:J s x x x x Upper Oligocene fossils. Massif du Nord. Borgne. s ? x ? s x tQ) x ......., .=cocos +;I 4> ..... t> r::; =' 0 4> co z l"'"4 lO s x x > :; 0 g x x ...... = (0 CQ 8 x Borders .... -<1> ...... = OS x s x x of Las -di CQ x Central Plain. Caho bas. ? x t x x 'C s z ==' 'C 4> .t ...... > ..... &Q t- x Montagnes du Trou d'Eau. Port au.Prince. I Mirebalais. 00 Q) x t-1.0 co O'.) x l"'"4 lO Q) x x 9 ,_ tr.I lO O> x x 00 lO x O> Q) 0) x x x x l"'"4 lO a; x IQ x Found al., in middle and aplJltl' OJtpiaen of l'blda and Qeomla. CQ O'.) 0) x x -di 0) x x x x i x >-' 01 O') 0 t-4 0 0 0 l?j t::d ...... 0 0 ":r:J p:: > ...... j

PAGE 183

SEDIMENTARY ROCKS 157 9449 (W 119 F). Arrondissement of Mirebalais, trail from Cornillon to St.Pierre, cliff along east side of trail about 1 kilometer east of Comillon. W. P. Woodring, collector. December 3, 1920. 9450 (W 120 F). Arrondissement of Mirebalais, float at same locality as 9449. W. P. Woodring, collector. December 3, 1920. 9559 (W 121 F). Arrondisscment of Mirebalais, trail from C9rnillon to St. Pierre, float about 2.5 kilometers east of Comillon, altitude 1,035 meters above sea level. W. P. Woodring, collector. December 3, 1920. 9451 (W 122 F). Arrondissement of Mirebalais, trail from Cornillon to St. Pierre, about 5 kilometers southeast of Cornillon, altitude 1,055 meters above sea level. W. P. Woodring, collector. December 3, 1920. 9452 (W 123 F). Arrondissement of Mirebalais, trail from Cornillon to Thomazeau and Mirebalais, float about 2.5 kilometers northwest of Cornillon, altitude 900 meters above sea level. W. P. Woodring, collector. December 4, 1920. 9453 (W 124 F). Arrondissement of Mirebalais, trail Cornillon to Mire balais, about 3.5 kilometers northwest of Cornillon, altitude 980 meters above sea level. W. P. Woodring, collector. December 4, 1921. 9560 (W 125 F). Arrondissement of Mirebalais, trail from Cornillon to Mire balais, float about 3.6 kilometers northwest of Cornillon, altitude 980 meters above sea level. W. P. Woodring, collector. December 4, 1920. 9454 (W 126 Arrondissement of Mireba lais, trail from Cornillon to Mire balais, crest of ridge about 5 kilometers northwest of Comillon, altitude 1,115 meters above sea level. W. P. Woodring, collector. December 4, 1920. 9561 (W 128 F). Arrondissement of Mirebalais, trail from Comillon to Mire balais, float on north slope of mountains about 10 kilometers northwest of Cor nillon, altitude 870 meters above sea. level. W. P. Woodring, collector. December 4, 1920. MIOCENE SERIES. Rocks of Miocene age are widely distributed in the plains and lowlands of the Republic and have a maximum thickness of more than a thousand meters. Probably all of them are lower and middle Miocene. The Miocene fossils in the different regions are so diverse that a short account of them is given for each region. GENERAL FEATURES. Areal distribution. The largest areas of Miocene deposits are in the Central Plain and the Artibonite Valley. Other areas are shown on Plate I. The larger areas shown are in plains or lowlands, and in other regions the Miocene rocks are restricted to lowlands; at few places are they found at altitudes of more than 400 meters above sea level. They were not seen on the crests of any of the mountain ranges, although they formerly ex tended over some ranges. Some lowlands, such as the Cul-de-Sac Plain, the lower part of the Artibonite Plain, and most of the Arbre Plain, con tain Miocene rocks, but they are almost wholly concealed by Quaternary alluvium. Stratigraphic relations. The transgression of the Miocene sea was extensive, and at places the Miocene overlaps the Eocene or even older rocks. Around the borders of the Central Plain and at other places where

PAGE 184

158 GEOLOGY OF THE REPUBLIC OF HAITI. the Miocene rests on upper Oligocene there is no unconformity between them and the change in lithology is gradual. In the northwest Peninsula, where Miocene rocks rest on Eocene rocks, there is little or no evidence of unconformity, the strike and dip of the two series being nearly the sa .me; neither is there any conglomerate at the base of the Miocene, although the change in lithology is abrupt. At a few places the Miocene appears to rest in marked unconformity on rocks older than Eocene. Lithology. The Miocene rocks consist principally of detrital deposit.a. Most of them are fine grained, consisting of clay, siltstone, sandstone, and marl. Unweathered surfaces of these rocks at most places are bluish or greenish, but on weathered outcrops they are yellowish brown or gray. The marl is highly ca lcareous, as is shown by the analysis on page 502 of a sample from a locality north of St.-Marc. On a moisture-free basis the sample contained more than 57 per cent of calci11m carbonate. Beds of impure yellowish limestone from a few centimeters up to a meter in thickness are interbedded with the marl at some localities. At other places there are thicker beds of coralliferous limestone. Close to old land masses the rocks are coarser, consisting of conglomerate and coarse sandstone. In the Central Plain, the Asile Valley, and the lowland at Camp Perrin the Miocene deposits contain beds of lignite. Although the unweathered Miocene rocks are firmly consolidated most of them are soft and crumble readily in the hand when weathered. Most of the Miocene d e posits are of shallow-water marine origin, but some of the deposits in the Central Plain were laid down in coastal-plain swamps and on flood plains. The Miocene rocks of the Asile Valley and the lowland at Camp Perrin are entirely nonmarine, consisting of lake, swamp, delta, and flood-plain deposits. Thickness. The thickness of the Miocene rocks differs greatly in different localities. In all the large areas it reaches a t l east several hundred meters. The estimated thickness in the Central Plain is 1,400 meters. Near La Chapelle in the Artibonite V:alley a thickness of about 700 meters is exposed. In the area northeast of St.-Marc 300 meters of Mio cene rocks are exposed, but the base of the series was not seen. Near Camp Perrin the thickness is at least 500 met.era. Structure. The Miocene rocks are folded and in many localities crumpled. Structurally the two largest areas of Miocene rocks, the Central Plain and the Artibonite Valley, are synclines Some of the details of the structure are given in the following pages. The structural features of the Central Plain are described on pages 488-492. DESCRIPTION BY REGIONS. JEAN RABEL v AIJLEY. Much if not all of the lowland called the Jean Rabel Valley is 11nderlain by soft Miocene rocks, which are easily eroded. In the northern part of the lowland, however, at least near Jean Rabel, the Miocene rocks are covered

PAGE 185

SEDIMENT.ARY ROCKS. 159 by Quaternary alluvi11m. Good exposures are fo11nd in the plateau region to the south. The Miocene rocks of the area were examined hurriedly near Jean Rabel and in the vicinity of Riviere Cadet and of Riviere de Jean Rabel, to the southwest and southeast, respectively, of the town. They consist principally of thin-bedded marl or sandy marl, generally bluish, but at a few places of a gray or yellow tint. The marl commonly has the typical conchoidal fracture shown in Plate XIV, 0, page 168. !Most of the sandy beds are :fine grained, but beds of coarse sandstone were seen on Riviere de Jean Rabel about a kilometer southeast of Jean Rabel. Beds of coarse sand stone and conglomerate are conspicuous near the southern border of the lowland. They are yellowish and rather firmly consolidated. Just west of Jean Rabel the Jv.Iiocene beds are sharply tilted and gen erally dip 30-40 N. Farther south, however, the dips flatten and become variable. The general inclination along Riviere Cadet seems to be slightly to the north, but along the ridge to the east the beds of conglomerate for a considerable distance dip a few degrees to the south. Along Riviere de Jean Rabel the general strike is northeastward, varying, however, with in the limits of north and east. The dips are 15 to 30, generally north westward so far as observed. The southward extension of the areas of Miocene beds along these two river valleys as reentrants into regions flanked by upper Eocene limestone and the divergence of the strikes at numerous places from the general east and west elongation of the lowland suggest that the valleys may be the troughs of minor synclines whose axes trend approximately northward. At a large spring about 5 kilometers south by east from Jean Rabel along the trail to Anse Rouge, near Riviere de Jean Rabel, the contact of Miocene with upper Eocene limestone is unusually well exposed. Both series of rocks dip northward at about the same angle, 15 or 20, but the change in lithology is abrupt and complete. The upper Eocene consists of hard white limestone containing numerous Forarninifera. The Miocene is a bluish sandy marl, but there is no trace of a basal conglomerate. The relations appear to be similar on Riviere Cadet about 4 or 5 kilo meters southwest of Jean Rabel, but the actual contact could not be found. The upper Eocene limestone, which contains many Foraminifera, dips northeastward at an angle of about 15 beneath fine-grained sandy and marly Miocene beds that dip gently northward. No estimate was obtained of the thickness of the Miocene beds exposed in this area, but it probably reaches some hundreds of meters. So far as observed the Miocene beds of the Jean Rabel Valley are unf ossilif ero11s. They are considered of Miocene age on stra tigraphic and Ii thologic grounds. Tho1s RmEREs VALLEY. Rocks similar to the Miocene rocks of the Jean Rabel Valley extend up the Trois Rivieres Valley for an undetermined distance. Their relations

PAGE 186

160 GEOLOGY OF THE REPUBLIC OF HAITI. to the Oligocene deposits of this valley are not known. Exposures in high bluffs along the lower part of the river show that they are made up of marl, sandstone, and siltstone. The beds generally strike parallel to the valley and dip to the east or west. Sandstone and shale, probably of Miocene age, crop out along the road leading westward from Port-de-Paix to the crossing of Les Trois Rivieres Similar detrital rocks extend along the north coast from Port-de-Paix southeastward to Cap Rouge at the eastern boundary of the arrondisse ment of Port-de-Paix. They dip to the northeast toward the se'a and over l ie limestone of upper Oligocene age. NEAR MOLE ST.-NICOLAS. The Riviere du Mole south of Mole St.-Nicolas occupies a deep, cany on like valley cut mainly in Quaternary corallifero us limestone. At La Gorge, about 3 kilometers south of Mole St.-Nicolas, the valley widens and is floored with Miocene marl. The average width of the area underlain by the Miocene beds probably is about half a kilometer. The area is at least 2 kilomet e rs and very likely 3 or 4 kilometers in length. The soft marl disintegrates readily into greenish clayey soil The bes t exposure was found in an excavation along the trail leading to Bombar dopolis, about 2 kilometers south of La Gorge, at an altitude of about 95 meters above sea level and 25 meters above the river bed. The marl is greenish in color and contains a few fragments of mollusks. The bedding was not visible, as the rocks in the exposure had sl11mped. Marl from this excavation was being tried in making bricks at La Gorge. (See p. 506.) The contact of the Miocene with the overlying coralliferous limestone is about 100 meters above sea level along the trail just above the excavation but is not well exposed. The Miocene beds generally seem to extend to elevations from 15 to 30 meters above the river bed. The coralliferous limestone generally forms steep bluffy slopes above the contact. The Miocene rocks exposed on this stream are probably continuous under the cover of Quaternary limestone with the extensive area of similar rocks in the Jean Rabel Valley. ARBRE PLAIN. The Arbre Plain, a lowland on the south side of the northwest Penin sula, resembles in many features the Jean Rabel Valley. Rocks of Miocene age 11nderlie probably the entire plain but are concealed by alluvium in its lower part. They consist of marl, conglomerate, and limestone. A Mio cene coral was collected from a conglomerate containing poorly preserved corals and mollusks (station 9846). The conglomerate is about 2 meters thick and is overlain and underlain by bluish marl. At this locality the beds are approximately horizontal. Just north of this exposure there is a low hill of weathered, pitted limestone, probably massive Eocene lime stone, which crops out from beneath the Miocene rocks.

PAGE 187

SEDIMENTARY ROCKS 161 Miocene marl was found in a low bluff along a dry stream channel on the trail from the Sources Chaudes to Anse Rouge, a little more than a kilometer southwest of the Sources Chaudes and just north of a promi nent low hill called Morne Meron. The marl is bluish and clayey and showed no bedding planes. The contact between Miocene marl and thin-bedded limestone and simi lar ro cks of middle Oligocene age is exposed at the northern edge of the plain on the trail from Anse Rouge to Jean Rabel. The two series of rocks seem to be conformable. At the Sources Chaudes there is evidence that the contact between Miocene and older lime s tone, of supposed Eocene age, is a fault. (Seep. 561.) Fossils. The single collection of fossils from the Miocene rocks of the Arbre Plain is given in the following table. The coral Psammocora n. sp. was obtained also from the upper part of the Thomonde formation in the Central Plain. Station in Arbre Plain (Miocene). 9846 (B 213 F). Arrondiss ement of Gonaives, trail from Sources Chaudes to l'Arbre, about 3 kilometers west-northwest from Sourc es Cha. udes. J. S. Brown, collector. February 4, 1921. Fossil (coral) from station 9846. Psammocora n. sp. b, also from Miocene of Trinidad. CENTRAL PLAIN. Artibonite group. The Miocene rocks of the Central Plain are better known than those of any other region in the Republic. Tippenhauer 1 and Jones 2 have de scrib e d some of the Miocene deposits of the plain, and a preliminary report setting forth the results of the re connaissance has been published..' Structurally the plain is a deep southeastward-plunging syncline modified by s econd ary synclinal and anticlinal folds. It contains an estimated thickness of 1, 4 00 meters of Mio cene ro cks, for which Woodring' has recently proposed the name Artibonite group, derived from Riviere Arti bonite. The Artibonite group is the equivalent of the lower part of the Yaque group of the Dominican Republic. It consists of the following formations : 1 Tlppenhauer, L. G., Beitrlige zur Geologfe Hatti, VI, Das Llgnltlager von Malssade und d e r Aufstteg zum Zentralplateau von Gonatves und von Norden aus : Petermanns l\Iitt., Band 47, pp. 193-199, pis. 15, 16 (map and sections), 1901; Neuer Beftrag z11r Topographie, Bevolkerungskunde und Geologie Haitls : Petermanns Mitt., Band 55, pp. 49-o7, pl. 5 (map), 1909. 2 Jones, W. F., A geological reconnaissance in Hatti ; a contribution to Antillean geology: Jour. Geology, vol. 26, pp. 728-752, pl. 5 (map and sections), 10 text figs., 1918. 1 Woodring, W. P., Stratigraphy, structure, and possible oil resources of. the Miocene rocks of the Central Plain, 19 pp., map, RP.p. Haiti Geol. Survey, 1922. 'Woodring, W. P., op. cit., p. 6. 11

PAGE 188

162 GEOLOGY OF THE REPUBLIC OF HAITI. Formations comprised in the Artibonite group. Approximate thickness in meters. Las Caho bas formation. . . . . . . . . . . . . . . . . . 600-700 Thomonde formation (including tongue in north-western part) . . . . . . . . . . . . . . . . . . . . 300-600 Madame Joie formation. . . . . . . . . . . . . . . . . . 100 The distribution of these formations is shown on Plate XXXVI (p. 488). MADAME JOIE FORMATION. The name Madame Joie formation has been proposed for the lowes t group of Miocene rocks in the plain.1 The name is derived from Madame Joie, a village about 10 kilometers west of Ma!ssade, and the type locality is a t the foot of Morne Madame Joie, about 1 kilometer southwest of the Metres Tmj 500 '+00 FIGURE 9. Diagrammatic section of foothills at Morne Madame Joie. Tp, Pliocene (?) clay; Tmj, Madame Joie formation, lower Miocene; Tos, upper Oligocene limestone. Tp village. It crops out at the base of the mountains along the western and southern margins of the plain. The Madame Joie formation overlies the upper Oligocene limestone, but the actual contact was not seen. Apparently there is no unconformity between them, but rocks that probably are the equivalent of the Madame Joie :formation elsewhere directly overlie Eocene deposits. At the type locality the formation is made up of a lower part consisting of bluish-gray siltstone and an upper part consisting of coralliferous lime stone, sandstone, and marl. The following section was measured at the type locality. Each of the cora .llif erous limestones in the upper part of the formation crops out in a prpminent ledge that has a long dip slope facing toward the plain, as shown in Figure 9. The soft lower part of the formation underlies the gap between the ridge formed by the coralliferous limestones and the main part of Morne Madame Joie. i Woodring, W. P., op. cit., p. 6, 1922.

PAGE 189

SEDIMENTARY ROOKS. 163 Section of Madame Joie formation at foot of Marne Madame Joie Est. thickness in meters. 12. Limestone, corallif erous, cream-colored; station 9720. . . . . 5 11. Sandstone, calcareous, yellow . . . . . . . . . . . . . . . 2 10. Siltstone, sandy, yellowish brown. . . . . . . . . . . . . . 1 9. Marl, light yellow . . . . . . . . ... . . . . . . . . . . . 1 8. Limestone, coralliferous, dirty white; station 9937..... . . . 5 7. Siltstone, marly, light brown. . . . . . . . . . . . . . . . 2 6. Marl, yellowish brown . . . . . . . . . . . . . . . . . . 1 5. Limestone, coralliferous, dirty white; station 9938............ 5 4. Marl, light yellow . . . . . . . . . . . . . . . . . . . . 1 3. Siltstone, marly, yellowish brown........................... 2 2. Limestone, coralliferous, light yellow. . . . . . . . . . . . 5 1. Siltstone, bluish gray, not well exposed. . . . . . . . . . . 30 60 The beds strike N. 50 W. and dip 25-33 NE. At the foot of the ridge lie clays of Pliocene or Quaternary age that probably conceal higher beds in the formation. On the Thomonde anticline 1 the Madame Joie for1nation consists of the same kinds of rocks. Limestone and yellowish-white marl in the upper part of the formation form a low ridge that arches over the plunging crest of the anticline about 2 kilometers northwest of Thomonde. Bluish-gray siltstone in the lower part of the formation crops out along a small stream that flows nearly on the crest of the anticline inside the ridge. The mollusks listed on page 164, station 9933, were collected from the siltstone along this stream. Similar mollusks (see list, p. 164, station 9784) were collected from yellowish-gray marl, probably in the upper part of the formation, on the south limb of the Thomonde anticline. At the south edge of the plain the Madame Joie for1nation was examined only along the trail from Belladere to Savanette. Here the formation seems to consist entirely of marl, from which the pteropods listed on page 164, station 9912, were obtained. The marl crops out in the mo11ntain slope facing the plain at an altitude of 460 meters above sea level. Fossils. The coralliferous limestones in the upper part of the forma tion at the type locality contain huge heads of Orbicella canalis Vaughan and Orb icclla altissima Duncan ( ?) 0. canalis bas not heretofore been recorded from rocks yo11nger than upper Oligocene, but it was fo11nd in Miocene deposits at a number of localities in the Republic of Haiti. The molluscan fa11na of the Madame Joie for1nation is meager, consisting principally of pi:eropods of the genera Cavolina and Styliola. At some localities, particularly station 9912, the pteropods are very abundant but poorly preserved. The presence of these pteropods and of the bivalve mollusks Bathyarca and Limopsis and the absence of common shoal-water 1 See pp. 488-492 and Pl. XXXVI for description and location of anticlines tn the Central Plain.

PAGE 190

164 GEOLOGY OF THE REPUBLIC OF HAITI. mol lusks indicate that these sediments were deposited in clear water at some distance from land. The depth of the water may have been greate r when the lower part of the for1nation was being laid down than when the upper part and the later formations were being deposited. The Madame Joie formation is either Aquitanian or Burdigalian. Al though the evidence is not very conclusive it s e ems to be lower Burdi galian. It marks the beginning of the change from the limestones of the Eocene and Oligocene to the detrital deposits of the Miocene. 'fl1e .Madame Joie fossils and the localities at which they were collecte d are given in the following table. Stations in Central Plai11r-Madame Joie formation (Miocene). 9720 (W 257 F). Arrondissement of Hinche, lower slopes of Morne Madame Joie, about 1 kilometer southwest of Madame Joie, from uppermost coralliferous l imestone, altitude 400 meters above sea level. W. P. Woodring, collector. February 3, 1921. 9937 (W 258 F). Arrondissement of Hinche, lower slopes of Morne Madame Joie, from middle coralliferous limestone, altitude 430 meters above sea level. W. P. Woodring, collector. February 3, 1921. 9938 (W 259 F). Arrondissement of Hinche, lower slopes of Morne Madame Joie, from lower coralliferous limestone, altitude 465 meters above sea level. W. P. Woodring, collector. February 3, 1921. 9933 (W 239 F). Arrondissement of Las Cahobas, crest of Thomonde anticline, about 3 kilometers west-northwest of Thomonde, altitude 315 meters above sea level. W. P. Woodring, collector. January 29, 1921. 9784 (W 213 F). Arrondissement of Las Cahobas, south limb of Thomonde anticline, trail from Thomonde to Mirebalais, of ravine about 4 kilo meters southwest of Thomonde, altitude 305 meters above sea level. W. P. Woodring, collector. January 17, 1921. 9912 (W 199 F). Arrondissement of Las Cahobas, trail from Belladere to Savanette, north slope of mountains about 7 kilometers southwest of Belladere, altitude 460 meters above sea level. W. P. Woodring collector. January 13, 1921. Miocene fossils from Central Plain (Madame Joie formation) Species. Oorals: Orbicella canalis Vaughan .............................. Orbicella altissima Duncan f ........................... Mollusca: Pteropoda: Sty liola sp. .......................................... Oa volina sp. cf. 0. bisulcata ( Gabb) ................. Gastro poda : Bullaria f sp .. Natica ? sp ..... Pelecypoda : Limopsia sp. ........................................... Bathyarc a sp. cf. B. hendersoni Dall........ ........... Hinche. Las Cahobas. 9720 9937 9938 9933 978' 9912 x x x . . . . .. x x x x x x x x x

PAGE 191

SEDIMENT.ARY ROCKS. 165 THOMONDE FORMATION AND MAISSADE TONGUE. The name Thomonde ''beds,'' derived from the name of the town of 'fhon1onde, was given by Jones 1 to the rocks overlying the J.\iladame Joie formation. The type 10cality is in the vicinity of Thomonde. In the southeastern part of the plain the Thomonde formation crops out at the foot of the mountains. The width of the outcrop increases as the rocks arch over the crests of the plunging anticlines. The outcrop is wider along its northeast edge, because the formation is thicker there and the struc ture is different. In the entire northwestern part of the plain the Thomonde for1nation lie,s at the surface or is covered only by a thin layer of Pliocene or Quaternary stream deposits. The Thomonde formation conformably overlies the Madame Joie for mation, which along the northeast edge of the plain seems to be com pletely overlapped by it. Southeastern part. Along the west and south edges of the southeastern part of the plain the Thomonde formation consists principally of soft rocks, and its outcrop is marked by a depression along the foot of the mountains and on the plunging anticlines. At the type locality the forma tion is about 400 meters thick and consists principally of siltstone, either sandy or clayey. There are a few thin beds of sandstone near the base and a few beds of coarser sandstone and conglomerate near the top. The siltstone is similar to the siltstone in the lower part of the Madame Joie formation. The unweathered rock is bluish gray, but surfaces that have been oxidized and leached are rusty brown or yellowish brown. The upper part of the formation is well exposed on the south limb of the Thomoude anticline on the slope leading down to Riviere Thomonde, along the trail from Las to Thomonde, and on the north limb along the trail from Thomonde to Thomassique. Good exposures of the upper part of the formation can be seen on Riviere Thomonde below Thomonde. Numerous collections of fossils were ob tained from the beds in the uppermost 100 meters of the Thomonde for n1ation, on the crest and limbs of the 'l1homonde anticline. They are listed on pages 178-190. Detailed descriptions of the localities are given on pages 175, 177 (stations 9782, 9781, 9780, 9929, 9778, 9779, 9791, and 9785). The best fossils were collected on Riviere Thomonde at station 9782. Along the trails where the beds lie above the level of ground water the shell substance of many specimens bas been removed by leaching, leaving only casts and impressions. Casts are particularly common in the porous beds at the top of the formation. The lower part of the Thomonde for mation on the Thomonde anticline also consists of bluish-gray siltstone but contains only a few fossils. (See list, pp. 178-190, station 9935.) The middle part was examined on Riviere l' Ayaye, on the crest of the l' Ayaye anticline along the west edge of the plain. The following section 1 Jones, W. F., op. cit., p. 736, 1918.

PAGE 192

166 GEOLOGY OF THE REPUBLIC OF HAITI. is exposed in the lower third of the first high bluff on the left bank of Riviere l' Ayaye above its mouth: Section of beds in middle part of Thomonde formation on Riviere l'Ayaye (station 99(Jl). Est. thickness in meters. 7. Siltstone, bluish . . . . . . . . . . . . . . . . . . . . . . 3 6. Siltstone, sandy; fragments of lignitized wood. . . . . . . . 1 5. Siltstone, bluish . . . . . . . . . . . . . . . . . . . . . . 2 4. Siltstone, sandy (lens) . . . . . . . . . . . . . . . . . . . .5 3. Siltstone, yellowish . . . . . . . . . . . . . . . . . . . . 1 2. Siltstone, sandy . . . . . . . . . . . . . . . . . . . . . . .5 1. Siltstone, bluish . . . . . . . . . . . . . . . . . . . . . . .5 Stream level. 8.5 The fossils from this locality (station 9907; see list, pp. 178-190) were collected from the beds given in the section and higher beds, some of the best specimens being obtained from loose blocks at the foot of the bluff, which is about 25 meters high. The fossils listed on pages 178-190 (station 9908), were collected from a layer in sandy siltstone in the next bluff on the left bank downstream. The Thomonde formation is not very well exposed along the south edge of the plain near Las Cahobas. At the exposures examined it consists of sandy and clayey siltstone. A collection of fossils (station 9904; see list, pp. 178-190), indicates that the beds were laid down in relatively deep water. On the east side of the road from Mirebalais to Las Cahobas and immediately north of the small stream at the foot of the mountains, on the north side of the gap, soft sandy and clayey beds in the Thomonde formation strike N. 30 W. and dip 32 NE. At the same locality but on the west side of the road, there is :finely laminated buff or yellowish clay, from which the plants and fish listed on page 206 (station 7 544) were collected. The clay dips about 5 SW. At. the top of the hill about 60 meters farther northeast, toward Las Cabobas, similar clay seems to rest unconformably on beds belonging to the Thomonde formation. The plants are strand plants, and one of the species was obtained from beds of known Miocene age in the Artibonite Valley (see p. 215), but the fish be longs to a fresh-water genus. The clay can hardly belong to the Thomonde formation because of its stratigraphic relations and because the Thomendc formation not far distant along the strike carries a relatively deep-watr.r fauna. It may represent some part of the Miocene series not known else where in the Central Plain. Its stratigraphic relations led to the supposition in the field that it was of Pliocene age, but overwhelming evidence shows that the sea withdrew from the a1ea now embraced by the Central Plain before Pliocene time. It is unfortunate that the age of these beds is not known, as they furnished the largest collection of fossil plants obtained in the Republic.

PAGE 193

SEDIMENTARY ROCKS. 167 On the Belladere anticline, at the southeastern extremity of the plain, the Thomonde formation was seen only on the north limb along the trail from Las Cahobas to Belladere and its prolongation eastward beyond Belladere toward Commendador. Siltstone and thin beds of sandstone dip from 12 to 24 almost due north. Along the northeast side of the plain the lithology of the Thomonde formation is entirely different, for in this region it consists largely of coarse detrital sediments, principally conglomerate and coarse sandstone. (See Pl. XIV, A.) Most of the coarse seuiments are of nonmarine origin and are apparently delta and flood-plain deposits laid down along the old landmass that stretched along this side of the plain during Miocene time. Toward the southwest, under the cover of younger rocks, the coarse non marine sediments probably interfinger with the fine-grained marine sedi5.0 N.E 0:7, r;:ii F1GuRE 10. Diagram showing lateral change in lithology of the Miocene rocks of t he Central Plain and the transgressive overlap of the Thomonde formation. The total thickness of the Las Cahobas formation is not sho"' n. ments that crop out along the west and south edges of the plain, as is shown in Figure 10. Thin wedges of fine-grained marine sedi1nents extend northeastward into this area of coarse rocks, indicating continuous move ments of the strand and breaks in the deposition of sediments along the margin of the sea. These marine wedges contain typical Thomonde fos sils. The largest collections made and the largest collections from the entire Thomonde formation were obtained on Ravine Roche Salee at the crossing of the trail from Hinche to Thomassique, about 3 kilometers southeast of Los Palos. (See lists, pp. 178-190; stations 9945 and 9946.) Many of the shells are closely packed in pockets, which indicates sorting by waves or currents. Fossils typical of the upper part of the Thomonde formation were col lected from sandy siltstone less than half a kilometer north of Thomas sique on the trail to Cerca-la-Source. (See list, pp. 178-190; station 9947.)

PAGE 194

168 GEOLOGY OF THE REPUBLIC OF HAITI. Countless specjroens of Orthaulax aguadillensis are weathered out of or partly embedded in the rock at this locality. Most of the specimens are casts, except Orthaulax, Ostrea, and Pecten, as is common in beds above the level of ground water. Fossils similar to those in the middle part of the Thomonde for1nation on the west side of the plain were coll ecte d on the same trail at the first crossing of Riviere Ronde. (See list, pp. 178-190; station 9948.) The Thomonde formation, which is cl early transgressive, probably overlaps the coarse marginal deposits of the Madame Joie formation on this side of the plain, as is shown on the map (Pl. XXXVI) and in Figure 10. Beds of fine sandstone and siltstone that contain no marine fossils and that are considered nonmarine are interbedded with the conglomerat e s and coarse sandstones, but t11ey are buff, reddish, or greenish. Most of the marine beds are bluish gray, but some of them are green i sh. Toward the mountains the conglomerates contain an increasing proportion of diskshaped pebbles of argillite derived from the area of argillite near Cerca la-Source. Conditions seem to be favorable for a quantitative study of the gradual elimination of pebbles of argillite in terms of increasing distance from the landmass. Northwestern part. In the northwestern part of the plain, northwest of Hinche, the stratigraphic relations of different parts of the Thomonde formation are more complex, and the relations presented in this report are based primarily on a study of the fossil mollusks. In this part of the plain the lower part of the Thomonde formation consists of siltstone and thin b e ds of fine-grained sands tone res embling the lower part of the formation in the southeastern part of the plain and containing the same fauna. This part of the formation is about 2 5 0 meters thick. It crops out along the foot of the mountains immediately above the Madame Joie formation and on the Fond Bleu dome. Fossils were collected from it on the south limb of the Fond Bleu dome (station 9719) and northeast of Madame Joie near Riviere Blanche (station 9939; see lists on pp. 178-190). The middle part of the formation, which is about 100 meters thick, consists principally of conglomerate and coarse sands tone res embling those in the overlying Las Cahobas formation. These beds form the pine covered '' rim rock '' around the Fond Ble u dome. Riviere Blanche flows across them in a short, narrow gorge. They probably r epresent some of the beds in the middle part of the Thomonde formation as developed in the southeastern part of the plain. Above these conglomerates and sandstones lies a series of lignite-bearing beds to which Jones 1 gave the name Maissade ''beds,'' derived from the name of the town of Maissade. Extensive collections of fossils were made from this series of beds at exposures on Riviere Fond Gras and Riviere 1 Jones, W. F., op. cit., p. 739, 1918.

PAGE 195

BEPUBLTC OF HATTI GEOLOGICA L SUR \ 'EY A. TITO:\IO:NDE ... \ t' 1\ I I PLATE XIV FORMATION ON RIVIERE RONDE, ON TIIE NORTHEAST IDIU OF 'l'IIE CEXTR.AI.J 1LAI:N. B L .. \S CATTOR1\S NE .. \R CAIIORAS, ON' THE SOUTIT SIDE OF I.1 I,LAIX 0. NEAR LA CI-lAPELLE.

PAGE 196



PAGE 197

SEDIMENTARY ROCKS. 169 Blanche (or Rio Blanco), northeastward-flowing tributaries of Riviere Canot northwest of Maissade. These collections show that the Ma1ssade ''beds'' consist of a remarkable alternation of beds, some of which contain a marine fauna, others a mixture of marine and brackish-water faunas, and others a brackish-water fauna. The collections are listed on pages 194-197 (sta. tions 9717, 9713, 9714, 9715, 9722, 9726, 9711, 9716, 9723, 9724, 9725, 9727, 9728, 9729, 9732, 9733, 9710, 9718, 9712, 9730, 9731). The beds containing the marine mollusks occur in the lower and middle parts of the lignite-bearing series. The largest marine faunule was col lected at the northwest end of the Fond Bleu dome, on Riviere Fond Bleu, from beds near the base of the series (station 9 717). These fossils are typical of the middle part of the Thomonde formation in the southeastern part, and show that the lower and middle parts of the Ma!ssade '' beds '' are a coastal-swamp facies of the middle part of the Thomonde formation. The series of lignite-bearing rocks is therefore called the Ma1ssade tongue S.E. N.O oo.o.0o o 0 o o o . . . . . . . ... . . . . . t 0 o -0. o .o 0 o.oo0o-0o o0 0 0 o :00 9 '?. o'o .. c)9.0o .. .o. o .. o. o :_:g';f:) O .Q .. ....,T ... O . . ,. , '. '"' . -.. . . ...... .. .. . . .. ----. . . . .. . . . . . . . -. ....... .. .-=. ........ ..... . --"' FIGURE 11.-Diagram showing stratigraphic relations of the Malssade tongue of -the Thomonde formation of the Thom on deformation. The fauna of the upper part of the Thomonde formation was not discovered in this part of the plain, and the higher beds of the Ma!ssade tongue, which are wholly nonmarine, seem to be the equivalent of the upper part of the Thomonde. Toward the southeast the Maissade tongue probably inter:fingers with other parts of the Thomonde forma tion, but the outcrops were not traced. This inf erred relation is shown on Plate XXXVI and :in Figure 11. Beds near the base of the Ma1ssade tongue are exposed in roadcuts along the road from Maissade to Hinche, up the hill south of the crossing of Riviere Frio. They consist of beds of sandy siltstone containing crystals of gypsum and beds of coarse, pebbly sandstone and dip 45 NE. Near the top of the hill lies a bed of carbonaceous clay 82 centimeters thick. Farther to the south and southeast there are soft cross-bedded sandstone containing Ostrea and Scapharca (station 9926) and con glomerate. These beds probably lie at the top of the series of coarse sand stones and conglomerates below the Maissade tongue.

PAGE 198

170 GEOLOGY OF THE REPUBLIC OF HAITI. The Ma!ssade tongue is about 200 meters thick. The lowermost beds, about 50 meters thick, consist of siltstone, sandstone, and clay. In this part of the formation some of the clay is carbonaceous. Higher beds contain more carbonaceous material, and most of the beds of lignite, which are described on pages 481-483, occur at about the middle of the formation. The uppermost beds are entirely nonmarine and consist of sandstone, clay, and siltstone, most of which are colored in shades of red and green. A conspicuous bed of red clay containing flakes of gypsum extends across the road from Ma1ssade to St.-Michel de l' Atalaye, about 200 meters north of the crossing of Ravine Reparadere, near Ma1ssade. The following sections of the Ma1ssade tongue show the rapid alterna tion of beds. The region in which they were deposited was a coastal swamp that was repeatedly invaded by the sea, but finally the sea withdrew. The carbonaceous material in the beds of carbonaceous shale and lignite prob .. ably represents debris of coastal-swamp vegeta tion, but no determinable plants were obtained. The Ma1ssade tongue is well exposed on Riviere Blanche. The following sections of beds about 20 meters above the base of the tongue was measured in the first bluff on the left bank of Riviere Blanche below the gorge formed by the underlying conglomerates and sandstones: Section of lower part of M aissade tongue on Riviere Blanche. lleters. 7. Clay, sandy . . . . . . . . . . . . . . . . . . . . . . . .50 6. Clay, black, carbonaceous. . . . . . . . . . . . . . . . . .11 5. Clay, sandy, brown; fragments of brackish-water mollusks... .20 4. Siltstone, sandy, bluish; fragments of marine mollusks...... .50 3. Oyster bed . . . . . . . . . . . . . . . . . . . . . . . 1.75 2. Siltstone, sandy, bluish . . . . . . . . . . . . . . . . . 10 1. Siltstone, sandy, fragments of marine mollusks............. 2 15.06 Beds in the following section are exposed in the first long bluff on the right bank of Riviere Blanche below the gorge. The base of this section is about 75 meters higher stratigraphically than the top of the preceding section. At this locality the beds dip northeastward 45a to 58. Section of middle part of M aissade tongue on R ivwre Blanche. Meters 47. Siltstone, bluish, marine fauna; station 9722.............. 5.40 46. Siltstone, mixed marine and brackish-water fat1na; station 9723 . . . . . . . . . . . . . . . . . . . . . . . . .25 45. Siltstone, bluish gray, sandy; small fragile bivalves........ .80 44. Siltstone, gray, mixed marine and brackish-water fauna; station 9724 . . . . . . . . . . . . . . . . . . . . . .60 43. Siltstone, gray; mixed marine and brackish-water fauna.; station 9725 ................................. . . . . .35

PAGE 199

SEDIMENTARY ROCKS. 42. Siltstone, dark gray, slightly carbonaceous ............... Lignite, dirty; streaks of pure lignite .................. 41. Lignite .............................................. Lignite, dirty . . . . . . . . ....................... 40. Siltstone, dark gray, carbona c eous; thin streaks of dirty lignite . . . . . . . . . . . . . . . . . . . . . . .. 39. Siltstone, gray ......................................... 38. Siltstone, gray; streaks of dirty lignite ................... 37. Not exposed ........................................... 36. Siltstone bluish gray; marine fauna; station 9726 ......... 35. Siltstone, gray; oyste r b e d .............................. 34. Siltstone gray; mixe d marine and brackish-water fauna; station 9727 ......................................... 33. Siltstone, gray, thin irregular Reams of lignite at base ..... 32. Siltstone, gray; mixed marine and brackish fauna; nests of Scapharca; station 9728 ............................ 31. Siltstone, gray; oyster bed .............................. 30. Siltstone, bluish gray; carries Potamides ................. 29. Siltstone, gray ; extends as ledge across stream ; mixed marine and brackish-water fauna; masses of Scapharca chiriquiensi,s; station 9729 ............................ 28. Siltstone, bluish gray .................................. Clay, carbonaceous ................................... L . igm te ............................................... 27. Lignite; two sandy partings that contain fragments of My tilopsis ......................................... L . di 1gn1te, rty ......................................... 26. Clay, greenish blue ..................................... 25. Siltstone; layers of hard sandy material and softer clay e y material ............................................ 24. Clay, black, carbonaceous; a thin band contains brackish-water mollusks; station 9730 .......................... 23. Siltstone, sandy, dark gray ............................. 22. Clay, black, carbonaceous; streaks of sandy siltstone ..... 21. Siltstone, slightly carbonaceous; brackish-water fauna; station 9731 ......................................... 20. Clay, dark gray to black; carries Potamide s .............. 19. Siltstone, greenish blue ................................. 18. Oyste r bed ............................................ 17. Siltstone, gre enish blue ; not well exposed ................ 16. Oyster bed ............................................ 15. Siltstone, bluish green; carries Scapharca ................ 14. Clay, carbonaceous .................................... 13. Sandstone; thin seams of lignite ......................... 12. Lignite, very dirty ...................................... 11. Siltstone, dark gray; seams of carbonaceous clay ......... 10. L . di igrute, rly ........................................ L . ign1 te . . . . . . . . . . . . . . . . . . . . . . . 9. Siltstone, sandy, dark gray .............................. 8. Lignite, dirty .......................................... 7. Siltstone, bluish gray .................................. 6. Siltstone, gray; mixed marine and brackish-water fauna.; Meters 3 45 .30 .13 .05 .60 .16 .55 3 .35 .65 .13 1.40 .20 2.50 .50 1.75 .15 .20 60 .30 5 1.14 .07 .27 .80 .80 1 1.25 .10 4 .50 .35 .45 1 .15 .75 4 .06 .48 .32 .05 3 l!ta ti on 9732 . . . . . . . . . 3 171

PAGE 200

172 GEOLOGY OF THE REPUBLIC OF HAITI 5. Oyster bed ............................................ 4. Siltstone, bluish ; carries Scapharca ...................... 3. Oyster bed ............................................ Meters .50 4 .50 2. Si ltstone, dark gray; pieces of carbonaceous material; mixed marine and brackish-water fauna; station 9733. . . . . 2 1. Siltstone, bluish gray . . . . . . . . . . . . . . . . . . 1 60.91 The fallowing section of beds in the middle part of the Ma1ssade tongue was measured at a long bluff on the right bank of Fond Gras about 2 kilometers above the junction of Riviere Fond Gras and Riviere Canot. The beds dip 70 to 80 NE. Section of middle part of M a"issade tongue on Riviere Fond Gras. Meters. 15. Oyster bed . . . . . . . . . . . . . . . . . . . . . . 1 14. Clay, bluish green; mixed marine and brackish-water fa11na; station 9711 . . . . . . . . . . . . . . . . . . . . 2 13. Lignite, dirty . . . . . . . . . . . . . . . . . . . . . .22 12. Siltstone, bl11ish; brackish-water fauna; station 9712.... . 5 11. Oyster bed . . . . . . . . . . . . . . . . . . . . . . 1 10. Siltstone, sandy . . . . . . . . . . . . . . . . . . . 3 9. Lignite, dirty ........................................ L . 1gni te .......................................... .50 .43 8. Siltstone, brackish-water fauna. . . . . . . . . . . . . . 2 7. Oyster bed . . . . . . . . . . . . . . . . . . . . . . 1 6. Siltstone, sandy; marine fauna; station 9713 ............. L t 1gn1 e ......................... Clay, carbonaceous ................................... L . igrute ............................................... L . d' 1grute, 1rty ......................................... 5. Clay, carbonaceous ...................... ; ............ L . d' 1gn1te, 1rty ........................................ L . ignite ............................................... L . d. ignite, 1rty ......................................... Clay, carbonaceous ................................... 4. Siltstone, bluish; marine fauna; station 9714 ............. 3. Siltstone, sandy; marine fauna; station 9715 ............. 2. Lignite, dirty .......................................... 1. Clay, carbonaceous; fragments of M ytilopsis ............. 1.50 .14 .03 .06 .07 .05 .05 2 .07 .08 .12 5 .09 .08 25.49 At a bluff on the right bank of Riviere Frio, about 100 meters abo\1e the crossing of the road from Ma1ssade to Hinche, the following beds are exposed:

PAGE 201

SEDIMENTARY ROCKS. Section of beds in middle part of M a"issade tongue on Riviere Frio. 10. Siltstone, grayish brown; carries Potamides. L . ign1 te . . . . . . . . . . . . ....................... Clay, brown, carbonaceous ............ ............... 9 L 't 1gn1 e ............................................... L . d' 1gn1te, irty ......................................... Clay, black, carbonaceous; fragments of mollusks ....... 8. Siltstone and clay, bluish green .......................... 7. Clay, chocolate-brown .................................. 6. Clay, black, carbonaceous; streaks of lignite ............. 5. Clay, grayish brown .................................... 1t1eters. .17 .43 .14 .18 .55 5 .50 .44 2 4. Siltstone, bluish green . . . . . . . . . . . . . . . . . 1 3. Not exposed . . . . . . . . . . . . . . . . . . . . . 30 2. Lignite, dirty streaks . . . . . . . . . . . . . . . . . . 2.50 1. Clay, brown, brackish-water fauna; station 9710. . . . . . 1 43.91 173 The upper part of the Ma1ssade tongue was examined on Riviere Ca.not below the crossing of the road from Ma1ssade to St.-Mi chel de l' Atalaye. It is made up of reddish and greenish siltstone, coarse rusty-brown sand stone, greenish sandstone, and conglomerate containing small pebbles. None of these beds is fossilif e rous. On the northeast side of the northwestern part of the plain the Tho monde formation res e mbles the middle part of that formation on the southwest side of that part of the plain, but is much thicker and consists principally of conglomerates and c oars e sandstones. These b eds are sim ilar to those in the Thomonde formation along the northeast side of the south eastern part of the plain, but no marine fossils w ere found in the northwestern part and the beds there probably consist entirely of flood -plain and delta deposit s They are indistinguishable from similar beds in the Las Cahobas formation, and wh e n the Ma1ssade beds were named it was supposed that they and the c onglom erates underlying the Ma1ssade tongue on the southeast side of the plain belonged to the Las Cahobas. Fossils, Thomonde formation. Large collections of fossils w ere obtained from the Thomonde formation, which can be divided into three faunal zones corresponding to the low er, middle, and upper parts already described. Corals are abundant in the middle and upper faunal zones, but none of them are of reef facies. The Recent species Bolenastrea bournoni Milne Edwards and Haime and Siderastr e a s iderea (Ellis and Soland er) give the fauna a modern aspect. Several species of Stylophora and a new species of Asterosmilia are common in the middle faunal zone at stations 9907 and 9908 on Riviere l'Ayaye. Stylophora minor Duncan, collected from the middle fa11nal zone at station 9948, is characteristic of younger beds ( Gurabo forrnation and Mao Adentro limeston e ) in the northern part of the Dominican Republic. A.ntillia dubia D11ncan, found in the upper I

PAGE 202

174 GEOLOGY OF THE REPUBLIC OF HAITI. faunal zone, is known also only from younger beds ( Cercado and Gurabo formations) in the Dominican Republic A new species of Psammocora was obtained from Miocene rocks in the Arbre Plain. The same species has been obtained also from the lower Miocene rocks of Trinidad. The middle part of the formation near Las Cahobas contains a Deltocyathus similar to the Miocene and Recent relatively deep-water species D. italicus (Michelotti), indicating that along the south edge of the plain the for1na tion was deposited in deeper water than in the central and northern partq Abundance of mollusks is a characteristic feature of the Miocene de posits of the West Indies and regions near by where these deposits consist of detrital rocks Large faunas have been obtained in the Dominican Republic, Jamaica, Trinidad, Colombia, Panama, and Costa Rica. In Cuba and Porto Rico, where rocks of the same age are more calcareous, mollusks are less ab11ndant and poorly preserved. The mollusks of the Cercado and Gurabo for1nations of the Dominican Republic are well known through Maury's 1 work, although many additional species were obtained during the reconnaissance made under the supervision of the United States Geological Survey.' About 500 species are known from the Cercado formation and about 400 from the Gurabo formation. A description by Woodring of the Bowden fa11na of Jamaica, comprising almost 600 species, is awaiting publication by the Carnegie Institution of Washington. Olsson has recently described more than 300 Miocene species from Costa Rica. The total molluscan fa11na of the Thomonde formation, including the Ma1ssade tongue, numbers more than 350 species. The upper fa11nal zone, and probably the whole formation, is the equivalent of the Baitoa forma tion of the Dominican Republic. As only 60 species were obtained from the Baitoa formation, the Thomonde fauna fills a gap in the succession of Miocene fat1nas of the West Indies. The lower faunal zone contains only a few mollusks, principally ptero pods of the genera Styliola and Cavolina, indicating clear and relatively deep water and resembling the small faunule in the lower part of tl1e underlying Madame Joie formation. The middle and upper fa11nal zones contain a rich shoal-water fa11na. Some of the characteristic Thomonde mollusks are shown on Plate XV. Orthaulax aguadillensis Maury (Pl. XV, fig. 3) is the most striking and one of the most common of the mollusks in the upper faun.al zone. Interesting specimens of this species have been described by Woodring.' It was originally described from beds of upper Oligocene age in Porto 1 Maury, C. J., Santo Domingo type sections and fossils: Bull. Am. Paleontology, vol. 5, pp. 165-459, pls. 27-68, 1917. 2 See lists In A. geological reconnaissance of the Dominican Republic: Dominican Rep. Geol. Survey Mem., vol. 1, pp. 116-129, 137-151, 1921. 1 Oleson, A. A., The Miocene of northern Costa Rica: Bull. Am. Paloontology, vol. 9 pp. 179-418, pls. 4-35, 1922. 'Woodring, W. P., Tertiary mollusks of the genus from the Republic of Hatti, Porto Rico, and Cuba : U. S. Nat. Mus. Proc., vol. 64, art. 1, 12 pp., 2 pls. 1923.

PAGE 203

SEDIMENTARY ROCKS. 175 Rico but has been found also in deposits of the same age (Cevicos lime stone) in the Dominican Republic and in deposits of Miocene age in Porto Rico and the Dominican Republic.1 Perhaps the most curious Thomonde mollusk is a Terebra-Iike columbellid which is considered a new subspecies of Strombinella acuf ormis Dall, described from beds of Miocene age near Potrero, Province of Santiago, Dominican Republic! Many specimens of this mollusk were collected from the upper faunal zone at stations 9945 and 9946. The following are the most striking species confined to the upper fa11nal zone: Oonus veatchi Olsson (Pl. XV, figs. 1, 2), Gonus n. sp. (also from Baitoa formation), Xancus rex Pilsbry and Johnson, Pkos semicostatus Gabb (Pl. XV, figs. 6, 7), Strombinella acuf ormis Dall n. subsp., Ort}iaulax aguadillensis Maury (Pl. XV, fig. 3), and .Scapharca corcupidonis Ma .ury n. subsp. Most of these and other species are identical with or similar to species from the Baitoa formation. Other species are similar to species from the Cercado formation. Many specie-s are similar to species from the Chipola marl of Florida, particularly some Pleurotomids, Marginellas, and Fusimitras. A few species are similar to those in the younger Gurabo formation and the Bowden marl. The Thomonde formation apparently is the equivalent of part of the Y aque group of the San Juan Valley and the valley of Rio Y aque del Sur in the Dominican Republic, although except for Orthaulax aguadillensis the molluscan fa11nas are not very similar. It is regarded as of Burdigalian age, probably middle Burdigalian, although no direct comparison has j-et been made with European faunas. Stations in Central Plain, Thomonde formation (Miocene). 9945 (W 320 F). Arrondissement of Hinche, trail from Hinche to Thomassique, left bank of Ravine Roche Salee at crossing about 3 kilometers southeast of Los Palos, upper part of Thomonde formation. W. P. Woodring, collector. March 12, 1921. 9946 (W 321 F). Arrondissement of Hinche, trail from Hinche to Thomassique, right bank of Ravine Roche Salee at crossing about 3 kilometers southeast of Las Palos, upper part of Thomonde formation, 1 meter higher stratigraphically than 9945. W. P. Woodring, collector. March 12, 1921. 9947 (W 322 F). Arrondissement of Hinche, trail from Thomassique to Cerca la-Source, 0.5. kilometer north-northeast of Thomassique; upper part of Thomonde formation. W. P. Woodring, collector. March 13, 1921. 9782 (W 211 F). Arrondissement of Las Cahobas, crest of Thomonde anticline, left bank of Riviere Thomonde, about 4 kilometers east-southeast of Thomonde; upper part of Thomonde formation. W. P. Woodring, collector. January 16, 1921. 9779 (W 208 F). of Las Cahobas, south limb of Thomonde anticline, on trail from Las Cahobas to Thomonde, about 2 kilometers south southeast of Thomonde, at an altitude of 380 meters above sea level; upper part of Thomonde formation. W. P. Woodring, collector. January 15, 1921. 1 Cooke, C. W., Orlhaulaa:, a Tertiary guide fossil : U. S. Geol. Survey Prof. Paper 129, pp. 25, 30-31 ; pl. 4, figs. 2-6; pl. 5, figs. la, lb, 1921. 2 Guppy, R. J., and Dall, W. H., Descriptions of Tertiary fossils from the Antillean region : U. S. Nat. Mus. Proc., vol. 19, p. 312, pl. 29, fig. 6, 1896.

PAGE 204

PLATE XV. SoME CHARACTERISTIC MOLLUSKS OF THE THOMONDE FORMATION. FIGURES 1, 2. Conus veatchi Olsson. Two specimens from the same locality, X 1. U.S. G. S. station 9782. U. S. N. M. catalogue No. 350578. FIGURE 3. Orthau/,ax aguadillensis Maury. View of small specimen showing sculpture and apertural callus, X 1. U.S. G. S. station 9945. U.S. N. M. catalogue No. 350577. FIGURES 4, 5. Phos costatus Gabb. Apertural and dorsal views of same specimens, X 1. U.S. G. S. station 9945. U.S. N. M. catalogue No. 350574 FIGURES 6, 7. Phos semicostatus Gabb. Apertural and dorsal views of the same specimen, X 1. U.S. G. S. station 9908. U.S. N. M. catalogue No. 350575. FIGURES 8, 9. Cymia henekeni Maury. Apertural and dorsal views of same speci-men, X 1 U.S. G. S. station 9946. U.S. N. M. catalogue No. 350579. FIGURES 10, 11. Potamides cahobasensis Pilsbry. U. S. G. S. station 9908. U. S. N. M. catalogue No. 350576. Fig. 10. Adult specimen, X 1. Fig. 11. Y 011ng specimen, X 2, showing sculpture of early whorls. 176

PAGE 205

REPUBLIC OF HAITI GEOL OG I CAl .. S l J R \ E Y 9 5 3 4 ( 1 0 PL.ATE XV 2 7 X2 1 1 MOLT ... OF TTl R FOR1\ [A'l'IO N

PAGE 206



PAGE 207

SEDIMENTARY ROCKS. 177 979 1 (W 315 F). Arrondissement of Las Cahobas, same locality as 9779. T. W. Vaughan and W. P. Woodring, collectors. March 1, 1921. 9778 (W 207 F). Arrondissement of Las Cahobas, south limb of Thomonde antic line trail from Las Cahobas to Thomonde, about 2.5 kilometers south southeast of Thomonde, just north of ridge farmed by base of Las Caho bas f orma tion, top of Thomonde formation. W. P. Woodring, collector. January 15, 1921. 9785 (W 214 F). Arrondissement of Las Cahobas, south limb of Thomonde anticline, trail from Thomonde to Mirebalais, about 7 kilometers southwest of Thomonde, altitude 395 meters above sea level, upper part of Thomonde forma tion. W. P. Woodring, collector. January 17, 1921. 9780 (W 209 F). Arrondissement of Las Cahobas, north limb of Thomonde anticline, trail from Thomonde to Thomassique, about 2.5 kilometers northeast of 'l'homonde, just south of ridge formed by base of Las Cahobas formation, top of Thomonde formation. W. P. Woodring, collector. January 16, 1921. 9781 (W 210 F). Arrondissement of Las Cahobas, north limb of Thomonde anticline about 2.8 kilometers east-northeast of Thomonde, upper part of Tho monde formation. W. P. Woodring, collector. January 16, 1921. 9929 (W 235 F). Arrondissement of Las Cahobas, trail from Hinche to Tho monde, ridge formed by base of Las Caho bas formation about 4 kilometers north northwest of Thomonde, top of Thomonde formation. W. P. Woodring, collector. January 26, 1921. 9907 (W 190 F). Arrondissement of Las Cahobas, first high bluff on left bank of Riviere l' Ayaye above mouth, about a kilometer above trail crossing near mouth, altitude 185 meters above sea level, middle part of Thomonde formation. W. P. Woodring, collector. January 10, 1921. 9908 (W 191 F). Arrondissement of Las Cahobas, low bluff on left bank of Riviere l'Ayaye, about 0.8 kilometer above trail crossing near mouth, altitude 175 meters above sea level, middle part of Thom.onde formation. W. P. Woodring, collector. January 10, 1921. 9926 (W 231 F). Arrondissement of Hinche, road from Ma.lssade to Hinche, about 3.5 kilometers southeast of Maissade, altitude 330 meters above sea level, middle part of Thomonde formation, or Ma1ssade tongue. W. P. Woodring, col lector. January 25, 1921. 9948 (W 323 F). of Hinche, trail from Thomassique to Ccrca la-Source, first crossing of Riviere Bonde, middle or lower part of Thomonde formation. W. P. Wood1 ing, collector. March 13, 1921. 9935 (W 243 F). Arrondissement of Las Cahobas, north limb of Thomonde anticline, left bank of Riviere Blanche, about 2.5 kilometers northwest of Thomonde, altitude 310 meters above sea level, lower part of Thomonde formation. W. P. Woodring, collector. January 29, 1921. 9904 (W 186 F). Arrondissement of Las Cahobas, trail from Las Cahobas to Petit-Fond, about 2 kilometers northwest of Las Cahobas, altitude 180 meters above sea level, lower or middle part of Thomonde formation. W. P. Woodring, collector. January 9, 1921. 9719 (W 256 F). Arrondissement of Hinche, west limb of Fond Bleu dome, trail from Maissade to Madame Joie, about 6 kilometers west of Mai'ssade, altitude 310 meters above sea level, lower part of Thomonde formation. W. P. Woodring, collector. February 3, 1021. 9939 (W 261 F). Arrondissement of Marmelade, trail from Madame Joie to Riviere Blanche, about 3 kilometers southeast of Riviere Blanche, lower part of Thomonde formation. W. P. Woodring, collector. February 5, 1921. 12

PAGE 208

-Miocene fossils from Thomonde formation in C e ntral Pla i n. U pp e r zon e Sp ec ie s Hinche. Las C a h o b as Mid d l e z o ne Hinc he. .g .c= = 0 Lower zone ., 'O ., ., s = .g d Ql I i:::l I 00 aj .... d : I , 0 I I ..,;:j . Foraminifera : H o m otrema sp. ,. Sor ites sp. cf. S. americana (Cushman) Orbi culina sp. Corals: cf. D. Deltocyathus sp. Parac y athus ? s p ..... italicus (Mic h e l otti) Stylopb ora minor Duncan ........ ...................... Styl o phora n. s p ........................................... Stylophora Stylopbora sp. Stylophora sp. Stylophora sp. A aterosmilia sp. n. a. b. c. d sp. Antillia dubia (Duncan) ... ............................... Antillia sp., apparently new .............................. S olenaatrea bournoni Milne Edw ards a n d Haime ........... Siderastrea sider e a (Ellis and Solander) ................... Psammocora n. s p. b, also from Miocene o f Trinida d ...... Porite s sp. aff. P. furcata Lamarck ........................ Porites sp., apparently new ...... Porites sp. Porites s p. ind e t ........................................... Goniopora sp cf. G. jacobiana Vaughan ......... 9946 I 9946 I 994 7 x x x x x x x x x . x x 9782 9779 9791 9780 9781 x x x x 9785 9 929 9778 . x 990 7 x x x x x x x x x 9908 9926 x x x x 99!8 9935 9719 9939 9904 x x . x ...... 00 0 Q 0 l:s:j q bj c 0 tJ:1

PAGE 209

Bryozoa: CJupuladria canarfensis Busk ........................ Oallopora d11merillii Savigny-Audouin ................ Mamillopora tuberosa Canu and Bassler ................... Metrarabdotos sp. Mollusca: Pteropoda: Styliola sp. Cavolina sp. . . . . . . . . . . . . . . . . . . . . . Ga stropoda : Actaeon ? sp .......................................... Acteocina n. sp. a cf. A. subbullata Pilsbry and Johnson Acteocina n, sp. b, also from Oercado formation of Dominican Republic Acteo cina sp. Oylichnella n. sp. a cf. C trictumtritonis Maury ...... Oyli cbnina n. sp. a cf. C. chipolana (Dall) ........... Retusa sp. cf. R. fos sil i s Pilsbry and Johnson ........ Volvula sp. cf. V. eylindrica Gabb ................... Ringicula n. sp. a cf. R. semilinata Dall ........... Bullaria paupercula (Sowerby) ? B11llaria sp. Terebra (Strioterebra) sp. a cf. T. (S.) berlinerae Maury Terebra (Strioterebra) sp. b cf. T. (8.) b erlinerae Maury Terebra (Strioterebra) sp. c cf. T. (S.) cambiars oi Maury Terebra (Strioterebra) sp. d cf. T. {S.) cambiarso i Maury Terebra (Strioterebra) sp. e cf. new species from Cer- cado formation Terebra ( Strioterebra) Oonus veatchi Oleson, sp. also from Baitoa forn1ation of Dominican Republic Conus sp. a cf. 0. furvoides bracbys Pilsbry and Johnson Conus n. sp. a cf. C. williampbbi Maury .............. Conus sp. b cf. new species from Oercado formation ... Conus ep. c cf. 0. burckhardti B
PAGE 210

Miocene fossil.s from Thomonde formation in Central Plain-Continued. Species. Surcula (Pleurofusia) sp. cf. S. (P.) paraservat.a Gard ner (Mss.) Turris (Pleuroliria) sp. a cf. new species from Cercado formation Turrie (Plcuroliria) sp. b cf. T. (P.) barretti (Guppy). '' Drillia '' henekeni (Sowerby) ............ '' Drillia '' sp. a cf. D. dorsuosa Pilsbry and Johnson .. ,, '' ,, Drillia Drillia Drillia Drillia Drillia '' sp. b cf. new species from Cercado formation '' sp. c cf. new species from Cercado for1nation '' sp. d cf. D. foveolata Pilsbry and Johnson . '' ,, n. sp. a cf. D. jamaicensis (Guppy) ........ '' ,, i "ll maon sr1parum Jll.aury ..................... '' Drillia '' n. sp. b cf. D. calligona Maury ............. '' Drillia '' sp. e cf. D. wincbesterae Pilsbry and Johnson '' Drillia '' sp. f cf. D. venusta (Sowerby) ............. '' Drillia '' n. sp. c .. '' D '' d r1 1a n. sp. ......................... '' Drillia ,, '' Drillia '' '' Mangilia '' Mangilia '' Mangilia '' Mangilia n. sp. e. sp. g ......... ............................ '' sp. a cf. M. ramondi Maury ........... ,, sp. b cf. M. psi la Bush .................. '' sp. c cf. unnamed Cercado sp .. '' sp. d cf. unnamed Cercado sp . '' Mangilia '' sp. e cf. unnamed Gurabo sp .. Hinche. 9945 x x x x x x x x x x x x 9946 x x x x x x x . .. x x x 9947 Upper zone. 9782 9779 9791 9780 x ? Las Cahobas. 9781 9785 9929 9778 9907 x x x x x x x x x x x Middle zone. Hinche. 9908 9926 9948 x x x x :a .g a> CIS 9936 Lower zone. ,.Q Q '"4 t= 9719 .., "" CIS a ... )1 9939 f = .g d 99()4 00 0 al toC 0 0 b:j t:rJ C1 bj H c 0 b:j H ::l

PAGE 211

'' llangilia '' '' Mangilla '' '' Mangilia '' '' Mangilia '' '' Mangilia '' '' Oythara '' '' Oythara '' '' Oythara '' IP I ..................................... sp. g. sp. b. ep. 1. sp. J. sp. a cf. C. cercadica Maury .............. ep. b cf. C. mucronata Guppy ............. sp. c. '' Oythara '' sp. d ..................................... '' Oytbara '' sp. e cf. 0. gibba Guppy .................. Glyphostoma sp. cf. G. dentifera Gabb ................ M.icrodrillia Oancellaria sp. sp. n. cf. C. rowelli Dall, also from Baitoa formation Oliva n. sp. cf. 0. gatunensis Toula, also from Bait.oa formation Oliva sp. a cf. 0. dimidiata Pilsbry and Johnson ... Oliva sp. b cf. 0. cylindrica Sowerby .................. Olivella (Lamprodroma) n. sp. a cf. 0. muticoides Gabb Olivella (Lamprodroma) n. ep. b cf. 0. indivisa Guppy. Olivella (Lamprodroma) sp. a ......................... Marginella nugax Pilsbry and Johnson ................ Margin el la n. sp. a cf. M. maoensie Maury ............. Marginella n. sp. b cf. unnamed Baitoa sp ...... ...... Marginella n. sp. c cf. M. oryzoides Gardner (Mss.) .. Marginella n. sp. d cf. M. eowerbyi Gabb .............. Marginella sp. a cf. M. amina Dall ..................... Marginella (Gibberula) sp. a cf. M. (G.) cercadensfts Maury Marginella (Gibberula) sp. b cf. M. (G.) cbondra Gard-ner (Mss.) Clo sia n. sp. a cf. unnamed Gurabo sp ...... Mitra (Cancilla) n. sp. a cf. M. (C.) henekeni SowerbJ. Mitra (Cancilla) n. sp. b cf. M. (0.) longa Oabb, also from Baitoa formation ............................ Mitra (Fusimitra) Gardner (Mss.) n. sp. c cf. M. (F.) manrilopsis Mitra ? sp .. x x x x x x x x x x x x x x x x x x ? x x . .. x x x x x x x x x x x x x x . x x x x x x x x x x x x x x x x x x ? x x x ? x x x x f x x x x '(fJ s t-3 0 Pi 00 ..... 00

PAGE 212

Miocene fossil.a from Thomonde fo1nation in Central Plain-Continued. Species. indet. Mitra ? sp. Strigatella ? sp .... Xancus rex Pilsbry and Johnson, also from Baitoa formation Xancus sp. cf. X. praeovoidee Maury ................ Fasciolaria sp. cf. F. kempi Ma11ry ... .......................................... Fasciolaria sp. Peristernia sp. cf. P. insula Olsson. Fusinus ? sp ..... Melongena consors (Sowerby) .......... Melongena orthocantha Pilsbry and Johnson ........... Solenosteira n. sp. cf. S. inornata Dall ................ Phos costatus Gabb, also from Baitoa formation ....... Phos semico status Gabb, also from Baitoa formation .. Phoe sp. a cf. P. elegans Guppy ...................... Phos sp. b cf. P. chi polanus Dall ...................... Phos sp. .Alectrion Alectrion c ............................................. cercadensis Maury ..... n. sp. cf. A cercadeneis Maury. n. sp ................... Nassarina Metula sp. Oolumbella cf. M. cancellata Gabb. (Conidea) n. sp ........................ Hinche. 9945 9946 I 9947 x x x x x x x x x x x Upper zone. 9782 9779 9791 9780 x x x x x x x I I I Las Cahobas. 9781 9785 9929 x x x .. _.... .. x 9778 ? x 9907 x x x x x x x x x x Kiddle zone Hinche. 9908 9926 9948 x x x x x t t e A x x x x .g 0 9936 _. ... ... ... 0 Lower zone. 41 -5 .E = 9719 9939 = .g d 99M ....... 00 t-t 0 Q 0 t;J 0 0 ej

PAGE 213

Atilia n. so. cf. unnamed Cercado sp Atilia (Oolumbellopsis) Atilia (Col11mbellopsis) e.xilis (Gabb). n. sp. a cf. A. (C.) exilis (Gabb) Atilia ( Columb e llopsis) n. sp. b cf. unnamed Cercado and Gurabo s p ..... Atilia (Columbellop s is) Atilia ( Columbellopsis) Strombina n sp. a cf. n. c. sp. sp .. s. caribaea Gabb, also from Baitoa formation Strombina n. sp. c cf. S. cypbonotus Pilsbry and Johnson Strombina sp. a ........................................ Strombinella acuformis Dall n. subs p ............ Murex sp. a cf. M me ssorius Sowerby ............ Mure x (Phyllonotus) compactus Gabb ................ O cinebra sp. Muri cidea s p. Typhis sp. cf. T. alatus Sowerby ................... Oymia henekeni Maury, als o from Baitoa formation ... Cymia henekeni tectiformis Pilsbry ................ Morum sp. cf. M domingen s e (Sowerby) ............ Mal ea elliptica Pilsbry and Johnson ................... Cypraea n. sp. cf. C. raymondrobertsi Pilsbry .... Oypraea sp. ind et ..................................... S. proximus Sow erby. Strombus sp. cf. Strombus s p. Orthaula x aquadillens i s Maury ..................... Cerithium (Vulgocerithium) n. sp. cf. 0. (V.) russeli M aury Cerithium ( V ulgo cerithi um) sp. Cerithium (Ptyc ho cerithium) n. sp. Cerithium (Ptyc ho cerithium) n. sp. a. b. Clava alajuela Ol s son, als o from Baitoa formation ..... Con o c erithium sp. Potamides dentilabris ( Gabb) ....................... Potamides n. sp. a cf. P. gastrodon Pilsbry and Johnson Potamides caobasen s ia P ilsbry ......................... Ceritbidea sp. a ................. .... ................. x x x x x x x x x x x x x x x x x x x x x x x x I t x x x x x t I .... ? x x x I x x x x . ,. x e I x x I e t I x x x x ..... x . . ,. . . ? ... t I t t x x . x x t I . ,. x x x I x x x x ? x x x x x x x x x x .... x x x x x x .... x x t t I t x x . m ti;j t:=' z t:d 0 0 m ......,. 00

PAGE 214

184 GEOLOGY OF THE REPUBLIC OF HAITI. nqoquo 8-S'J 41 = apw1amJ'8Ji s g ... QJ aq:>tiJH s B'8'J 0) 00 -d4 x G> ..d C.> G> i:: = fO ..... 0 l:I1 N G> "O "O x x x x x ..... )t Cl) i x x x x x x ::I s:= ..... 0 0 ,.g -.... 00 t- x x s x ..0 0 s 0 I'll t- .5 0 s QJ i:: 0 x x ,.. s QJ Q, Q, t:J t- x x s 0 &] x x x x s ] 0 0 t- g QI C.> x x x x x x x x x .e = x x x x x x x x x x x x x x 0 .... ..es 0 :! (.) Cl 0 I I I .= I I I I d 0 Cl I I I Q, "O - I Q, 0 = I I I l>l ... .... clS I I Q, I l>l ..... Q, l>l I ::s "O "d ... 0 s:::: = .0 I'll '-" clS I ..... ltl l>l l>l I I ::s rn ,... I I I ,.Q ,.. Q.> ... ... I I I - .!:9 ..... Ul Q.> ..0 ..... Q, ...... .... I I'll .... C.> f.ID .... ... .g I P-4 QJ ..0 Q) tl1 I > a 0 clS !! )t c s .... CllS o:S 1112 s s G> I I C.> o:S ,.Q c.> "O "O I 't 0 CV 0 I ..0 t> .... "O fE' 0 r.o f: fE' c.. clS .0 C.> e 0 0 0$ ... .0 0 c.. G'l llM Q,) tO -t> Q, Q, C'.IQ ::s o:S 5 m I O'} O'} a> m c.. c.. ..d .... c.. C1J ;; fE' Q, 0. ..... .... ..... .... CiO r.o C.> f: o:S m & I'll rJ.) = r.o .... r.o tlQ .c s:l o:S clS Q.) Q, c.. o:S o:S cd .... "O "C 0 0 0 to l'tl ... g -s ... ... ,.. ::s 0 .... ..... .... ... .... ..... 0 j _g d ::s $ $ :S :S :S :S .d '3 .... 5 +J o:S ::s ::s -a .... ...... ..... ..... ...... ,.. .... c.. 0. .c Q, t: ... .... ..... .... ..... ..... ..... Q, "O "O ,.. t ... t ... ... .,.... o:S ... Q.> .... ..... 0 0 ... Q) ..... ..... .... ..... ..... QJ CV QJ G> E:: ...... QJ 41 ::3 d 0 0 i:Q i:Q i:Q 0 0 0 0 00 88 rn tl.t 8 8

PAGE 215

Caecum ap. Jleioceras ap. cf. Meioceras n. sp. M. constrict11m (Gabb) a ............. Hemisinus sp. cf. H. truncatus (Gabb) ................ Architectonica sp. cf. A. quadri9eriata (Sowerby) ..... Architectonica stonemanae Maury ................ Architectonica sp. a cf. 11nnamed Cercado sp ......... Architec'tonica si;>. b ............ Fossa.rue (Isapis) ap. a ..................... Risaoina crassilabris ( Gabb) ..................... Rissoina sp. a cf. R. striatacostata d'Orbigny .......... Riseoina ep. b cf. R. multicostata Adams ............ Rissoina sp. c cf. R. sagriana d'Orbigny .............. Riasoina sp. d .......................... -........ Tryonia sp. a ............................... T 'ryonia sp. b ..................................... Tryonia sp. c ................................... Orepidula ep. Crucibulum sp. cf. 0. chipolan11m Dall ................ Orucibulum sp. cf. C. piliferum Guppy ................ Orucibulum sp. Oalyptraea sp. Oal,yptraea ? sp .................. Natica sp. cf. N. canrena Linnaeus ............... Natica sp. a. Natica sp. b ........................ ...... ... Natica ep. indet ...................... Natica (Stigmaulax) sp. cf. N. (S.) sulcata Born. Polinicea subclausa (Sowerby) ...................... N everita nereidis Maury ..... Amauropsis sp. cf. A. guppyi Gabb .................. sp s. chipolanum Dall. Amauropsis ? Sin um sp. cf. Sinum sp. Epitonium Epitonium Melanella Melanella sp. ep. sp. sp. a. b. a. b .. .. x x x x x x x x x x x x x x x x x x x x . x x x x x x x x x x x x x x x . x x x ? x x x x x x x x x x x x x x .. x x x x x x x . f x x 00 l:tj t:i H J:( t=rJ z td 0 c 00 r-' 00 ..

PAGE 216

Miocene fossils from Thomonde formation in Central Plain-Continued. Species. Melanella Melanella Melanella sp. Strornbif orrnis s p. Strombiformis sp. sp. c. sp. d. e ... a. b P,yram idella sp. a .......................... P,yramidella sp. b ......................... Pyramidella sp. Turbonilla sp. a ....................................... Turbonilla sp. b ..................................... Turbonilla sp. Turbonilla Turbonilla Turbonilla Turbonilla Odostomia Od ostomia sp. sp. sp. sp. sp. sp. c. d. e. f. a ................ b t e e e e I t t t t t t t t t t t t t t t t t t t t t t t t e t t t e t Turbo sp. cf. Astraea s p. Phasianella punctata T. dominic ensi s Gabb t t t t I t t t t t t e e t t t I I t (Gabb) Phasianella sp. Neritina (Smaragdia) viridimaris Maury ...... Neritina (Puperita) figulopi cta Maury ............ Solariella ? sp ..... Hinche. 9945 9946 I 9947 x x x x x x x x x x x x x x x x x x x x x x x x x x x Upper zone. Laa Cahobas. 9782 9779 9791 9780 9781 0785 9929 9778 9907 x x x x x x x I t t e x x x .. x x Middle zone. Hinche. 9908 9926 9948 x ? x x x x x x x x .0 0 9936 Lower zone. G> .ci CJ c 9719 . 9939 x = .g 0 9904 t I lo-4 00 a> 0 0 0 bj td 0 0 bj P:1

PAGE 217

Liotia sp. a. . . . . . . . . . . . . . . . . . . . . Lfotia sp. b ............................................ Liotia sp. c. Liotta sp. d ............................................ Oirculua pentagona (Gabb) sp. depressum Solariorbis Teinostoma Teinostoma Teinostoma sp. Teinostoma sp. Pseudorotella sp. (Gabb) sp. a. b. c. Discopsis derbyi Maury n. subsp ...................... Fissuridea sp. Fiasuridea sp. Scaphopoda : a. b. Dentalium dissimile Guppy ........................... Dentalium pyrum Pilsbry and Sharp .................. Dentalium sp. group of D. antillarum d'Orbigny ...... Dentalium sp. Dentalium ? iP .............. Oadulus sp. a cf. 0. phenax Pilsbry and Sharp ......... Cadulus sp. b cf. C. colobus Pilsbry and Sharp ........ Cadulus sp. c cf. C. elegantissirnus Pilsbry and Sharp .. Cadulus n. sp. cf. C. simrothi Pilsbry and Sharp ...... Pelecypoda : Nucula tenuisculpta Leda n. sp. a cf. L. Gabb .... extricata Pilsbry and Johnson. Leda T sp ............................................. Arca sp. cf. A. yaquensis Maury ....................... Arca yaquensis Maury n. eubsp ....................... Arca sp. cf. A. bowdenensis Dall ................. Barbatia sp. cf. B. marylandica (Conrad) ............. Barbatia (Acar) sp, cf. B. (A.) domingensis (Lamarck) Scapharca tolepia Dall n. subsp., also from Baitoa f f e f t t e e f e e e t f e a I a e formation Scapharca corcupidoni11 a .... Scapharca sp. Scapharca sp. b cf. Maury n. subsp. s. actinophora Dall. x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x . . . .. . x x . . .. .. .... ... ... x ? e t t I . .. . . . . t I x x x x x x . x x x x x x x . x . x x ? x X . x x x ? x x I e t t x T f T . .. . x ... . . x ... . ' x x . . x f/J ttj tj ..-4 t-3 a t:'d 0 a f/J )o-L 00

PAGE 218

Miocene fossils from Thomonde formation in Central Pl.ain-Continued Upper zone. Middle zone. Species. Hinche. Las Cahobas. Hine he. = .g Lower zone. Q> -s c .... = I> )1 994-0 I 9946 I 9947 9'182 9779 9791 . . .... .... Scapharca sp. Scapharca sp. c cf. S. hypomela Dall. d .. Scapharca cibaoica Maury n. subsp .................... Scapharca sp. e cf. S. golfoyaquensis Maury .......... Scapharca sp. f cf. S. halidonata Dall .............. Scapharca chiriquiensis ( Gabb) ..................... Scapbarca chiriquiensis websteri (Pilsbry) ..... Fossularca sp. cf. F. adamsi Dall .............. Olycymeris acuticostata (Sowerby) n. subsp .......... Olycymeris sp. Pinna sp. Perna sp. Ostrea haitensis Ostrea sp. cf. 0. Sowerby. folium Linnaeus ........... Ostrea sp. Pecten (Pecten) soror Cblamys ( Chlamys) Oabb. sp Oblamys (Aequipecten) sp. a cf. unnamed Oercado sp .. Ohlamys (Aequipecten) sp. b ........................ Chlamys (Aequipecten) sp. c cf. 0. (A.) thompsoni (Maury) Ohlamys sp. d cf. Cercado sp .. Amusium (Amusium) sp ........ Plicatula sp. x x x x x x x x x x x x x x x x x x x x ? x ? x 9780 9781 9785 9929 9778 9007 9908 0026 9948 9936 9719 9939 x x x x x x x x x x ? x x x x x x ? x .. x x \ x x .... = i 3 9904 00 00 t'4 0 0 t:;j 0 b:I s:: a 0 1-:i:j t--4

PAGE 219

Anomia sp. ct. A. indecisa Dall ........................ M.ytilopeis sp. Orassinella microdelta Pilsbry and Johnson ... Orassinella n. sp. a .. Oraseinella sp. cf. 0. bowdenensia (Dall) Oyrena (Pseudocyrena) n. sp ..... Obama sp. cf. 0. involuta Guppy ............... Ood.akia ? sp .......................................... Lucina sp. cf. L. cbrysostoma Philippi .......... Phacoides (Parvilucina) n. sp. a cf. P. (P.) yaquensis Gabb, also from Baitoa formation ............... Phacoides (Parvilucina) n. sp. b ...................... Divaricella sp. cf. D. prevaricata Guppy .............. Diplodonta sp. Montacuta sp. Sportella ? sp ......................................... Alveinus n. sp. cf. A. rotundatus Dall, also from Bait.oa formation Alveinus Cardi um Cardi um sp. (Cardium) n. sp ..... (Trachycardium) sp. a. - Cardium (Trachycardium) sp. b. Cardium (Trachycardium) sp. c. Oardium (Trigoniocardia) sp. cf. 0. (T.) aminense Dall Protocardia ? sp ..................................... Dosinia sp. cf. D. chipolana Dall ...................... Dosinia ? sp ........................................... sp. Olementia ? Transennella Macrocallista Oallocardia ? sp. ? sp so ...... Pitar ? (Lamelliconcha Pitar (Hyphantosoma) ?) sp. # sp. cf. P. (H.) car base us (Guppy) Oytherea sp. cf. C. tarquinia (Dall) ................... Antigona (Ventricola) sp. cf. A. (V.) blandiana (Guppy) Ohione sp. cf. 0. woodwardi (Guppy) x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ..... ? x ...... ? x x t x x x x ? ? x x x x x x x x ? t x x t x x x x x x x ? ... 00 t!tj t=' ttj z t:d 0 c f7J 00 "'

PAGE 220

Miocene fossi,l3 from Thomonde fonnation in Central Plain-Continued. Upper zone Species. Hinche. Laa Cahobas. Middle zone. Hincbe. as .g a Lower zone. S ., r--1 .d ., t.> s = .g ., -., ., a:ilS:: . Chiona sp. cf. 0. socia Pilsbry and Johnson ........... Anomalocardia ? sp ................................... Tellina sp. Tellina sp. Tellina sp. Tellina sp. Macoma sp. Macoma ? a .. b. c ........................................... indet. sp. ind e t .................................... Mac oma (Psammacoma) sp. cf. M. (P.) olivella Dall . Metis ? sp ............................................. Strigilla sp. cf. S. pisiformis (Linnaeus) Tagelus sp. Sphenia sp. Ervilia sp. . . . . . . . . . . . . . . . . . . . . . . Corbula sp. a cf. 0. cercadica Maury. Corbula sp. b cf. 0. sericea Dall ...................... Oorbula sp. c cf. 0. knoxiana fossilis Pilsbry ... Corbula sp. d cf. O. domini c ensis Gabb ...... Corbula (Aloi dis ) vieta Guppy ............... Corbula ( B othrocorbula) viminea Guppy ............. Rocellaria sp. cf. R. r otunda (Dall) ............... Decapod Crustacea : Portunus (Portunus) baitensis Rathbun ................... .. sp. ind et ........................................ Panopeus Panopeus Partbenope ? sp .... ? sp . I I t t \ 1 9945 9946 x x x x x x x x t x x x x x x x x x 9947 9782 x ? x x 9779 t 9791 9780 9781 9785 9929 ..... x x x x . x 9778 9907 x x ? x x x x x x x x x x x x x x x 9908 x x ? x x x x x x x x x 9926 . 9948 9935 9719 9939 9004 x x x x r t . x x c.o 0 t-4 0 Q 1-4 0 b;1 d b:1 0 0 .....

PAGE 221

SEDIMENTARY BOCKS 191 Fossils, Matssade tongue. Mollusks and decapod Crustacea are the most abundant fossils obtained from the Ma1ssade tongue of the Thomonde formation. No corals were fo11nd in the Ma!ssade tongue, as the deposits were laid down near the mouths of streams that flowed across coastal-plain swamps. Some of the characteristic mollusks are shown on Plate XVI. The collections of mollusks are divided into three lists to show that some of the m contain a marine fauna, others a mixed marine and brackish fauna, and others a brackish fauna. The list on pages 194-195 shows that the ma.rine fauna is ess entially the same as the fauna in the middle part of the Thomonde formation in the southeastern part of the plain, 44 per cent of the spe c i e s b eing identical. The most striking identical sp e cies are several species of Tectibranchs and Pleurotomids, several small species of Marginella, Melongena orthocantha Pilsbry and Johnson (Pl. XVI, fig. 5), Phos costatus Gabb (Pl. XV, figs. 4, 5), Turritella sulcigyrata Pilsbry and Brown, and Scapharca chiriquiensis Gabb (Pl. XVI, figs. 6-8). Some of the species in the Maissade tongue are known also in the upper part of the Thomonde formation in the southeastern part of the plain, but Xancus rex Pilsbry and Johnson, Phos semicostatus Gabb, and Or thaula x aguadill e n s is Maury are not found in the Ma!ssade tongue. The absence of some of the se species and of .the Cones may be due to environ mental conditions, but the evidence seems to indicate that the lower and middle parts of the Maissade tongue, which contain the marine fauna, are the equivalent of the middle part of the Thomonde formation. The collections containing a mixed marine and brackish fa11na a1e listed on page 196. The dwarf race of Phos costatus Gabb and species of Hydrobia and Tryonria indicate a brackish element. The same dwarf race of Phos costatus was found in the marine beds, but the collection f1om station 9717, the largest marine faunule, includes typical specimens of this species. The collections containing a bracki s h fauna are listed on page 197. Potamides tipp enh.aue ri Woodring and Mansfield (see p. 611 and Pl. XVI, figs. 3, 4), M elongena orthocantha Pilsbry and Johnson (Pl. XVI, fig. 5), S capharca chi1iquiensis ( Gabb) (Pl. XVI, figs. 6-8), and subspecies websteri ( Pilsbry) (Pl. XVI, figs. 9-11), and Ostrea sp., are the most abundant mollusks in the Ma!ssade tongue. Whole beds consist of Scapharca chiriqui e nsis, and others are made up of oyster shells. Pota mides roumaini Pilsbry (Pl. XVI, figs. 1, 2) is remarkably abundant at station 9710. Stations in Central Plain, M aissade tongue (Miocene). 9710 (W 246 F). of Hinche, right bank of Riviere Frio about 100 meters above crossing of road from Maissade to Hinche. W. P. collector. February 1, 1921. 9711 to 9715. Arrondissement of Hinche, right bank of Riviere Fond Gras, about a kilometer in air line above junction with Riviere Canot. (See p. 172 for

PAGE 222

PLATE XVI. SoME CHARAcrERISTIC MOLLUSKS OF THE MAissADE TONGUE AND OF THE ARTIBONITE GROUP. FIGURES 1, 2. Potamides roumaini Pilsbry. Maissade tongue, U. S. G. S. station 9710. U. S. N. M. catalogue No. 350580. Fig. I. Adult specimen, X 1. Fig. 2. Young specimen, X 2, showing sculpture of early whorls. FIGURES 3, 4. Potamides tippenhaueri Woodring and Mansfield (p. 611). Ma!ssade tongue, U. S. G. S. station 9725. Fig. 3. Tyoe, X 1. U.S. N. M. catalogue No. 350581. Fig. 4. Young specimen, X 2, showing sculpture of early whorls. U. S. N. M. catalogue No. 350582. FIGURE 5. Melongena orthocantha Pilsbry and Johnson, X 1. Maissade tongue, U. S. G. S. station 9732. U.S. N. M. cat&.!3gue No. 350583. FIGURES 6, 7, 8. Scapharca chiriquiensis (Gabb). Three views of same specimen, X 1. Thomonde formation U.S. G. S. station 9908. U. S. N. M. catalogue No. 350584. FIGURES 9, 10, 11. Scapharca chiriquiensis websteri (Pilsbry). Three views of same specimen, X 1. Las Cahobas formation, U. S. G. S. station 9930. U. S. N. M. catalogue No. 350585. 192

PAGE 223

REPUBLIC OF HAITI SUR\'EY l x2 7 6 X2 l ( ) PLATE XVI ') - 9 11 8 CHARAC1. 'ERISTIC OF TIIE l\lAiSS .\DE TONGUE AND OF TIIE ARTIBONITE GROUP.

PAGE 224



PAGE 225

SEDIMENT.ARY ROCKS. Miocene marine fossils from Central Plain (M a"issade tongue). Hinche. I Species. 193 Marme lade. 9717 9713 9714 9715 9722 9726 Mollusca: Gastropoda : Actaeon sp. cf. A. riomaensis Maury .................... Acteocina n. sp. a cf. A. subbullata Pilsbry and Johnson. Acteocina n. sp. b, also from Cercada formation ......... Oylicbnella n. sp. a cf. 0. trictum-tritonis Maury ....... Ringicula n. sp. b cf. R. dorninicana Maury ............. Bullaria sp. indet ....................................... Turris (Pleuroliria) sp. a cf. new species from Cercado formation ........................................... '' D '' f b ri 1a sp. c n. sp. .............................. '' Drillia '' sp. b cf. new specie1::1 from Cercado formati on .. '' Drillia '' sp. c cf. new species from Cercado formation .. '' Mangilia '' sp. t ....................................... '' M '' ang1 1a sp. g ................... '' M '' d t angi 1a sp. 1n e ................................... Olivella (Lamprodroma) sp. cf. 0 limonensis Olsson ..... Marginella nugax Pilsbry and Johnson ................. Marginella n. sp. a cf. M. maoensis Maury ............... Marginella n. sp. c cf. M. oryzoides Gardner (Mse.) .... Marginella sp. b cf. M. latissima Dall ................... Marginella sp. c ..................................... Marginella (Gibberula) sp. b cf. M. chondra Gardner ( M ss.) ......................................... Marginella (Gibberula) sp ......................... Mitra (Cancilla) n. sp. a cf. M. (C.) henekeni Sowerby .. Mitra (Fusimitra) n. sp. c ct. M. (F.) mangilopsis Gard-ner (Mss.) .......................................... Mitra (Fusimitra) sp .................................. Fusinus sp. ........................................... Melongena ortbocantha Pilsbry and Johnson ............ Pbos cos ta. tus Gab b ...................................... Phos costatus Gabb (dwarf race) ...................... Strombina n. sp. b, cf. S. caribaea Gabb ................. Murex sp. a cf. M. messorius Sowerby ................... Cerithium (Vulgocerithium) sp. cf. C. (V.) dominicense Gabb ........................................... Oeri thi um ? sp ........................................ Clava alajuela Olsson ................................. Potamides tippenhaueri Woodring and Mansfield ........ Cerithidea sp. a ......................................... Bittium sp. b ................................... ; ....... Bittium sp. ............................................ Cerithiopsis sp. d ................................. .... Triphora sp. a ..................................... Vermicularia sp. cf. V. spirata (Philippi) .............. Turritella cal ostemma Pilsbry and Brown ............. Turritella sulcigyrata Pilsbry and Brown ............... Caecum sp. cf. 0. anellifer Pilsbry and Johnson ......... Meioceras sp. cf. M. constrictum (Gabb) ............... 13 x ? x x x x x ? x x x x x x x x x ? x ? x x x x x x x x x x x x x x . . . . . .. ... x x x x x x x x x x x x x x x x .. x x x x x x x x x x x x x x x

PAGE 226

194 GEOLOGY OF THE REPUBLIC OF HAITI. Mioc e ne marine fossils from C e ntral Plain (Matssade tongue)-Continued. Species. Mollusca-(Cont'd) : : a .. s p. b. Me ioceras n sp. Fos s arus (Isa p is) Hydrobia sp. I f f f f f f f f I e f f f f f f f f f f f f f f I f f f f f f f f f f f f Tryonia sp. d ........................................... Orepidula sp. Epitonium sp. Melanella sp Turbonilla sp. e. Turbonilla sp. g ........................................ Turbonilla sp. h ......................................... Odost omia sp. c ....................................... Od os 'tomia sp. d ...................................... Odostomia sp. Odostomia sp. e. t ...... N eritina ( Smaragdia) viridimaris Maury ................ Neritina (Puperita) figulopicta Maury ................. Liotia sp. a .............................................. Oirculus sp. Vitrinella sp. Teinost oma depre ssum (Gabb) Dis copsis sp. Scaphopoda : Cadulus sp. b c f C. colobus Pilsbry and Sharp .......... Pelecypoda : Nucula tenuisculpta Gabb ........ ...................... Leda n. sp. a cf. L. extricata Pilabry and Johnson ....... Leda n. sp. b ......... S capharca chiriquiensis (Gabb) S capharc a ep. O strea sp. Amusium (Pseudamusi11m) sp. cf. A. (P.) guppyi (Dall) Anomia sp. Modiol us ( Brachydontes) sp. n. sp. a. Mytilop sis sp. Crassinella Cyrena sp. Phacoide s (Parvilucina) n. sp. b ...................... Obi one sp. cf. 0. socia Pilabry and Johnson ........... Ervilia sp. Tellina sp. Ma c oma ? sp ....................................... f ossllis Pilsbry .. Corbula sp. c cf. C. knoxiana Rocellaria l!lp. cf. R. rotunda (Dall) . Decap o d Crustacea : Portunus (Portunus) baitensis Rathbun .................. 9717 x x x x x x x x x ? x x x x x Hinc he. 9713 x x x x x x 9714 x x x x x x x x x x x x x x x x .. 9715 x x x x x x Marme lade. 9722 x x x x x x x x x x x x x x x x x x x x x x

PAGE 227

SEDIMENT.ABY ROCKS. 195 Miocene marine and brackish-water f ossil.s from C en tr al Plain (M a'issade tongue). Species. Mollusca: Gastropoda : sp. b. cf. R. Ringicula n. cana Maury domini-. . . ........... Ringicu]a sp. ................... Olivella (Larnprodroma) sp. a ... Melongena consors (Sowerby) ... Melongena orthocantba Pilsbry and Johnson Phos costatus Gabb (dwarf race) Atilia sp ... Strombina n. sp. b cf. S. caribaea Gabb Strombina sp. b ................. Murex sp. a cf. ll. mesaor1us Sowerby Potamides tippenhaueri Woodring and Mansfield Bittium sp. b ................... Bittium sp. d .................... Bittium sp. Bittium e .................. sp. f. Bittium sp. ................... Hydrobia sp. Tcyonia sp. d .................... Crucibulum sp. ............... Turbonilla sp. N eritina (Puperita) figulopicta Maury Pelecypoda: Leda n. sp. b ..................... Scapharca cbiriquiensis Scapharca cbiriquiensis (Gabb). websteri (Pilsbry) Ostrea sp. ?tfytilop sis sp. Phacoides (Parvilucina) n. sp. b. Cbione sp. cf. C socia Pilsbry and Johnson Tellina sp. Corbula sp. c cf. C knoxiana f os-sils Pilsbry ................ Hinche. 97!1 x x .. . x x x x x x x x x 9716 x x x x x x x x x x x 9723 x x x I I x . x x x x Marmelade. 9724 9725 9727 9728 ? x x x x x x x x x x x x x x x 9729 x x x 9732 x x x x x x x x x x x 9733 x x x x x

PAGE 228

196 GEOLOGY OF THE REPUBLIC OF HAITI. section.) W. P. Woodring, coll e ctor. F e bruary 2, 1921. The beds in the section that yielded the fo s sil s are 9711 (W 248 F), b e d 14; 9712 (W 249 F), bed 12; 9713 (W 250 F), b e d 6; 9714 (W 251 F), b e d 4; 9715 (W 252 F), b e d 3. 9716 (W 253 F). Arrondiss e m ent of Hinche, right bank of Riviere Fond Gras, about half a kilomet e r below junction with Riviere Fond Bl e u. W. P. Woodring, collector. February 2 1921. 9717 (W 254 F). Arrondi s s e m ent of Hinche right bank of Rivier e Fond Bleu, about half a kilometer above junction with Riviere Fond Gras. W. P. Woodring, collector. February 2, 1921. 9718 (W 255 F). Arrondissement of Hinche, right bank of Rivi ere Fond Bleu, about 2.5 kilometers above junc tion with Riviere Fond Gra s. W. P. Woodring, coll e ctor. February 2, 1921. 9722 to 9733. Arrondi s s ement of Marme l a de first high bluff on right bank of Riviere Blanche below gorge form e d by conglom e r a t e s in middl e p art of Thomonde formation. (S ee pp. 170-172 for s e ction ) W. P. Woodring collector. F ebruary 5, 1921. The b e ds in the section tha t yi eld e d the foss il s are 9722 (W 262a F), bed 47; 9723 (W 262b F), bed 46; 9724 (W 262c F), b e d 44; 9725 (W 262d F), b e d 43; 9726 (W 262e F), b e d 36; 9727 (W 262f F), b e d 34; 9728 (W 262g F), bed 32; 9729 (W 262h F), bed 29; 9730 (W 262i F), b e d 24; 9731 (W 262j F), b e d 21 ; 9732 (W 262k F), b e d 6; 9733 (W 2621 F), b e d 2. Miocene brackish-water fossils fro m C entra l Plai n (Ma"issade tongue ) Species. Mollus ca: G astro p oda: Planorbis sp. Bullaria s p. .... .............................. .. . Potamides roumaini Pilsbry ............... ....... ........ Potamides tippe nhau eri W o odri n g and Mansfield ............. A.labina sp. .............................................. .. Hemisinus truncatus (Gabb) .................... . ........... H .. em1s1n\1S sp. ....................... ...................... Hydrobia sp. ................................................ Tryonia sp. e . ................................................ P elec ypoda : Scapharca chiriquie n s i s w e b steri (Pilsbry) .............. Mytilopsis sp. ........................................ LAS CAHOBAS FORMATION. Hinche. Marme lade. 9710 9718 9712 973 0 973 1 x x x ? x x x x x x x ........ x ? x x Jones 1 gave the name ''Las Caho bas beds,'' derived from that of the town of Las Cahobas, to the ro c ks that overlie the Thomonde formation. The type locality is north of Las Cahobas. The Las Cahob a s formation is the youngest Miocene formation in the Centra l Plain and the most exten' sive surface formation in its southeastern part, but at many places in the interior it is concealed by younger stream deposits. It conformably over. 1 Jones, W. F., op. cit., p. 737, 1918

PAGE 229

SEDIMENTARY BOCKS. 197 lies the Thomonde formation but was laid down along a retreating shore line. Its estimated maximum thickness is 700 meters. Along the west and south edges of the plain the Las Cahobas formation is strikingly different in lithology from the underlying Thomonde forma tion, as it consists almost entire l y of coarse detrital sediments, principally conglomerates and coarse cross-bedded sandstones. These deposits are lar gely nonmarine and apparently represent delta and flood-plain deposits that extended far seaward beyond the shore line of Thomonde time. As these ro c ks are harder than the Thomonde rocks they form ridges with steep mountain-facing cuestas Such a ridge formed by the basal rocks of the Las Cahobas formation is a striking feature along the south edge of the plain. (See Pl. XIV, B.) North and northeast of Las Cahobas there are two coralliferous lime stones at and near the bas e of the formation The lower limestone, which is considered the base of the formation, crops out in a low ridge about half a kilometer north of Las Caho bas. It is yellowish and rather soft. Heads of Orbicella and branches of Stylophora and Pocillopora weather out of the rock and lie on the ground. The corals listed on page 20 3 ( station 9906) were collected at this locality. This limestone is exposed on the main trail l eading northeastward from Las Cahobas, about a quarter of a kilometer from the town, where it overlies thin-bedded sa ndstone and sandy siltstone at the top of the Thomonde formation. About 50 meters farther northeast the trail turns eastward and passes over a cora llif erous limestone about 20 meters higher stratigraphically. This limest one is hard and grayish but w ea thers yellowish. It contains many corals, especially Stylophora, Pocillopora, Orbicella, and Cyathom orpha. (See list, p. 203; stations 9909 and 9773.) This limestone is over la .in by shaly b eds and poorly exposed coarse sandstone and conglomerate consisting of small pebbles. Farther to the northeast a conspicuous bed about 40 centimeters thick, consisting of the large massive valves of Ostrea cahobasensis, crops out west of the trail (station 9910). At the place where the Belladere trail branches off the Thomonde trail a bed of dark-gray siltstone contains the mollusks listed on page 204 (station 9772). The overlying beds consist of soft rusty-brown sandstone, white poorly consolidated cross-bedded sandstone, and conglomerate The higher of the two corallif erous limestones forms the c r est of the high ridge north of Las Cahobas. The ruins of Fort Anglais, built by the British during their occupation, stand on this ridge. The wa. lls of the fort are constructed of blocks of this lim estone The corals listed on page 203 (station 9905) were collected loose a long the south slope of the ridge, having weathered out and rolled down the slope. The lower part of the Las Cahobas formation is exposed at many other localities near Las Cahobas. Most of the b eds above the coralliferous limeston es are congloll:lerates or coarse sandstones. The conglomerates contain pebbles of limestone, argillite, and various igneous rocks, all derived

PAGE 230

198 GEOLOGY OF THE REPUBLIC OF HAITI. from the present region of the Massif du Nord. Ostrea bolus was col l ected from coarse poorly consolidated sandstone exposed along the trail from Las Cahobas northwestward to Riviere Artibonite (station 9988). The trail from Las Cahobas to Belladere approximately follows the strike of the rocks for the greater part of the distance. Conglomerate and sandstone crop out in many gullies and ravines. The beds at the base of the Las Cahobas formation form a'' rim rock'' curving around the north westward-pl11nging crest of the Belladere anticline. Large valves of Ostrea cahobasensis were collected from a bed near the base of the formation on the south limb, where the dip is 55 The coralliferous limestones at the base of the formation are not so prominent here as near Las Cahobas. Along the trail from Las Cahobas to Thomonde there are many expos ures of beds higher in the Las Cahobas for1nation. Like the lower beds they consist principally of conglomerate and sandstone, but they contain no fossils. Beds at the base of the f or1na ti on form the '' rim rock '' on the southeastward-plunging crest of the Chamouscadille anticline, which is visible from the trail. Oysters and other marine fossils were collected from beds in the lower half of the formation. (See lists, pp. 203-204; stations 9774, 9775, 9776, and 9916.) Successively lower beds appear at the surface on the ascent of the high ridge overlooking Thomonde. This ridge is part of the '' rim rock '' formed by the basal beds of the formation that curve around the southeastward plunging crest of the Thomonde anticline. Westward and eastward from the trail long, gentle dip slopes on the south limb of the anticline are visible. They are formed by beds of conglomerate. The crest of the ridge consists of conglomerate. Above and below the conglomerate lies a coralli ferous limestone in the same stratigraphic position as the limestones at the base of the formation at Las Cahobas. Here the septa of the corals are dissolved, and the only specimens obtained are casts of Stylophora from the upper limestone (station 9777). Below the lower limestone lies a conglomerate containing small pebbles and below the conglomerate lies sandy siltstone that contains Thomonde fossils (station 9778). The lower part of the Las Cahobas formation is well exposed on the north limb of the Thomonde anticline, where it forms a '' rim rock '' that is a continuation of the ridge on the south limb described above. On the trail from Thomonde to Hinche the beds dip from 70 to 80, or even 90 The following generalized section was measured along this trail: Generalized section of lower part of Las Cahobas formation and upper part of Thomonde formation on trail from Thomonde to Hinche Las Caho bas formation: Est. thickness in meters. 29. Coarse sandstone . . . . . . . . . . . . . . . . . . 1 28. Oyster bed . . . . . . . . . . . . . . . . . . . . . .5 27. Conglomerate . . . . . . . . . . . . . . . . . . . 1 26. Oyster bed . . . . . . . . . . . . . . . . . . . . . .5 25. Sandy siltstone . . . . . . . . . . . . . . . . . . . 2

PAGE 231

SEDIMENTARY ROCKS. 199 Las Caho bas formation Continued. Est. thickness in meters. 24. Oyster bed . . . . . . . . . . . . . . . . . . . . 23. Conglomerate ...................................... 22. Oyst-er bed ......................................... 21. Sandy siltstone ..................................... 20. Oyster bed ......................... ............... 19. Sandstone ......................................... 18. Conglomerate ...................................... 17. Siltstone, marine fossils; station 9927 ................. 16. Conglomerate ...................................... 15. Coralliferous limestone ............................. 14. Conglomerate ...................................... 13. Corallif erous limestone; station 9928 ................. 12. Conglomerate ...................................... 11. Sandstone .......................................... 10. Conglomerate ..................................... .5 1 .5 1 .2 1 .5 5 3 1.2 3 1.6 5 5 3 9. Sandy siltstone ...................................... 2 8. Conglomerate ...................................... 7. Sandy siltstone ..................................... 6. Conglomerate ...................................... Thomonde formation: 5. Siltstone; numerous casts of mollusks; station 9929 .... 4. Conglomerate ...................................... 3. Sandy siltstone ..................................... 2. Conglomerate ...................................... 1. Siltstone 1 2 3 10 .5 2 .5 50 107.5 In this section the corallif erous limestones seem to lie higher above the highest Thomonde fauna than elsewhere. The collection from station 9927 (bed 17 of preceding section) contains fragments of dicotyledonous leaves, probably representing the genus Ficus, in addition to marine mollusks. On the trail from Thomonde to Thomassique a yellowish corallif erous limestone lies above beds of heavy conglomerate and coarse sandstone at the base of the forrnation. The corals listed on page 203 (station 9783) were collected at this locality. Farther northeast on this trail the beds flatten abruptly in the trough of the Central Plain syncline. They consist of alternating beds of conglomerate, sandstone, and siltstone. They are irregularly bedded and seem to be flood-plain and delta deposits. The siltstones have a reddish or greenish color, which is in striking contrast to the bluish color of the marine silt.stones. These are the highest Las Cahobas beds examined. Similar beds crop out in ravines and gullies along the trail from Tho monde to Hinche, north of the ridge described on page 198, but on the upland they are concealed by younger stream deposits. As these higher Las Oahobas beds are stream deposits they can hardly be distinguished from the younger stream deposits in the regions where they lie fiat

PAGE 232

200 GEOLOGY OF THE REPUBLIC OF HAITI. At Hinche, on the right bank of Riviere Guayamouc, a high bluff ex poses in ascending order rusty-brown sandstone, siltstone, and a bed of sandy clay about 60 centimeters thick, which contains fragments of leaves and thin streaks of lignite. These beds seem to l i e horizontal, but the river flows along the strike and they probably dip gently southeastward. Sandy clay containing a few small pebbles overlain by buff clay is exposed in bluffs on the left bank of Riviere Guayamouc opposite Hinche, about 15 0 meters back from the river. Ponderous valves of Ostrea cahobasensis were collected at the foot of the bluff. These beds dip 3 to 4 SW. On the northeast side of the plain the Las Cahobaa formation is lithologi cally indistinguishable from the Thomonde formation, as both consist principally of coarse d etrital sediments and are included in the thick wedge of delta and flood-plain deposits that tapers southwestward away from the old landmass. Thin wedges of sandstone and siltstone near the base of the formation carry a marine fauna (see pp. 20 2, 205; stations 9930 and 9931), showing that in early Las Cahobas time the sea at times extended northeastward over the area where delta and flood -plain deposits were usually being laid down. These marine deposits probably are the littoral equivalent of the coralliferous limesto n es which are not found on this side of the plain. The boundary between the Thomonde Las Cahobas formations on this side of the plain is drawn above the highest beds carrying a Thomonde fauna. Fossils. The Las Cahobas formation inc ludes deposits of different origin, and the collections of fossils obtained from them are correspond ingly diverse. The coralliferous limestones near the base of the formation on the west side of the plain contain an extensive coral fauna of reef facies and apparently represent a horizon from which a reef fauna has not been collected in the Dominican Republic. For this reason the corals of the Las Cahobas formation have a peculiar aspect, some of the species heretofore being known only from older beds and others only from yo11nger beds. Orbicella imperatoris Vaughan was formerly thought to be an upper Oligocene species, and the genus Cyathomorp}ia was formerly thought to be extinct before Miocene time. In the northern part of the Dominican Republic Stylophora monticulosa Vaughan (MS.) and Pocil lopora crassoramosa Duncan are confined to the Gurabo formation and Mao Adentro limestone. Both of these species are very abundant in the Las Cahobas formation. \ Oysters are the most abundant mollusks on the west side of the plain. Whole beds are composed of the large, heavy valves of Ostrea cahobasensis Pilsbry and Brown. Other beds contain the smaller species 0. bolus Pilsbry and Johnson. Both of these species are confined to the Las Cahobas formation. The large plicate oyster 0. haitensis Sowerby was found in both the Thomonde and Las Cahobas formations. Some beds contain large numbers of Sropharca chiriquiensis Gabb (Pl. XVI, figs. 6-8), which also is a common Thomonde fossil, especially in the Maissade tongue. This

PAGE 233

SEDilvIEN'.fARY ROCKS. 201 species, though found in strictly marine beds, seems to have been adapted to living in brackish water. Beds near the base of the formation contain a meager marine fa 11na. The mollusks listed under stations 9930 and 9931 were obtained from beds slightly higher than beds carrying Orthaulax aguadillensis and other mollusks typical of the upper fauna! zone of the Thomonde formation. Many of the species, such as Conus n. sp., the Marginellas, M e longena orthocantha Pilsbry and Johnson (Pl. XVI, fig. 5), and Scapharca chiri quiensis websteri ( Pilsbry) (Pl. XVI, figs. 9-11), are common Thomonde species, and if the stratigraphic position of these beds were not known they probably would be included in the Thomonde formation. They are supposed to be the littoral equivalent of the coralliferous limestones at the base of the Las Cahobas formation on the west side of the plain. 'l' he inclusion of these beds in the Las Cahobas formation weakens the fauna! evidence for considering the Ma!ssade tongue a part of the Thomonde formation, as the mollusks from these Las Cahobas beds are very sim ilar to those in the marine beds of the Ma1ssade tongue. The Las Cahobas formation is the equivalent of the Cercado fonnation of the Dominican Republic, although the faunas of the two formations are not at all comparable because of dissimilar environmental conditions. The corals indicate that the Las Cahobas formation is the equivalent of an unknown part of the Y aque group of the valley of Rio Y a que del Sur, Domini can Republic. Because of its correlation with the Cercado forma tion the Las Cahobas formation seems to be of upper Burdigalian age. Statwns in Central Plain, Las Cahobas formation (Miocene). '\Vest side ot plain. 9909 (W 193 F) Arrondissement of Las Cahobas, trail from Las Cahobas to Thomonde and Belladere, about 0.25 kilom,eter northeast of Las Cahobas. W. P. Woodring, collector. January 11, 1921. 9773 ( W 314 F). Arrondissement of Las Caho bas, same locality as 9909. T. W. Vaughan and W. P. Woodring, coll e ctors. March 1, 1921. 9906 (W 188 F). Arrondiss ement of Las Cahobas, low ridge about 0.6 kilometer north of Las Caho bas, at an altitude of 260 meters above sea level. W. P. Woodring, collector. January 9, 1921. 9905 (W 187 F). Arrondissement of Las Cahobas, float along south slope of high ridge about a kilometer north of Las Caho bas, corallif erous limestone in place at crest of ridge. W. P. Woodring, collector. January 9, 1921. 9910 (W 194 F). Arrondissement of Las Cahobas, on trail from Las Cahobas to Thomonde and Bell a dere, about a kilometer northeast of Las Caho bas. W. P. Woodring, collector. January 11, 1921. 9772 (W 195 F). Arrondissement of Las Cahobas, on trail from Las Cahobas to Thomonde, at place where Belladere trail branches off. W. P. Woodring, col lector. January 11, 1921. 9988 (W 189 F). Arrondissement of Las Cahobas, on trail leading north northwestward from Las Cahobas to Riviere I' Ayaye, about 5 kilometers from Las Cahobas. W. P. Woodring, collector. January 10, 1921

PAGE 234

I 202 GEOLOGY OF THE REPUBLIO OF HAITI. 9911 (W 198 F). Arrondissement of Las Cahobas, on trail from Las Cahobas to Belladere, 80Uth limb of Belladere anticline, about 4 kilometers west of Belladere. W. P. Woodring, collector. January 12, 1921. 9774 (W 204 F). Arrondissement of Las Cahobas, on trail from Las Cahobas to Thomonde, about 1.5 kilometers north of cemetery at Savane Cha.mouscadille W. P. Woodring, collector. January 15, 1921. 9775 (205a F). Arrondissement of Las Cahobas, on trail from Las Cahobas to Thomonde, about 2 kilometers north of cemetery at Sa vane Chamouscadille. W. P. Woodring, collector. January 15, 1921. 9776 (W 205b F). Arrondissement of Las Cahobas, same locality as 9775 but from underlying bed. W. P. Woodring, collector. January 15, 1921. 9777 (W 206 F). Arrondissement of Las Cahobas, south limb of Thomonde anticline, on trail from Las Caho bas to Thomonde, crest of ridge formed by base of Las Cahobas formation, about 3 kilometers southeast of Thomonde, at an altitude of 395 meters above sea level. W. P. Woodring, collector. January 15, 1921. 9916 (W 205c F). Arrondissement of Las Cahobas, same locality as 9776 but from underlying bed. W. P. Woodring, colle )tor. January 15, 1921. 9927 (W 233 F). Arrondissement of Las Cahobas, north limb of Thomonde anticline, on trail from Thomonde to Hinche, north slope of ridge formed by base of Las Cahobas formation, about 4.5 kilometers north-northwest of Thomonde, at an altitude of 350 meters above sea level. W. P. Woodring, collector. January 26, 1921. 9928 (W 234 F). Arrondissement of Las Cahobas, no1ih limb of Thomonde anticline, on crest of ridge formed by base of Las Caho bas formation, 0.5 kilometer west of trail from Thomonde to Hinche. W. P. Woodring, collector. January 26, 1921. 9783 (W 212 F). Arrondissement of Las Cahobas, north limb of Thomonde anticline, on trail from Thomonde to Thomassique, north slope of ridge formed by base of Las Cahobas formation, about 2.8 kilometers east-northeast of Thononde. W. P. Woodring, collector. January 16, 1921. 9944 (W 318 F). Arrondissement of Hinche, float at foot of bluffs on left bank of Riviere Guayamouc opposite Hinche, about 150 meters northeast of upper crossing. W. P. Woodring, collector. March 12, 1921. East aide of plain. 9930 (W 236 F). Arrondissement of Las Cahobas, right bank of Ravine Boucan Toureau, about 0.5 kilometer southwest of Thomassique. W. P. Woodring, col lector. January 27, 1921. 9931 (W 237 F). of Las Cahobas, right bank of Riviere Honde about 2 kilometers above crossing of trail from Thomassique to Thomonde. W. P. Woodring, collector. January 28, 1921.

PAGE 235

Miocene fossils from Central Pwin, Las C ahobas formation. Species. Foraminif era : Sorites sp. cf. americana ( C'UShman) .................................. Corals: Vaughan (MS.) .... Stylophora monticulosa Stylophora n. sp., also from station 8610, Yaque group, Dominican Republic Stylophora sp. Pocillopora cressoramosa Duncan. Pocillopora sp. Dichocoeni a stokes i Milne -Edwards and Haime Antillia dubia (Duncan) ............................................... Antillia sp. cf. A. walli Duncan ........................................ Orbicella imperatoris Vaughan ........................................ Orbicella sp. cf. 0. irnperatoris Vaughan .............................. Orbicella sp. cf. 0. annularis (Ellis and Solander), smaller secondary septa e I I I I I I I I e I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I Orbicella cavernosa (Linnaeus) var .................................... Solenastrea bournoni Miln e-Edwards and Hai me ..................... Solena -strea byades (Dana) ....................................... Mycet opbyllia sp. a. Mycet opby Ilia sp. b .................................................... Agaricia sp., also from Gurabo formation and Mao Adentro limestone .. Oyathomorpba n. sp ................................................ Genus ? fungid coral, cf. Cyathomorpha, but reproduction by fission .... P s amrno cora n. sp. P sammocora n. sp. c. d. e e e e e e e e e I e e e e I e I I Porites sp. aff. P. furcata Lamarck ..................................... Goniopora jacobiana Vaughan ..................................... Goniopora Goniopora n. sp ..... sp. indet. West side of plain. Las Cahobas. 9909 9773 9906 9905 9910 9772 9988 9911 9774 9775 I t t x x t I t x x I I I I I e x x x I a I a . x x e I t t I t x x x x x x I t t x I I I I .. x x I I t t x x x x x x x I I I I I t t t x x I t I I I I I t t x x 9776 9777 9916 9927 x x x 9928 9783 x x x x x -5 .e Ill 0044 00 t:tj tj z t:d 0 a tll 0 v;

PAGE 236

Miocene fossil.s from Central Plain, Las Cahobas formation-Continued. West Specie s Brachiopoda : Lingula s p. Mollusca: Gastropoda : Melongena consors Sowerby ? .. Ale ctrion sp. cf. A. cercadensis Maury ............................. Ceri thium ? sp ................................................. sp .................................................... Potamides ? Turritella sp. Fissuridea gp. indet. Pelecypoda : Barbatia sp. Scapharca chiriquiensis ( Gabb) ............................ Ostrea cahobaseneis Pilsbry and Brown ........................... O strea bolus Pilsbry and Johnson ............................... Ostrea haitensis Sowerby ......................................... Pecten (Pecten) sp .... sp ...... Oblamys (Aequipecten) Spondylus ep. cf. S. bostrychites Guppy. Mytilopsis sp. s p. ? Miltha Tellina Ma c o ma Tagelus Sol e n ? Labiosa Oorbula Corbula sp. indet. ? sp. indet. ? sp. indet indet ... indet. sp. ? sp. sp ( Bothrocorbula) v1m1nea Guppy. aide of plain-Continued. Las Cah o bas. 9909 9773 9906 9905 9910 9772 9988 9911 977 9775 x x x x x x x x x x x x 9776 9777 9916 9927 x x x x x x x x x x x x x x x x x x ? 9928 9783 x Q,I -5 .9 = 994' x 0 0 t_:rj 0 t1 0 Q 0 q b:J t'4 ...... c 0 P:t > ......

PAGE 237

SEDIMENTARY ROCKS. 205 Miocene f ossil,s from Central Plain, Las Ca ho bas formation Continued East side of plain. Species. Mollusca: Gastropoda : Cylicbnella n. sp. a cf. C. trictumtritonis Maury ....................... Ringicula n. sp. c ...................................................... Conus n sp., also from Baitoa formation ............................ Marginella nugax Pilsbry and Johnson ................................. Margin ell a n. sp. a cf. M. maoensis Maury .............................. Marginella (Gibberula) sp. b cf. M. chondra Gardner (MSS.) ........... Closia n. sp. a cf. unnamed Gurabo sp ................................. Melongena ortbocantba Pilsbry and Johnson ............................. Alectrion cercadensis Maury ............................................ Cerithium (Ptychocerithium) sp ........................................ Potamides n. _,p, a cf. P. gastrodon Pilsbry and Johnson ................ Potamides sp. ......................................................... Cerithidea sp. ........................................................ Bittium sp. g cf. B. asperoides Gabb ................................... Bittium sp. b .......................................................... Alabina sp. ......................................................... Meioceras sp. cf. M. constrictum ( Gabb) ................................ Fossarus (Isa pis) sp .................................................... Rissoa sp. ............................................................. Hydrobia sp. a .................. Hydrobia sp. b ......................................................... Tryonia sp. ........................................................... Epitonium sp. Turbonilla sp. ............................................... Odostomia sp. a ......................................................... Phasianella pun eta ta ( Gabb) .......................................... Neritina (Puperita) ftgulopicta Maury .................................. Circul us sp. ......................................................... Teinostoma depressum (Gab b ) .......................................... Scapbopoda : Cadulus sp. ...................................................... Pelecypoda: Nucula tenuisculpta Gabb ........................................ Leda sp. ............................................................... Scapharca chiriquiensis websteri (Pilsbry) ....................... Ostrea sp. ................................... .. ...... Pbacoides (Phacoides) sp. ...................................... Phacoides (Parvilucina) sp. ...................................... Callodardia ? sp ........................................ ..... ..... Corbula sp. b cf. C. sericea Dall .................................... Las Cabobas. 9930 9931 ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x Fossils of A rtiboni te group. The collection from station 7 544, ap parently from the Artibonite group, includes 6 new species of plants, described by Berry,1 and a new species of fish, described by Cockerell.1 The fish belongs to Oichlasoma, a common genus in the streams of South and Central America and north Africa. Six species of Oichlasoma are now 1 Berry, E. W., Tertiary fossil plants from the Republic ot Haiti: U. S. Nat. Mus. Proc . vol. 62, Art. 4, 10 pp., 1 pl., 2 text figs ., 1922. 2 Cockerell, T. D. A., A fossil cichJid fish from the Republic of Haiti: U. S. Nat. Mus. Proc., vol. 63, Art. 7, 2 pp., 1 pl., 1923.

PAGE 238

206 GEOLOGY OF THE REPUBLIC OF HAITI. living in Cuba, but none in the other West Indian islands. The Cuban cichlids are evidently remna.nts of a once widely distributed West Indian cichlid fa11na. Station in Central Plain, Artibonite group (Miocene). 7544 (W 185 F). Arrondissement of Las Cahobas, in cut a.long road from Mirebalais to Las Cahobas, on north side of ravine at foot of mountains on north side of gap. W. P. Woodring, collector. January 8, 1921. Miocene fossils from Central Pl,ain, Artibonite group (station 7:544). Fish: Plants continued: Cichlasoma woodringi Cockerell. Mespilodaphne hispaniolana Berry. Plants: Mimusops preparvifolia Berry. Gymnogramme woodringi Berry. Chrysophyllum cahobasensis Berry. Simaruba haitensis Berry. Bumelia cuneata.folia Berry. ARTIBONITE v ALLEY AND CHAINE DES MATEUX. The Miocene beds in the Artibonite Valley and on the Chaine des Mateux were once probably continuous with those in the Central Plain over the Montagnes Noires. The name Artibonite group is here applied to them. As the outcrop of these beds is almost continuous the fossils obtained from them are given in one list, on pages 217-218. Artibonite Valley. Artibonite Valley, as the term is here used, embraces the Artibonite Valley proper and its southeastward prolongation, the valley of Riviere Fer-a-Cheval. In many features it resembles the Central Plain, as it is a deep plunging syncline, modified by secondary folds, containing Miocene rocks. Although the essential stratigraphic features of the Artibonite group of the Artibonite Valley are known, naming of the recognizable rock units would only result in a cumbersome nomenclature. In the lower part of the Artibonite Valley the Miocene rocks on the northeast side of Riviere Artibonite seem to be concea led by Recent floodplain deposits and Quaternary dissected terrace gravels downstream from the vicinity of La Chapelle. Miocene rocks lie at the surface on the southwest side of Riviere Artibonite downstream from La Chapelle, and generally on both sides of the river upstream from La C 1 hapelle to Mirebalais except where they are covered by flood-pla .in deposits. In the foothills 3 or 4 kilometers southwest of La Chapelle Miocene marls rise to an altitude of 500 or 600 meters above sea level and conformably overlie white, chalky, unfossiliferous limestone of supposed Oli gocene age. Both series of rocks dip northeastward at angles of 45 to 7 5 A transition from limestone to marl takes place by the intercalation o! shaly beds that gradually increase in number and thickness as the beds of white or yellowish unfossilif erous limestone din1inish. The thickness of these transition beds is perhaps 100 meters. Above the transition beds are several h11ndred meters of marl and very fine grained soft sandstone.

PAGE 239

I SEDIMENTARY ROCKS. 207 The prevailing colors are blue, gray, or yellow. The marl has the typical conchoidal fracture shown in Plate XIV, 0. The beds are thin but gen erally very even. These rocks underlie a long lowland extending from Morne Saut d'Eau northwestwa .rd beyond Desarmes, parallel to the Artibonite flood plain. In the upper part of this marl-sandstone series there are thin beds of yellowish limestone containing indeterminable casts of corals. The beds of limestone probably increase in number a .nd thick ne s s up to the base of the overlying coralliferous limestone member. The coralliferous limestone member is massive and has a thickness of perhaps 50 to 100 meters. It forms the ridge that extends from Morne Saut d'Eau northwestward to Marche Desarmes and an 11nknown distance beyond, and that separates the lowland 11nderlain by the lower marl and sandstone from the flood plain of the Artibonite. (See Pl. XXX, B, p. 386.) The coralliferous limestone appears to be at the top of the Mio cene section below Morne Saut d'Eau, but farther southeast it is overlain by marl beds like those in under lying part of the section. s o --. - 0 I 3 ---Hauteurs doublees FiouBE 12.-Section of Miocene beds on the south side of Artibonite Valley near La Chapelle, as exposed a.long Riviere Delean. Total thickness of Miocene beds about 7fi0 meters. In the lower valley the strike of the rocks averages about N. 50 W., parallel to the general elongation of the trough. The of the Miocene beds on the southwest side of the valley is illustrated in Figure 12, which is based on excellent and almost continuous exposures along Riviere Delean. This limb of the major syncline is modified by a single prominent anticlinal arch, the crest of which has been eroded to the soft marl and sandstone and forms the lowland between the coralliferous lime stone ridge and the base of the mo11ntains. The anticline probably rises to the southeast, forming the crest of Morne Saut d'Eau, where the older limestones apparently have been exposed by erosion. The folding on the north flank of Morne Saut d'Eau is far more complex than that in the region to the northwest. The dips at some places are steeper, the strikes are variable in direction, and the beds here and there dip to the southwest. Both on Riviere Delean and on the trail to Couyau the dips steepen at the edge of the mo11ntains, where they range from 45 to 75 NE.

PAGE 240

208 GEOLOGY OF THE REPUBLIC OF HAITI. .. . The thickness of the Miocene beds to the top of the corallif er ous lim:estone member, as estimated from a paced traverse along the Riviere Delean, is about 725 meters. This measurement includes the transition beds at the base All the determinable fossils collected in this area were found on the north flank of Morne Saut d'Eau, near the highest part of the trail between La Chapelle and Mirebalais, at an altitude of a little more than 300 meters above sea level. They include fossil leaves, crinoid stems, corals, and mollusks. (See lists, pp. 217-218, stations 7 542, 9523, 9494, 9496.) Rocks of the same age underlie the southeastern part of the Arti-bonite Valley, but along the south edge of the valley they differ in lithology. The structure of the part of the valley immediately be low Mirebalais is shown in Figure 13. The lowest beds at the edge of the mountains northeast of Dufailly consist of white and yellow ish marl. A coral listed on page 217 (station 9920) was collected from float derived from a coral liferous limestone probably a hundred meters above the base of the Miocene series. This corallifer ous limestone apparently is the same as that described on page 20? although the thjckness of marl below it is less. Upon the corallif erous limestone rest beds of reddish-brown sandy marl and a few thin beds of conglomerate. All these beds dip southwestward

PAGE 241

SEDIMENTARY ROCKS. 209 at angles of 30 to 60. Southwest of Dufailly the trail to Mirebalais crosses the southeastward-pl11nging crest of an anticline paralleling the main synclinal trough. Northwest of the trail a narrow ridge, probably formed by the corallif erous limestone in the lower part of the Miocene series, encircles the plunging crest. Inside this ridge there is a depression and inside the depression a high jagged ridge composed, probably, of Oligocene or Eocene limestone. Beds of white marl, reddish-brown sandy marl, and marly clay overlie the coralliferous limestone and arch over the crest along the trail. On the south s i de of the river a similar anticline pl11nges southeastward. The corallif erous limestone is on the crest of the anticline immediately south of Mireba lais, where it forms the ridge on which Fort Anglais stands. In a road cut about a kilom eter south of Mirebalais the limestone weathers to a y e llowish marl c.:ontaining harder lumps of corall t ferous limestone. The corals listed on page 217 (station 9563) were collected at this lo c ality, where the beds dip 20 SW. At Fort Anglais, southeast of Mirebalais, the coralliferous limestone seems to dip northeastward. 'l,he corals listed on pages 217-218 were collected on the crest of the ridge at Fort Anglais (station 9922) and on the north slope of the ridge (station 9921). The southeasternmost exposure of this limestone is about 2 kilometers southeast of Mirebalais, on the road to Las Cahobas, where the corals listed on pages 217-218 (station 9901) were obtained. The beds along the south side of the valley, as exposed along the road between Mirebalais and Port-au-Prince, contain detrital material. Above the coralliferous l i mestone lie reddish-brown siltstone, rusty brown sandstone, and conglomerate consisting of small pebbles. Similar beds crop out farther southwest, between Mirebalais and Trianon, where they lie in the trough of a syncline. Between Trianon and the mountains the beds have a general northeastward dip, but near the mountains they are crumpled, and some exposur e s show that the softer beds have been squeezed out of place. The corall i f erous lime stone crops out about half a kilometer south of Trianon, but h ere it contains only pieces of Porites firmly em bedd e d in hard rock. Farthe r south the 11nderlying thin beds of clay, sandstone, and conglomerat e are w ell exposed in cuts along the recently relo cated road. Miocene rocks crop out along the road from Mirebalais to Las CahobaR east of the crossing of Riviere Fer-a-Cheval. The upper beds consist of sandy shale, sandstone, and conglomerate. The lower beds, made up of marl and thin beds of limestone, are crumpled near the foot of Morna Michel. At Savanette, in the valley of Riviere Fer-a-Cheval, the Miocene beds correspond to those in the lower part of the series in the Artibonite Valley proper, consisting of thin beds of marl, marly shale, calcareous sandstone, and limestone. Along the lower slopes of the mountains northeast of 14

PAGE 242

210 GEOLOGY OF THE REPUBLIC OF HAITI. Savanette, on the trail to Belladere, these beds are folded in a very narrow syncline. Immediately west of Savanette they are arched in an asymmetric anticline (see Pl. XXVI, B) that extends across the narrow valley and then bends southeastward along the south side of the valley. About 3 kilometers west of Savanette Riviere Fer-a-Cheval :flows in a narrow gorge of limestone of supposed upper Oligocene age, but farther northwest the Miocene rocks reappear and beds of sandstone and conglomerate overlie the marl and limestone. These detrital rocks form the northwestern end of the ridge between Riviere Fer-a-Cheval and Riviere Gascogne. The Miocene series in the valley of Riviere Gascogne consist almost entirely of rather coarse detrital rocks. The lowermost beds exposed, which confonnably overlie upper Oligocene limestone on t .he north :flank of the Montagnes du Trou d'Eau, consist of calcareous sandstone containing Stylophora (station 956 '2) and sandy clay. Above these beds lie beds of coarse sandstone, containing lenses of conglomerate, and beds of conglom .. and sandy clay. The pebbles in the conglomerate are small and consist of limestone, chert, and basalt. The mollusks listed on page 218 (station 9455) and an undetermined shark tooth were collected from the con glomerate. On the south side of the valley the rocks are arched in two secondary anticlines. Ohatne des Kateux near St.-Karc. The rocks of the Artibonite group bend southeastward aro11nd the plunging northwest end of the Chaine de Mateux and :flank most of the southwest slope of this range of mountains. They were examined at the following localities near St.-Marc: M orne des Guepes. The Mornes des Guepes comprise some low ridges and intervening lowlands of northwestward trend that lie between St. Marc and the Artibonite Plain. The hills border the northeast side of St.-Marc Bay and extend southeastward for several kilometers, probably continuing without interruption up the Artibonite Valley. Along the road from St.-Marc to Gona1ves the hills consist of two ridges separated by a lowland about 2 kilometers wide. To the northwest these ridge s unite, thus inclosing tl1 e lowland in that direc t ion. The drainage escapes to the southwest through a narrow gap into St.-Marc Bay. The entire lowland is underlain by blue, br<>wn, or yellowish soft Mio cene marls. A few thin beds of limestone interbedded with the marl contain casts of corals. Good exposures occur in the principal stream course, a usually dry ravine that is about 10 meters below the lowland surface. The marl is very fine grained and highly calcareous. (See analysis, p. 502.) The ridge on the northeast side of the lowland is composed of rather soft corallif erous limestone, doubtless the same as that which forms the prominent ridges in the Artibonite Valley farther southeast. (See pp. 207-208.) The rock is fairly well exposed in a cut along the road. The northeast slope of this ridge leading down to the Artibonite Plain is covered by alluvial gravel and boulders.

PAGE 243

SEDllIENTARY ROCKS. 211 The ridge southwest of the lowland is compose d mainly of the same coralliferous limestone. In this lo cality it contains many pebbles of the older Eocene limestone, some of them as much as 10 centimeters in diame ter, and also beds of calcareous sandstone. At both localities a transition from marl to limestone seems to take place by gradual intercalation of limestone beds. The northwest end of the lviornes de Guepes, particularly the slopes leading down to St.-Marc Bay, is covered with a veneer of Quaternary coralliferous limestone, which rests 11nconformably on the Miocene. This Quaternary limestone probably forms most of the ridge northwest of the lowland area. The structure of the Mornes de Guepes is rather simple. The whole region forms the nose of a northwestward-pl11nging anticline, probably the s.o &de tLe Qc Tm Tern Kilometres 0 i ............., hi I Hauteurs douolees .... 2 5 4 t FIGURE 14. Section across the Mornes Guepes. N.E Qal, Quaternary alluvium; Qc, Quate1nary coralllferous limestone; Tcm., Miocene coralliferous limestone; Tm, Miocene marl. continuation of the anticline on the south limb of the syncline in the Artibonite Valley. (Seep. 207.) The lowla .nd underlain by marls is on the crest of this anticlinal nose; the ridges on either side are inward-facing cuestas whose gentler dip slopes ]ead outward. Figure 14 is a cross section of the fold along the line of the road. The anticline is modified on the southwe.st limb by a secondary arch. As shown in figure 14, the base of the Miocene is not exposed, but the thickness of the expooed beds is estimated roughly at 315 meters. Fossils are more or l ess abundant in all the exposed rocks but most of them are poorly preserved, especially those in the marl beds. At the bridge over the dry ravine near the base of the exposed rocks fossil leaves and mollusks (stations 7543 and 9801) and fossil corals (stations 9897 and 9531) were collected. Mollusks were obtained from the intercalated lime stone beds farther southwest (stations 9532, 9533). These fossils are listed on pages 217-218. The corals listed on pages 217-218 (station

PAGE 244

212 GEOLOGY OF THE REPUBLIC OF HAITI. 9794) were collected from the coralliferous limestone on the ridge north east of the lowland. The cora ls in this limestone weather out in loose heaps on the surface of the ground. South of St.-Marc. South of St.-Marc there is a region comparable in several features to the Mornes des Guepes. It comprises an interior lowland inclosed on the southwest, northwest, and northeast by ridges com posed principally of coralliferous limestone, through which the drainage flows southward and northward in narrow gaps. In the western part of the low land is the small undrained lake called Etang Bois-N euf. The lowland appears to be 11nderlain entirely by }rliocene marl, sand stone, and conglomerate. The ridge to the southwest and northwest is capped mainly if not entirely by Quaternary coralliferous limestone, which unconformably overlies the Miocene beds. To the northeast, how ever, Miocene coralliferous limestone forms part of the inclosing ridges. The corals listed on page 217 (station 9550) were collected from this limestone. In the northern part of the lowland the beds dip northeastward. At the southern border of the lowland beds of sandstone and marl, which are well exposed, strike N. 35 W. and dip 35 SW. A few harder beds of sandstone are etched into relief by weathering. This sandstone, when examined under the microscope, is found to contain about 20 or 25 per cent of angular to subangular quartz grains, perhaps 15 per cent of green ish weathered fragments of igneous rock, mostly chloritic material, feld spar, or epidote, and the rest li1nestone fragments . Gypsum is abundant in this exposure, forming veins along bedding planes and joints, and is also scattered through the sandstone as cementing material. It occurs more sparingly in the exposures farther north. Loose weathered blocks of coralliferous limestone indicate that a cap of this rock farmer ly cov .. ered the ridge here as it now does a short distance to the west. The Miocene rocks in this area are similar to those in the Mornes des Guepes and appear to lie on the nose of an anticline, the crest of which trends northwestward. This anticline appears to plunge rather steeply to the northwest. St.-Marc Valley. Between the Mornes des Guepes on the northeast and the similar area just described to the south of St.-Marc an intervening lo,vland extends southeastward from St.-Marc. For about 2 kilometers southeast of St.-Marc the lowland is covered with a lluvium, but farther southeast soft Miocene rocks, chiefly marl, appear to lie at the surface. Bordering the lowland on the northeast n .nd southwest are bills capped by Miocene corallif erous limestone. Corals were collected from a bluff in this limestone at the stream crossing about 4 kilometers east by south from St.-Marc. (See list, pp. 217-218, station 9497.) The region 11ndoubtedly is occupied by a synclinal trough that plunges northwestward 11nder St.-Marc Bay between plunging anticlines.

PAGE 245

SEDIMENTARY ROCKS. 213 Southwest slope of Chatne des Mateux. Miocene rocks underlie the foothills on the southwest slope of the Cha!ne des Mateux from a lo cality near the northwest end of the Arcahaie Plain southeastward to the place where the trail from Saut d'Eau through Fonddes-Orangers to Port-au-Prince emerges on the Cul-de-Sac Plain. In the Arcahaie Plain the Miocene rocks are covered by Quaternary alluvial deposits. Southeast of Prince, a railway station about 2 kilometers southeast of Boucassin, the Miocene deposits extend to the sea or are separated from it by a very narrow fringe of alluvi11m. Southeast of the Sources Puantes the Miocene extends along the northern border of the Cul-de-Sac Plain. These !1iocene rocks no doubt formerly extended over the crest of the Chaine des Mateux and were continuous with the Artibonite group of the Artibonite Valley, but they have since been completely removed from the crest of the mountains by erosion. To the south they probably are continuous under the cover of alluvium with the Miocene rocks along the south edge of the Cul-de-Sac Plain. At Prince a fine-grained yellowish sandstone is exposed in a roadcut . It contains Orbiculina and many broken and indeterminable fragments of mollusks (station 9478). About 200 meters southeast of this exposure a railway cut has been made by digging down the side of a sea c liff. At the base soft yellowish sandstone and clay strike N. 10 E. and dip 40 SE., the strike being nearly at right angles to the shore These b eds contain abundant flakes of gypsum, some parallel to the bedding, others filling joints transverse to the bedding. Along the trail from l' Arcahaie to Marche Desarmes beds of Miocene age are exposed from the border of the plain, at an altitude of about 100 meters, up to an altitude of about 400 meters above sea level on the mountain side. Figure 5, page 128, shows the rather complicated structural features and stratigraphic relations at this locality. A prominent foothill ridge at the border of the plain is composed of rather fine-grained yellowish sandstone containing layers of clay. The first beds seen dip to the northeast, away from the plain, but in a short dis tance the dip changes to the southwest. Northeast of the sandstone ridge stands another ridge, composed of black basaltic rock, considerably weathered. Whet.her the basalt is solid rock or is detrital rock composed entirely of igneous fragments is not evident The field observations sug gest that the material is solid rock. Northeastward beyond the basalt and apparently dipping beneath it are beds of yellowish marl. In a high river bluff below the trail this marl is seen to overlie a considerabl e thickness of bluish marl such as is typical of the Miocene at numerous localities. On the mountain slope the prevailing dip is steep to the southwest up to an altitude of about 300 meters above sea level, and successively lower beds are exposed. The marl grades

PAGE 246

214 GEOLOGY OF THE REPUBLIC OF HAITI. down through a series of interbedded layers of blue clay and limestone into rather thick beds of limestone separated by thin shale partings. Higher up on the trail the dips reverse, so that two short anticlines, close together, are crossed, and there the be'ds dip northeastward into the face of the mountain for a considerable distance, so that the section on the lower slope is repeated, and near the summit of a ridge that ha.s an altitude of 410 meters beds of marl dip beneath basalt, which probably is the same as that at the foot of the mountain. The exposure of basalt is not more than 50 or 100 meters in width and caps a narrow ridge. The rock appears to be solid basalt but it is greatly weathered, and the appearance may be deceptive. Immediately northeast of the basalt lies a hard white limestone, prob ably of Oligocene age. The bedding at first is indeterminable, but farther up the mo11ntain the dips are to the south or southwest. It seems rea sonably certain that the limestone has been thrust over the Miocene as shown in Figure 5. The fault indicated at the border of the plain is not so well established, and this structure might be explained as an over turned fold. The same Miocene rocks were seen along the trail from Saut d'Eau through Fond-des-Orangers to Port-au-Prince. The crest of the mo11ntains on this trail has an altitude of about 680 meters above sea level. To the south the first Miocene beds crop out on the south slope immedi ately below the crest. They consist of limestone and yellowish marl con taining harder 111mps of limestone. These beds apparently are high in the series. Some of the beds of limestone contain corals, and specimens weathered out and lying along the trail were collected. (See list, p. 217, station 9460.) The underlying beds consist of marl and sandy limestone arched over the crest of an anticline south of Source Marissel. About 2 kilometers southeast of Source Marissel the trail crosses a bed of igneous rock, apparently interbe dded with the marls and limestones. The rock is greenish and is weathered. It seems to consist entirely of igneous material but contains small, smooth pebbles of rock that have the same appearance as the matrix. It contains also large angular pieces of vesicular basalt. The basalt is apparently overlain by a limestone containing the corals listed on page 217 (station 9461). This limestone probably is the same as the coralliferous limestone near the crest of the mountains. Toward the edge of the Cul-de-Sac Plain thin beds of marl and sandy marl, which dip steeply northeastward, underlie the coralliferous limestone Fo1sll1. The list given on pages 217-218 shows that many corals were collected from the coralliferous limestone, or limestones, in the middle part of the Artibonite group in the Artibonite Valley and on the crest and flanks of the Chaine des Mate11x. Several of the species were also found in the

PAGE 247

SEDIMENTARY ROCKS. 215 coralliferous limestones at the base of the Las Cahobas formation in the Central Plain, and the coralliferous limestones in all these regions prob ably lie at about the same horizon. The collections, like collections of corals from Miocene rocks in the Central Plain, contain Orbicella impera toris Vaughan and Orbicella canalis Vaughan, which formerly were thought to be restricted to upper Oligocene deposits. The collection from station 9494 includes fragmentary crinoid stems that have been examined by Mr. Frank Springer, of the United States National Museum. He has described these remains as Bala nocrinus haitensis n. sp.1 The genus Balanocrinus is most ab11ndant in Jurassic and Cretaceous deposits, but one species has been described from the Eocene of the Anglo-Parisian basin and another from the Miocene (Helvetian) of Italy. This Haitian crinoid is the first stalked Tertiary rrinoid described from the western hemisphere. '!'he molluscan fa11na of these beds is very meager. For some unknown reason, probably environmental conditions, the rich shoal-water fauna of the Thomonde formation was excluded from these regions. A collection of plants obtained in the Mornes tle Guepes contains the species Gymn:ogramme woodringi Berry, also collected in the Central Plain. Another collection from the same region contains Pisonia conditi Berry, described from beds of Miocene or Pliocene age on the north shore of Samana Bay, Dominican Republic. Stations in Artibonite Valley and Chaine des Mateux (Miocene). 7542 (B 139 F). Arrondissement of St.-Marc, trail from La Chapelle to Mire bala.is, ravine s. kilometer west of divide on trail, about 10 kilometers southeast of La Chapelle, at an altitude of about 300 meters above sea level. J. S. Brown, collector. December 18, 1920. 9523 (B 137 F). Arrondissement of St.-Marc, crest of hill near divide on trail from La Chapelle to Mirebalais. J. S. Brown, collector. December 18, 1920. 9494 (B 138 F). Arrondissement of St.-Marc, divide on trail from La Chapelle to Mirebalais. J. S. Brown, collector. December 18, 1920. 9496 (B 140 F). Arrondissement of St.-Marc, float on slope of hill on trail from La Chapelle to Mirebalais about a kilometer west of divide. J. S. Brown, collector. December 18, 1920. 9920 (W 222 F). of Mireba.lais, float on trail from Thomonde to Mirebalais, about 2 kilometers north of Dufailly, at an altitude of 360 meters a.hove sea level. W. P. Woodring, collector. January 18, 1921. 9563 (W 134 F). Arrondissement of Mirebalais, road from Mirebalais to Port- au-Prince, about half a kilometer south of Mirebalais. W. P. Woodring, collector. December 7, 1920. 9921 (W 223 F). Arrondissement of Mirebalais, Boat on north slope of ridge on which Fort Angla.is stands, 0.5 kilometer southeast of Mirebalais, at an altitude of 140 meters above sea level. W. P. Woodring, collector. January 19, 1921. 1 Springer, Frank, A new Tertiary crinold in the West Indies: U. S. Nat. Mus. Proc. (Awaiting publication.)

PAGE 248

216 GEOLOGY OF THE REPUBLIC OF HAITI. 9922 (W 224 F). Arrondissement of Mirebalais, crest of ridge at Fort Angla.is about 0.75 kilometer southeast of Mirebalais, at an altitude of 190 meters above sea level. W. P. Woodring, collector. January 19, 1921. 9901 (W 180 F). Arrondissement of Mirebalais, road from Mirebalais to Las Cahobas, float about 2 kilometers east-southeast of Mirebalais, at an altitude of 165 meters above sea level. W. P. Woodring, collector. January 8, 1921. 9941 (W 183 F). Arrondissement of Mirebalais, road from Mirebalais to Las Cahobas, about 4 kilometers east of crossing of Riviere Fer-a-Cheval. W. P. Woodring, collector. January 8, 1921. 9455 (W 131 F). Arrondissement of Mirebalais, trail from Cornillon to Marche Canard, 200 meters south of first crossing of Riviere Gascogne, at an altitude of 325 meters above sea level. W. P. Woodring, collector. December 3, 1921. 9562 (W 129 F). Arrondissement of Mirebalais, trail from Comillon to Marc.he Canard, north slope of mountains, about 11 kilometers northwest of Cornillon, at an altitude of 545 meters above sea level. W. P. W ood1ing, collector. December 4, 1920. 9801 (K 210 F). Arrondissement of St.-Marc, road from St.-Marc to Gonaives, near bridge across dry ravine about 4 kilometers northeast of St.-Marc. W. S. Burbank, collector. March 30, 1921. 7543 (K 210 F). Same collection as at station 9801. 9897 (B 346 F). Arrondissement of St.-Marc, float at same locality as station 9801. J. S. Brown, collector. March 30, 1921. 9532 (K 51 F). Arrondiss ement of St.-Marc, same locality as station 9801. W. S. Burbank, collector. December 8, 1920. 9533 (K 52 F). Arrondissement of St.-Marc, road from St.-Marc to Gonaives, about 3 kilometers northeast of St.-Marc. W. S. Burbank, collector. December 9, 1920. 9531 (K 50 F). Arrondissement of St.-Marc, road from St.-Marc to Gonaives, just south of crest of divide about 4.5 kilometers northeast of St.-Marc. W. S. Burbank, collector. December 8, 1920. 9794 (W 346 F). Arrondissement of St.-Marc, road from St.-Marc to Gona1ves, north slope of divide about 6 kilometers northeast of St.-Marc. W. P. Woodring, collector. March 27, 1921. 9550 (B 124 F). Arrondissement of St.-Marc, float on road from St.-Marc to Port-au-Prince, south end of gap about 6 kilometers south of St.-Marc. J. S. Brown, collector. December 7, 1920. 9497 (B 142 F). Arrondissement of St.-Marc, trail from St.-Marc to Verrettes, crossing of stream a .bout 3 kilometers east-southeast of St.-Marc. J. S. Brown, collector. December 22, 1920. 9478 (B 118 F). Arrondissement of Port-au-Prince, road from St.-Marc to Port-au-Prince, cut about a kilometer east of Boucassin. J. S. Brown, collector. December 6, 1920. 9460 (W 145 F). Arrondissement of Port-au-Prince, trail from Fond-des Orangers to the Cul-de-Sac Plain, float on south slope of mountains about a kilometer north of Source Marissel, at an altitude of 635 meters above sea level. W. P. Woodring, collector. December 8, 1920. 9461 (W 148 F). Arrondissement of Port-au-Prince, trail from Fond-desOrangers to the Cul-de-Sac Plain, south slope of mountains 3 kilometers southeast of Source Marissel, at an altitude of 530 meters above sea level. W. P. Woodring, collector. December 8, 1920.

PAGE 249

Miocene fossils from Artibonite Valley and C haine des M atew;. St.-Marc. Mirebalaia. Species. Foraminifera : Orbiculina sp. ................... ........................ Corals: Stylophora n. sp., also from station 8610, Yaque group, Domini can Republic .. Stylophora s p. Stephanocoenia intersepta (Esper) .......................................... Dichocoenia tuber<>ea Duncan ................................................ Dichocoenia stokesi Milne-Edwards and Haime. Orbi cella imperatoris Vaughan .............................................. Orbicella sp. cf. 0. tmperatoris Vaughan. Orbicella canalie Vaughan ................................................... Orbicella n. sp., also from Miocene of Trinidad .............................. Orbicella sp. Orbicella sp. cf. 0. annularis (Ellis and Solander) limbata (Du:ican) ................................................. Orbicella sp. atf. 0. alti. s sima (Duncan) ..................................... cavernosa var. endothecata (Duncan) ............................. c avernosa (Linnaeus) var . e e e I e I e e e I Orbicella Orbicella Orbi cella ? sp . . ........................................................... Solena strea bournoni Milne-Edwards and Haime ............................. S olenastrea sp. atf. S bournoni Milne Edwards and Hai me ................. Solena s trea hyades (Dana) .................................................. a .......................................................... Myc e tophyll ia sp. Siderastrea radians (Pallas) A d v han gar1c1a om1n1cens1s aug .............................................. Agari cia sp. also from Gurabo formation and Mao Adentro lime stone ....... Agaricia sp. Psammocora n. sp. b ......................................................... Porites sp. cf. P. porites (Pallas) -:?; t- l.O Q) x x x x ------------------- x x x x x x O> ? x x x O> x x C"I 8! O> 0) x x x 0) 0) x O> t- 00 Q) x St.-Marc. C"I C") ,... CQ x x Q) s x x x 0 lt') O> ? x x Q) x Port-au Prince. O> x x x x x x 8 ... 0) x 00 tzj t--4 z pj 0 a -1

PAGE 250

, Miocene fossils from Artibonite Valley and Chaine des Mateux Contint1 e d. Sp ec ies. Corals-COntinued. Porite-s furca'ta. Lamarck .............................. ...................... Porites sp. a:ff. P. furcata Lamarck .......................................... ns treoides Lamarck ........................................ from Mao Ad entro limestone .............................. G. jacobiana Vaughan. Porites s p. cf. P. Porites sp. cf. sp. Gon i opora sp. cf. Goniop ora sp. indet ..................... Orinoid: Balanocrinus baitensis Springer .............................................. Mollu sca: Pteropoda: Styliola sp. Ca volina sp. Gas tropoda : ? sp indet. Planorbis Con us Pho e sp. ? sp .. Pelecypoda : Nucula sp. Ostrea sp. Pecten sp. Pecten sp. cf. P. pittieri Dall ............................................ Amusium (Pseudamusium) sp ............................................ Spondylus sp. sp. Phacoides (Parvilucina ?) sp. Plants: Gymnogramme woodringi Pisonia conditi Berry .... Berry. Simaruba baitensis Berry ................................ ... t-- x ? St.-Marc. 0) x """ 0) x x x x 8 ""' 0) x .-4 Mirebalais. &l 0) 0) x 8 g x x :;f1 x x x x r8 0) i x <: "!ti t-- x St. Marc. I 0) <: .-4 fC5 Ol x x x x x x x -.tC 0) s x 0 '(') 0) 0) x x Portau Prince. 00 0) g ""' O> 00 s Q 0 d to ..... 0 0 bj .....

PAGE 251

SEDIMENTARY ROCKS. 219 SOUTHERN EDGE OF THE CUL-DE-SAC PLAIN. General features. The surf ace f ea tu res of the Cul-de-Sac Plain are shown in Plate XXXIX (p. 516), and are described on pages 395-397 Most of the plain, which is relatively smooth throughout, is covered with alluvi11m, but Miocene rocks are exposed in the foothills that border the mountains on the south. The Miocene rocks are at most places unconform a bly covered with dissected gravels, probably of Pleistocene age, and are exposed only in ravines and along roads and trails. The exposures were examined at several places, most of them near Port-au-Prince and Petionville. Besides containing characteristic fossils, the Miocene series has certain distinctive lithologic and structural features at nearly all the localities examined The series consists largely of soft, fine-grained grayish sandstone and of bluish marl, in thin, even beds. Beds of harder yellowish s.o 0 3 -------------- ----- He,.1teurs doublees N.E P1 .AINE. DU L-D!:-5'C '+Kil ornetres FIGURE 15.-Composite generalized section of Miocene beds along the south side of Cul-de-Sac Plain. Qal, Quaternary alluvium ; Qp, Pleistocene conglomerate ; Tm, Miocene ; Te, upper Eocene limestone. limestone and of conglomerate consisting of a mixture of boulders of black basalt and of upper Eocene limestone are conspicuous at some places. Fossils, especially small mollusks, are moderately abundant in the sandy beds but most of them are poorly preserved. Some of the beds of limestone contain numerous fossils, many of which, however, are only casts. In nearly every exposure the beds are tilted at steep angles and appear to be cr11mpled. There is clear evidence of faulting at some places in the foothills where the upper Eocene limestone of the adjacent mountains appears to have been thrust northward over the Miocene beds. (See Fig. 15.) The thickness of the beds ca n not be estimated accurately but is doubt less several h11ndred meters. Although the line of exposures extends nearly parallel to the strike of the beds, a thickness that appears to be at least 200 meters is exposed on the Petionville road. The ''Rasco'' well, near Port-au-Prince (seep. 525), penetrated the Miocene to a depth of 141 meters, which, allowing for the dip, indicates a thickness at that place of 100 meters.

PAGE 252

220 GEOLOGY OF THE REPUBLIC OF HAITI. Vicinity of Port-a'U-Prince and Petionville. The exposures near Portau-Prince and Petionville, which are in hillside cuts made in road grading and are unusually good, are most numerous along the main road from Port-au-Prince to Petionville, llarticularly in the first 2 kilometers out of Petionville, along the southward bend of the road. These and other exposures are shown diagrammatically in Figure 36 (p. 568), a sketch map of the geology of the region near Port-au-Prince. They are not continuous but are isolated. In these exposures the beds strike N. 45 W. to N. 75 W. and dip northward at angles of 30 to 79 The beds consist of sandstone, conglomerate, and limestone a .nd contain the fossils listed on page 222, stations 9564, 9574, 9573, and 9572. Miocene beds are well exposed in the Ravine Bois de Chene just below Source Plaisance at intervals for several hundred meters downstream from the spring, but the full extent of the exposures was not determined. All the dips observed were southward. The beds, which are intimately related to the origin of the spring (seep. 572), a .re chiefly bluish marl and brownish sandstone. They contain the fossils listed on p. 222 (station 9468). Three exposures of the Miocene were seen near Source Turgeau. Grayish sandy beds, apparently dipping southward, are exposed in the bed of a deep ravine about 300 meters east of Turgeau. The fossils listed on p. 222 (station 9571) were collected at this locality. At a point about 100 meters east of Turgeau, on a trail, is a poor exposure of sandy beds, apparently like those just mentioned. About 300 meters a little south of west of Turgeau, on a trail leading up Morne Hopital, is a poor exposure of clayey marl, whose bedding is indistinct but appears to dip southward. Miocene beds are exposed on the low hill where the houses of the Haytian-American Sugar Co. ('' Hasco '') are situated, at the north edge of Port-au-Prince and just east of the Arc 9.e Triomphe. The slopes of the hill are covered with soil, but good exposures are found in the ditches along the road leading up to the houses and in pits on the crest of the hill south of the houses. The beds exposed along the road consist of sandstone, clay, marl, and limestone. Fossils were collected at several places along the right side of the road going up the hill. (See list on p. 222, stations 9462, 9463, 9464, and 9750.) Valves of the large plicate oyster Ostrea haitensis are strewn on the ground below the first house, and ma11y bar nacles may be seen in pockets in the weathered limestone. A pit near the crest of the hill southeast of the club house exposes yellowish limestone interbedded with poorly consolidated conglomerate containing pebbles, as large as a man's fist, of hard white limestone, chert, and weathered basalt. At all these exposures the beds dip steeply southwestward. A well on this hill penetrated Miocene rocks to a depth of 141 meters. At Drouillard, a railway station on the majn road out of Port-auPrince about 2 kilometers northeast of the Arc de Triomphe, Miocene beds consisting of soft bluish marl and yellow clayey limestone are expose(l for about 60 meters in a roadside cut. The beds dip 50 southwestward

PAGE 253

SEDIMENT.ARY ROCKS. 221 Beds of Miocene age are exposed on the right bank of the Grande Riviere du Cul-de-Sac for a distance of about 400 meters above the Bassin General. The rocks exposed consist of bluish marl, grayish sand, and thin beds of limestone The prevailing dip is 30 to 60 N. At the south end of the exposure the beds are in fault contact with upper Eocene lime stone, and near the contact they are crumpled. The mollusks listed on page 222 (stations 9578 and 9590) were collected from these beds. There is every reason to believe that the Miocene beds underlie the alluvium throughout the plain. Wells at La Moriniere and Dessources penetrate more than 100 meters of beds classified from the logs as Miocene (see pp. 521-522). Fossils. Tl1e fossils collected from the Miocene beds along the south 8ide of the Cul de-Sac Plain comprise a variety of orga nisms. The corals are poorly preserved but include typical Miocene species. Echinoneus cyclostomus L e ske, picked up loose on the hill where the houses of the Hay tian-American Sugar Co. stand, is the only post-Eocene echinoid obtained in the Republic. This Recent species has been found in deposits as old as the Anguilla formation.1 The largest collection of mollusks made, obtained from a bed of limestone on the same hill (station 9464), contains many specimens of Ostrea haitensis, the large plicate oyster found also in the Thomonde and Las Cahobas formations of the Central Plain and in the Gurabo formation of the Dominican Republic. Several Pectens and V eneroids found at this locality are very similar to Bowden species. Pteropods are abundant in some of the beds exposed on the Petionville road and on the Grande Riviere du Cul-de-Sac. Some of the pteropods listed as '' T entac ulites '' sp. are remarka bly similar to the Paleozoic Ten-taculites. The y have not been studied critically but are probably similar to the so-called Tentaculites obtained from the Oligocene of Germany by Ludwig and by Blankenhorn. These Miocene beds seem to be younger than the Artibonite group of the Central Plain and Artibonite Valley and are correlated with the Gurabo formation of the Dominican Republic, which is considered Helve tian. The C e rros de Sal formation, at the east end of the trough of which the Cul-de-Sac Plain is a part, seems to be even younger. Stations along south edge of Cul-de-Sac Plain (Miocene). 9654 (W 12 F). Arrondissement of Port-au-Prince, road from Port-au-Prince to Petionville, about 1.5 kilom e ters from Petionville. W. P. Woodring and J. S. Brown, coll e ctors. October 9, 1920. 9574 (B 18 F). Arrondissement of Port-au-Prince, road from Port-au-Prince to Petionville, about 2 kilometers from Petionville. J. S. Brown, collector. Octo ber 12, 1920. l Jackson, R. T Fossil Echlni of the West Indies: Carnegie Inst. Washington Pub. 306, pp. 5455, pl. 9, figs. 4, 5, 1922.

PAGE 254

222 GEOLOGY OF THE .REPUBLIC OF HAITI. Miocene fossils, south edge Cul-de-Sac Pl sp. ..................... sp I e e t e e I e e e e e a e 4 t e t t ? sp ............................ Modiolaria Lithopbaga sp. Obama sp. cf. C. macerophylla Gmelin ...... Cardium (Tracbycardium) sp ................ Cardium (Papyridea) sp. cf. 0. (P.) bulbosum Dall Antigona (Ventricola) sp. cf. A. (V.) blandi-ana (Guppy) Ohione (Lirophora) sp. cf. C. (L.) bendersoni Dall Pi tar ? Petri cola sp . ? sp .... Macoma ? sp ................................. Semele sp. cf. S. sulcata Dall ............ Oirripedia : Balanus sp. cf. B. eburneus Gould ............ Balanus concavue esepatus Pilsbry ....... Balanus polyporus Pilsbry ...................... x x x x x x x x x x x x x x 0.) I x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x ? x x x x x x x x ? x

PAGE 255

SEDIMENTARY ROCKS. 223 9573 (B 17 F). Arrondissement of Port-au-Prince, road from Port-au-Prince to Petionville, about 100 meters farther from Petionville than 9574. J. S. Brown, collector. October 12, 1920. 9572 (B 14 F). Arrondissement of Port-au-Prince, road from Port-au-Prince to Petionville, about 300 meters farther from Petionville than 9573. J. S. Brown, collector. October 12, 1920. 9468 (B 21 F). Arrondissement of Port-au-Prince, Ra.vine Bois de Chene, just below Source Plaisance. J. S. Brown, collector. October 12, 1920. 9571 (B 13 F). Arrondissement of Port-au-Prince, 300 meters east of Source Turgeau. J. S. Brown, coll ect or. October 11, 1920. 9462 (W 149 F). Arrondissement of Port-au-Prince, road lea.ding up hill to houses of Haytian-American Sugar Co.; altitude 60 meters above sea level. W. P. Woodring, collector. December 14, 1920. 9463 (W 150 F). Arrondiss ement of Port-au-Prince, road leading up hill to houses of Haytian-American Sugar Co.; altitude 70 meters above sea level. W. P. Woodring, collector. December 14, 1920. 9464 (W 151 F). Arrondissement of Port-au-Prince, road leading up hill to houses of Haytian-American Sugar Co., in long cut on right side of road just below first house; altitude 75 meters above sea level. W. P. Woodring, collector. December 14, 1920. 9750 (B 358 F). Arrondissement of Port-au-Prince, float on hill on which houses of Haytian-American Sugar Co. are located. J. S. Brown, collector. April 11, 1921. 9590 (W 29 F). Arrondissement of Port-au-Prince, right bank of Grande Riviere du Cul-de-Sac about 300 meters above Bassin General. W. P. Woodring, col lector. October 15, 1920. 9578 (B 26 F). Arrondissement of Port-au-Prince, right bank of Grande Riviere du Cul-de-Sac, about 400 meters above Bassin General, at fault contact. J. S. Brown, collector. October 15, 1920. NORTH COAST OF SOUTHERN PENINSULA. Miocene rocks were recognized at a number of places on the north coast of the Southern Peninsula, chiefly in the region between Port-au-Prince and Petit-Goave, along the mai n highway following the coast. The rocks consist generally of soft sandy or clayey beds, but grade into conglomerate at one extreme or into limestone at the other. As a rule they are tilted and folded. They doubtless form part of a series of marginal marine beds that were continuous westward from the Cul-d e -Sac Plain, but are now either submerged or concealed by later deposits along most of the coast. As only small isolated exposures were observed at most localities the details of their stratigraphy can not be given. M orne-d-Bateau. teau, about 2 kilometers east of the vil lage of Gressier and midway between Port-au-Prince and Leogane, is a small hill at the base of low limestone mountains. In a road cut on the point of the hill there is an exposure of soft yellowish sandy and clayey beds which dip a few degrees westward. The beds contain numerous speci mens of the few mollusks listed on page 2 25, station 9480. The age of the beds is not evident; they may be Pliocene or Quaternary rather than Miocene.

PAGE 256

224 GEOLOGY OF THE REPUBLIC OF HAITI L'Acul. At the southwest corner of the Leogane Plain, less than a kilometer from a little valley and settlement, both called l' Acul, the road crosses a low ridge in which a poorly consolidated conglomerate is exposed. This conglomerate contains boulders of Eocene limestone and large lumps of material carrying corals and other fossils that apparently resemble Miocene species. At the little stream of l' Acul a collection of similar fossils, clearly of Miocene age, was taken from a pile of loose limestone boulders by the roadside (see list, p. 225, station 9481). The source of the boulders is unknown, but they doubtless came from no great distance. Grand-Goave. Some low hills rise from the little alluvial pla .iu of Grand-Goave just south of the road and about a kilometer east of the town. Here soft yellow i sh and brownish sandstone is exposed in a road side cut. The bedding is not apparent, but the rocks are cut by numerous fracture planes, which are filled with an unidentified salt, probably alka line. The rocks contain the fossils listed on p. 225, station 9540. Blocks of cavernous coralliferous limestone have slumped down from the slope above. They may possibly be Miocene but more probably belong to the Quaternary limestone that is well developed farther east. Ta pion du Petit-Goave. Miocene rocl{s underlie a narrow trough between Tapion du Petit-Goave a .nd the mountains to the south. The eastern part of the trough drains eastward and the western part drains westward from a divide which has an altitude of about 182 meters above sea level. On both sides of the divide, for a distance of about 2 kilometers along the road, exposures of the Miocene are common. The first exposure at the eastern entrance to the trough-like valley is a bluff of coralliferous limestone, probably of Miocene age. The rock is greatly weathered, so that the bedding is not apparent. From this exposure to the top of the divide, a distance of about a kilometer, the rocks in numerous roadcuts consist mainly of brown calcareous sandstone, sbaly beds, and angular boulders of basalt and limestone. These beds are unfossiliferous and appear to be nonmarine, at least in part. The pre vailing strike is about N. 65 W. and the dip is 40 to 60 NE. West of the divide the surface is covered by gravel for a few hundred meters, but farther west there are more exposures of the sandy and con glomeratic beds. The beds are folded and show minor faulting and slick ensiding. About a kilometer west of the divide limestone is interbedded in the conglomerate in beds striking about N. 60 W. and dipping 30 NE. The corals listed on page 225 were obtained from the conglomerate (station 9953) and from the limestone interbedded in the conglomerate (station 9954). In this locality the Miocene is unconformably overlain by a younger coralliferous limestone, probably of Pliocene age, which at some places is difficult to distinguish from the limestone in the Miocene series.

PAGE 257

SEDIMENTARY ROCKS. 225 As the road in this area is nearly parallel to the strike of the rocks no estimate of the thickness of beds can be made; probably only a small part of the section is exposed. Near Bwraderes. Near Baraderes, on the trail between Petit-Trou de I Nippes and Baraderes, there are a few exposures of rocks that are doubt-fully referred to the Mio c ene. Exposures of a fairly consolidated bedd e d tilted conglomerate were noted 5 or 6 kilometers east of Baraderes, in an area underlain chiefly by basalt. The conglomerate contains pebbles of basalt and of the limestone that was assumed to be the upper Eocene lime stone of this vicinity. Beds of soft chalky, unf ossiliferous mar 1 are exposed for 200 meters or more along the trail in a small valley about 3 kilometers east of Baraderes. If these beds are Miocene they represent remnants preserved by faulting or folding during the uplift and erosion of the surrounding mo11ntains. Fossils. The few fossils obtained from the scattered outcrops of Mio cene beds a long the north coast of the Southern Peninsula indicate that these beds are of about the same age as the Miocene along the south side of the Cul-de-Sac Plain. Miocene fossils from north coast of Southern Peninsula. Species. 9480 9481 9540 9953 9954 Foraminifera: Orbiculina sp . . . . . . . . . . . . . . . . . . . . . . . . . . . ... x Corals: Stylophora montic ulosa Vaughan (MS.)............................ ... x Sty lophora sp. a. . . . . . . . . . . . . . . . . . . . . . . . . x . . . . ... Orbicella annularis (Ellis and So lander) . . . . . . . . . . . . . . . . .. x Yycetophy Ilia sp. . . . . . . . . . . . . . . . . . . . . . . . . . . . x . . . .. Porites sp. indet... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x ... Mollusca: Gastropoda : Conus sp. c f. unname d Cer cado species ......................... . . x . . . . ... M urex sp. ...................................... . . . . . . . . . . . x . . .. Stro mbus s p cf. S. chipolanus Dall ............. .............. Cerithium sp. ................................................ Torinia sp. cf. T. rotunda G abb ............................... Pelecypoda : x x x Scapharc a sp. .................................... . .. . x . . .. O strea sp. ............................................. .. . x . . . . . . .. Chlamys (A e quipe cten) sp. cf. C. (A.) u s elmae Pilsbry and Johnso n ................................... ... . . X Ohlamys (A e quipecten) sp ........................ ....... X Cardi um ( Lae vicardium ?) sp ........................... . . . Metis sp. cf. M. chipolana Dall. . . . . . . . . . . . . . . .. Semele sp. cf. S. sulcata Dall... . . . . . . . . . . . . . .. Oirripedia: Balanus a1nph itrite Darwin ............................ ..... . X 15 x x x

PAGE 258

226 GEOLOGY OF THE REPUBLIC OF HAITI. Stations on north coast of Southern Peninsula (Mioc e ne). 9480 (K 2 F). Arrondissement of Leogane, cut along road from Port-au-Prince to Leogane at foot of Morne-a-Bate au, about 2 kilom eters east of Gre ssier. W. S. Burbank, coll e ctor. October 21, 1920. 9481 (K 3 F ) Arro ndiss e m ent of Leoga ne, road from Port-au-Prince to Leogane, from boulde rs at cro s sing of Riviere l Acul. W. S. Burbank, coll e ctor. October 22, 1920. 9540 (B 31 F). Arrondi ssement of L eoga ne, road from L e og a n e to Miragoane, about a kilometer e a s t of Grand-Goave. J. S. Brown, collector. Octob e r 23, 1920. 9953 (W 348a F). Arrondi s s e m ent of Leo gane, road from L e ogane to GrandGoave, west sl ope of Tapion du P etit-Goave; from conglomerat e W. P. Woodring and J. S. Brown, coll e ctors. April 10, 1921. 9954 (W 348b F). Arrondissement of Leogane, same locality as 9953 but from limeston e interbedde d with conglomerate. W. P. Woodring and J. S. Brown, collectors. April 10, 1921. COM?\iUNE OF J EREMIE. Grande Rivie r e and Bras-a-D1oit. Miocene rocks occupy two areas that form small but promin ent interior lowlands southwest of Jeremie, one along the course of the Grande Rivi e re de Jeremie and the other on its tributary, the Bras-a-Droit. The c ente r of the larger area, the one on the Grande Riviere de Jeremie, is about 7 kilometers southwest of Jeremie, near the village of Tissier. In the southwest part of the area is the large market c alled Mafron. This lowland m e asures about 3 kilometers from north to south and at l east 7 or 8 kilometers from east to west. The smaller area is abou t 22 kilom e ters west-southwest of Jeremie and con tains the large m arke t call e d Chamb e llan. Its length from ea.st to west is about 4 kilomet e rs, and its width is probably less than 2 kilometers. The two lowlands w e re traversed hurriedly and were examined only near the trail. u c h of the are a in e ach low land is covered by thin allu vial deposits, so only a few exposur e s of the underlying rocks were found. The full ext ent of the Miocene, which undoubtedly underlies most of each lowland, is inferre d from the surface features, which contrast strongly with those of the surro11nding lime stone areas. From the f e w exposur e s examin e d it appears that the Miocene rocks in both areas are very similar and con sist largely of highly calcareous brown marl and of brown or y e llow sandstone, together with some conglomerate. Many conspi c uous low hills and ridges in the large area, whi c h trend n early east and west, appear to be composed of the harder conglomer atic beds. Marine fossils and strand plants of Miocene age were obtained in the larger area (se e lists on p. 228; stations 9992 and 7541). Similar but indeterminable remains of fossil plants were obtained in the smaller area. These fossils indic ate that the Miocene sea extended far up the region that now constitutes the vall e y of the Grande Riviere de Jeremie and perhaps covered a great part of the peninsula. Wherever the bedding is apparent the Miocene beds are steeply tilted. All the observed dips are to the north or northwest. It seems likely that

PAGE 259

SEDIMENTARY ROCKS. 227 the larger area is bounded on the north by a thrust fault. At this northern boundary a very steep and regular escarpment of upper Eocene limestone rises from 200 to 300 meters above the lowland. The nearest observed M j ocene beds, several hundred meters south of the escarpment, dip north ward, apparently beneath the ridge. Although this is only a single ex posure it appears to represent one of the prominent east and west ridges, the profile of which shows that it is a cuesta having a dip slope to the north. This ridge and other northward .. dipping ridges do not reappear between the observed exposure and the base of the escarpment, as would be expected if the rocks are in a synclinal trough. The inference is that they plunge beneath the limestone. The structural rela tions of the smaller area are more obscure. It, too, is bordered by a steep limestone slope on the north, and the single observed dip, at a point not more than 500 meters south of the ridge, is 37 N., suggesting a relation similar to that in the larger area. However, the dips observed in the Eocene limestone just east of the area seem to indicate a simple syncline along the axis of this lowland, and possibly the Miocene beds merely occupy the trough of a steeply folded syncline. South of Les Roseaux. Numerous apparently authentic reports ob tained at Jeremie indicate that there are beds of lignite at a place a few kilometers south of Les Roseaux. As lignite in the Republic of Haiti, AO far as known, is found only in the Miocene formations there is probably an area of Miocene rocks in that locality. Possibly, however, it may be merely the eastward extension of the lowland on the Grande Ri\iere de Jeremie. Fossils. The marine fossils obtained at station 9992 are poorly pre served but indicate that the beds are of Miocene age, probably lower or middle Miocene. The strand plants Gymnogramme woodringi Berry and Guettardia cookei Berry, were collected at station 7541. Gymnogramme woodringi is also recorded from the lower part of the Artibonite group of the Artibonite Valley and from the Artibonite group of the Central Plain. Guettardia cookei was described from beds of Miocene or Pliocene age on the north shore of Sarnana Bay, Dominican Republic. Stations in Commune of Jeremie (Miocene). 9992 (B 96 F). Arrondissement of Grand 'Anse, trail from Jeremie to Moron, about a kilometer east of Tissier and about 6 kilometers from Jeremie. J. S. Brown, collector. November 18, 1920. 7541 (B 95 F). Arrondissement of Grand 'Anse, valley of Grande Riviere de Jeremie, trail between Mafron and Fondelet markets, half a kilometer south of Mafron and about 10 kilometers southwest of Jeremie. J. S. Brown, collector. November 18, 1920.

PAGE 260

228 GEOLOGY OF THE REPUBLIC OF HAITI. Miocene fossils from Commune of Jeremie (station 9992). Foraminif er: Sorites sp. cf. S. americana (Cushman). Coral: Porites sp. indet. Mollusca: Gastropoda : Oliva sp. cf. 0. testacea Lamarck. Olivella sp. Bittium ? sp. Liotia ? sp. N eritina sp. Pelecypoda : Arca sp. cf. A. bowdeniana Dall. Cardi um (Laevicardium) sp. cf. C. (L.) venustum Gabb. Miocene fossil plants from Commune of Jeremie (station 7541). Gymnogramme woodringi Berry. Guettardia cookei Berey. BETWEEN PoRT-8ALUT AND PoRT-1-PrMENT. Three exposures of limestone, probably of Miocene age, were noted on the trail between Port-Salut and Port-a-Piment. Two exposures are on either side of Anse-a-Drick; the third is about a kilometer southeast of Damassin. The rock is a rather massive limestone, less dense than the common massive upper Eocene and at most places containing numerous poorly preserved corals, although only one identifiable specimen, a Pocil lopora similar to P. arnoldi Vaughan, was obtained (station 9477). The limestone north of Anse-a-Drick contains undetermined calcareous algae. These exposures seem to indicate a belt of Miocene limestone intervening between the Quaternary coralliferous limestone and the higher mountains of upper Eocene limestone. Station between Port-Salut and Port-a-Piment (Miocene). 9477 (B 72 F). Arrondissement of Coteaux, trail from Port-Salut to Port-8. Piment, about a kilometer southeast of Damassin. J. S. Brown, collector. N ovem-ber 11, 1920. Miocene coral from station 9477 Pocillopora sp. cf. P. arnoldi Vaughan. ASILE v ALLEY. The Asile Valley is an interior lowland similar in many respects to the lowlands of the commune of Jeremie. The valley stretches from east to west and is about 3 kilometers wide and about 12 kilometers long. On its west side it rises into a plateau of 11ndetermined extent. The town of l' Asile is in the central part of the valley.

PAGE 261

ROC:KS. 229 Virtually all the Asile Valley seems to be underlain by Miocene rocks, although a large part of the surface is covered with alluvial deposits and the Miocene crops out only in strea m bluffs, along trails, or on isolated hills. In all the valley proper, that part east of the pla teau, the exposed 1'iio cene rocks consist of soft, fine-grained sediments tha. t were laid down in lakes or fresh-water swamps. White or cream-colored marl is the pre dominating ro c k near l' Asile. In the east and west ends of the valley the sediments c ontain more clay and in places consist almost entirely of greeni s h or yellowish clay. Thin seams of black carbonaceous clay are interbedded with the clay and marl. In the western part of the valley the carbonaceous beds contain less clay and consist of impure lignite. These beds of lignite in the Asile Valley have been mentioned in many publi c ations and have been the object of s e veral mining concessions and some special investigations, but none of the beds observed could ever be of comm e rcial value, and the existence of more promjsing beds seems very doubtful. Beds of g rav e l crop out in the western part of the valley, and coarse angular conglomerate underlies the plateau to the west. The conglom erate and gravel appear to be composed mainly of fragm ents of limestone, and none of the basalt that is common in the vicinity was found in the ex posures obs e rved. The beds of conglomerat e are believ e d to represent a d elta deposit formed at the border of the lakes and swamps in which the fine-grained beds were depo s ited and probably grade into the finer beds on the east. The conglomerate may, however, be really younger than the Miocene beds. The fine-grained clayey and calcareous sediments are very fossiliferous at certain hori z ons and contain many small fresh water mollusks. The beds in the central and eastern part of the area are generally slightly folded and dip to the south or southwest. Those in the western part of the valley are n e arly horizontal. The conglomerate in the plateau se e ms to dip gently eastward. The relation of the Miocene to the older rocks was not deterrnined, most of the contacts being covered by alluvium. Undoubtedly the Miocene was deposited on a floor of Eoc ene limestone and consi sts partly of material erod e d from it. It doubtl ess also overlies the basalt, although no basaltic debris was observed in the Miocene. The broader structural r e lations of the Asile Valley also are obscure. The only observed contact of the Miocene and the older rocks, on the trail to Aquin at the southeast edge of the valley, probably is a fault contact. On the ascent from the Riviere Serpent the Miocene beds, which dip to the southwest, are succeeded by supposed upper Eoc ene limestone striking N. 65 E. and dipping 72 NW.; these in turn are succeeded abruptly by a narrow band of Miocene and finally by older limestone, most of it of upper Eocene age. West of this locality the southern border of the

PAGE 262

230 GEOLOGY OF THE REPUBLIC OF HAITI. valley is formed by a steep escarpment of limestone several hundred meters high. The northern border of the valley is less precipitous and is apparently largely occupied by basalt. The mot1ntains south of this valley have probably been elevated since the Miocene beds were deposited. The structural relati ons may then be comparable to those suggested for the lowland at Ca mp Perrin (see p. 236), and the mountain wall to the south may have been uplifted either by folding or faulting, more likely both. An exposure of the Miocene beds on the south bank of Riviere Serpent, between Riviere Gaudet and Source Paul, about 5 kilometers southeast of the town of l' Asile, shows the following section : Section of fresh-water Miocene beds on Riviere Serpent. Meters. Alluvium . . . . . . . . . . . . . . . . . . . . . . . . 2 Unconformity. 6. Clay, yellowish . . . . . . . . . . . . . . . . . . . . . 3.30 5. Clay, black, carbonaceous. . . . . . . . . . . . . . . . . .16 4. Clay, yellowish, containing subangular pebbles as large as peas and a little carbonaceous material . . . . . . . . . . . .53 3. Clay, black, carbonaceous. . . . . . . . . . . . . . . . . .18 2. Clay, grayish buff with darker streaks of carbonaceous mate-rial containing carbonized fragments of wood and stems.... .20 1. Not exposed . . . . . . . . . . . . . . . . . . . . . . . .30 Stream level. Thickness of Miocene beds. . . . . . . . . . . . . . 4.69 The beds strike N. 45 W. and dip 25 SW. Beds of soft buff to white marl containing numerous fresh-wa .ter mol lusks were examined at several localities farther northwest along Riviere Serpent. (For list of fossils see p. 231 ; stations 9519, 9618, and 9619.) Similar fossiliferous marl is exposed on the south bank of Riviere des Pins just north of l' Asile (stations 9617 and 9655). The following section was measured on the west bank of Riviere la Hotte about 4 kilometers west of l' Asile on the trail to Les Cayes : Section of fresh-water Miocene beds on Riviere la H otte. Top of bluff. Meters. 7. Clay, greenish . . . . . . . . . . . . . . . . . . . . . . 4.50 6. Marl, soft, buff colored; station 9621. . . . . . . . . . . . . .36 5. Lignite, soft . . . . . . . . . . . . . . . . . . . . . . . .03 4. Marl, soft, yellowish. . . . . . . . . . . . . . . . . . . . .15 3. Impure lignite grading downward into black clay. . . . . . .15 2. Clay, black, grading downward into green clay. . . . . . . 1 I. Concealed . . . . . . . . . . . . . . . . . . . . . . . . 2 Bed of stream. 8.19

PAGE 263

SEDIMENTARY ROCKS. 231 The beds are approximately horizontal. Similar beds crop out fartl1er southwest along the stream. In the plateau w est of the lower valley, near the edge of the mountains, on the trail to Les Cay es, b e ds of coarse reddish conglomerate are ex posed in the ste ep bluffs of Rivi e re la Hotte. Limestone cobbles from 15 to 30 centimet e rs in diameter are abundant. Fragme nts of chert are common, and there is some fine sand. The beds seem to dip several de grees eastward. Similar conglomerate is found along the trail for some kilometers, but is very poorly exposed. Fossils. The fossils obtain e d from the Miocene beds of the Asile Valley are fresh-water mollusks and fruits of Chara, described by Berry as Cliara woodringi. A similar 1.1ioc ene fresh-water molluscan fauna is not known in the West Indies. The two species of Planorbis are abundant at sta tions 9519 and 9619. At s e veral lo c alities the marls contain innumerable specimens of H ydrobia. Stations in Asile Vall e y (non marine Miocene). 9519 (W 96 F). Arrondissement of Nippes, trail l e ading eastward from l'Asile, betw een Riviere Goudet and Source Paul. W. P. Woodring, coll e ctor. November 16, 1920. Also Unite d States Geological Survey fossil plant station 7555. 9618 (B 42 F). Arrondissement of Nippes, north bank of Riviere S erpent a. little more than a kilom e t e r east of l Asil e J. S. Brown, collec t or. October 31, 1920. 9619 (B 44 F). Arrondiss ement of Nippes, east bank of Riviere Serpent just above junction wit h Riviere d e s Pins, about a kilom ete1 northeast of l Asile. J. S. Brown, coll e ctor. Octob e r 31, 1920. 9617 (B 41 F). Arrondissem ent of Nippe s, south bank of Riviere d e s Pins, just north of l Asile, at crossing of trail to Anse-a-Veau. J. S. Brown, collector. Octob e r 29, 1920. 9555 (W 99 F). Arrondissem ent of Nippe s, same locality as 9617. W. P. Woodring, collector. November 17, 1920. 9620 (B 45 F). Arrondissem ent of Nippes, south banl{ of Riviere des Pins, about a kilometer west of l'Asile. J. S. Brown, collector. November 1, 1920. 9621 (B 46 F). Arrondiss e m ent of Nipp e s, w est bank of Riviere la Hotte, about a kilometer above junction with Riviere d e s Pins, and about 3 kilometers west of l Asil e J. S. Brown coll e ctor. Nove mb e r 1, 1920. N onmarin e Mioc e n e f ossi,ls from A si l e Vall e y, arrondiss e m ent of Nippes . Species. 7555 9519 9618 0019 9617 9555 9620 9621 Mollu sca: Gastropoda : Planorbis sp. a ............................... x x x x Planorbis sp. b ............................... x x Hydrobia sp. x x x x Physa sp. x x Pelecypoda : Pisidium sp. x x Plant: Ohara woodringi Berry ........................... x .

PAGE 264

232 GEOLOGY OF THE REPUBLIC OF HAITI. CAMP PERRIN. Non.marine lignite-b earing beds of Miocene age under lie a large interior lowland near Camp P e rrin. In many features this lowland resembles the Asile Valley and the smaller lowlands in the commune o:f Jeremie. The beds consist of conglomerate, sandstone, shale, lignite, marl, and lime stone. Their distribution with r e f e rence to the older rock formations is shown in Figure 16. The Miocene rocks are covered by alluvial gravels over almost the entire lowland and are exposed at places along the prjn cipal streams and dee p ravines. The most complete section examined is on La Ravine du Sud, where good exposures extend almost continuously from a locality about 600 meters north of the diversion dam of the Canal 0 I ECHEI .I.E K i.loznetres 1 2. 3 "V. -,-HAuiES -.. FIGURE 16.-Sketch map of the lignite area near Camp Perrin. N Base from map of engineers of the Department ot Public Works. See Fig. 17 tor section along line a b o a. d' Avezac downstream to a locality about 250 meters south of the dam. A section of the beds measu1ed is given on page 234. The thickness of most of the lignite beds was measured, but the thickness a ssigned to the other beds is based on the width of outcrop as deterxnined by pacing. Figure 1 shows the structural relations of these beds. The exposures along La Riviere and some tributary ravines were ex amined for a distance of 2 kilometers at a locality about 4 or 5 kilomete1s northwest of Camp Perrin. The beds consist of sandy clay, marl, and lignite and closely resemble the beds in the upper part of the section on La Ravine du Sud. As the strike of the beds is approximately parallel to the course of the river, no gre at thickness is exposed. The base of the series was not found in this area.

PAGE 265

S N Recent " 0 o __ o Grarier c b '! a r .. Miocene .. ---------------------------. . . . ..j) .... 1 ----... .oeo o ... . . -I I I I I . oo ;-..p .. --0 I " .. ........ 0 o . .. Houi.lle (lignite) Argile CaJcajre Sableet&avier Conomemt 75 100 o '25 50 Metres I Eocene Calcaiire CretaceC?> .C,alca :ire metamorpbiqt1 e FIGURE 17. Section of Miocene lignite-bearing beds exposed north of Camp Perrin on La Ravine du Sud. See fig. 16 for line of section a--b c d. Structure below the surface beds from a to o is largely hypothetlcaL ... 00 tlj t1 .... z t:d 0 0 f;j CJ:l

PAGE 266

234 -No. of bed. '4 43 42 41 40 39 38 37 36 85 34 33 82 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 e 5 8 2 2 3 1 GEOLOGY OF THE REPUBLIC OF HAITI. Section of nonmarine Miocene beds on La Ravine du Sud. Kind of bed. Strike and dip. Approximate thick ness (meters). Clay, poorly exposed ............................. Green clay ..................................... N. 60 E. ; 45 SE. lnterbedded shale and lignite .................... Gravel .......................................... Shale containing a poorly exposed bed of lignite. Shale and lignite ................................ Shale ........................................... Good bard ligni"te .............................. Thin-bedded bard shale .......................... . . . . . . . . . . Gravel ...................................... East-west; 75 S ... Shaly limestone ................................. Good bard lignite ; sample 1 ................... Shale with lignite ............................... Thin-bedded limestone .......................... Green clay ..................................... Coarse gra vel, pebbles, 2 to 3 cm. in diameter ... Green clay ............................... N. 70 W.; 76 SW. Sandy clay with shells and foesil wood ; st.ation 9624 Green clay .................................... Blue marl and shale ............................. Sbaly limestone ................................. Ligtlite and clay ................................. Blue marl ....................................... Hard I igni 'te .............. Dark-green clay ................................. N. 70 E. ; 40 SE. Shale grading into lignite ....................... t t e t t e I e I t t e t f Jreen and red clay .............................. rhinbedded shaly limestone ..................... Clay and thin seams of lignite ................... Hard lignite poorly expose d ..................... Chiefly clay, some limestone ..................... Limestone ...................................... Blue marl, containing fossil wood at base........ East-west; 45 S ... Impure limestone ............................... Blue marl and green clay ........................ Coarse gravel with fossil wood ................... Green clay with thin beds of calcareous sandstone Shaly limestone carrying thin seams of lignite ... Olay .......................................... Gravel .......................................... Chiefly green clay, some gravel .................. Diversion dam, Canal D' Avezac .................. Principally sandy clay. In lower part of section some gravel ; in upper part some limestone and traces of carbonaceous shale ; in midde of sec tion fossil wood. Station 9640 about 15 meters above base ; station 9625 about 20 meters from N. 70,, E. ; 40 SE. 35 5 1 1 7 0.6 2 0.3 2 2 5 0.9 1 0.5 9 1 9 0.5 S2 4 0.7 0.8 3 0.5 2 1 2 1.6 2 0.7 7 0.7 7 2 28 2 9 0.8 2 1 80 top . . . . . . . . . . . . . . . . . . . . . . N. 55 E. ; 50 SE. 6' Coarse gravel, some pebbles 10 cm. diameter ..... Total tbicknesa of beds exposed upstream from this locality ................................... Coarse gravel, same bed as No. 2 ............. Sandy clay, etc., same bed that overlies No. 2 ... Fault ; about 80 meters concealed ............... At base very coarse and poorly stratifted con glomerate with some boulders mere than a meter in diameter, resting unconformably on meta morphosed limestone. Grades into finer and better stratified material above. Lenticular beds N. 55 E. ; oo NW. 3 8 25 of gravel and sand near top..................... N. '5 E.; so SE. 200

PAGE 267

SEDIMENTARY ROCKS. 235 The estimated thickness of the beds in the unbroken southward or southeastward dipping section above the fault on La Ravine du Sud is 289 meters. As neither the top nor the base of the section is exposed, the total thickness of this lignite-bearing section exceeds 300 meters. The apparent thickness of the coarse basal conglomerate to the north of the fault is about 200 meters, but a part of it may possibly correspond to certain beds in the lignite-bearing section to the south. The MioceD:e beds probably have a total thickness of about 500 meters. The structure of the beds, as can readily be seen from the strikes and dips in Figure 16, is very complicated. The whole lignite-bearing part of the section appears to be crumpled into clos-e folds, which are very likely overturned to the north. The general strike of the beds is east and west, parallel to the length of the lowland, but locally the strikes are di,r ergent. The northward strikes found along a ravine in the section examined on La Riviere probably indicate faulting, although more work would be re quired to determine the structure in that locality. In the section on La Ravine du Sud there seems to be clear evidence of an extensive fault at the contact of the conglomerate and the lignite-bearing beds. As the anticlinal fold south of the contact is slightly overturned to the north and as the beds of conglomerate plunge with 11ninterrupted dip beneath this fold, the contact is interpreted as a thrust fault. (See Fig. 17.) Positive correlation of the coarse i1nfossiliferous conglomerate with the finer fossiliferous, lignite-bearing beds to the south is impossible, but as the finer beds in the conglomerate closely resemble the coarse gravel beds in the other part of the section and as the degrees of consolidation and deformation are comparable, they probably belong to the same series. If this assumption is true the conglomerate represents a basal and marginal phase of the series, and certain beds in the lignite-bearing series may pos sibly be the equivalent of some of the finer beds in the conglomerate, as shown in Figure 17. When the !fiocene beds were deposited there was a land mass on the north which supplied sediment to a lowland, probably a plain, on the south. Near the border of this lowland the sediment was coarse and was probably deposited on land. Farther south the sediment became progres sively finer, and some of it was deposited in fresh-water swamps, forming beds of clay, limestone, and lignite. A similar change in the character of the sediments can be observed at the present time on the Cayes Plain, where coarse gravel at the northern border of the plain grades into clay and even into peaty material in the swampy land near the sea. Still farther south, beneath Cayes Bay, there are marine beds. The beds of lignite and marl at Camp Perrin may grade into marine beds farther south, as Miocene marine beds are exposed in hills on the Cayea Plain.

PAGE 268

236 GEOLOGY OF THE REl>tJBLIC OF ltA1Tt. The low mountain range south of the dissected lowland has been ele vated since the Miocene beds were deposited. The upper Eocene limestone may have been thrust northward over the Miocene beds in the same manner in which the lignite-bearing beds appear to have been northward over the basal and marginal conglomerate Evidence of several kinds indicate that the lignite-bearin g beds are of Miocene age, although the fresh-water and land mollusks, lignified steins, and large pieces of wood that were collected from them do not positively establish their age. The beds are lithologically and structurally similar to the Miocene beds in similar dissected lowlands of the Southern Peninsula. Moreover, they are probably the equivalent of the marine Miocene beds that crop out on the Cayes Plain and at one time may have been continuous with those beds before the ridge of upper Eocene limestone was elevated and the Miocene beds removed from it by erosion. Fossils. An interesting fauna of land and fresh-water mollusks was obtained from the lignite-bearing beds at Camp Perrin. The land mollusks were examin e d by the late Mr. John B. Henderson, of the United States National Museum. Although most of the specimens are crushed it seems that the genera Thysanophora and Orocidopoma are represent ed. 'rhese two genera have not heretofore been found in beds as old as Miocene. Pachyclieilus is rather abundant at station 9624. The molluscan faunas of the nonmarine beds in the Asile Valley and at Camp Perrin are entirely different, although they seem to be of about the same age. Stations near Camp Perrin (nonmarine Miocene). 9624 (B 57 F). Arrondissement of Cayes, la Ravine du Sud, about 300 meters below intake of Canal d'Avezac. J. S. Brown, collector. November 5, 1920. 9625 (B 60 F). Arrondissement of Cayes, la Ravine du Sud, about 150 meters below intake of Canal d Avezac. J. S. Brown, collector. November 6, 1920. 9640 (B 61 F). Arrondissement of Cayes, la Ravine du Sud, just above intake of Canal d'Avezac. J. S. Brown, collector. November 6 1920 Non marine Miocene fossils from Camp Perrin. Species. 9624 9625 9640 Mollusca: Gastropoda : Tbysanophora sp. .................................................. x Crocidopoma sp. ................................................... x Pacbycheilus sp. ................................................... x x Hydro bia ........................................................... x Pelecypoda: Cyrena ? sp. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... x

PAGE 269

SEDIMENTARY ROCKS. 23'1 CAYES PLAIN. A considerable number of low rounded hills rise from 10 to 30 meters above the alluvial deposits in the northern and northwestern parts of the Cayes Plain. These hills were examined at only one locality along one of the main trails between Les Cayes and Camp Perrin, about 18 kilometers up the trail northwest of Les Cayea and 3 kilometers southeast of a habi tation called La Vie, which is at the border of the plain. The locality is only a few hundred met.era east of La Ravine du Sud. The trail passes between two hills which form part of a low chain that trends north westward. The rocks exposed along the trail and in an adjacent ravine consist principa lly of soft limestone and a little interbedded clayey marl. 'l,he beds are approximately horizontal. The limestone contains poorly pre served corals and mollusks. (See list below; station 9626.) Similar mar i ne Miocene beds probably form the other hills, and l\fio cene beds very likely underlie all or most of the Cayes Plain, the hills being either erosional remnants not covered by alluvium or eminences formed by uplift, either through folding or faulting. The suggestion also seems warranted that these marine beds are the seaward equivalent of the nonmarine lignite-bearing Miocene at Camp P e rrin. Fossils. Only two fossils, one coral and one mollusk, were collected from the limestone in the Cayes Plain, but the coral is similar to a Mio cene species from Cuba. Station in C ayes Plain (Miocene). 9626 (B 63 F). Arrondissement of Cayes, ma.in trail from Les Cayes to Camp Perrin, about 3 kilometers southeast of habitation La Vie. J. S. Brown, collector. November 7, 1920. Miocene fossils from station 9626, Cayes Plain. Coral: Goniopora sp. cf. G. jacobiana Vaughan. Mollusca: Lithophaga sp. GoNAVE IsLAND. Roc ks of known or supposed Miocene age cover a large part of Gonave Island, but as they consist entirely of limestone it is difficult to dis tinguish them from the older rocks, which also are limestone. The Mio cene limestone is massive, white on unweathered surfaces, and usually gray on weathered surfaces except beneath overhanging cliffs and in other sheltered places, where it weathers yellowish. Although no accurate estimate of its thickness is possible the Miocene is probably less than 200 meters thick. It rests with transgressive overlap on the upper Eocene limestone, at some places on the upper massive beds and at others on the lower chalky beds. Where it rests on the massive upper Eocene limestone

PAGE 270

238 GEOLOGY OF THE REPUBLIC OF HAITI. it is almost impossible to distinguish them 11nless fossils are fo11nd. 'fhe supposed stratigraphic relations of the Miocene limestone is shown in Figure 8, page 138. The Miocene limestone forms a fringe around the north and south coasts of the southeastern half of the island. South of Anse-a-Galets, on the north coast, it crops out on a steep slope leading up to a dissected coastal plateau. The rock there is jagged and deeply pitted by solution cavities. The corals listed on page 239, station 9661, were collected near the top of the slope a short distance east of the trail leading southwestward from Anse-a-Galets to the spring that supplies the village. This trail follows the channel of a ravine along which the Miocene limestone crops out at intervals. About 2 kilometers southwest of Anse-a-Galets the limestone is exposed in a cliff about 50 meters high, on the east side of the ravine. Corals and mollusks were collected in a shallow cavern at the foot of this cliff. (See list, p. 239, station 9663.) In the cavern t .he rock weathers to a soft powdery marl containing hard lumps of yellowish li mestone, but on the outside of the cliff it is gray and jag. ged. Near this place calcareous rind on the surface of the limestone contains specimens of the Recent land snails Urocoptis (Autocoptis) gruneri Dunker and Pleurodonte. Another collection of corals was obtained along the trail leading from Anse-a-Galets northwestward along the coast to Etroit, at a point about 4 kilometers from Anse-a-Galets, where the massive pitted limestone extends to the shore line. (See list, p. 239, station 9680.) The surface rock over the entire northwestern half of the island consists of similar massive limestone etched into fantastic shapes by solution. It is supposed to be of Miocene age, but the only fossils obtained are the impressions of the few mollusks listed on page 239, stations 9675, 9677, and 9678. Erosion remnants of this limestone resembling ruined walls rest on the chalky upper Eocene limestone about a kilometer south of Fond-Negre on the trail to Grande Ravine. On the same trail, about 3 kilometers northwest of Grande Ravine and near the foot of the slope leading up to the interior plateau, massive limestone of supposed Miocene age rests on an uneven surface of the chalky upper Eocene limestone that under lies the plateau. Fossils. The corals collected from the limestone on Gonave Island have a Miocene aspect, and some of them are similar to species obtained from Miocene beds on the mainland. The mollusks have no stratigraphic significance. Stations on Gonave Island (Miocene). 9661 (W 153 F). Gonave Island, short distance east of trail leading southwest ward from Anse-a-Galets, about a kilometer southwest of Anse-a-Galets; altitude 60 meters above sea level. W. P. Woodring, collector. December 17, 1920. 9663 (W 155 F). Gonave Island, trail leading southwestward from Anse-a Galets, cliff on right side of dry ravine about 2 kilometers from Anse-A-Galets. W. P. Woodring, collector. December 17, 1920.

PAGE 271

SEDIMENTARY ROCKS. 239 9680 (W 175 F). Gonave Island, trail from Etroit to Anse-a-Galets, about 4 kilometers northwest of Anse-a-Galets. W. P. Woodring, collector. December 25, 1920. 9675 (W 169 F). Gonave Island, trail from Grande-Ravine to Fond-N egre, about 5 kilometers northwest of Grande Ravine; altitude 410 meters above sea level. W. P. Woodring, collector. December 22, 1920. 9677 (W 171 F). Gonave Island, trail from Fond-N egre to Dandeville, about 8 kilometers west-northwest of Fond-N egre; altitude 190 meters above sea level. W. P. Woodring, collector. December 23, 1920. 9678 (W 172 F). Gonave Island, trail from Fond-N egre to Dandeville, about 15 kilometers west-northwest of Fond-N egre; altitude 275 meters above sea level. W. P. Woodring, collector. December 23, 1920. Miocene fossils from Gonave Island. Species 9661 9663 9680 9675 9677 9678 Corals: Stylophora sp. . . . . . . . . . . . . . . . . . . . . . . . x x Stephanocoenia intersepta (Esper)....... . . . . . . . . . . . . ... x Meandrina sp. apparently new. . . . . . . . . . . . . . . . x Orbicella sp. cf. 0. altissima (Duncan). . . . . . . . . . . . . x Mollusca: Gastropoda : Conus sp. . . . . . . . . . . . . . . . . . . . . . . . . . . x . . . . . . ... Xancus ? sp.............................................. . . x . . . . . . ... Cerithium ep. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x .... Emarginula ? sp. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Pelecypoda: Ohlamys ( Aequipecten) sp. . . . . . . . . . . . . . . . . . . . ... x Lucina ? sp x . . . . . . ... PLIOCENE SERIES. Faunal evidence and the degree of deformation of the rocks are the c1iteria used in recognizing marine deposits of Pliocene age in Haiti. Judged by these criteria marine Pliocene deposits appear to be confined to relatively small areas in the Southern Peninsula. Some of the corallif erous limestones and other ma rine beds called Quaternary may really be Pliocene, but all the beds classed as Quaternary are undeformed or only slightly deformed and have approximately the same fauna. N onmarine deposits of Pliocene age are known in some of the larger valleys and 7 plains. They are more dissected than similar Quaternary beds and may consist of different material. MARINB DEPOSITS. v ALLEY OF RIVIERE GAUCHE. The lowland that is drained by Riviere Gauche, extending northwest ward from J acmel for at least 17 kilometers, contains marine deposits of Pliocene age. On the northeast side of the lowland these beds rest 11nconformably on upper Eocene limestone, although at some localities they

PAGE 272

240 GEOLOGY OF THE REPUBLIC OF HAITI. have about the same strike and dip as the upper Eocene rocks. The relation to the older rocks along the foot of the steep, cliff-broken slope on the southwest side of the lowland are not known. 'fhe Pliocene beds are folded, -and along the trail fol lowing Riviere Gauche they dip steeply southwestward They consist principally of conglomerate and marl. The conglomerate is firmly consolidated, except certain very coarse beds, and contains pebbles and cobbles of different kinds of limestone, which have a maximum diameter of half a meter. It contains also cobbles of chert and basalt, which, however, are much less numerous. At some places the conglomerate is made up of pebbles of uniform size, 5 or 6 centimeters in diameter. Interbedded with the conglomerate are beds of gray and yellow marl containing sandy and clayey layers. Some of the beds of marl carry perfectly preserved fossils. Fossils were collected from thin marl beds at two localities. (See list, p. 242, stations 9529 and 9530.) There are also a few beds of lim estone containing poorly preserved corals and mollusks. The strike of the beds along Riviere Gauche below Bou cicaut is parallel to the longer diameter of the lowland and the dip is southwestward at angles of 40 to 50 The thickness of these beds is not known but is probably more than 100 meters. Near the mouth of Riviere Gauche and on Riviere Gosseline, at the southeast end of the lowland, the Pliocene beds dip more gently south westward at angles of 10 to 25, having virtually the same strike and dip as the underlying upper Eoc ene limestone. bluff on the right bank of Riviere Gosseline exposes marly sandstone and sandy marl containing casts of the mollusks li s t e d on page 242, sta .tion 9604. T APION DU PETIT-GOA VE. The beds in the gap south of Tapian du Petit-Goave seem to consist chiefly of the Miocene rocks that are described on page 224. Along the road on the west slope a thick coralliferous limestone overlies the Miocene rocks. It is more massive than the coralliferous limestones interbedded with the Miocene conglomerate but seems to have about the same dip. An extensive series of corals (see list, p. 243, station 9793) collected along the road at an altitude of 150 meters above sea level contains only genera still living in the West Indies, and this limestone is believed to be either Pliocene or Quaternary probably Pliocene, because the beds are more greatly deformed than those classed as Quaternary. NONMARINE DEPOSITS. Nonmarine deposits of Pliocene age probably 11nderlie similar Quaternary deposits in most of the larger valleys and plains, but detailed work would be required to separate them. Most of the dissected nonmarine deposits in these regions are arbitrarily called Quater nary.

PAGE 273

SEDIMENTARY ROCKS. 241 CENTRAL PLAIN. Hinche f orma.tion. In the Central Plain b e ds of silt, clay, sand, and gravel, usually uncon solidated or imp e rfectly consolidated, rest on the eroded surface of the folded rocks of the Artibonite group. These beds were apparently laid down along a drainage system that emptied southeastward into the San Juan Valley of the Dominican Republic. Jones 1 gave the name '' Hinche beds,'' d e rived from the town of Hinche, to these stream deposits. So far as observe d these ro c ks are undeformed. They closely resemble imperfectly consolidated ro c ks of sim ilar orjgin in the Las Cahobas formation, and where the Las Cahobas rocks lie flat, as in the trough of the Central Plain syncline, the two for1nations can hardly be distinguish e d. The gravels of the Hinche beds contain a greater proportion of limestone pebbles than is found in poorly consolidated conglomerates in the upper part of t he Las Cahobas formation, but they do not contain the large bould e rs of limestone and brown chert that are common in the Quaternary stream gravels. The Hinche beds underlie the extensive savanna in the northwestern part of the plain and patches of them are found at succes sively lower altitudes in the southeastern part. The estimated maximum thickness of this formation is 25 m ete rs. The Hinche formation is considered Pliocene because of its strati graphic position. It is the equivalent of the Las Matas formation of the San Juan Valley in the Dominican Republic, and the two formations are probably continuous. FOSSILS. 'l'he corals obtained from the Pliocene beds on Riviere Gau che have a more modern aspect than the Miocene corals collected in the Republic. Only five genera are represented, but these genera a1e all still living in the \Vest Indian region. Maeandra labyrinth if ormis (Linnaeus) and Acropora muricata (Linnaeus) have never been recorded from West Indian Mio cene deposits. The molluscan fauna seems to be intermediate between MiQcene and Quaternary faunas. Species like Fusinus sp. cf. F. ulcimus ru.shii Dall, Astrea sp. cf. A. caelata Gmelin, Oalliostoma sp. cf. 0. zonamesta Reeve, and Torinia sp. cf. T. cylind1ica ( Gmelin) are more modern than any Miocene species. The Sconsia is inter1nediate between the Miocene B. laevigata (Sowerby) and the Recent S. striata (Lamarck). Torinia rotundata Gabb, recently :figured by Pilsbry,1 is the only Torinia recorded from West Indian Miocene deposits, although a similar species is listed on page 225, station 9481, from the Miocene at the west end of the Leogane 1 Jones, W. F., A geological reconnaissance in Haiti; a contribution to Antillean geology : J our. Geology, vol. 26, p. 7 48. 1918. s Pilsbry, H. A., Revision or W. M. Gabb's Tertiary Mollusca of Santo Domingo: Acad. Nat. Sci. Philadelphia Proc., vol. 73, p. 379, pl. 34, figs. 19-20, 1922. 16

PAGE 274

242 GEOLOGY OF THE REPUBLIC OF HAITI. Plain. The species from the Pliocene beds is more similar to the Recent T. cy lindrica ( Gmelin). Oth e r mollusks from thes e b e ds res e mble Bowden and Gurabo (Miocene) species. Marine deposits of Pliocene age have not been recognized in other West Indian islands, with the doubtful exception of Cuba, but beds of this age are known at Port Limon, Costa Rica. Most of the corals from the limestone on the west slope of Tapion du Petit-Goave are Recent species, but the l i mestone is considered of doubt ful Pliocene age because of its structural relations Stations in arrondi ssement of Jacmel (Pliocene). 9529 (K 30 F). Arrondissement of Jacmel, trail from Jacmel to Boucicaut, right b ank of Riviere Gauche, about 7 kilometers northwest of Jacmel. W. S. Burbank, coll ecto r. November 16, 1920. 9530 (K 31 F). Arrondi ssement of Jacmel, trail from Jacmel to Boucicaut, left bank of Riviere Gauche, about 9 kilometers northwest of Jacmel. W. S. Burbank, collector. November 16, 1920. 9604 (W 68 F). Arrondissement of Jacmel, trail from Jacme l to Carrefour, right bank of Riviere Gosseline 100 mete rs above second crossing above J acme!. W. P. Woodring, collector. November 3, 1920. Pliocene fossils from b e ds on Riviere Gauche, arrondissement of J acmel. Species. Corals: Solenastrea hyades (D8lla) ............................................... Maeandra labyrinthilormis (Linnaeus) ................................... Maeandra ? sp .................... Siderastrea sp. afJ. S. siderea (Ellie and Solander) ....................... Agari cia sp. ............................................................ Acropora muricata (Linnaeus) .......................................... Mollusca: Pteropoda: Cavolina sp. cf. C. tridentata. Forskal ............................... Gastropoda : Oliva sp. cf. 0. reticularis Lamarck ................................ Olivella sp. cf. 0. jaspidea r otunda Dall ............................ Fusinus sp. cf. F. ulcimus rusbii Dall ............................ Phos sp. cf. P. moorei Guppy ........................................ Murex sp. cf. M. domingensis Sowerby ............................. Sconsia sp. cf. S. laevigata (Sowerby) ............................ Col ubraria sp. ............................................... .... Turritella sp. cf. T. submortoni Maury ............................... Astrea sp. cf. A. caelata (Gmelin) ................................. Calliostoma sp. cf. 0. zonamesta Reeve .............................. Torinia sp. cf. T. cylindrica (Gmelin) .......................... Pelecypoda: Leda sp. ........................................................... Glycymeris sp. cf. G. pennacea (Lamarck) .......................... Chlamys (Aequipecten) sp. cf. 0. (A.) uselmae Pilsbry and Johnson. Cardium (Fragum) medium Linnaeus ................. 9529 9530 960, x x x x x x x x x x x x x ? x x x x x x x x x

PAGE 275

SEDllfENTARY ROCKS. 243 Stations on west slope of Tapion du Petit-Goave (Pliocene r). 9793 (W 317 F). Arrondissement of Leogane, road from Leogane to Miragoane, west slope of Ta pion du Petit-Goa ve, altitude 150 meters above sea level. T. W1 Vaughan and W. P. Woodring, collectors. March 6, 1921. 9470 (B 33 F). Arrondissem ent of Leogane, road from Leogane to Miragoane, west slope of Tapion du P etit-Goavc, altitude about 100 meters above sea level. J. S. Brown, collector. October 23, 1920. Pliocene ( 1) fossils from west slope of Tapion du Petit-Goave. Species. Corals : Orbicella annularis (Ellis and Solander) ............ S olenastrea b ournoni Mil ne-Edwards and Haime ........ ........ Solenastrea byades (Dana) ........................................................ M aeandra labyrinthiformis (Lin naeus) ........ .............. Siderastrea sp. aff. S. sidera (Ellis and Solander) ........... Agaricia agaric i tes var. purpurea (Le Sueur) .................. Porites sp. cf. P. porites (Pallas) ............................................. Porites sp. cf. P a streoides Lamarck ............................................ Mollusca: Gastropoda : 9793 9470 x x x x x x x x Bullaria ? sp......... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X Pelecypoda : Arca um bona ta Lamarck .............................. Chlamys (Aequipecten) sp. cf. C. (A.) phrygium Dall ............. Cblamys (Plagioctenium) sp. cf. 0. (P .) gibbus Lamarck ...... Oatrea sp. ................... . Venus ca.mpecbiensis Gmelin ? Chi one ? ep Ma c roc.allista f sp QUATERNARY SYSTEM. GENERAL FEATURES. x x x x x x x x x The effects of the rise and decline of glaciation in the temperate regions are apparently not r e cognizable in the Republic, either directly or indi rectly, and the limits of Plioc e ne, Pleistocene, and Recent time are poorly defined, depending on several criteria, none of which can be rigidly applied. Deposits that are poorly consolidated, that are undeformed or only slight ly d e formed, that are dissected by present streams, and that contain a fauna identi c al with or very similar to the living fa1ma are here considered Quaternary. Quaternary sed i mentary deposits are rather widely distributed in the Republic, although most of the areas in which they fonn the surface rocks are small. (See Pl. I.) They may be divided into marine and non marine deposits. MARINE DEPOSITS. Marine Quaternary deposits are common only along the coast, where they form narrow fringes, at few places more than 1 or 2 kilometers wide.

PAGE 276

244 GEOLOGY OF THE REPUBLIC OF HAITI. However, they spread over virtually all the Bombardopolis Plateau, thus extending back many kilometers from the coast. Small patches are fo11nd along the north and south borders of the Cul -de-Sac Plain to and beyond the international boundary. This probably is the greatest distance inland at which marine Quaternary deposits have been found, but when they were deposited the sea occupied the trough now known as the Cul-de Sac Plain, and these deposits, like the others, are the ref ore coastal fringes. These marine Quaternary deposits cover the emerged coastal terraces that stretch along nearly one-half the coast of the Republic. STRATIGRAPI-IIC RELATIONS. The Quaternary deposits everywhere rest on the underlying formations unconformably. The unconformity is angular as well as erosional, even with the Pliocene and Miocene, but at some places where there are no Pliocene deposits the discordance with the Miocene deposits may be slight. LITHOLOGY AND LOCAL DETAILS. The greater part of the marine Quaternary deposits are reef deposits very similar to those now forming at many localities around the shore of the Republic. They consist of reef rock, coralliferous limestone, a11d molluscan limestone, but reef rock and corallif erous limestone are the most common, a fact indicating that fringing reefs were :flourishing during probably all of Quaternary time. If some of these rocks are indeed of Pleistocene age, as seems probable, there seems to be no basis, at least so far as the Republic of Haiti is concerned, for the contention that the vigorous growth of Pleistocene reefs was retarded by the supposed lowering of the temperature of tropical seas during Pleistocene time. Although detailed work should be done to warrant definite statements, it seems that the maxima of Pleistocene glaciation can not be recognized by their effect on the Pleistocene reefs. The reefs of supposed Pleisto cene age are similar in many features to the Recent reefs. The list on page 250 shows that Orbicella annularis (Ellis and Solander) was the most common reef builder in the emerged Quaternary deposits, just as it is the most common coral on the living West Indian reefs. At few places does the weathered surface of the limestone give a cor rect impression of its lithology, which can be observed only in very fresh stream-cut bluffs or in artificial excavations. In such favorable exposures the rock is seen to consist of thick-bedded, poorly assorted, and very poorly consolidated calcareous and detrital materials, in most places containing a large proportion of heads of corals and shells of mollusks. There are all gradations of rock from pure coral and pure coquina to gravel derived entirely from land debris. Ordinarily the purer calcareous rock is found near the sea or on the outer margins of terraces, and the detrital material is fo11nd at the landward margin of the deposits. The charac I

PAGE 277

SEDl1'ENTARY ROCKS. 245 teristic gradation is well shown along the Riviere des Roseaux near Roseaux, a village about 7 kilometers east of Jeremie. The terrace near the shore is composed very largely of heads of corals and shells of mol lusks. To the south the organic remains become rarer and rarer, and the detrital materials become more and more abundant, until in the bluffs about a kilometer south of Rosea11x only dissected water-laid gravels can be found. The zone of gradation ordinarily, however, is not conspicuous. At some places where the initial slope was steep a narrow reef is plastered almost directly on the basement of the older rocks. This relation is shown at St.-Marc Bay, where relatively wide terraces veneered with reef rock or coralliferous limestone cover the promontories north and south of the bay. At the head of the bay, just south of the place where the road to Gona1ves leaves the coast and enters the ravine, a narrow fringe of coralliferous limestone rises to an altitude of 15 meters above sea level. The limestone grades laterally and vertically into marl and poorly con solidated gravels. The fossils lis ted on pages 250-254 (station 9551) were collected here. At some places the zone of gradation is obscured by later alluvial deposits or is indistinguishable from them. In places it seems t.hat the deposits in the zone of gradation are softer than the more cal careous rock and have been eroded away, as a t the rear of the coastal es carpments near Anse Rouge and north of Jean Rabel. The erosion in both these localities, however, has undoubtedly been due primarily to the presence of soft Miocene beds back of the Quaternary limestone. A lateral gradation in constituent material somewhat like that just described was observed at some places directly on the shore, where the prevailing corallif erous limestone of the lowest coastal terraces is replaced for an interval by stratified gravel or sand that generally contains marine fossils in considerable numbers. This type of material doubtless repre sents the delta deposits of streams which at their mouths interrupted the continuity of the reefs by depositing a large amount of sediment. Such a gradation is particularly well shown just west of the mouth of Riviere de Jean Rabel, along the trail between Mole St.-Nicolas and Jean Rabel. The corallif erous limestone typically developed 2 or 3 kilometers west of that stream grades eastward into stratified sand and coarse gravel containing few fossils, such as that found in bluffs along the Rivi ere de Jean Rabel. Again, between Petit-Riviere de Nippes and Grande-Riviere de Nippes the prevailing coralliferous limestone is replaced by fine-grained soft brown sand containing shells of mollusks. (See list, pp. 250-254; station 9520.) At these two localities, as probably at most others, the main streams of Quaternary time seem to have occupied approximately the same valleys which they now occupy, and Quaternary delta deposits may normally be expected at the mouths of all the larger streams where the coast has emerged during Quaternary time. Where they have been long exposed to weathering the Quaternary rocks assume an appearance very different from that which they show in fresh

PAGE 278

246 GEOLOGY OF THE REPUBLIC OF HAITI. excavations. The fossils are commonly dissolved by preferential leaching, leaving the surface of the rock pitted, and only casts or impressions of the fossils remain. A veneer of travertine deposited over the rock surface in many places accentuates its massive appearance. There is also a pro nounced hardening of the outer crust, which commonly extends to a depth of severa l meters. This hardening is probably due mainly to firm cementation of the particles by redeposited calcium carbonate. The density of the rock also may be increased to some extent by this introduced material, some of which was probably brought to the surface by capillary water, which having penetrated the rock to a shallow depth returned to the surface and evaporated, leaving its dissolved material behind. As a result of these changes the purer calcareous Quaternary rock may closely resemble phases of the massive upper Eocene and other Tertiary limestones, as for instance, in the cliff shown in Plate XVII, A. Some of the Quaternary limestone also yields a red clay soil almost identical with that characteristic of areas of the massive Tertiary limestones. The following section of Qua ternary deposits is exposed in a sea cliff 7 meters high just west of the colonial fort on the west side of the entrance to the harbor of Port-de-Pai:x: Section of Quaterna ry beds expos e d near Port-de-Paix. Meters. Reef rock, yellowish buff, soft, containing large heads of Orbi-cel'la annularis and other corals; station 9765. . . . . . . . 3.6 Waterwom pebbles, poorly consolidated, in matrix of sand; lenses of sand containing worn shells and small pieces of corals ; station 9764. . . . . . . . . . . . . . . . . . . . 1.5 Sand, rusty brown, poorly consolidated, fragmentary shells in pockets . . . . . . . . . . . . . . . . . . . . . . . . . 2 Sea level. 7.1 The lenses of sand in the middle bed furnished the la.rgest collection of Quaternary mollusks obtained in the Republic. (See list, pp. 250-254, station 9764.) The overlying reef rock contained only a few mollusks. The remarkable emerged coastal terraces of the Northwest Peninsula are described on pages 371-375, and the fossils obtained from the coralliferous limestone covering them are listed on pages 250-254 (stations 9841, 9840, 9844, 9838, 9837, 9836, 9835, 9834, 9833, 9832, and 9831). In the Cul-de-Sac Plain the most extensive areas of Quaternary lime stone lie along the south side. The best collection of corals was obtained at Balan, near the southwest corner of Etang Sa11matre. (See list, p. 250; station 9659.) Along the trail from Gantier to Fond-Parisien cavernous limestone crops out near Pont Quinet at an altitude of about 150 meters above sea level. This limestone is the surface rock for most of the distance from this locality eastward to Fond-Parisien and probably forms the ridge that exi"iends northward to the lake. Partly dissolved corals were collected about 3 kilometers northwest of Fond-Parisien at an altitude of 90 meters above sea level. (See list, p. 250; station 9591.)

PAGE 279

C OJ<' IT .\ l J l LATE XVII _4. PITTED QT. ATERX ARY I..1J:\IEST(JXE (ROCIIE-A-RA \TET) IX l 'Ili''l'II SI:.i\. CLII1"'I 1"' OX 1'IIR Tl{AJL li'RQ:\[ l)E1"IT JlAl{AJ)J S 1_ro B ,.\.IE fl. Pf.1EIS1"'0CENE Tilflj ROAD TO FORT N 1)01{1'-.-t\. U -IJl IX('E. Xote th(l hnrflenecl ltrface cr1 1 t

PAGE 280



PAGE 281

SEDIMENTARY ROCKS 247 At numerous other l ocalities where Quaternary deposits fringe the coast they consist of one of the types of deposits that have just been de scribed. The terraces on which they rest are fully described in Part III. THICKNESS. Because of the great unevenness of the floor on which it was deposited and the unequal amount of erosion at the surface the thickness of the marine Quaternary is decidedly variable It probably exceeds 50 meters at but few places, although on the Bombardopolis Plateau the maximum thickness may be 100 meters. At La Gorge, on Riviere du Mole, south east of Mole St.-Nicolas, the contact of the Quaternary deposits with the underlying Miocene beds is at an altitude of less than 100 meters above sea level, and t he Quaternary deposits rise to heights of abo11t 400 meters above sea level on the plateau on either side of the river valley. It can not be assumed, however, that the intervening interval represents the thickness of the Quat e rnaIJ ', b eca use on each of the terraces the Quater nary is represented only by a thin lens of deposits that gradually tapers landward and more rapjdly seawa rd. STRUCTURE. The structure of the Quaternary formations is very simple Although they have undergone arching, as indicated by the unequal amounts of emergence of the coastal terraces, especially in the Northwest Peninsula, they are not at all crumpled, as are the underlying formations At most places they were deposited with a pronounced seaward dip, commonly from 5 to 15, which is generally visible in large excavations or in deep stream channels. As the direction of the coast line changes the direction of dip determined by the initial slope usually changes correspondingly. FOSSILS. Some of the collections here classified as Quaternary may really be Pliocene. It is very difficult to distinguish Pleistoc e ne from Pliocene faunas, a difficulty which probably accounts for the apparent scarcity of Pliocene deposits in the American tropics. No adequate attempt has yet been made in the American tropics to sepa .rate Pleistocene and Recent faunas, much less to recognize faunal zones in the Pleistocene. Corals are the most abundant marine Quaternary fossils, as the most common rock is coral reef rock or coralliferous limestone. Virtually all the species are identical with living ones. Orbicella annularis (Ellis and Solander) was obtained at almost every locality wh ere collections were made from rocks of this type Solenastrea bournoni Milne-Edwards and Haime, Siderastrea siderea (Ellis and Solander), and Acropora muricata (Linnaeus) are other common Quaternary corals. The lists of Miocene corals (see pp. 178, 203, 217) show that the following Recent species were living in Miocene time: Orbicella annularis, Dichocoenia stokesi, Stephanocoema intersepta, Solenastrea bournoni, Solenastrea hyades, and Siderastrea siderea. A small collection of Miocene corals may therefore

PAGE 282

248 GEOLOGY OF THE REPUBLIC OF HAITI. resemble a Quaternary fauna, unless it includes Stylophora, Pocillopora, Antillia; Oyathomorpha, or Goniopora.1 The molluscan fa11na of the Quaternary fringing reefs is rather meager, but the total molluscan fa11na is extensive, because large collections were obtained at localities where the deposits consist of sand and gravel or of limestone that contains hardly any corals. Strombus gigas Linnaeus, Arca umbonata Lamarck, Oodalcia orbicularis (Linnaeus), and Phacoides pennsylvanicus (Linnaeus) are the common mollusks of the fringing reefs. Not enough time was given to the determination of the mollusks to sl1ow clearly whether any of them are different froin living West Indian species. Most of those that are not determined specifically are poorly preserved. Stations in marine Quaternary deposits. 9764 (W 301a F). Arrondissement of Port-de-Paix, on the coast, 100 meters west of old fort on west side of entrance to Port-de-Paix harbor; from middle bed of section on page 246. W. P. Woodring, collector. February 21, 1921. 9765 (W 301b F). Arrondissement of Port-de-Paix, same locality as 9764; from upper bed of section on page 246. W. P. Woodring, collector. February 21, 1921. 9841 (B 210 F). Arrondissement of Mole St.-Nicolas, trail from Mole St. Nicolas to Jean Rabel, about 2 kilometers northeast of Mole St.-Nicolas. J. S. Brown, collector. January 30, 1921. 9840 (B 209 F). Arrondissement of Mole-St.-Nicolas, cliff at northeast end of bay. J. S. Brown, collector. January 30, 1921. 9844 (B 211a F). Arrondissement of Mole St.-Nicolas, trail from Mole St. Nicolas to Bombardopolis, about 2 kilometers south of Mole St.-Nicolas. J. S. Brown, collector. January 31, 1921. 9838 (B 207 F). Arrondissement of Mole St.-Nicolas, trail from Mole-St. Nicolas to Bombardopolis, about 8 kilometers northwest of Bombardopolis. J. S. Brown, collector. January 29, 1921. 9837 (B 206 F). Arrondissement of Mole St.-Nicolas, trail, from Mole St. Nicolas to Bombardopolis, about halfway between the towns. J. S. Brown, col lector. January 29, 1921. 9836 (B 205 F). Arrondissement of Mole St.-Nicolas, trail from Mole St. Nicolas to Bombardopolis, about 7 kilometers northwest of Bombardopolis. J. S. Brown, collector. January 29, 1921. 9835 (B 204 F). Arrondissement of Mole St.-Nicolas, trail from Mole St. Nicolas to Bombardopolis, about 5 kilometers northwest of Bombardopolis. J. S. Brown, collector. January 29, 1921. 9834 (B 203 F). Arrondissement of Mole St.-Nicolas, trail from Mole St. Nicolas to Bombardopolis, about a kilometer northwest of Bombardopolis. J. S. Brown, collector. January 29, 1921. 9833 (B 200 F). Arrondissement of Mole St.-Nicolas, trail from Bombardopolis to Baie de Henne, about 3 kilometers west of Baie de Henne and halfway up slope of Morne Chien. J. S. Brown, collector. January 27, 1921 . 9832 (B 199 F). Arrondissement of Mole St.-Nicolas, trail from Bombardopolis to Baie de Henne, 2 kilometers west of Baie de Henne, foot of Morne Chien. J. S Brown, collector. January 27, 1921. 9831 (B 197 F). Arrondissement of Gonaives, trail from Terre-N euve to Anse Rouge, gap in low range of coastal hills about 5 kilometers southeast of Anse Rouge. J. S. Brown, collector. January 25, 1921. 1 There are other West Indlan Miocene genera now confined to the IndoPaclftc region, or extinct. (See Vaughan, T. W., U. S. Nat. Mus. Bull. 103 p. 222, 1919.) The genera listed are knQwn in the Miocene fauna from the Republic of Haiti. I

PAGE 283

SEDIMENTARY ROCKS 249 9551 (W 2 F). Arrondissement of St.-Marc, road from St.-Marc to Gona!ves, steep slope on east side of t'oad at outskirts of St.-Marc. W. P. Woodring, col lector. September 29, 1920. 9479 (B 120 F). Arrondi ssement of St.-Marc, road from St.-Marc to Port-auPrince, 4 kilomet e rs northwest of Mont Rouis. J. S. Brown, collector. December 7, 1920. 9659 (W 110 F). Arrondissement of Port-au-Prince, Balan, southwest corner of Etang Saumatre. W. P. Woodring, coll e ctor. November 24, 1920. 9504 (W 30 F). Arrondissement of Port-au-Prince, trail from Gantier to FondParisien, 4 kilomet ers east of Gantier, at an altitude of 150 meters above sea level. W. P. Woodring, collector. October 20, 1920. 9591 (W 31 F). Arrondissement of Port-au-Prince, trail from Gantier to FondParisien, 3 kilometers northwest of Fond-Parisien, at an altitude of 90 meters above sea level. W. P. Woodring, collector. October 20, 1920. 9539 (B 28 F). Arrondissement of Leogane, road from Leogane to Miragoane, about 2 kilomet.ers west of l'Acul. J. S. Brown, collector. October 22, 1920. 9576 (B 29 F). Arrondissement of Leogane, road from Leogane to Miragoane, Source Mahot. J. S. Brown, coll ec tor. October 22, 1920. 9469 (B 30 F). Arrondissement of Leogane, road from Leogane to Miragoane, short distance west of Source Mahot. J. S. Brown, collector. October 22, 1920. 9613 (B 35 F). Arrondissement of Nippes, trail from Miragoane to Anse-a Veau, about a kilometer east of village of Charlier. J S. Brown, collector. October 28, 1920. 9520 (W 103 F). Arrondissement of Nippes, trail from Miragoane to Anse-a Veau, about halfway b e tween Petit-Riviere de Nippes a11d Anse-A-Veau. W. P. Woodring, collector. November 18, 1920. 9614 (B 36 F). Arrondissement of Nippes, road s ide cut descending steep slope in town of Anse-a-Veau. J. S. Brown, collector. October 29, 1920. 9484 (B 115 F). Arrondissement of Nippes, about 300 meters east of PetitTrou de Nippes. J. S. Brown, collector. November 22, 1920. 9637 (B 100 F). Arrondissem ent of Grand'Anse, trail from Les Roseaux to Corail, about 2 kilometers east of Les Roseaux. J. S. Brown, collector. N ovember 19, 1920. 9638 (B 101 F). Arrondiss ement of Grand'Anse, shore line at Les Roseaux. J. S. Brown, coll e ctor. November 19, 1920. 9627 (B 74 F). Arrondissement of Coteaux, cliffs along coast 5 kilometers northwe s t of Port-a-Piment. J. S. Brown, collector. November 13, 1920. 94 73 (B 67 F). Arrondissement of Cay es, sea. cliff a kilometer northwest of Port-Salut. J. S. Brown, collector. November 11, 1920. 9474 (B 68 F). Arrondissem ent of C a y es, sea cliff 1.5 kilometers northwest of Port-Salut. J. S. Brown, collector. November 11, 1920. 9652 (K 22 F). Arrondissement of Aquin trail from Aquin to Cotes-de-Fer, a.bout 9 kilometers west of Cotes-de-Fer. W. S. Burbank, collector. November 12, 1920. 9650 (K 29 F). Arrondissem ent of J acmel, trail from Bain et to J acmel, from surface of lowest terrace. W. S. Burbank, collector. November 14, 1921. 9603 (W 63 F). Arrondissement of J acme!, east side of J acm e l Bay about 2 kilometers southeast of Jacmel, at. an altitude of 1 meter above sea level. W. P. Woodring, collector. November 1, 1920. 9514 (W 62 F). Arrondissement of Jacmel, east side of Jacmel bay, about a kilometer southeast of Jacmel, at an altitude of 50 meters above sea level. W. P. Woodring, collector. November 1, 1920. 9602 (W 61 F). Arrondissement of Jacmel, same locality as 9514, at an altitude of 60 meters above sea level. W. P. Woodring, collector. November 1, 1920. I

PAGE 284

Marine Quaternary f ossi.ls. GIS I C) ... ;:> OS I MO le St.Nicolas. G> 'O I 0 Spe c ies. ... 0 r:n 0 0 i I t-i ... Q) c:-:1 .-4 ,.... 0) O> i Cl) r-8 t-s 00 ..... 0) 0) 0) 0) 0) O> 0) 0) 0) 0) Foraminifera : Orbiculina adunca Fichte! and Moll ? x Corals: Madracls decactis (Lyman) ........... x x Steph a nocoenia intersepta (Esper) .... x x x Eusmilia fastigiata (Pallas) ...... ... x Dic ho c c enia stokesi Mil ne Edw ards a n d Hai me x x Dic hocoenia sp., apparently new ...... x Meandrina maeand.rites (Linnaeu s ) ... Orbicellaannularis (Ellis and Solander) x x ><. x x ? x x Orbicella cavernosa (Linnae u s) ... ... x x x Solenastrea bournoni Milne -Edwards and H a i m e x Solenastrea hyades (Dana) ........... Maeandra labyrinthiformis (Linnaeus) x ? x Maeandra areola ta (Linna eus) ........ x x x Ma eandra strigosa (Dana) ...... ...... x x x Ma eandra clivosa (Ellis and Solander) x x Mani cina gyrosa (Ellis and Soland er) x Siderastrea radians (Pallas) .......... x x Siderastrea s iderea (Ellis and Soland e r ) x x x Agaricia agaricites (Linnaeus ) ....... x x x Agarici a agarici t e s var purp urea ( Le S u e ur) x x Acrop ora muricata (Linnaeus) . ...... x x Acropora palmata (Lamarck) .... .... x x Porites porites (Pallas) .............. x x Porite s f urcata Lamarc k .............. x P orites astre oides Lamarck . . ....... x x Porites sp. cf. P. a s t reoides Lamarck. 8 G> s= f.111 .... G> ... a> c 6> .. I 'O :::s s:a. c al ..... z OS ,.. ... 0 0 .-4 O> Cf,) 0 -.fl t-0) Cf,) to 8 C'l 8 8 l.l) lQ l(!) 0) 0) O> 0) CD 0) x x x x x x x x x x ? ? x x x ? all 4> .... oS :::3 ;;..... .8 O' oS 0 < 8 00 t-c:-:1 8 t--d4 C') CD 0) 0) I x x x x x x x x x ? x x x x x Jacmel. 0 8 ,.... 0) x x x x x ? x x x x x x x x Ol 0 0 t-4 0 0 td t-4 H c 0 > H

PAGE 285

M .... OS GI "' Species. ... 0 I ..-4 ..... ..... ..... t-s I 00 00 O> O> O> Hydrocorallinae: Millepora alcicornis Linnaeus ......... Brachipoda : Liothyrina sp. x Mollusca: Gastropoda : Acteocina candei (d'Orbigny) x Bullaria stria ta (Bruguiere) ... Con us sp cf. c. pygmaeus Reeve x Oliva reticular is Lamarck ... x Olivella jaspidea Gmelin ? . Olivella ep. x Marginella sp. cf. M. n1vosa Hinds Mitra sp. cf. M. sulcata Gmelin . x I 0 Oolumbella mercatoria Linnaeus. x Murex (Phy llonotus) po mum Gme-. lin ? e e a a a I I a a I I I I I I I e I 0 Mor um (Linnaeus) . x OD18CU9 Oypraea lurida Linnaeus . x x Oypraea exanthema Linnaeus . Trivia pedicula Linnaeus .... x Trivia sp. x Strom bus gigas Linnaeus ..... x Strom bus pugilis Linnaeus ... Ceritbium caudatum Sowerby Cerithium sp. cf. c. fiorida11um MBrch x Cerithium sp. cf. c. literratum Born x - Marine Quaternary f ossil.8 Continued. c:.> ..... .... > i::::: ; MO le St. Ni c olas. )l =' I OS 8 "Q,) 00 ... 0 t-(0 ...... f2 ,.-4 r-"4 O> i ..... r-"4 O> (0 &3 t.O s O> I:-' O> O> O> OI O> O> O> OI x I x I x I I I o I I I I I I o I o I t I 0 I I I x x Q) llQ r:: GQ Q> < .. "' ..... c z OS ... 0 0 Cl') 0 .... I:' .... 8 O> O> O> O> O> O> ? ? x I O x x I I x x r:: ..... & 0 < Cl') ..... t-.... O> O> O> I o ? I I Jacmel 0 ..... 8 r-"4 O> O> x I I x ... 00 ltj li-4 t:( l::rj z 0 c CfJ l-.:> 01 ,_.

PAGE 286

252 u1nbv ....... 0 (.) ..... z I tI96 80 96 0996 Z996 tLt6 8Lt6 LZ96 8 896 Lf.'96 t 8t6 t196 OZ9 6 8 196 69t6 9L96 689 6 1696 f 096 6996 6Lt6 1 9 96 Z886 f %'86 9886 LS86 8886 t t86 Of86 tt8 6 99L6 t9l6 GEOLOGY x OF x x ------------ x THE x REPUBLIC x x x x OF x HAITI. x x x x x ---------------- ------------------------------------------------- ---- x x x --------------------------- ---------- --------------------------------------------------------- c .... 0 x c-.. x c. Ga ti> = .... "O -..... 0 fj CJ ..... x r:::: ..... Q,) s 0 - x x Z c. cl9 Q,) c .... al ..... CJ ..... > .,... ....... Q,) Cl> zz x x x Q.I a 0 .... "O CIS .... 0 0 x x -c ... --

PAGE 287

Marine Quaternary tossil,s Continued cu c:> aS rn () Q,) ... ;:;.. :a I cu MO le St. Nicolas. I "C I I cd t! 0 I Species. 0 0 r.n ... 0 I re ...-1 00 t-cc M C'l ...-1 0) 0) 'di ...-1 Ci) Ci) M f2 M f2 CQ lO t-0 0) s 00 00 00 I lO ""'4 I lO 0) Cl) Cl) Cl) Cl) Q) o:i Q) Cl) Cl) 0) I 0) Cl:) Scaphopoda : Den tali um sp. cf. D. semistriolatum Guilding x Pelecypoda : Arca um bona ta Lamarck .............. x x Barbatia candida (Gmelin) .... x ? Scapharca auriculata (Lamarck) .. x Bcapbarca deshayesi (Lamarck) ... Glycymeris pennacea (Lamarck) ... x x Glycymeris sp. cf. 6. pectin a ta (Gme lin) x Glycymerie sp. x Atrina sp. x Ostrea folium Linnaeus ............... x Ostrea s p. x x Chlamys (Ohlamys) imbricatus (Ornelin) x Ohlamys (Aequipecten) sp .. x Cblamys (Plagiocte nium) sp, cf. o. (P.) gibbus (Linnaeus) .... Spondylus amer1cana Lamarck ..... ? x Lima sea bra B o rn ..................... x Mytilus sp. x x Ohama macerophylla Gmelin .......... x Cocakia (Codakia) orbicularis (Lin-naeus) ? x ? x Oodakia (Jagonia) sp ................. x Lucina sp. Phacoid es (Here) penns ylvanicus (Lin-naeus) x cu t.o Q,) < 8: 'tj 0 z aS ... 0 I Cl) i -Pelecypoda Continued : Pbacoides (Oallucina) radians (Oon rad) x Phacoidee (Parvilucina) sp. cf. P. (P.) lintea Conrad ..... x Divaricella quadrisulcata d'Orbigny . Diplodonta sp. cf. D. candeana d'Orbigny Cardi um (Fragum) mediu m Linnaeus x Oardi11m (Papyridea) sp. cf. c. (P.) Meuschen x sp1nosum Protocardia sp. Dosinia elegans Conrad ............... x Pi tar ep. x Oytherea (Ventricola) rigida (Dill wyn) x Obi one cancella Linnaeus x Chi one (Lirophora) paphia Linnaeus . x Coralliophaga ep. x Tellina (Tellina) interrupta Wood .. x Tellina ( Eurytellina) Dall. x georgiana Tellina (Angulus) sp. cf. T. (A.) promera Dall Tellina ( Cyclotellina) fausta Donovan x Tellina ep. x Strigilla rombesgii lll::Srch ............ Yacoma ? IP .. ......... Oorbula sp. Decapod Crustacea: Mithrax sp. x Ma1ine fossils Continued. G> (J c GD C) .... G> ,.. ,.. = ;:.. OS OS MO le St.-Nicolas. '(; ::s c OS 0 I 0 (/). M ,.. 0 p.,. i i > ... .... ....... O> O> g ...... O> O> Ct) gg !:--8 O> t-rd ct U') lt) O> O> 0) O> 0) O> O> O> O> O> O> O> x x x x x ? Q,) c.o a> Q,) < g: OS .. "O .8 Q -8 z OS ,.. 0 "Of4 "Of4 .t'--i t-8 8 8 O> ? ? x x x x x x x x c tO -Q,) ::s r::1' 0 < Ct) .. e-.1 t-to-U') .. "lf4 8 O> O> x x JaCJnel. 0 Ct) "Of4 U') 0 r-1 8 8 O> x i l-.:> Ot fil 0 t-4 0 0 0 ,.., l2j Cl td a 0

PAGE 289

SEDil!ENTARY ROCKS. 255 N ONMARINE DEPOSITS. The nonmarine Quaternary deposits consist of stratified but generally poorly sorted silt, clay, sand, or gravel deposited by streams. The material of which they are composed is usually identified as derived from adjacent highlands They are unconsolidated, and although the beds in many places have a slight initial dip, equivalent to that of the slope on which they were deposited, they are not deformed. No attempt was made to separate Pleistocene and Recent deposits. In general the older (prob ably Pleistocene) deposits are more dissected than the yo11nger deposits. It is equally difficult to di s tinguish the Quaternary deposits from similar deposits of Pliocene age. The degree of dissection and the general geologic history have b e en utilized in atte mpting this separation. The presence of nonmarine Quaternary gravels at the landward margin of the marine Qua ternary deposits and the reasons for their apparent or actual absence at many places have been mentioned on page 245. Exte nsive areas of sands and gravels of Quaternary age are found on some of the plains adjoining the sea, particular 1 y in the eastern part of the North Plain, east of the longitude of Fort Liberte. Here they floor a dissected plain, and a thin veneer of gravels covers parts of the rock platform that extend along the mountain front to the south. (For de s c ription of surface features see pp. 356-358.) The deposits consist of 11ncons olidated gravels, sands, and clays, derived chiefly from eroded and weathered products of the quartz diorite, which crops out in the mol1ntains and on the platform on which the deposits lie. The beds have a gentle seaward slope, parallel to the surface of the plain. The beds of sand and gravel are lenticular and cross-bedded. The clay is fine and dark, and at places it contains indeterminable fragments of plants. On the plain northeast of Acul Samedi, where the deposits are well exposed, the sands and gravels cap small mesas. Near the surface the sandy beds are at places cemented into relatively hard sandstones by iron and manga nese oxides. Residual concretionary masses of these oxides are scattered on the surfa ce. (See pp. 4 77-4 78.) The deposits probably were laid down by streams during the retreat of the s e a after the cutting of the rock platform. More careful study might show that some of the beds are marine. Coral reefs and coralliferous limestones, probably of Quaternary age, were found on the eastward prolongation of the plain near Copey, in the Dominican Republic.1 Other unc onsolidated gravels and sands of Quaternary age occupy con siderable areas in the interior, generally at the foot of high mol1ntain ranges or in interior valleys and plains. Quaternary deposits cover stream terraces along t he large streams. The most extensive terrace deposits are in the drainage basin of Riviere Arti-. 1 A geological reconnaissance of the Dominican Republic : Dominican Rep. Geol. Survey Mem., vol. 1, pp. 165, 174-175, 1921.

PAGE 290

256 GEOLOGY OF THE REPUBLIC OF HAITI. bonite. The terraces are described on pages 380-385. The separation of the Pliocene Hinche formation from the Quaternary terrace deposits in the Central Plain may be artificial. In general the Quaternary deposits here contain a greater proportion of large cobbles and boulders of limestone and brown or black chert derived from the surro11nding mountains. Similar terrace deposits are widespread in the Artibonite Valley. Along the trail from Mirebalais to Saut d'Eau the deposits on the west side of Riviere Tombe consist of gravel, coarse sand, and lenses of white marl. At places boulders of chert and basalt larger than a man's head are strewn on the surface. These deposits are as much as 30 meters higher than the stream and are well dissected. The land gastropods listed on page 25 7 (station 9457) were collected from coarse sa .nd 25 meters above the stream at a locality about 5 kilometers southwest of Mirebalais. Fresh water and land gastropods were collected from silt a kilometer southeast of Mirebalais, on the road to Las Caho bas. (See list, p. 257, station 9902.) Some of these stream deposits may be of the same age as the Hinche for mation of the Central Plain, but more detailed work would be required to separate Pliocene and Quaternary deposits. In the lower part of the Artibonite Va lley Quaternary terrace gravels extend along the north side of the river. At places the surface is strewn with boulders and cobbles. (See Pl. XXX, A, p 386 ) On the south side of the river a narrow band of gravels extends downstream from Les Verrettes. Northwest of the road from St.-Marc to GonaYves the gravels grade into conglomeratic coralliferous limestone. The relations of the Quaternary gravels near Port-au-Prince to the water supply of the city are discussed on pages 571-573. The nonmarine Quaternary deposits are similar to the marine in showing a peculiar hardening near the surface, which causes them to resemble much older and more consolidated rocks. The process is essentia lly the same as with the marine deposits. Near the surface the bedding usually is destroyed by this process, which is assisted by slump and creep. A typi cally hardened surface is shown in Plate XVII, B. This effect is not noticeable more than 2 or 3 meters below the surface of the ground. The thickness of the nonmarine Quaternary deposits is perhaps even more variable than that of the marine. Only at a few places, probably, does it exceed 50 meters. FOSSILS. Land and fresh-water mollusks are the only invertebrate fossils obtained from nonmarine deposits of Quaternary age. Most of the land mollusks, which were examined by the late Mr. John B. Henderson, of the United States National Museum, are similar to species now living in Haiti. Stations near M irebalais ( nonmarine Quaternary). 9457 (W 135 F). Arrondissement of Mirebalais, trail from Mirebalais to Saut d'Eau (Ville Bonheur), about 5 kilometers southwest of Mirebalais. W. P. Woodring, collector. December 7, 1920

PAGE 291

I ROCKS. 257 9902 (W 181 F). Arrondissement of Mirebalais, road from Mirebalais to Laa Caho bas, about a kilometer east-southeast of Mirebalais. W. P. Woodring, col lector. January 8, 1921. Non marine Quaternary fossils collected near M irebalais. Species. Gastropoda: Limnnea sp. ................................ Oepolis (Plagioptycbus) indisti ncta Ferussac .............................. Pupo ides marginatus nitidulus Pfeiff er .................................... Obeliscus ep. ......................................................... Obeliscus (Dolicholestes) sp., apparently diff erent from any described West Indian s pecies .................................... Pleurodonte (Parthena) sp. cf. P. (P.) angusta Ferussac ................ ...... Pleurodonte (Parthena) undulata Ferussac ............................... Pleurodonte (LuquilJa) sp. cf. P. (L.) audelbardi Pfeiffer ................. EXTINCT QUATERNARY MAMMALS AND BIRDS. 9457 9902 x x x x x x x x Extinct Quaternary mammals, prjncipally rodents, ground sloths and bats, have been found in caves and kitchen middens in Cuba, Porto Rico, and other West Indian islands. Miller 1 has described three rodents, be longing to the genera Isolobodon, Plagiodontia, and Brotomys, from kitchen middens at San Pedro de Macorfs and San Lorenzo in the Do miniran R e public. During the reconnaissance of the Republic of Haiti a preliminary examination was made of two caves in the arrondissement of Marmelade in order to determine whether a similar fauna could be discovered in the Republic. The results fully justified the preliminary exploratjon and indicate that more extensive exploration of these and other caves in the Republic is warranted. The caves examined are on the south slope of the mountains northeast of St.-Michel de l' Atalaye and northwest of the plantation of the United West Indian Corporation. The mountains here consist of limestone of middle Eocene age. In this region there are a large number of caves, formed during a period when the drainage systems were different from the present. Many of the caves stand high on the mountain slope and have no apparent relation to the present drainage, either surface or subterra nean. There is little evidence of active solution. The caves are dry and are apparently being filled with residual clay, rain-washed debris, and other material. Many of them contain a thick floor cover of guano left by the thousands of bats that now inhabit or recently inhabited them. 1 Miller, G. S., jr., Bones of mammals from Indian sites in Cuba and Santo Domingo: Smithsonian Misc. Coll., vol. 66, no. 12, 10 pp., 1 pl., 1916. 17

PAGE 292

258 GEOLOGY OF THE REPUBLIC OF HAITI. The larger of the two caves explored is about 3 or 4 kilometers northeast of St.-Michel de l' Atalaye and the same distance northwest of the plantation. It stands about 600 meters above sea l evel, nearly 200 mete1s above the Central Plain, which adjoins the mountains to the south. Its length is about 40 meters and its width and height 10 to 20 meters. Several large columns, formed by the union of stAlactites and stalagmites, extend from the floor to the roof. The cave has two large openings, separated by a pillar, and a third small opening on the sloping hillside, which affords entrance nearly on the plane of the floor. Near the rear of the cave there is an opening or skylight, about 5 meters in diameter, through which long roots of figuier trees extend. Fragments of rock and surface wash falling down the skylight have built up a small cone of coarse debris beneath it. An excavation nearly 2 meters deep and a little more than a meter in diameter was made in the middle of one of the largest open spaces. Only firm, dry, reddish dirt was enco11ntered. The rock floor appears to be very deep here and was not approached in this pit. Another hole was made near the rear end of the cave, about a meter from the wall and 5 meters from the cone of debris near the skylight. The excavation was less than a meter deep and about a meter in diameter. Rocks and boulders were embedded in the cave earth. From very near the surface downward this hole yielded bones. The smaller cave examined is about 2 kilometers north-northwest of the plantation and perhaps 2 kilometers east of the first cave, on the south side of a deep dry ravine. The diameter of the opening is about 30 meters. The roof is arched, all in one chamber, and the floor is convex, the rear half being nearly bare ro c k, partly covered by a thin deposit of bat guano The mouth, which was formerly much larger, is choked by a pile of debris from the cliff that rise s above it. This debris has rolled inward as well as outward, covering the floor of the front part of the cave An excavation was made at the low est part of the cave adjacent to one of the vertical rock walls, following down the wall to a depth of nearly 2 meters. At the bottom the rock wall sloped inward steeply, and the entire floor of the excavation was on rock. The material excavated consisted of loose stones, between which lay dirt and guano. Bones were found from a depth of 0.6 meter to the bottom, increasing in number downward. The bon es obtained are the remains of mammals and birds. The mammals have been examined by Mr. G. S. Miller, jr., of the United States National Muse11m, and the birds by Mr. Alexander Wetmore, of the Bio logical Survey, United States Department of Agriculture. Papers containing descriptions of the remains have recently been published.1 The remains identified by Miller and Wetmore are tabulated below. 1 Miller, G. S., jr., Remains ot mammals from caves in the Republic of Haiti: Smith sonian Misc. Coll., vol. 74, no. 3, 8 pp., 1922. Wetmore, Alexander, Remains of birds from caves In the Republic of Haiti: Smithsonian Misc. Coll., vol. 74, no. 4, 4 pp., 2 figs., 1922.

PAGE 293

SEDIMENTARY ROCKS. 259 Mammals: Rodents: Isolobodon portoricensis Allen. Aphaetreus montanus Mille1 n. gen. and n. sp. Ithydontia levir Miller n. gen. and n. sp. Brotomys voratus Miller? Ground sloth : Megalocnus? Mammals :-Continued. Man: Homo sapiens (head of left femur). Unidentified fragments. Birds: Chaemepelia passerina (Linnaeus). Crotophaga ani Linnaeus. Tyto ostologa Wetmore n. sp. Tolmarchus gabbii (Lawrence). Rodents of the genus Isolobodon furnished the most abundant remains in both caves. The new genus Aphaetreus is represented by a mandible with a full set of cheek teeth, and the new genus Ithydontia by isolated teeth, all of which were collected in the larger cave. A small ground sloth, probably resembling the extinct M egalocnus of Cuba, is represented by a nearly perfect cauda l vertebra, an imperfect probably dorsal vertebra, a fragment a pparently of the proximal end of the radius of a young animal, all obtain e d in the larger cave; and by two imperfect caudal vertebrae and the proximal end of a fragment of the shaft of a rib, collected in the smaller cave. The human remains consist of the head of a left femur found at the same level with the remains of the ground sloth in the smaller cave. The bone substance is lighter and less infiltrated with mineral matter than the bones of the sloth. A chipped chert, identified by Mr. Walter Hough, of the United States National Museum, as a human artifact, was found at an undetermined level in the same cave The human remains are probably yo11nger than the others. The unidenti fied fragments include parts of a foot of perhaps a large rodent and a piece of a large bone, probably of a ground sloth Three of the birds a dove ( Ohaemepelia), an ani ( Orotophaga), and a petchary ( T olmarchus) are now found in the Republic. The fourth, Tyto ostologa, is a very remarkable extinct gigantic barn owl, represented by the head of a metatarsus and other fragments. Evidently this bird 'vas the marauder that brought into the caves the rodents whose bones are found there in such large numb e rs. Some o.f these rodents are so large that no living owl of the Republic could handle them. Aside from the interesting discovery of the unknown gigantic owl, the remains obtained from the excavations increase our knowledge of the extinct early Quaternary mammalian fa11na of the West Indies. Although no carnivores or ungulates have yet been found, the known fauna is be coming more and more diversified with each exploration. The extinct Haitian mammals support Miller's suggestion 1 that the West Indian early Quaternary mammals descended directly from a South American fauna probably not older than Miocene. This suggestion hannonizes with the view that during late Miocene time the West Indian land s were more extensive than now and probably were directly connected with Central America, and that they were separated into islands during Pliocene time. 1 Miller, G. S., jr., op. cit., p. 4, 1916. -

PAGE 294

260 GEOLOGY OF THE REPUBLIO OF HAITI IGNEOUS ROCKS. By WILBUR s. BURBANK. GENERAL DISTRIBUTION. Igneous rocks crop out in probably a little more than one-fifth of the total area of the Republic. The largest areas are in the Massif du Nord and in the ranges of the Southern Peninsula. Relatively few exposures are found in the central ranges and intervening plains. Most of the igne ous rocks are older than the T ertiary sedimentary rocks and are now ex posed principally on the erod e d crests of anticlinal arches and in deep valleys where the sedimentary rocks have been removed. The areal distribution of the igneous rocks is shown on the geologic map (Pl. I.) The igneous ro cks are discussed unde r the heading of extrusive rocks, or those that have poure d out on the surface of the earth, and of intrusive rocks, or those that have been injected into or have penetrate d between other rocks and solidified with out reaching the surface. As it is not always possible in the field, especially among the older rocks, to distinguish between extrusive and intrusive rocks, certain rocks of doubtful origin are either classified on t extural and other features, or are discussed with other rocks of known origin that accompany them. The geology of the igneous rocks is relatively complex. Notable differences are found both in the rocks and in the igneous history of the northern and southern parts of the country. The early igneous history of the central part of the country is obscured by the overlying Tertiary forma tions, but its later Tertiary history has some distinctive features. Both because of convenience in discussion and because of petrologic considera tions, the igneous rocks of the Republic will be discussed under three geographical divisions: 1. The northern region (see Pl. XXVII), comprisjng Tortue Island, the North Plain, the Massif du Nord, the Northwest Peninsula, the northwestern part of the Montagnes Noires, the northern part of the Central Plain, and the northern extremity of the Artibonite Plain. 2. The central region, comprising the southern part of the Central Plain, the southeastern part of the Montagnes Noires, the southern part of the Artibonite Plain, the Montagnes du Trou d'Eau, and the Cha!ne des Ma te11x. 3. The southern region, comprising the Massif de la Selle and the Massit de la Hotte. The general boundaries of these regions are shown on the geologic map, although naturally they can not be sharply drawn. A summary of the igneous activ ity and of the tectonic history of these three regions is given in the following table:

PAGE 295

IGNEOUS ROCKS. 261 Summary of the igneous activity and tectonic history. Time Northern Central subdivisions. region. region. Pli'>Cene . Emergence Emergence .. Southern region. Minor submergence and continuation of fold ing. Upper Miocene .. Emergence and folding. Intrusion of quartz di orite in Monts. de Terre-N euve. Emergence and intense Emergence and folding. folding. Lower and middle Miocene. Marginal submergence. Mar ls and \!orallif erous limestones. Nearly complete subMarginal submergence. mergence. Minor erupUpper Oligocene Middle Oligocene Lower Oligocene . Upper Eocene tions of basalt, inter-bedded in marl. Partial submergence folExtensive submergence. lowed by emergence. Lime s tones. Eruption of nephelitl! basalt. Partial emer gence. Partial submergence . Extens ive submergence .. I Emergence and folding. I Probable emergence ... Extensive submergence. Complete submergence. Limestones. Limestones. Emergence. Local submergence. j Emergence and folding. Complete submergence. Limestones. Middle Eocene I Partial submergence I Emergence . . . . I Emergence. Lower Eocene ... Emergence and extensi v e erosion. ( ? ) Minor eruptions of basalt, andesite, diabase, es sexite Exact age un known. Upper Cretac eous. Batholitbic intrusions of quart z diorite. Intense folding. Local submergence Lime stones. Lo\ver or middle Partial submergence. Cretaceous. Sandstones, shales, ar gillites, non marine and marine. Jurassic ........ Emergence. volcanic Eruption of and dac ites. Extensive activity. andesites Early Jurassic Triassic ( ?) or Eruptions of basaltic lavas and minor in trusions of basic and ultrabasic rocks. Emergence and extensive erosion. Basalts and tufts, prob ably of this age. Probably partial sub mergence. Emergence and extensive \ erosion. F oldin g. Extensive fis sure eruptions of ba salt, partly submarine. Partial submergence. Limestones. Partial submergence Argillaceous limestones and shales Emergence ( ?). No evi dence of igneous ac tivity.

PAGE 296

262 GEOLOGY OF THE REPUBLIO OF HAITI. OUTLINE OF THE IGNEOUS GEOLOGY. Northern region. The northern system of mountains in the Republic of Haiti, here called the Massif du Nord, is characterized by a core of large bodies of massive quartz diorite, probably of middle or late Cre taceous age. These granitic rocks intrude and metamorphose lavas of Mesozoic age and some older metamorphic sediments, possibly of Paleo zoic age. The mountain ranges of the Northwest Peninsula also contain some bodies, of unknown size, of similar granitic intrusives. Apparently the oldest Mesozoic lavas are basalts. Associated with them are some pyroxenites, peridotites, and diabasic intrusives, which are probably of nearly the same age. Near the quartz diorite batholith these older rocks have been subjected to contact metamorphism and a re con verted locally into amphibolites and serpentinous amphibolites or into amphibole, talc, and chloritic schists. The alteration of the pyroxenites and peridotites is somewhat in contrast to that of similar rocks in Cuba, where the f orm ation of serpentine was the most common alteration. lf ost of the earlier basaltic rocks have suffered more or less metamorphism, a .nd many are converted to '' greenstones.'' Olivine-free and hypersthene basalts and porphyritic basalts rich in pyroxene and containing some olivine seem to be the more common types. The age of these basalts is possibly Jurassic or older. The yo11nger and more extensive Mesozoic lavas are chiefly andesites but include some dacites. They are found throughout nearly the lengtl1 of the Massif du Nord and are remarkably l1niform in composition and texture. The apparent lack of more diverse types of lava in association with the andesites and dacites is noteworthy. Where they have been intruded by the quartz diorite these lavas have been locally metamorphosed but not to the extent of the older basaltic rocks. The more common varieties seem to be pyroxene andesites, hornblende-augite andesit e s, and hornblende andesites; hypersthene andesites and hornblende-biotite ande sites are less common. The quartz-bearing varieties of lava include pyrox ene dacites, hornblende-augite dacites, and hornblende-biotite dacites. Some of the lavas that contain only a little quartz should possibly be classified as quartz-bearing pyroxene andesites. The eruption of the andesites and dacites in the central part of the Massif du Nord certainly took place prior to the deposition of certain sedimentary rocks that are considered Lower or Middle Cretaceous, but in the Montagnes Noires the lavas appear locally to have buried similar sediments. However, no lava flows have been found definitely interbedded in the Cretaceous ( ?) sedimentary rocks. Probably the eruptions occurred largely in the Juras sic period, but in some regions they extended at least into the lower Cre taceous, and minor undiscovered eruptions may be even later. Basaltic lavas and intrusive rocks, some of them alkaline, have a rather local distribution in the western part and along the southwestern

PAGE 297

lGN'EOUS :aoc:KS. 263 border of the Massif du Nord. At several localities these rocks seem to be younger than the andesites, but their relation to the quartz diorite was not detennined at any place. Calcareous sediments containing Foraminif era of 11nknown age are in places associated with these lavas and are locally engulfed or included in pillow lavas of this series. The varieties of rock include amygdaloidal basalts, olivine basalts and diabases, analcite andesites, analcite olivine andesites, and essexite. Some of the rocks are much zeolitized. The more or less alkaline varieties are rather common and are widely distributed. The age of the eruptions may have been late Cretaceous or early Eocene, but the evidence warrants only very general conclusions. The major igneous intrusion occurred in rnjddle or late Cretaceot1s time and consisted of large bodies of quartz diorite of rather uniform composition. In the eastern part of the Massif du Nord, where it is ex tensively exposed, the quartz diorite contains hornblende and chlorite. In the Northwest Peninsula altered porphyritic facies of the intrusive rock locally contain mica. Some lamprophyric dikes and veins of quartz hornblendite cut the main body of the batholith. Intrusions of minor stocks of quartz diorite, granodiorite, and asso ciated porphyries occurred probably during the Miocene along the central arch 0 the Montagnes de Terre-N euve. These rocks invaded the Mesozoic andesitic lavas and the overlying upper Eocene limestones. The intrusives where exposed are principally the fine-grained porphyries. There are also some associated dikes of porphyry and veins of pegmatite. In a general way the igneous rocks of the northern part of the Republic are similar in many features to those of the Western Cordillera of South America and to those of the Pacific Coast ranges of North America. The pre valence of ande sites and dacites among the volcanic rocks is a char acteristic f e ature of the Western Cordillera of South America. An analy sis of a pyroxene andesite from the Terre-N euve region (p. 276) shows relatively higher silica and lower potassium oxide than the average andesite. The high percentage of quartz and the practical absence of orthoclase in the quartz diorite from the central part of the Massif du Nord is also particularly noteworthy. The Cretaceous quartz diorite of the Northwest Peninsula and the Mioc e ne ( ?) intrusives of the Montagnes de Terre .. N euve are higher in potash feldspar than the principal body of quartz diorite. All the rocks are rather high in titanium. Quartz and hornblende bearing granitic intrusives are widespread in the West Indies and are found in the islands of Cuba, Porto Rico, Vieques, St. Thomas, and St. Martin.1 They a re both pre -Cretaceous and post-Cretaceous in age.1 Central region. In the central ranges and plains of the Republic the pre-Tertiary igneous rocks are rarely exposed. Small patches of glassy 1 Vaughan, T. w., unpublished notes.

PAGE 298

264 GEOLOGY OF THE REPUBLIC OF HAITI. hypersthene basalts are exposed along the central arch in the southeastern part of the Montagnes Noires. These lavas are older than upper Eocene. Small areas of early or pre-Tertiary igneous rocks are exposed in tl1e central part of the Cha.lne des Mate11x. Nephelite basalts are found at several localities in the Montagnes du Trou d'Eau. Flows and explosive eruptions of these lavas took place probably from isolated central vents. In composition the rocks are largely or entirely nephelite basalt . Some varieties contain haiiynite or melilite and others are much zeolitized. Rocks of similar composition (p. 316) have been found in Grenada 1 and in Uvalde County, Texas! The nephelite basalts northeast of Thomazeau are of middle or upper Oligo cene age; basalts near Saut d'Eau are post-middle Oligocene and may be younger than the others, but their relations to the Miocene beds of the Artibonite Valley are not known. In the Dominican Republic limburgite, probably of Pleistocene age has been found .8 Minor amounts of basaltic lavas and debris are interbedded in the Miocene sedimentary rocks on the western flank of the Chaine des Mateux and in the southwest ern part of the Montagnes du Trou d'Eau. Southern region. The pri ncipal igne ous rocks found in the mountain ranges comprising the Southern Peninsula of Haiti are basaltic lavas. These lavas are exposed in the central arch of the mountains or in deep cut in the Tertiary limestones. The lavas have buried older lime stones and argillaceous rocks, probably partly of Cretaceous age, and together with these comprise a bas ement on which the Tertiary l i mestones lie. Probably only a part of the original extent of the eruptions is included by the present outline of the Southern Peninsula. The eruptions were doubtless for the most part of the fissure type, as pyroclastic debris is found only locally. Ordinary basalts are the most common variety of lava, but olivine basalts, spilitic basalts, diabase porphyries, and augite ande sites are also found. Some of the lavas are amygdular and many have pillow structure. Large amounts of tuff and agglomerate are exposed in the northwestern part of the Massif de la Selle. An analysis of an ordinary basalt from the Mas sif de la Selle (p. 325) shows that the lava is unusually high in calcium oxide and rather high in titanium. One of the noteworthy features of the analysis is the very high ferrous iron, showing that the lavas were probably fluid.4 The eruptions doubtless occurred in Upper Cretaceous time, as the lavas have buried argillac e ous limestones that are considered Lower or Middle Cretaceous, and pillow lavas have locally buried or intruded unconsolidated calcareous deposits that are 1 Harrison, J. B., Rocks and soils of Grenada and Carriacou, p. 10, London, 1896. 1 Osann, A., Melilite-nephelite basalt and nepheline basanite from Southern Texas: Jour. Geol., vol. 1, pp. 341-346, 1913. Vaughan, T. W., and C1oss, W., U. S. Geol. Survey Geol. Atlas, Uvalde Folio (No. 64), pp. 3-5. 1900. 1 Condit, D. D., and Ross, C. P., Dominican Rep. Geol Survey Mem., vol. 1 p. 203, 1921. Washington, H. S., Deccan traps and other plateau basalts: Geol. Soc. America Bull., vol. 33, pp. 765-804:, 1922.

PAGE 299

IGNEOUS ROCKS. 265 supposed to be Upper Cretaceous (pp. 96-97). The lavas lie uncon formably beneath upper Eocene limestones. Hypersthene and hornblende andesites are found near Baraderes in the Southern Peninsula. They comprise only a small group of rocks, and their relations to the basalts are unknown. They are probably, like the basalts, of late Cretaceous age. NORTHERN REGION. EXTRUSIVE RocKs. GENERAL FEATtJ RES AND DISTRIBUTION. The extrusive rocks of the northern region are principally lavas. Al though they include some interbedded breccias and tuffs, no thick de posits of such rocks were seen except in the western part of the Northwest Peninsula, where some tu:ffaceous and agglomeratic rocks of unknown thickness underlie limestone of supposed upper Eocene age. A large part of the detrital volcanic rocks have probably been reworked and deposited in water. The lavas form the larger part of the igneous rocks in the northern region and in general range in composition from basalts to dacites, but may include more basic or more acidic rocks. The thickness of the extrusive series is unknown but is probably varia ble in different parts of the region, and at some places is undoubtedly very great. No reliable estimate of the thickness is possible at present, as the basement on which the lavas rest has not been definitely recognized in outcrops. In the western part of the Massif du Nord the thickness probably is more than 1,000 meters. The original thickness of the series may have considerably exceeded 1,000 meters in some places, as in addi tion to their erosion in the present cycle these rocks underwent erosion during long periods in both Cretaceous and early Eocene time, parts of the later volcanic accumulations probably having been removed during each period. In some regions a part of the series has also been engulfed in a Cretaceous batholitbic intrusion of quartz diorite. In the eastern part of the Massif du Nord the volcanics flank the intrusive quartz diorite along its entire southern boundary, and a band of metamorphic volcanic rocks extends along the north side from GrandeRi viere du Nord eastward to Les Perches. East of Les Perches they were largely engulfed in the quartz diorite, although small patches are still preserved and in places crop out through the alluvial deposits of the North Plain. (See Fig. 18, A and B, p. 311.) In the western part of the Massif du Nord, west of the Grande Riviere du Nord, the volcanic rocks comprise 80 to 90 per cent or more of the exposed igneous rocks. A small area of volcanic rocks is found at the border of the Central Plain north of Hinche. Northwest of the plain, near St.-Michel de l' Atala ye and Ennery, small patches of the underlying lavas are exposed in the areas of Eocene limestones.

PAGE 300

266 GEOLOGY OF THE B.EPUBLIC OF HAITI. In the northern part of the Montagnes Noires and outlying parts of the Artibonite Plain, lavas apparently constitute the bulk of the igneo11s rocks and are exposed in the deeper valleys and along the tilted northwest front of the mountains. The igneous rocks in the Montagnes de Terre-N euve are p1edominantly lavas In the extreme western part o:f the Northwest Peninsula small areas of the lavas and pyroclastic rocks underlie the upper Eocene limestone. AGE OF ERUPTIONS. The longest periods of volcanic activity were clearly pre-Tertiary, and as lavas of these periods near Grande-Riviere du Nord and Dondo11 are i1nconformably overlain by argillites of supposed Lower Cretaceous age and by Upper Cretaceous limestone the earlier periods can be ref erred back to pre-Cretaceous time. Little or no evidence was found to show how long before Cretaceous time the volcanic activity may have existed. Much of the activity was probably o:f Jurassic age, but at least the later part of it may have been of Lower Cretaceous age, as hornblende andesites in the Montagnes Noires southwest of St.-Michel appear to cover sedimentary rocks tentatively correlated with the Cretaceous series. No positive evidence of igneous activity in middle or upper Eocene time was found. Except for usually thin basal beds, composed largely of detrital igneous material, the Eocene sedimentary rocks are principally lime stones. The basal conglomerate of the Plaisance limestone (middle Eocene) exposed at the northeast base o:f Mont Puilboreau and between Ennery and St.-Michel de l' Atalaye contains rounded pebbles and boulders of the 11nderlying volcanic rocks (see pp. 272, 102, and Fig. 18, 0, p. 311). This material was derived by erosion from the 11nderlying volcanic basement. In the western part of the Montagnes de Terre-Neuve the basal part of the Eocene limestone, for a thickness of several hundred meters, contains interbedded sandy layers of igneous material. '11his material appears to be waterworn sand and pebbles and was derived by erosion from an igneous landmass above sea level probably in what is now the central part of the Montagnes de Terre-N euve. (See pp. 113-115.) In the extreme western part of the Northwest Peninsula, at Morne Chien, between Baie de Henne and Bombardopolis, a small amount of sandy material derived from igneous rocks is interbedded in the basal parts of limestone of sup posed upper Eocene age. The sandy basal beds immediately overlie tuffaceous and agglomeratic volcanic debris. The igneous material in the sandy limestone apparently was not a product of contemporaneous vol canic activity but was derived by erosion from the underlying tuff. About 10 to 12 kilometers farther west, near Plateforme, according to Liitgens,1 ash, tuffs, and basalt flows are interbedded with metamorphosed marly 1 Ltttgens, R., Geographische und geologische Beobachtungen in Nordwest-Haiti: Geog. Gesell. in Hamburg Mitt., Band 32, pp. 72-75, 1919.

PAGE 301

IGNEOUS ROCKS. 267 limestone. Although Liitgens has apparently misjudged the age of the limestone and basalt, which, from lack of paleontologic data, he calls Pliocene, his sections and descriptions appear to indicate that basal parts of the upper Eocene limestone and the llnderlying volcanic debris are exposed near Plateforme. (See also pp. 112 and 370.) However, great thicknesses of interbedded tuffs are found farther west in the upper Eocene limestone of the Sierra Maestra of Cuba. The upper limits of age of the principal periods of volcanic activity can thus be placed at middle Eocene and in most places at upper Cre taceous. There is little or no evidence to show how far back in the Mesozoic the earlier volcanic eruptions took place. The intense local meta morphism produced during the batholithic intrusion of quartz diorite near the close of Cretaceous time renders the criterion of metamorphism of rather doubtful value. Certain schists that possibly consist of remnants of the basement upon which the lavas were erupted show a degree of metamorphism much greater than that found in any of the volcanic rocks or other rocks in which the metamorphism can be definitely attributed to the period at the end of Cretaceous time. (See pp. 84, 85.) If it is a ssumed that these schists are of early Mesozoic or of late Paleozoic age, the principal periods of volcanic activity must be referred to early or rnj ddle Mesozoic time, to the Triassic, Jurassic, or Lower Cretaceous. In certain areas later basaltic eruptions may be as young as late Cretaceous or early Eocene. (Seep. 280.) ORDER OF ERUPTION. The Mesozoic igneous activity was so complex that only the general features of its history can be determjned. At critical localities the older rocks may be covered by the Tertiary and Cretaceous sedimentary rocks, and erosion during Cretaceous and Eocene time has removed many of the later volcanic accumulations. Detailed work would be required to establish accurately the relations of the rocks in these series, but a few of the more important features of the igneous history are recognizable. The major eruptions apparently were marked by enormous outflows of lava from a large number of central vents or :fissures, which at times probably covered much of the northern and central part of the island of Haiti, including what is now the northern part of the Republic. The landmass at the time of these eruptions was presumably very much greater than at present. The unifor1nity of the lavas in the northern part of the Republic indicates large reservoirs of a common magma from which they were derived. The earliest known eruptions were basaltic. At many places the older basaltic rocks are associated with metamorphic rocks and schists, some of which are pres11mably the result of local metamorphism of the igneous rocks or of associated tuffs and sediments. The schists at some localities

PAGE 302

268 GEOLOGY OF THE REPUBLIC OF HAITI. may, however, be much older. Peridotites and pyroxenites ( ?) are found at some localities with the basalts and probably are intrusive into them. The rocks of these earlier periods of activity are generally characterized by much metamorphism and alteration, as they were buried to consider able depths l1nder accumulating volcanic and sedimentary rocks and were in tensely affected by the folding and ba tholi thic intrusions at the close of Cretaceous time. These ear lier periods of basaltic eruptions may have been of Jurassic age or even older. Later eruptions consisted largely of pyroxene andesites and dacites. Events separating these later eruptions from the earlier periods of activity are obscure. Neither accumulations of sedimentary material nor evidences of erosion were definitely recognized. In the Montagnes de Terre-N euve the pyroxene andesites apparently overlie a basement containing shales and metamorphic limestone, but the older basalts are not present in the section Whether the eruptive periods were continuous or were separated by an interval of quiescence, the lavas of the later eruptions are uniformly andesitic or dacitic, presumably indicating important changes in the parent magmas from which the eruptive rocks were derived. Probably many of the later lavas of tl1is series were dacites. The age of the andesitic and dacitic lavas is considered Jurassic or possibly in part Lower Cretaceous. Hornblende and pyroxene andesites and some olivine-bearing andesites that are found in the northwestern part of the Montagnes Noires are of unknown age, but from their petrographic character they are tentatively assigned to the period during which the andesites were erupted in the Massif du Nord and the Montagnes de Terre-Neuve. At the southern border of the northern region, in a belt extending from St. Michel northwestward through Ennery, including the Montagnes de Terre-Neuve and the region around Gros-Morne, there is a series of younger basaltic lavas and intrusive rocks. These rocks are younger tl1an the pyroxene and hornblende andesites of the Terre-N euve mountains and are clearly younger than the hornblende andesite bet\veen Gona1ves and Ennery. They may be of Cretaceous or early Eocene age. Among the later eruptives that reached the surface in the northern region is a series of dacite porphyries, which engulfed and possibly over flowed Cretaceous argillites southwest of St.-Michel in Section Paul. Because of their petrographic character and possible intrusive relations these rocks are tentatively correlated with the ql1artz-diorite intrusion of the Massif du Nord and hence may be of very late Cretaceous age. EARLIER BAS.ALTIO ROOKS. DISTRIBUTION AND STRUCTURAL RELATIONS. Although the earlier basaltic rocks are widely distributed and their exposures cover considerable areas in some localities their structural relations are somewhat obscure, and the basement on which they rest was

PAGE 303

I IGNEOUS ROCKS. 269 not definitely recognized. As float of old schists is found on the surface of the volcanic rocks in the North Plain, it is believed that they originally rested on a basement of early Mesozoic or Paleozoic schists. (See p. 84.) The old basalts are exposed on the North Plain at a n11mber of places, and are found as float in the central and western parts. The northward extending spurs and foothills of the Massif du Nord between Grande.; Riviere du Nord and Les Perches are composed of volcanic rocks, at least a part of which are the older basalts. These hills, except for the no1ihern most extension of bedrock east of Le Trou, were not examined during the reconnaissance. Along the road from Cap-Ha1tien to Ouanaminthe, about 6 kilometers east of Le Trou, these rocks crop out and are exposed inter mittently from this locality eastward to the Fort Liberte road, a distance of about 10 kilometers. They are exposed in low ridges or :fiat domes that barely rise above the alluvial covering of the plain. At many places actual outcrops were not seen, although abundant float of angular joint blocks of the bedrock indicates that the covering of alluvium is thin. These exposures east of La Trou are cut by dikes and small stocks of quartz diorite porphyry. Elsewhere in the eastern part of the North Plain these oider volcanic rocks seem to have been largely engulfed in the quartz diorite and are now preserved only in small much metamorphosed patches. Small areas of the volcanic rocks were seen in the quartz diorite south of Les Perches, and small included fragments of botl1 volcanic rocks and schists were noted at several places in the quartz diorite. (See Fig. 18, B, p. 311.) Most of the amphibolitic and talcose schists at Morne Beckly were prob ably formed by intense dynamic and contact metamorphism of rocks of this series At Morne du Cap the altered basaltic rocks comprise a large part of the basement on which lies the upper Eocene limestone that caps the mountain. The basaltic rocks and some patches of Cretaceous ( ?) sedimentary ro c ks form the lower slopes and foothills of the Morne du Cap. The Cretaceous ( ?) cherts and sediments rest on the basaltic rocks, and certain conglomeratic beds in them contain pebbles derived from the basalts. Small bodies of quartz diorite porphyry are associated with the volcanic rocks here and presumably are intrusive into them. Altered volcanic rocks, probably large ly basalts, form a belt of low hills that extends southward from the igneous foothills of the Morne du Cap to the outlying hills of the Massif du Nord west of Milot. These volcanic rocks are intruded at a few places by small dikes or stocks of porphyry. West of Grande-Riviere du Nord, and in Section Cormiers, where copper veins have been prospected, the country rock consists of basalts and some associated chloritic schists. The igneous rocks of Limbe Mountain consist in part of much altered basaltic rocks of this series. The metamorphism of the rocks in this mountain is so intense that their exact classification is difficult. A large

PAGE 304

270 GEOLOGY OF THE REPUBLIC OF HAITI part of Plaisance Mountain also is composed of altered volcanic rocks. The metamorphosed basaltic rocks crop out on the crest and north slope, but the south slope contains some yo11nger andesitic and dacitic lavas. Intrusive stocks of quartz diorite, which proballly caus e d much of the metamorphism, cut the older rocks and are exposed south of Limbe. The basalts are exposed also at the base of the middle Eocene limestone in the small valley northeast of St.-Michel in Section Las Lomas. Here they are associated with chloritic schists exactly similar to those west of Grande-Riviere du Nord. Associated with the basalts at many places there are ultra. basi c rocks, such as p eridotite s and pyroxenites, and muc h-metamorpho se d ro c ks, su c h as amphibolit e s and epidiorites. Some of these rocks may be intrusive into the basalts. PETRO O RAPHY. The earlier basaltic rocks are genera .lly characterized by intense altera tion and metamorphism. Most of the1n are greenish gray to dark green or nearly black, although som e are reddish brown, purple, or gray. The green color in most of the ro c ks is due to secondary minerals, such as hornblende, epidote, and chlorite, and as a field designation most of the rocks could appropriately be call e d greenstones. Some of the m are tough and :fine-grain e d, and most of them are porphyritic or amygdular. The phenocrysts are generally altered plagioclase and augite and, more rare ly, altered hypersthene or olivine. The amygdul e s are compos e d of chlorite, epidote, quartz, or zeolites. The ro c ks at some places are closely and irregularly jointed, and in road cuts and other outcrops they crumble away into small, irregular blocks Som e of the altered basalts are r e ddish brown or purplish gray. The weather e d surface of the normal greenston e s may be somewhat bleached to gray or brown. The following type s of roc k s are the more common among the earlier basalts. Olivine-free basalts. Basalts of the olivine-free type, found west of Grande Riviere du Nord in Section Cormi e rs, on the North Plain, and in the vicinity of Limbe, are dark green to greenish gray and general!) .. are porphyritic. Plagioclase when p1'E.1sent as phenocrysts is generally found in clusters of pris ms, 2 to 3 millimeters jn diameter. So far as could be determined these basalts contain no alteration products derived from olivine. The groundmass is dens e and in thin section the texture is intergranular that is, the rock consi sts of thin plagio c lase laths, the largest about 0.1 millimeter in length, b e t\ve e n which lie crowd e d small giains of augite. The plagioclase, both of the. phenocrysts and groundmas s, is dull megascopically and in thin section is invariably seen to be more or less completely altered to albite and clouded with alteration products. In some rocks the alteration products accompanying the albite can be recognized as epidote and zoisite.

PAGE 305

IGNEOUS ROCKS. 271 The relatively unaltered augite is light greenish or yellowish and is more generally found 11naltered than the plagioclase. In some rocks it is partly or completely replaced by chlorite or uralitic hornblende. Grains of magnetite between the plagioclase and augite are generally recrystal lized in the altered rocks, some of them to tabular hematite. A peculiar nonporphyritic basaltic rock of amy.gdular texture was found as a cobble in the Riviere Limbe, just south of the town of Limbe. The rock is purplish brown and contains amydules of white and greenish silicates, some as much as 4 centimeters in diameter. The texture is intergranular, the plagioclase laths, 0.05 to 0.10 millimeter in length, being surrounded with granules of augite about 0.05 millimeter in diameter. The somewhat dusty plagioclase has been recrystallized and consists largely of albite. The colorless or pale greenish a ugite is practically unaltered except for a small amount of secondary chlorite. The magnetite, a primary accessory, has been recrystallized in part to a platy ore, probably hematite, which in places cuts across the amydular fillings. The peculiar amygdules are bordered with albite, the centers consisting of blue-green and yellow-green minerals, possibly epidote. The :filling of the vesicles or replacement of the original amygdules with the albite and epidote ( ?) is clearly a result of the regional metamorphism of these basalts. Hypersthene basalt. At only one loca .lity, in Section Cormiers west of Grande-Riviere du Nord, was hypersthene basalt found, but as it was fo11nd there associated with the ordina1y basalts, basalts of this type may occur elsewhere. The rock is dark greenish gray, but weathers rusty brown. The plagioclase appears as clusters of phenocrysts and as thin laths in the groundmass largely altered to albite and clouded with p-rehnite, chlorite, and sericite. The hypersthene occurs as phenocrysts, which are now completely alte1ed to a greenish pleochroic bastite. The augite in the groundmass is relatively fresh but bas been in part altered to chlorite. The magnetite has been partly recrystallized. Chlorite and prebnite are present in amygdules and replace the plagioclase of the rock. Other basaltic rocks. Much-altered basaltic rocks containing numerous large phenocrysts of pyroxene were found on the N ortb Plain. The constituents of these rocks are largely altered to amphiboles and to talc, so that there is some uncertainty as to the original character of the rocks. The porphyritic texture is not promjnent on freshly broken surfaces, but on weathered surfaces the phenocrysts out in relief, giving the rock a warty appearance. The pyroxene, which in some of the rocks is near diopside, is largely altered to a pale fibrous amphibole and chlorite, or to serpentine or talc. The secondary amphibole in many of these rocks is very pale and scarcely pleochroic. In some rocks talc is an important secondary constituent, having nearly completely replaced phenocrysts of diopside or forming part of the groundmass. A network of actinolitic

PAGE 306

272 GEOLOGY OF THE BEPUBLIO OF HAITI. hornblende needles has replaced the pyroxene in some rocks along cleavage lines. In one specimen a few remnants probably of olivine were asso ciated with flaky serpentine. Because of the abundance of large pheno crysts of pyroxene, many of the rocks are believed to be transitional toward the pyroxenites and peridotites. Extreme metamorphism and more com plete recrystallization of basaltic rocks of this type may have produced the amphibolites and schists found at some localities on the North Plain. (See pp. 306-309.) Au!ERATION AND METAMORPHISM. These earlier basaltic rocks, which are at the base of the series of lavas, have been extensively intruded and partly engulfed in the quartz diorite batholith. This intrusion in Cretaceous or early Eocene time was prob ably the cause of their intense metamorphism, which is regional in extent. The most common alteration is the albitization of the calcic plagioclase, which is generally accompanied by the separation of epidote or uralitic hornblende. Albite and epidote were formed in some rocks without appreciable alteration of the augite. In nlore advanced metamorphism the albitization is accompanied by complete uralitization of the augite, and finally the rock is converted to an amphibolite. Some of the albite is altered along cleavage lines to sericite. Talc or serpentine may be important products of alteration in rocks consisting predominantly of pyroxenes. At some localities, as near Grande-Riviere du Nord, the augite has been converted to chlorite and the conversion appears to be connected with the general metamorphism that accompanied the formation of the veins at that locality. Chlorite, calcite, and finally zeolites may more or less completely replace the plagioclase. Prehnite was noted as a replacement product of plagioclase and it forms amygdules in rocks from near GrandeRi viere du Nord. Epidote, chlorite, calcite, and zeolites are at many places found in amygdules, and zeolites encrust joint planes of an amphibolitized rock from Limbe Mountain. ANDESITES AND DACITES. D1sTRmuTioN AND STRUCTURAL RELATIONS. Andesitic and dacitic lavas are abundant in the northwestern part of the Massif du Nord and in the eastern and northeastern parts of the Montagnes de Terre-N euve. are also found on the south slope of the Massif du Nord as far east as Lamielle. In general andesitic lavas appear to be the predominating type, although quartz-bearing rocks are widespread. South of Plaisance Valley, at the foot of Mont Puilboreau, the dacites and andesites 11nderlie the middle Eocene (Plaisance) limestone. (See Fig. 18, 0, p. 311.) Near the foot of the mountain basal beds of the Plai sance limestone contain fragments of these lavas and also considerable

PAGE 307

IGNEOUS ROCKS. 273 quartz, which appears to have been derived from dacitic lavas outcropping near by. Just below the conglomeratic beds, at an altitude of about 530 a yellowish much altered dacitic lava, with included blocks of dark, fresher dacite is exposed. A kilometer or so farther down the slope, near the foot of the mountain, at an altitude of about 420 meters, there are exposures of altered pyroxene andesites. Faulting in the volcanic rocks obscures their original relative positions, although the position of the dacites close to the middle Eocene conglomerate indicates that the dacites were at or near the top of the volcanic series and probably overlie the andesites. The alteration of both the dacites and andesites gives evi dence of intrusive activity, although no intrusive rocks are exposed at the surf ace. Andesites and dacites were found north of Dondon and in the valley of Grande-Gille, which runs from a locality just east of the Citadelle north eastward to the valley of the Grande Riviere du Nord. Conglomeratic and tuffaceous beds in marine Cretaceous ( ?) argillites, about 3 kilometers north of Dondon (see p. 88), contain abundant angular fragments of altered dacitic lava and of quartz and plagioclase. The ab11ndance of quartz in the argillite indicates that quartz-bearing lavas were surface rocks during its deposition. Homogeneous tuffaceous-appearing material in a calcareous matrix containing small Foraminifera may indicate some contemporaneous eruptions during the deposition of the argillite, although rapid transportation or reworking of older tuffs might under certain conditions give similar results. No actual flows of volcanic rock were found in any part of the argillite series. The prevailing volcanic rocks along the trail southward from Milot to the Citadelle of Christophe and Dondon are pyroxene andesites. Small areas of hornblende and pyroxene andesites were found in the northwestern part of the Montagnes Noires. A small interior valley along the Riviere Coupe-a-l'Inde and back of the first limestone range east of Dessalines is floored in part with ho1nblende andesites. Southeast of Gona!ves pyroxene andesites, olivine-bearing andesite, and_ basaltic rocks are exposed along the mountain front and in the valleys east of Morne Grammont. Small patches of much altered hornblende andesites are associated with the Cretaceous ( ?) argillites exposed in the valley back of the first limestone range of the Montagnes Noires southwest of St.-Michel. Locally the andesites seem to have buried or intruded the Cretaceous ( ?) rocks. The alteration of the andesites appears to be due to intrusive bodies of dacite porphyry. The lavas in the Montagnes Noires are exposed in lowlands or valleys formed by removal of the upper Eocene limestone. Andesites and dacites constitute a considerable proportion of the lavas in the mountains between Gros-Morne and Le Borgne. At some places in this region the lavas are altered, apparently owing to the proximity of t1nderlying intrusive bodies of qua1 .. tz diorite. 18

PAGE 308

274 GEOLOGY OF THE REPUBLIO OF HAITI. The distribution of the andesites in the Montagnes de Terre-Neuve is discussed in the description of the Terre-N euve district. (See p. 430.) Hornblende andesite was seen at one locality northeast of Gona1ves, along the railroad track to Ennery, in a cut near Kjlometer 16. Here the hornblende andesite has been thrust up along a high-angle fault into contact with the overlying basaltic rocks. (See Pl. XVIII, A, p. 280.) Hornblende-mica andesite or dacite porphyries crop out southwest of Jean Rabel and are apparently intruded by the quartz diorite porphyry. They are bright reddish rocks showing in some places a conspicuous flow structure. The structural relations of these rocks are somewhat obscure, but owing to their texture and the apparent intrusive relations of the quartz diorite porphyry they are tentatively correlated with the effusive andesi tes and daci tes. Andesitic and dacitic appearing lavas were noted in the agglomeratic material underlying the upper Eocene ( ?) limestone on Morne Chien, northwest of Baie de Henne. PETROGRAPHY AND CHEMICAL COMPOSITION. Pyroxene ( oogite-hypersthene) andesites. The rocks called pyroxene andesites are fine grained to dense or glassy and are prevailing porphyritic in texture; an1ygdular types are rare. In general the fresher glassy or partly glassy lavas are black to green or greenish gray; the more crystalline and more altered lavas are reddish, purplish brown, or gray. Oxidation by weathering and by circulating solutions has changed the lavas locally to red ferric oxide tints. A mottling of green and reddish tints is found in some places. Weathering bleaches the dark-green lavas to brown or gray speckled with whitish feldspar phenocrysts. Most lavas of this group contain some glass; lavas that are predominantly glassy or predominantly holocrystalline a .re not so abundant as intermediate types. Phenocrysts of both plagioclase and pyroxene are generally present in the lavas. The plagioclase phenocrysts reach a length of 5 millimeters in some lavas, but their average length is between 1 and 2 millimeters. Greenish-black augite and in some lavas yellowish-brown grains of hypersthene can be recognized. In thin sections the p1agioclase is in euhedral to subhedral laths that range in length from 0.2 to 4 or 5 millimeters and comprise about 15 per cent of the vol11me of the rock, though their sizes and proportions differ in di:ff erent lavas. The crystals are zoned and show Carlsbad, alb ite, and in some specimens penetration twinning. Extinction angles on com bined twins and indices of refraction show that the plagioclase has a composition around Ab,0 An601 Strongly zonal phenocrysts range from bytownite (Abso An10) to sodic labradorite (Abcso AnGo). Both augite and hypersthene generally are present in varying proportions and together 1 Wright, F. E., A graphical plot for use tn the microscopic determination ot the pla gloclase feldspars: Am. Jour. Sci., 4th ser., vol. 36, pp. 540-542, 1913.

PAGE 309

IGNEOUS ROCKS. 275 constitute from 6 to io per cent of the rock. The hypersthene occurs in prisms, the largest of which are 1 to 2 millimeters in length, and the mineral has a distinct but faint pleochroism, which ranges from pale yellowish to gray-green. In the lavas that contain a considerable proportion of glass in the groundmass the hypersthene is practically llnaltered. As the groundmass becomes predominantly crystalline most of the hyper sthene is altered to bastite, which may be accompanied by calcite and by various forms of silica. A second generation of hypersthene, occurring as microlites, invariably altered to bastite, is found in places in the groundmass of the more crystalline lavas. The a ugite is in light-greenish or greenish-gray prisms, from 0.5 to 2 mjllimeters in length, that show no pleochroism. The angle Z /\ c is about 46. In the more glassy lavas it is unaltered, but in the more crystalline lavas it is here and there partly altered to calcite or quartz. It generally occurs in larger crystals than the hypersthene. The groundmass usually has a characteristic andesitic or hyalopilitic texture, consisting of a felted aggregate of plagioclase microlites in an interstitial light-brown glass base. In specimens in which the base is more crystalline small prisms and quadratic sections of plagioclase may predominate. The plagioclase microlites are generally 0.05 millimeter or less in size and vary in composition from oligoclase to sodi c labradorite in different speci mens. Small grains and specks of magnetite are attached to the plagio cla s e microlites. Apatite in thin needles is here and there an accessory mineral. Reddish-brown flakes of hematite are present in some specimens. The subordinate glass base usually has an index of refraction of 1.50 to 1.52. Holocrystalline pyroxene andesites are not abundant, although they may be more common than the specimens collected would indicate. In general they seem to have been confined to the hornblende-bearing types. A few specimens of holocrystalline pyroxene andesites were found, but they are somewhat altered, so that the original character of the plagioclase could not be determined. The hypersthene is altered to bastite or other alteration products, and the augite in one specimen partly to calcite, plagioclase, quartz, and iron oxides. 'rhe gro11ndmass of such rocks is a fine-grained irregular intergrowth of plagioclase grains and prisms. Pyroxene andesites, consisting predominantly of glass, were found on Morne Dumuraille in the Montagnes de Terre-Neuve, in the vicinity of Morne Macat, in the valley of the Riviere Lhormand west of Gros-Morne, and northeast of Gros-Morne. Most of the very glassy rocks have a sheeted or platy structure, caused probably by contraction due to sudden cooling after they were extruded. Plagioclase and pyroxene phenocrysts can be recognized with the unaided eye and range from a fraction of a milli meter to 1 or 2 millimeters in length. The augite is in greenish-black grains and the hyperstl1ene, where unaltered, is in somewhat smaller yellowish-brown grains.

PAGE 310

276 GEOLOGY OF THE REPUBLIC OF HAITI. The plagioclase in these glassy lavas generally ranges in composition from bytownite ( Ab25 An1i>) to sodic labradorite ( Abtso An50). In some specimens magnetite is rathe r abundant in o ctahedral grains, the largest half a millimete r in diameter. The inde x of r e fraction of the glass base generally is about 1.50 to 1.52. The glass base of some lavas has perlitic texture. (See Pl. XXII, B, p. 304.) In other features the lavas a re like the more crystalline ones and are presumably of similar composition. A chemical analysis of a glassy pyrox ene andesite from Morne Dumuraille, commun e of T erre-N e uve (Pl. XXII, B), is given below. This analysis, as w ell as those of other rocks in this report, was made by Dr. H. S. Washington of the Geophysi c al Laboratory, Carnegie Institution of Washington. This analysis, its calculation as water-free, and the average of some analyses of andesites for comparison are given in the following table : Chemical analy s is of pyroxene andesite Jrom M orne Dumuraille and average analyses of andesit e s . 1 1 a 2 3 .................................. .......... 61.41 64.23 60.35 59.92 Al20a 14.00 14.65 17.54 17.51 Fe20a 2. 5 6 2.6 8 8.87 2.98 FeO 2.69 2.81 8.17 8.70 MgO 8.01 3.15 2.78 8.31 OaO 5.29 5.53 5.87 6.66 Na20 8.39 8.55 8.63 8.44 K 2 0 1.56 1.63 2.07 1.65 H20 + 8.93 H 20- 0.55 TiOa 1.17 1.22 .78 .48 P20is 0.37 0.89 .26 .20 ){110 0.14 0.15 .18 .15 100.
PAGE 311

' IGNEOUS ROCKS. 277 In its relatively low alumina and high titanium the andesite corre sponds to quartz diorite of Cretaceous age (p. 292) and to the granodiorite of Miocene ( ?) age (p. 304). The silica is also higher than in the average given by Daly. As there is little secondary alteration the high proportion of water in the lava is noteworthy. Hypersthene andesites. Lavas containing predominant hypersthene or h3rpersthene alone as the ferromagnesian constituent are apparently less abundant than t11e a ugite-hypersthene type. A dark brownish-gray glassy hypersthene andesite was found along the trail between Gros-Morne and Pilate. In a dark-gray andesite from middle or upper Eocene con glomeratic beds at Morne Macat, hypersthene is the predominant ferro magnesian mineral. Except for the absence or the low proportion of augite these lavas are similar to the glassy pyroxene andesites. A ugite-hypersthene-h .CYrnblende andesite. Andesite containing hornblende in addition to augite and hypersthene was found only at Dola11, southwest of Terre-N euve village. This rock is holocryst alline, partly altered, and mottled reddish brown to green. The plagioclase phenocrysts range in composition from calcic labradorite (Ab35 An65) to sodic labradorite or andesine. Augite crystals together with bastite pseudomorphs after hypersthene form 10 or 15 per cent of the rock. The hornblende, generally largely resorbed to an opaque aggregate contai;ning iron oxides, has a pleochroism from yellow to reddish brown, probably due to a high proportion of ferric iron. The groundmass is plagioclase, largely andesi11e, and some magnetite and secondary iron oxides. Secondary quartz and chrysocolla are present as the result of alteration adjacent to mineralized fissures. Hornblende-augite andesites. Lavas of the type known as hornblendeaugite andesites were found at several places, although alteration and weathering makes most of the specimens unsuitable for petrographic study. The unweathered and unaltered rocks are generally gray, but the weathered or altered rocks may be reddish or rusty brown. They are usually speckled with rusty brown needles of hornblende, which have been largely resorbed. A comparatively fresh rock from a locality in the valley of the Riviere Coup a-l'Inde, a tributary of the Estere, just east of Dessalines, is repre sentative of these lavas. It is a .gray rock speckled with small altered needles of hornblende and whitish prisms of partly kaolinized plagioclase from 1 to 2 millimeters in length. The plagioclase phenocrysts occur in rather stout prisms, in part fragmental, and comprise about 15 to 20 per cent of the rock. They are zoned but not so markedly as in the pyroxene andesites, ranging from medium andesine to a sodic oligoclase at tl1e margins of crystals. Their average composition is probably around Ab65 An85 to Ab60 An40 Iron oxides and scattered flakes of kaolinite are alteration products. The hornblende needles comprise 5 to 8 per cent of the -

PAGE 312

278 GEOLOGY OF THE REPUBLIC OF HAITI. rock and are largely resorbed, although a few remnants of a reddish-brown hornblende remain. There are scattered prisms of a fresh colorless to pale greenisb augite. The ground.mass is apparently largely oligoclase but is clouded with specks of iron oxides and contains some chlorite. The plagioclase is in thin needles and stouter anhedral plates, which show no well-defined arrangement. Hornblende andesites. The lavas known as hornblende andesites are generally light gray or greenish gray where fresh, but where oxidized and altered may be reddish or rusty brown. Many of the reddish lavas in the Montagnes de Terre-N euve are probably altered hornblende andesites. The freshest andesite was obtained along the Gona!ves-Ennery railroad in a cut near Kilometer 16. The rock is light gray and carries pbenocrysts of plagioclase and thin needles of hornblende 1 to 2 millimeters in length. The plagioclase phenocrysts range in composition from calcic andesine to labradorite and comprise about 10 per cent of the rock. The horn blende has a thin resorption rim, probably consisting of iron oxides and augite, and is pleochroic from faint yellow to yellowish or brownish green. It comprises 5 to 8 per cent of the rock. The groundmass contains some glassy base and consists of a plexus of thin, needle-like prisms and larger quadratic and sections, chiefly andesine. There is some quartz in the ground.mass, mostly in small nests of anhedral grains. Octahedral grains of magnetite and stout prisms of apatite are accessories. Hornblende-mica andesites or dacites. Rocks of the type of horn blende-mica andesites or dacites were fo11nd southwest of Jean Rabel. They are conspicuously porphyritic, and some specimens have a promi nent :How structure. They are purplish and contain large whitish pheno crysts of plagioclase, from 1 to 6 millimeters in length. Hornblende occurs generally in small, inconspicuous needles, although some are 2 to 3 millimeters in length. There are a few scattered flakes of biotite. The phenocrysts of plagioclase are zonal and range from labradorite (Ab,5 An6s) to andesine, probably averaging aro1lnd Ab50 Anr;0 to Ab45 An55 They comprise 15 to 20 per cent of the rock. A few corroded and fragmental crystals of quartz show transition toward dacite. The hornblende is deep brown, and the smaller needles are generally resorbed to iron oxides and augite. The biotite, like the hornblende, is partly resorbed to magnet-1 ite. The ground.mass, like that in the other hornblende andesites, consists of andesine and contains a few scattered grains of prima ry augite. Magnet ite and stout prisms of brownish apatite are the usual accessories. Opal escent silica, a secondary mineral, obscures the structure of the ground mass and is probably formed by weathering . Pyroxene ( augite-hyperstkene) dacites. Pyroxene daci tes were found at the north base of Mont Puilboreau and in the region north of Gros Morne and Pilate. These rocks are greenish gray to gray where dense and relatively fresh, but many of them, either from oxidation or from alteration show dull shades of red or purple. The green or gray lavas are por

PAGE 313

IGNEOUS ROCKS. 279 phyritic, with phenocrysts of plagioclase ranging from 0.5 to 5 milli1neters in length and crystals of augite and hypersthene, both of which are commonly altered. Quartz is present in rounded bipyramidal crystals, some of them 5 to 6 millimeters in diameter. In most of the specimens studied the plagioclase had been altered by the Cretaceous quartz diorite intrusion to albite and sericite; in one relatively fresh rock it consisted of sodic andesine ( Abs5 Ans5) Except for the phenocrysts of quartz and for some quartz in the groundmass the rocks ate much like the pyroxene andesi tes in general textural features and accessory minerals. H orn b lende-augite dacites. Lavas of the type known as hornblende augite dacites were found in the Montagnes de Terre-Neuve north of Terre -Neuve village. They are purplish gray and contain prisms of whitish feldspar from 1 to 2 millimeters in length, scattered crystals of quartz, and dull-brown needles of hornblende that are largely resorbed. The phenocrysts of plagioclase are near andesine ( Ab60 An40), and the groundmass consists of sodic andesine in thin prisms with a trachytic texture. There are scattered crystals of augite partly replaced by calcite Hornblende-mica dacites. The hornblende-mica andesites near Jean Rabel contain some quartz, a .nd some varieties are dacites. Hornblende mica dacite porphyry was found in the Montagnes Noires, but because of its coarse porphyritic character and the accompanying alteration it prob ably do e s not belong to the andesite-dacite effusive group, and it is dis cussed under the heading ''Intrusive rocks.'' (See p. 299.) RELATIONS OF THE v ABIOUS TYPES. The variation in the mineral composition and texture of the lavas of this .group covers a rather wide range. The almost continuous gradations between the various types indicate their close relationship and their common origin. The pyroxene and hypersthene andesites constitute the more basi c members of this group; the hornblende andesites and the dacites the more salic members. Pyroxenes are found, however, in some of the dacitic lavas, and some hornblende also occurs i n the more basic andesites. Conditions such as the percentage of water or variations in pressure or in temperature probably controlled the crystallization of hornblende and pyroxene from the magmas. In the rocks nearest the salic end of the group, the hornblende-mica andesites or dacites, the pyroxenes are subordinate or absent, and the alteration of these more salic rocks as a whole indicates greater activity of the water in their magmas. The l1niversal resorption of hornblende to augite and iron oxides probably is the result of extrusion or intrusion to positions of reduced pressure, where the escape of water vapor rendered the hornblende unstable. In certain altere d hornblende-mica andesites near Jean Rabel small crystals and microlites of pyroxene crystallized in the groundmass, though hornblende was the stable mineral during the early or intratelluric period of crys tallization.

PAGE 314

280 GEOLOGY OF THE REPUBLIC OF HAITI. ALTERATION AND METAMORPHISM. The principal widespread alterations in this group of lavas are the alteration of hypersthene, resorption of hornblende, and some replace ments of the constituents by secondary forms of sil i ca, chlorite, serpentine, and probably hematite and other iron oxides. 'I,he hypersthene commonly alters to bastite or other forms of serpentine, which may be accompanied by quartz, chalcedony, or opal. These alterations are all apparently related to events occurring during the extrusion of the lavas, such as the action of expelled magmatic waters, or to the action of associated hot springs. More intense forms of local metamorphism, as alb i tization and sericitization of the plagioclase, replacement of augite by aggregates of calcite, quartz, plagioclase and iron oxides, and more complete cbloriti zation, are the results of circulating solutions genetically related to the quartz diorite intrusion. LATER BASALTIC ROCKS. DISTRIBUTION AND STRUCTURAL RELATIONS. Basaltic lavas that are yo11nger than the andesites and dacites prevj ously described appear to be confined to a zone which extends from the vicinity of Ennery northwestward at least as far as the village of Gros Morne and which includes parts of the Montagnes de Terre-N euve. This zone is rather indefinitely bounded, both because of the lack of informa tion and because the lavas are covered by the Tertiary limestones. Lavas of this group are exposed at a number of places along the Gona!ves-St.-Michel road southeast of Ennery. They are fo11nd under the upper Eocene limestone and fragments of them occur as cobbles or peb bles in conglomeratic beds near its base. Outcrops of the lavas at many places in this vicinity show pillow structure. Other exposures of these basalts 11nderlying the upper Eocene limestone were seen at the southern base of Mont Puilboreau along the Plaisance-Ennery road. Northeast of Gona!ves, approximately south of Morne De11x Mam elles, these basalts crop out both along the automobile road to Ennery and along the Gonaives-Ennery railroad. At Kilometer 16 on the rail road a cut shows a fault contact between the basaltic rocks of this series and the l1nderlying hornblende andesites. (See Pl. XVIII, A.) The an desite has been thrust up into contact with the basalt. The fault strikes about N. 50 E. and dips 57 SE. The basalt is amygdular and shows pillow structure. A reddish shaly limestone containing many small Foraminif era fills the interstices between the rounded to sub angular pillows of basalt. The evidence indicates that the basalt had intruded or flowed over beds of unconsolidated calcareous deposits. Essexite belonging to this group of rocks crops out about half a kilometer northeast of Potea u. It probably underlies the Eocene limestone unconforrnably. The texture

PAGE 315

REPUBLIC OF HAITI GEOf..,OGICAL SURVEY : I 1 F .\.l_;I.,T IX R .\IT .. .. \.D Cl.1 OCTII OF MORNE DEUX )!Al\IELLES, S11 Ob" GO:NA l \ ? J:S .. .\ncles ite (at l<1ft) is tbrt1Rt o (' r basalt ( I., 8Alr T D'EAU

PAGE 316

' I

PAGE 317

IGNEOUS ROCKS. 281 of this rock indicates that it may belong to a minor intrusive body of this series. Lavas petrographically s imilar to those in the vicinity of Ennery occur in the Montagnes de Terre-N euve. Particularly fresh diabasic and amygdular rocks are exposed along the trail from Gros-Morn e to Terre .N euve on the northeast slope of Morne Decouflay. Some of the rocks at this locality have a coarse diabasic texture with ophitic plate s of augite 8 or 10 centimeters in diameter and may be parts of minor intrusive bodies. Abundant float of this type of basalt is found along the valley of the Riviere Lhormand, part of which may come from the western part of Section Moulin. The basaltic rocks in the vall e y of the Riviere Bras-aDroit are of the Ennery type, and at one place they seemed to form an intrusive sheet parallel to the bedding of the Eocene limestone. As there were no inclusions of the limestone in the basalt and as the relations could be explained by strike faulting, this exposure is not considered as conclu sive evidence of the Eocene or post-Eocene age of the basalt at tl1is locality. Along the trail from Gros-Morne to Pilate, 2 or 3 kilometers east of Gros-Morne, these later basalts are exposed near reddish-brown shaly limestones of unknoWI} age. Although the region was not explored, rocks of t his group will probably be found in the area of igneous ro cks on the east side of the trough of the vall e v of Les Trois Rivieres from Morne D eux Mamell e s northw e stward ., to Gros-Morne. PETBOGRAPHY. Ge1teral features. This later series of basaltic lavas and intrusive rocks has some distinctive features which sharply separate them petrographi cally from the ear lier basaltic rocks. The y are characteriz e d by a purplish titaniferous augite that commonly shows a tendenc y toward idiomorphic development, in contrast to the interstitial pale greenish augite or diop side of the ordinary ba salts; some of the more alkaline rocks are further. characterized by sodic plagioclase, anorthoclase, and analcite or other soda zeoli tes. In chemical composition the magmas ranged from essexite to nearly normal olivine diabase. All the types probably have a high content of titanium, but no chemical analyses are ava i lable to determine the extent of the variation in the amount of this constituent from that in ordinary basal tic rocks. No primary sodalite or nephelite was identified in any of the rocks, but muc h of the clear analcite, if it is secondary, has probably replaced these minerals, as its relations in many places indicate that it is not a replacement of primary plagioclase. In some rocks intense zeolitization has resulted probably from the local circulation of highly alkaline solutions of extraneous origin caused by the general volcanic activity.

PAGE 318

282 GEOLOGY OF THE REPUBLIC OF HAITI. Essexite. Rock here classified as essexite was found at only one lo cality, about half a kilometer northeast of Poteau along the road from Gona1ves to Ennery, at the base of a small hill west of the road. BecauBe of it.s relatively coarse diabasic texture this rock has been classified as essexite, although its field relations do not show whether or not it is part of an intrusive body. The essexite is a medium-grained greenish-gray rock of diabasic texture in which prisms and grains of dark augite and of greenish plagioclase can be recognized with the unaided eye. The rock is spotted with pinkish glassy patches of analcite 2 to 3 millimeters in diameter. In thin section the texture generally is hypidiomorphic but locally may be typically ophitic. The earlier plagioclase is labradorite ( Ab,ts Anlits), but the crystals are very strongly zoned and most of these have borders of oligoclase or of albite. The albite is distinguished by its low refractive index and positive optical character. The average plagioclase of the rock is at least as sodic as andesine. Potash feldspar forms irregular borders surrounding the plagioclase, or occurs in clear interstitial patches between the plagioclase and augite, or borders miarolitic cavities which are filled with analcite and zeolites. The potash feldspar in some places shows idiomorphic outline against analcite in miarolitic cavities. The orthoclase probably is soda orthoclase, as the indices are higher than those for normal potash orthoclase (a=l.523; y=l.53), and the optic angle (2 V) is medium to small. The proportion of soda orthoclase or anortho clase is difficult to estimate because of its alteration and the presence of albite, but the ratio of orthoclase to plagioclase appears to lie between one :fifth and one-third. The pyroxene is a pleochroic augite, commonly somewhat zonal, and is found mostly in hypidiomorphic prisms with cross fractures and with the margins molded on the more calcic pla gioclase. A few of the crystals are more idiomorphic, but some of the augite fills the angular spaces between the plagioclase in a typical ophitic man ner. The maxim11m extinction angle of the zonal augite on 010 is 53, and the pleochroism is from yellowish gray to purplish brown. The crys t als are generally from 1 to 2 millimeters in length. The analcite is interstitial and fills the angular spaces between the feldspars and between the plagioclase and augite, or else it occurs in aggregates, some of which show crystal outline in miaroli tic ca vi ties in the rock. Analci te also re places the plagioclase along some of the cleavage cracks, and such analcite is undoubtedly secondary. Ilmenite, which is usually in peculiar skeletal growths, is quite common in grains from microscopic size up to 0.4 or 0.5 millimeter in diameter, and some of it is included in the augite. Some titaniferous magnetite may be present. Long fractured needles of a nearly colorless or slightly greenish apatite are very abundant as inclusions in the alkali and plagioclase feldspars but are less common in the augite. Small shreds oi a dark-brown strongly pleochroic biotite are

PAGE 319

IGNEOUS ROCKS. 283 present. Both the feldspar and the augite are replaced along cracks by chlorite. Calcite occurs in spherulitic or radial growths, some of them molded on the analcite crystals, and fills many of the cavities in the roclr. The mineral composition of the rock was calculated approximately by the Rosiwal method as follows: Mine ral compoS'ition of essexite from locality near Poteau. Percentage by volume. Feldspar . . . . . . . . . . . . . . . . . . . . . . . . . . 50 .0 Augite . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 .0 Analcite . . . . . . . . . . . . . . . . . . . . . . . . . . 15.0 Iron-ores . . . . . . . . . . . . . . . . . . . . . . . . . . 8.0 Biotite and apatite. . . . . . . . . . . . . . . . . . . . . . 0.5 The percentage of soda orthoclase in the rock could not be conveniently measured but was estimated at 8 to 12 per cent. The strongly zonal plagioclase probably averages at least as sodic as andesine in composition. 'l'he rock has some affinities with rocks which have been described as analcite monzonites, although in the relative proportions of the alkali feldspar and the plagioclase it is more comparable to the granodioritea among the quartz-bearing rocks. Johannsen, in his mineralogical classi fication, 1 has suggested the name monzodiori te for rocks of this char acter, and in this classification the rock would be an analcite monzodi orite with the symbol 2224. A nalcite andesites and analcite-olivine andesites. The analcite ande si tes and analcite-olivine andesites are dark-gray to greenish-gray rocks where unweathered, but where they aTe at the surface they may be a rusty or greenish brown. Their field relations do not definitely show whether these rocks are parts of thick flows or of intrusive bodies. They are gen erally spotted with growths of natrolite or other zeolites. Fine-grained roc ks of this type were found north of Ennery near the base of Mont Puilboreau and in the valley of the Riviere Bras-a-Droit north of '11erre-N euve. The plagioclase is strongly zonal and ranges in composition from labra dorite to sodic oligocla.se or albite. Because of the irregular borders of sodic plagioclase the inter lacing prisms are never strictly euhedral. Be tween the plagioclase laths a small amount of soda orthoclase is found in some of the specimens, but it is inconspicuous. The plagioclase generally amo11nts to 40 or 50 per cent of the rock. 1The purple pleochroic augite occurs in subordinate amo11nt, for1ning 20 to 30 per cent of the rock, and is exceedingly variable in its development. It may form subhedral to euhedral prisms, which are commonly hollow or skeletal, or it may occur in anhedral grains interstitial to the plagioclase. Olivine apparently i Johannsen, A., A quantitative mineralogical classification ot igneous rocks revised! Jour. Geol., vol. 28, pp. 50, 175, 177, 1920.

PAGE 320

284 GEOLOGY OF THE REPUBLIC OF HAITI. was a primary constituent in some of the rocks but is completely altered to serpentine in all those examined. The serpentine has migrated to some extent and is accompanied by chloritic minerals. In some specimens analcite and in others natrolite or other zeolites fill angular spaces be tween the plagioclase, and all the zeolites replace the feldspars to a variable extent. In one specimen analci te forms sea ttered round grains. The iron ore is probably large l y ilmenite and forms peculiar skeletal grains, but much of it shows no alteration. Some of it is accompanied by small flakes of biotite. Small microlites of a greenish mineral, probably augite, are ab11ndant in some specimens. All the rocks contain some calcite. These rocks clearly are transitional between the essexite and the more normal basalts described below. As the average plagioclase is andesine and the f erromagnesian minerals are subordinate, the rocks can not be classified as basalts or tephrites, although they have some of the structural characteristics of the normal basaltic rocks of this group. Amygdaloidal basalts. Conspicuous amygdaloidal basalts were found along the road between Ennery and St.-Michel, where they underlie the upper Eocene limestone. The amygdules generally consist of chlorite, calcite, or zeolites. The rocks are normally dark greenish gray but may be stained brown by weathering. One of these rocks from a locality about 3 kilometers southeast of Ennery is nonporphyritic in thin sec t ion and has a type of intergranular texture. The prisms of plagioclase and augite are from 0.2 to 0.5 millimeter in length, and the amygdules, which comprise about 20 per cent of the vol11me of the rock, are from 1 to 3 millimeters in diame t er. The plagioclase is labradorite and slightly zonal, ranging from Ab85 An65 to Ab.5 An55. The augite is purplish brown and zonal, with ''hour-glass'' structure, and has an extinction on 010 that ranges from 46 to 52 Some of the augite occurs in irregular grains interstitial to the plag ioclase and some forms thin nearly euhedral prisms, in plac e s grown parallel to the plagioclase. The iron ore s occur in part in skeletal growths and are presumably titaniferous. Chlorite is abundant in the rocks and is either interstitial to the plagioclase and augite or fills or lines the vesicles. Augite is the predominating con s t ituent of the rock and forms 40 to 50 per cent of the groundmass; chlorite makes up 12 to 15 per cent of the rock exclusive of that in the amygdul e s. Both calcite and zeolites form :fillings in the amygdules. Z eolitized olivine basalts and diabases. Olivine-bearing basaltic rocks, which are nearly nor1nal basalts although they still show some affinities fo the more alkaline members of this group, were fo11nd in the Montagnes de Terre-N euve. They are dark greenish gray and contain prisms of plagioclase 1 to 2 millimeters long, a few yellowish grains of olivine, and large plates of pyroxene. Some of the rocks are speckled with white radial growths of zeolites, mostly natrolite. The plagioclase is only slightly

PAGE 321

IGNEOUS ROCKS. 285 zonal, ranging from bytowni te ( Ab15 An85) to basic labradori te, and is in euhedral to subhedral laths from 0.5 to 2 millimeters in length. It comprises approximately 40 per cent of the rock. The augite is brownish gray, and some of it has a distinct purplish tinge, but it is not so st1ongly tinted as that in the more alkaline rocks. The extinction on 010 is about 45 In places it forms large plates, 4 to 6 millimeters in diameter, which are poikili tic and inclose prisms of plagioclase. The olivine occurs in ro11nded and partly or completely serpentinized granules, which range in diameter from 0.5 to 2 millimeters, and comprises from 5 to 10 per cent of the rock. Some of the olivine wraps partly around the end of the prisms of plagioclase like the augite. The principal accessory mineral is magnetite in irregular skeletal or octahedral grains, amo11nting to about 5 per cent. Thin needles of apatite, a few prisms of green and brown hornblende, and a few flakes of biotite partly altered to chlorite are minor accessories. The serpentine is a greenish flaky nonpleochroic variety and replaces plagioclase adjacent to the olivine along with some chloritic minerals. Analcite has replaced the basic plagioclase along cracks to some extent, and zeolites, mostly natrolite, are present as fibrous radial growths be tween the laths of plagioclase. A small amount of calcite is generally associated with the zeolites. The zeolites and subordinate analcite may constitute from 5 to 10 per cent of the rock. Olivine diabase. The type classified as olivine diabase is similar to the preceding rocks in texture and mineral composition, but the rock contains little or no analcite and only a few scattered growths of natrolite. It was found on the east slope of Morne Decouflay, along the trail be tween Gros-Morne and Terre-N euve. Its coarse texture and position indicates that it may bear intrusive relations to the pyroxene andesites. (See Fig. 27, section A-A'; p. 442.) ALTERATION. The principal alterations are the formation of zeolites, chlorite, serpen tine, and calcite. These rocks show no recognizable alterations that could be attributed to the Cretaceous regional metamorphism. The above mentioned secondary minerals can all be attributed either to later magmatic ( deuteric) replacements, or to the action of circulating alkaline solutions associated with the vulcanism. As indicated by the sodic plagio clase and soda ortboclase among the later minerals of crystallization of some of the rocks, there was a considerable concentration of alkaline con stituents in the residual magmatic liquids, which resulted in the formation of analcite and natrolite. Some of the rocks have probably been enriched by soda derived from alkaline solutions of extraneous origin. Chlorite and serpentine generally appear to have been formed before zeolite or calcite. Chlorite fills pores in the rocks and in many lavas lines vesicles that are filled with later calcite or zeolite.

PAGE 322

286 GEOLOGY OF THE REPUBLIC OF HAITI. RELATIONS OF THE v ARIOUS TYPES. Be cause of the close assbciation of the members of this series in their distribution, and because they show certain distin ctive mineralogic fea tures, they are believed to constitute a cogenetic group that is entirely separate from the older andesites and dacites. The distinct gradations that occur between the alkaline and the normal basaltic members of this group indicate genetic relations. In the group as a whole the order of crystallization of the essential minerals is generally olivine, bytownite and labradorite, titaniferous augite, sodic plagioclase, soda orthoclase, analcite, and natrolite. Sodic plagioclase and small amo11nts of orthoclase found in rocks where crystallization begins with olivine and labradorite furnish sufficient proof that the rocks of this group are closely r e lated and that olivine diabase magma was the parent magma of this group. The decided tendency of the magma toward an alkaline descent is appar ent even in the more basic members of the group. The strong zoning of the plagioclase and possibly a low percentage of water and the presence of carbon dioxide in the magma should be considered as contributing to this descent. Calcite occurs in all the rocks and appears to have been deposited from the solutions that deposited zeolites, although at a some what later stage. The surface lavas contain blowholes, many of which are filled with calcite. AGE OF THE LAVAS. At most places these rocks seem to be older than the middle and upper Eocene lim estones, and probably they are younger than most of the pyrox ene and hornblende andesites. This would place their period of eruption some time between early Cretaceous and mjddle Eocene. At several places they are associated with shaly calcareous sediments of unknown age, and at some places they have intruded or flowed over llnconsolidated calcareous muds. So far as known they were not intruded or altered by the late Cretaceous batholithic intrusion. They are there fore probably of ear ly Eocene or late Cretaceous age. INTRUSIVE ROCKS. GENERAL FEATURES AND DISTRIBUTION. Igneous intrusives, which represent at least three periods of intrusion, can be recognized in the northern part of the Republic. Little is known of the earliest period, evidences of which remain only in much meta morphosed intrusives associated with the older basalts. Most of the rocks of this group were basic types intruded probably during or soon after the earlier periods of basaltic eruptions. The most important period of intrusive activity is represented by the quartz diorite intrusions. These batholithic rocks, which are probably of late Cretaceous age, intrude the

PAGE 323

IGNEOUS ROCKS. 287 older Mesozoic volcanic rocks and the older schists. The intrusion and the preceding and accompanying folding resulted in widespread regional metamorphism. The latest known intrusions are represented by minor stocks of quartz diorite or granodiorite of Miocene ( ?) age. Minor intru sive bodies that accompanied the various volcanic eruptions have been mentioned in the description of the volcanic rocks. The largest area of intrusive rocks is exposed in the eastern part of the Massif du Nord, where the core of the quartz diorite batholith has been exposed by erosion. West of Grande-Riviere du Nord the quartz diorite is exposed only in smaller stocks or bosses. The oldest intrusive rocks associated with the metamorphosed volcanic rocks and amphibolites were not differentiated in the field. The yo11ngest intrusions are in the Mon tagnes de Terre-N euv:e and occupy only a narrow zone. Small bodies of porphyry, probably in part of intrusive origin, are exposed in the Mon tagnes Noires. EARLIER BASIC INTRUSIV'ES. DISTRIBUTION AND STRUCTURAL REI.ATIONS. In the field no attempt could be made to separate the earlier basic intrusive rocks from the older volcanic rocks, as their character and rela tions are obscured by extreme metamorphism. No large bodies of intrusive rock of this group were recognized. Minor intrusive bodies of composition similar to that of the extrusive basaltic rocks and minor bodies of basic differentiates, such as peridotites and pyroxenites, probably constitute most of the bodies of rock that make up this group. Altered gabbros, epidiorites, and peridotites or pyroxenites, which are described from the Province of Santiago in the Dominican Republic,1 may represent part of these old basic intrusives. The serpentine that covers large areas in the Cordillera Central of the Dominican Repub lic may represent altered peridotites, pyroxenites, or gabbros. Much altered and relatively coarse grained diabasic rocks, the structural relations of which are unknown, were found on Limbe Mountain and between Limbe and Plaisance. Float of similar rocks was fo11nd on the Plaine du Nord. Much-metamorphosed basaltic rocks and float of augite peridotite were found in the Valley of Las Lomas. Many of the highly metamorphosed rocks the amphibolites, talc schists, and epidosites were derived by alteration from basic or ultrabasic rocks, some of which may have been parts of minor intrusive bodies. PETROGRAPHY. A number of the rocks of these older igneous groups could not be definitely classified because of advanced metamorphism. They consist chiefly of uralitic hornblende and some dusty albite feldspar, generally 1 A geological reconnaissance ot the Dominion Republic: Dominican Republic Geol. Survey Memoirs, vol. 1, pp. 84-85, 1921.

PAGE 324

288 GEOLOGY OF THE REPUBLIC OF HAITI with some chlorite, epidote, and quartz. Some of the commoner types of the rocks showing more advanced metamorphism are described on pages 306-309. Metadiabase. Rocks that have probably resulted from the metamor-. phism of relatively coarse diabasic rocks were found on the Plaine du Nord a .nd in the mo11ntains near Limbe and north of Plaisance. These rocks are dark green to greenish gray and of rather unifor1n texture. They may be stained rusty brown by weathering. One of these rocks, from a locality n ear the crest of Limbe Mountain, shows in thin section traces of an op hi tic or dia basic texture in less altered parts. The plagioclase consists largely of dull-brown or dusty albite crowded with flakes of s ericite and microlit e s of the alteration products of the rocks, pale-green hornblende, chlorite, and epidote. Remnants of primary augite show an extinction angle on 010 of about 45, although the crystals are largely replaced by uralitic hornblende. The secondary hornblende is faintly pleochroic, from yellowish or nearly colorless to a faint gree n or blue green with the angle Z /\ c about 16 or 17. The amount of uralitic hornblende a .nd augite somewhat exceeds the amount of plagioclase. Chlorite is fo11nd as an alteration product parallel to the cleavage lines of the hornblende. The iron ores are pa1tly recrystallized and are coated with grains of ti tani te. A rock probably of similar origin fo11nd on the North Plain contains more epidote and quartz. In some of the specimens the sodic plagioclase is partly altered to sericite. Augite peridotite. Although p eridotite was found only in the Valley of Las Lomas, rocks of this type are believed to be generally distributed in minor quantities throughout the older series. Peridotites are reported from the basal complex of the Dominican Republic, as well as from Cuba, a .nd are the natural basic differentiates of the basaltic magmas of these early periods of igneous activity. The augite peridotite is a greenish-black rock of medium grain com posed of augite and other ferromagnesian minerals and minor quantities of greenish plagioclase. In thin section the rock is seen to be composed predominantly of olivine in anhedral to subhedral crystals, which form 60 to 70 per cent of it. It alters with the characteristic mesh structure to serpentine and iron oxides. The augite is brownish yellow, unaltered, and generally interstitial to the olivine crystals. Some of the larger plates of augite inclose small subhedral olivine crystals in a poikilitic manner. An orthorhombic pyroxene in small cross-fractured prisms and grains with faint pleochroism, from nearly colorless to pale yellowish, is a minor accessory. Plagioclase, probably bytownite, is an interstitial constituent and is in part altered to opaque brownish saussuritic patches and to ser pentine derived from the olivine. It constitutes about 10 per cent of the rock. The other acces s ories are grains of iron ore and flakes of a pale altered biotite

PAGE 325

IGNEOUS ROCKS. 289 EARLIER QUARTZ DIORITE GROUP. DISTRIBUTION AND STRUCTURAL RELATIONS. The rocks of the earlier quartz diorite group are most widely distributed in the eastern part of the Massif du Nord, where the batholith of quartz diori te is exposed by erosion over an area of 500 to 650 square kilometers. Small stocks and apophysal dikes are exposed on the Plaine du Nord. The rocks of this group clearly are intruded into the Mesozoic volcanic rocks and schists, but whether they are intrusive into the sedimentary rocks of supposed Lower Cretaceous age on the southern border of the Massif du Nord is not known. During the reconnaissance no intrusive contact between these rocks was found. A locality between Bahon and Ranquite,1 where they presumably are in contact, was not visited. Evi dence discussed on page 299 indicates, however, that the quartz diorite is younger than the Lower Cretaceous rocks. About 12 or 15 kilometers east of Le Trou the volcanic rocks are in truded and metamorphosed by small stocks or dikes of quartz diorite porphyry, and abundant float of the intrusive rocks is found on the sur face of the volcanic rocks. On Morne Beckly dikes of porphyry and felsite, which probably belong to this intrusive group, cut the schists. Veins of quartz carrying sulphides or iron oxides and pegmatitic veins have been considerably defor1ned. West of the Grande Riviere du Nord the intrusive rocks are exposed only in small stocks or dikes cutting the metamorphosed volcanics. Such bodies were seen on Morne du Cap north and west of Cap-Ha1tien and on the North Plain along the road from Cap-Ha1tien to Limbe. About 6 or 7 kilometers north of the town of Grande-Riviere du Nord, near the railroad station called La Tannerie a .nd west of the river, the1e is an exposure of a kaolinized porphyry in contact with fossiliferous limestone of late upper Cretaceous age. (See pp. 94-95.) The relations at this locality suggest faulting, and the limestone shows no evidence of contact meta morphism, but the relations of the quartz diorite to these Upper Cretaceous limestones are not known. South of Limbe, on the road to Plaisance, there are several exposures of quartz diorite. The surrounding older volcanic rocks are strongly meta morphosed. Farther west evidence of intrusive bodies was established by the finding of float boulders of quartz diorite in Section Margot, along the trail from Port Margot to Pilate. Exposures of quartz diorite porphyry, presumably related to this intrusive group, were found jn the Northwest Peninsula at several places in the Commune of Jean Rabel. The largest exposures examined are 7 or 8 kilometers southwest of the town of Jean Rabel, along the valley of the 1 Tippenhauer, L. G., Beitrilge zur Geologie Haltis: Petermanns Mltteilungen, vol. 47, map opposite page 198, 1901. 19

PAGE 326

290 GEOLOGY OF THE REPUBLIC OF HAITI. Riviere Cadet. The rocks are apparently intrusive into hornblende-mica andesites at this locality and are overlain unconformably by limestone of upper Eocene age. Cobbles of quartz diorite and of the porphyries are fo11nd in the basal upper Eocene conglomerate along the valley of the river. Near the crest of the mountains along the trail from Jean Rabel to Anse Rouge quartz diorite porphyry underlies the upper Eocene limestone. PETRoGRAPHY AND CHEMICAL COMPOSITION OF THE NoRMAL QUARTZ D10RITE. The normal quartz diorite, the dominant type of this intrusive group, is a speckled gray to greenish-gray medium to coarse grained granitoid rock. (See Pl. XIX, A.) The constituents recognizable with the unaided eye are prisms of whitish plagioclase showing albite twinning lamellae, greenish-black hornblende, generally subordinate to the light constituents, and abundant quartz, usually in grains interstitial to the plagioclase and hornblende. A little magnetite and some chlorite can be distinguished, and in altered rocks some secondary pyrite and epidote. The rocks are generally rather even grained in texture, and the plagioclase is here and there idiomorphic. The distinctly porphyritic rocks, which are less abun dant, contain phenocrysts of plagioclase, the largest of them 1 centi meter in length, which stand out prominently as white blotches on the weathered surface. (See Pl. XIX, B.) One rock was found in which large poikilitic plates of hornblende, 8 to 10 millimeters in diameter, inclose some of the smaller prisms of plagioclase. Along narrow zones the quartz diorite has been intensely sheared, although no general gneissoid texture was found. Some of the rock shows evidence of intense dynamic strain that has crushed the quartz and plagioclase a .nd drawn out the hornblende. In some of these crushed rocks quartz is distinctly bluish. Quartz diorite gneiss that apparently belongs to this group of int1'usives has been found in the Dominican Re public 1 in the Cordillera Central, a continuation of the Massif du Nord of the Republic of Haiti. In thin section the plagioclase shows a euhedral tendency and generally seems to have been the first mineral to crystallize except a few small grains of accessory minerals (See Pl. XXI, A.) The prisms are sub hedral to euhedral and are generally 1 to 4 millimeters long. The outside rims of the smaller crystals show growth interference with the larger crystals of hornblende and with a few of the crystals of quartz. The plagioclase shows Carlsbad and albite twinning, is zonal, and ranges in composition from calcic andesine or sodic labradorite in the centers of p1isms to sodic andesine or oligoclase on the borders (Ab45 Ancs5 to 1 A geological reconnaissance of the Dominican Republic : Dominican Republic GeoL Survey vol. 1, pp. 84, 202, 1921

PAGE 327

J t1:<; 1 t t: f I(' <11' 11,\.l'l'I 1; I:< 1 1 < J (; l l'. \ l :-> l It \ l. :l .t. QlJ c\It' l,7. DIOfiT'l'E J 4.,fl0:\f :\T.\DF:f,TXE. l,fi:8 11<.)l{l1lIYl{l' l 'IC QUAR'fZ DIORITE FROl\I MOR:XE <>I' J,Ji:'

PAGE 328



PAGE 329

I IGNEOUS ROCKS. 291 Ab1o Anso) Its average composition is probably around Ab6o An.ao Most of it is slightly altered to gray saussuritic patches, in so:ille specimens zonally developed, and small grains of colorless zoisite or clinozoisite. Some sericite generally occurs along the cleavage lines. The quartz occurs in irregular grains from 1 to 3 millimeters in diameter, interstitial to the other constituents of the rock. It is irregularly cracked and has a wavy extinction. Small bubbles containing gaseous and liquid inclusions are common along the cracks in the quartz. The hornblende is in subhedral prisms, generally 1 to 3 millimeters in length, less commonly 5 to 6 millimeters, or in irregular grains of smaller size molded on the more calcic plagioclase. The earliest hornblende gen erally is interstitial to the larger prisms of plagioclase, but the period of crystallization of the hornblende overlaps that of the plagioclase and to some extent that of the quartz. The hornblende ranges in color in different specimens of rock, or even in the same specimen, from colorless to olive green, blue-green, or brown-green. Colorless hornblende in a coarse por phyritic rock from Morne Madeleine, south of Les Perches, has a border of light olive-green. The colorless variety was of slightly lower refraction and somewhat higher double refraction than the green variety, but both varieties had about the same extinction angle (Z A c=l9 to 20). Magnetite is present in octahedral or irregular grains generally asso ciated with or inclosed in hornblende or chlorite. Small prisms of colorless a .patite inclosed in plagioclase and a few grains of titanite are found in most of the rocks. A few minute crystals of zircon were noted in some slides. Chlorite (pennjnite) is a secondary mineral that is everywhere present, either replacing hornblende or in separate flakes, and is in places interlaminated by opaque streaks of fine granular epidote or epidote and calcite. Possibly some of the chlorite may be secondary after biotite. Biotite altering to cblorite was fo11nd in a porphyritic phase of the intrusive rock in the Northwest Peninsula. Biotite is reported to be subordi nate to hornblende in specimens of quartz diorite from the Cordillera Central in the Dominican Republic.1 Orthoclase is usually absent in sec tions of the typical quartz diori te. One rather unusual type of quartz diorite was found near the chapel on Savane Longue, on the trail from Mont Organise to Ouanaminthe. In this rock large poikili tic plates of hornblende in close small prisms of plagioclase and crystals of magnetite. The hornblende is more abundant than that in most of the quartz diorite and constitutes about 28 per cent of the rock by volume. The quartz also forms large grains, which include and wrap aro11nd the plagioclase in a poikilitic manner. (See the follow ing table, specimen No. 3.) The plagioclase ranges from Abtscs An,1 to Ab,o Anao and the average is probably a calcic andesine. 1 Op. cit., p. 84, I ..

PAGE 330

292 GEOLOGY OF THE REPUBLIO OF HAITI. The approximate mineral composition of several thin sections of quartz diorite as determined by the Rosiwal method is as follows : Approximate volumetric mineral composition of quartz diorite.a Quartz ............................................... Andesine b ... Hornblende ........................................................... Magnetite ............................................................ Ohlorite Epidote Titanite e t t t e t e e t t e t e e e t t e t e I e e e e e e e e e e e e e e e a e e I e e e e e e 1 34.0 51.3 12.0 1.2 1.5 Trace 2 27.6 43.3 24.0 2.1 8.0 Trace s 22.4 41.0 28.0 2.7 4.2 1.2 0.6 a See also p .. 495 for mineral composition and physical properties of quartz diorite tested for road-building material. b Includes 2 to 3 per cent of sericite. 0 Includes some epidote. 1. Specimen from Riviere Marion, near Acul Samedi. Field No. K 185 L. (See Pl. XIX, B.) 2. Specimen from Morne Madeleine, south of Les Perches. Field No. K 186 L. 3. Specimen from Savane Longue between Mont Organise and Ouanaminthe. Field No. W 342 L. The chemical composition of a specimen of quartz diorite collected on the north slope of Morne Madeleine, along the trail between Les Perches and Valliere, is shown in the following table, which contains also one hitherto unpublished analysis of quartz diorite from the Virgin Islands and the average of twenty analyses of quartz diorite for comparison. Analyses of quartz diorite from M orne Madeleine and quartz diorite from Virgin Islands, and average analysis of quartz diorite. 120a A s F F M c N i02 e20s eO gO aO a20 K20 e I e I e ' I e ' I e e ' I e e I e I I I a e e e e e a t 20 H H T p + 20- i02 205 M nO Bao, 0.04; SrO, 0.06. 1 2 8 13.33 16.47 16.52 59.66 61.22 59.,7 5.12 2.97 2.63 6.58 3.47 4.11 4.50 2.54 3.75 5.96 6.18 6.24 S.l, 3.81 2.98 0.9'7 1.77 1.93 0.47 1.12 .. 1.89 0.08 0.19 .. 1.55 0.48 O.M 0.18 None 0.26 0.12 0.13 0.08 100.61 100.4f>G 1. Quartz diorite, north slope of Morne Madeleine, between Les Perches and Valliere, Republic of Haiti. H. S. Washington, analyst. 2. Quartz diorite, Virgin Islands, Island of Vieques, western edge of Mosquito. R. C. Wells, analyst; T. W. Vaughan, collector. 8. Average of 20 analyses of quartz diorite. Daly, R. A., Igneous rocks and their origin, p. 26, 191,

PAGE 331

IGNEOUS ROCKS. 293 The approximate mineral composition of the analyzed specimen, esti mated from determinations by the Rosiwal method and from chemical calculations, and the norm calculated according to the quantitative clas sification are given in the following table: Approximate mineral composition and norm of quartz diorite from M orne Madeleine Approximate mineral composition. Norm. Quar'tz ..................................... 23. o Quam .................................... 18. 89 Andesine (Abeo ............... .0 Orthoclase ................................. 6.95 Hornblende ................................ 23. 6 Albi'te ................................. 26. 62 Magnetite .................................. 4. 7 Anorthite ................................. 19.12 Chlori te .................................... 112. 7 Diopside . . . . . . . . . . . . . . . . . 7 57 Hyper s thene ............................... 11 10 Ma, gnetite . . . . . . . . . . . . . . . . 7. 47 Ilmenite . . . . . . . . . . . . . . . . . 2. 9"1 Apatite . . . . . . . . . . . . . . . . . 0.8' The rock is tonalose (II. 4 3. 4''). Includes sericite amounting to 2 to S per cent of the total. "Including some epidote. The high percentage of quartz in these rocks and the negligible per centage or lack of potash feldspar are striking. In the specimen analyzed no potash feldspar was fo11nd on the slide, although very small amounts may have escaped detection. The ch e mical analysis shows, however, that the rock is unusually low in potassium oxide, even for rocks in the quartz diorite family. All or nearly all of the potassium is probably in the plagioclase. The high content of titanium and the low content of alu mina are the only other notable features of the chemical analysis. Ferric iron possibly replaces alumina in some of its combinations. MINOR VARIATIONS. Minor quantities of the rocks of this group range from seemingly basic appearing types, in which the hornblende may constitute 35 to 40 per cent of the rock, to types in which the dark minerals may constitute 15 per cent or less. Most of these rocks are, however, probably classifiable as quartz diorite, although the more basic members are transitional toward quartz-hornblende gabbro, and the more acidic members are oligoclasc quartz diorites or transitional toward granodiorites. The structural and age relations of these different types are unknown. Quartz diorite. Rocks in which the plagioclase is strongly zonal, gen erally ranging from calcic andesine (Ab6o An60) to sodic oligoclase and albite, were found south of Limbe. The borders of oligoclase and albite are irregular, so that although zoning gives the plagioclase an appearance of euhedrism it is largely in subhedral to anhedral crystals. The albite may form irregular patches interstitial to the plagioclase. The plagioclase in these rocks generally is considerably altered to serjcite, to saussuritic

PAGE 332

294 GEOLOGY OF THE REPUBLIC OF HAITI. patches, and to epidote, probably owing to the position of the rocks near intrusive contacts at this locality. Ophitic quartz diorite. A fine-grained phase of the intrusive was found about 4 kilometers northeast of Mont-Organize. It is a gray speckled rather even textured rock that contains a very few scattered ill de:fined pbenocrysts of plagioclase. In thin section the texture is ophitic, the thin euhedral plagioclase laths being inclosed in grains of later crys tallized hornblende and quartz. The plagioclase, which occurs in thin laths from 0.5 to 1 millim e t e r in length, with only a few twinning lamellae, ranges in composition from sodic labradorite ( Abao An5o ) to andesine (Ab65 The mo1e massive hornblende is brown-green and incloses the plagioclase laths like augite plates in typical djabase. In addition a secondary blue-green hornblende in shre ddy prisms cuts across the structure of the rock. The quartz also is in grains inclo sing the plagioclase and is an abundant constituent, forming 25 p e r c ent or more of the rock. Magnetite, chlorite, epidote, and titanite are found in the rock. The plagioclase is partly clouded with saussurite and sericite. In mineral composition the rock apparently corresponds closely to the normal quartz diorite, of which it appears to be a textural variety. Dacite pophyries. Fine-grained contact porphyries were fo11nd near the southern border of the batholith north of Lamielle. They are greenishgray porphyritic rocks with phenocrysts of plagioclase, hornblende, and quartz in a fine-grained groundmass. In one section studied the subhedral plagioclase phenocrysts are stro ngly zon e d and range in composition from sodic bytownite to calcic andesine. The quartz occurs in somewhat cor roded bipyramida l crystals. The hornbl e nde is a pale-green fibrous variety, partly altering to chlorite and epidote. The groundmass con sists of a fine-grained granuliti c growth of andesine and quartz in whi c h the quartz apparently occurs in somewhat subordinate amounts. Rathe r large octahedral grains of magnetite are commonly associated with the hornblende. Kaolinized porphyry. A kaolinized porphyry with large phenocrysts of quartz, some of which are 4 or 5 millimeters in diameter, was collected north of the town of Grande Riviere du Nord, near the railroad station called La Tannerie. The phenocrysts of plagioclase have been completely replaced by kaolinite, and the gro11ndmass has recrystallized to an aggre gate of quartz, kaolinite, and some sodic plagioclase. The rock contains abundant pyrite, which apparently accompani e d the introduction of the kaolinite and quaitz. Quartz diorite porphyries. A greenish-gra y quartz diorite porphyry of medi11m grain crops out about 10 kilomet ers south of Jean Rabel. It is part of the basement under the upper Eocene limestone. In thin sec tion the texture is porphyritic, euh edral to subhedral prisms of plagioclase and hornblende, from 1 to 5 millimeters in length, and cor roded bipyramidal crystals of quartz, the largest of them 2 millimeters in

PAGE 333

IGNEOUS ROCKS. 295 diameter, in a fine-grained ground.mass. The phenocrysts comprise about 50 per cent of the volume of the rock. The plagioclase ranges in com position from andesine to sodic oligoclase and probably averages a sodic andesine (Ab10 Anao to Ab65 Ans5). The hornblende is in long twinned pale-olive or blue-green prisms and comprises about 20 per cent of the rock. The gro11ndmass of the rock consists of a :fine intergrowth of sodic plagioclase, quartz, and some orthoclase, although the proportions of the constituents are difficult to determine. Titanite is an ab11ndant accessory or secondary mineral. It accompanies the hornblende and occurs also in the groundmass. The orthoclase in these porphyries is much more abundant than in the quartz diorites exposed farther west in the Massif du Nord. Rocks like the porphyries just described but in which the plagio clase is more sodic, probably averaging oligoclase in composition were found southwest of Jean Rabel. Much of the porphyry here is pyritized. The pale-green color and shreddy appearance of the hornblende in these rocks from the Northwest Peninsula and the grains of titanite included in and associated with it suggest that it has been formed by re crystallization of a primary hornblende with the accompanying separation of titanite. These porphyries are intrusive into hornblende-mica andesites, which possibly accounts for their texture and alteration. D1xE RocKs AND VEINS. General features. }fost of the dike rocks that are definitely correlated with the quartz diorite are those which cut the main body of the batho li th itself in the Massif du Nord. A small number of dikes and veins of doubtful origin on the Plaine du Nord cut the schists and metamorphosed volcanic rocks but are believed to be genetically related to the rocks of the quartz diorite group. Fine-grained to aphanitic dikes a few meters across cut the batholith at several places. Some of them are extremely altered flinty rocks carrying considerable pyrite and other alteration products. A section of one from a locality north of Les Perches shows a microcrystalline growtl1 of secondary amphibole, chlorite, and magnetite, probably some quartz, and a few nests of coarser grained plagioclase, chlorite, or hornblende. The rock shows evidence of considerable brecciation. A somewhat coarser grained dike of this type from the Riviere Marion near Acul Samedi contains about 60 per cent of hornblende in euhedral b11t somewhat fibrous prisms, 0.2 to 1.0 millimeter long, in a gro11ndma.ss of plagioclase and quartz. The hornblende, a blue-green variety, apparently has been somewhat recrystallized. Tl1e plagioclase is labradorite and commonly shows micrographic intergrowth with quartz or occurs in sheaf-like groupings. There is about 20 per cent of quartz in the rock. Magnetite, epidote, and chlorite are present. The rock is considerably

PAGE 334

' 296 GEOWGY OF THE REPUBLIC OF HAITI. richer in the ferromagnesian minerals than any of the quartz and is probably related to the lamprophyres. Dikes of porphyry and felsite at Morne Beckly are extremely altered. A. porphyritic dike containing phenocrysts of plagioclase and quartz has been largely recrystallized. The plagioclase was r e crystallized to albite with the separation of epidote. The original character of the ro c k can not be determined. Large veins of pure quartz cut the quartz diorite and the surrounding rocks. The mineral-bearing veins related to this intrusion are described on pages 459-468. Veins of hornblende and of hornblende and quartz, sharp or blurred, cut the batholith along shear zones and joints. A quartz hornblendite of this type from Morne Madeleine consists o f hornblende, 50 to 60 per cent; quartz, 25 to 35 per cent; magnetite, 5 per cent; and chlorite and epidote, 8 to 10 per c ent. The hornblende is massive and pleochroic in colors ranging from yellowish to brownish gre en. The quartz grains are full of dustlike inclusions and give wavy extinc tion. The rock contains some titanite, which is associated with the chlorite and epidote, and a few grains of zircon. These hornblendite veins, as shown by their mineral composition, are the later products of crystallization of the quartz cliorite magma and presumably comprise a differentiate that has been transported outward from the zone of c rystallizat i on by aqu e ous solutions. ALTERATION AND METAMORPHISM. Both the contact phases of the intrusive itself and the invaded country rocks for a considerable distance from the intrusive contact are intensely altered and metamorphosed. Enormous quantities of highly alkaline aqueous solutions must have escaped from the solidifying magma, in places even at considerable depths and pre ssures. Alteration in the contact phases of the intrusive itself consisted in some recrystallization of the hornblende, ac c ompanied by separation of titanite and iron oxides, and albitization and seri c itization of the plagio clase. The epidote and calcite in the plagioclase and the quartz and epi dote veinlets that cut the quartz diorite itself were probably formed during this period of alteration. Lo c ally pyrite was d e veloped abundantly in the contact porphyries and a c companies quartz and kaolinite along later shear veins. The quartz veins carrying copper sulphides and iron oxides are later phases of the intrusive activity and are described in greater detail on pages 459-468. The deep-seated metamorphism of the basic eruptive rocks next to the quartz diorite contacts was very intense and gene rally resulted in the development of amphibolites or of rocks consisting essentially of amphi boles aud varying amounts of albite. Epidote, quartz, chlorite, sericite, and iron oxides are other products of the metamorphism. Where intense

PAGE 335

REPUBLIC OF HAITI GEOLOG I CAL SUR\'EY l DACITE PORPHYRY FRO:\I SAVANFJ I...1A CIDRA, l\IONTAGNES NOIRES. PLATE XX B. GRANODIOillTE FROl\1 l\1EME VALLEY, CO:\i:\I UNE OF UVE.

PAGE 336

,._ .. ... \ -. , ... I '- .; i. "' J.

PAGE 337

IGNEOUS ROCKS 291 shearing action has accompanied the intrusion hornblende, chlorite, and talc schists have been formed. The lines of schistosity developed in some of the schists clearly antedate some of the later quartz veins of the intrusion, showing that intense deformation was in progress before the intrusive period and during its earlier part. Some pegmatite veins intruded along the schistosity have a pinch and swell structure. The most intensely metamorphosed rocks were generally fo11nd near the outcrops of quartz diorite, and where intrusive bodies were not seen the quartz or pegmatite veins and intrusive dikes indicate the presence of larger intrusive bodies near by or concealed beneath the surf ace. At greater distances from the contacts albitization is not generally ac companied by the change of the pyroxenes to amphiboles, although some epidote is formed. The formation of chlorite from augite took place at some localities. In a few andesitic rocks orthoclase had partly replaced secondary albite. Rarely the augite in some of the less intensely altered andesites or dacites was partly recrystallized to a granular aggregate of calcite, plagioclase, iron oxides, quartz. Prehnite was found near Grande-Riviere and Lamielle as an alteration product of plagioclase and filling cavities in much altered andesitic rocks. Zeolites are found occasionally under similar circumstances, and at one place they were found along joint planes in an amphibolitized rock. Prob ably these minerals repr esen t a very late stage in the intrusive period, when the temperature of the rocks and of the aqueous solutions had fallen considerably. Although the distance from the intrusive cont.acts to which the more intense metamorphism reached can not be estimated because of the possi bility of concealed or undiscovered intrusive bodies, partly amphibolitized and albitized rocks were found from 2 to 4 kilometers from any known surface exposures of the quartz diorite. Rocks in which albite had been formed but in which the augite was left relatively 11naltered were found from 6 to 7 kilometers from known exposures of the intrusive mass. Rocks found in place in the eastern part of the de Terre-Neuve, probably from 15 to 20 kilometers from any outcrop of quartz diorite of this age, show no alteration attributable to the intrusion. This area and small areas of volcanic rocks in the extreme western part of the North west Peninsula and possibly in the northwestern parts of the Montagnes Noires are the only ones in the northern part of the country where older ro c ks are 11naffected by this intrusion. WEATHERING. In exposures of the quartz diorite that are not subjected to active erosion weathering has decomposed the rock to considerable depths. Excellent exposures of a section through the residual soil were fo11nd at the

PAGE 338

298 GEOLOGY OF THE REPUBT.1IC OF HAITI. crest of the trail between Les Perches and Valliere on Morne Madeleine The upland at this locality stands about 600 meters above sea level. It iA rolling and bas a relief of about 100 meters. Re1nnants of flat surfaces possibly represent an older Tertiary erosion surface which had been partly peneplained. The surface of this upland is covered, except where erosion is active, with a thick mantle of clayey soil derived from the qua .rtz diori te bedrock The upper 4 or 5 meters of the soil has a mottled appearance and is of a reddish color. It consists essentially of a mixture of clay, earthy hematite and limonite, brown earthy manganese oxides, and undecom posed fragments of quartz. At a depth of a few meters the clay is spotted with white blotches of nearly pure kaolinite, some of which are stained a delicate pink, probably by manganese oxides. A sample of the kaolinite was picked and separated from included qua .rtz fragments and was an alyzed with the following results: Analysis of kaolinite from residual soil on quartz dwrite at M orne Madeleine .1 Si02 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44.88 Al20s ....................................................... 41.03 Fe201 . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Mll20s ..................................................... Trace. H20 below 110 C .......................................... 1.40 H20 above 110 C. . . . . . . . . . . . . . . . . . . . . . 12.70 The analysis as well as the optical properties indicate that the material probably is kaolinite, which is derived from the plagioclase by altera tion. The earthy bro"rn to black manganese oxides, which form veins and irregular pockets in the upper part of the residual clay, are derived from the hornblende by decomposition. The upper reddish clayey sojl grades down through the speckled clay to partly decomposed quartz diorite and to solid rock. The concentration of the manganese in the upper part of the soil is very noticeable but not in such quantities as to form com mercial deposits at any place. A description of a section through weathered quartz diorite gravel and sand containing large concretions of iron and manganese oxides is given on pages 477-478. Between Lamielle and Mont Organise a red claj'ey soil similar to that on Morne Madeleine is found on the weathered surfa .ce of the quartz diorite. The red clay is very conspicuous at Mont Organise. At the northern part of the North Plain, in the vicinity of Les Percl1es, small pockets of white clay, probably largely kaolinite, were seen along the slopes of the ravines. They probably were concentrated by water in favorable places during Pleistocene or Recent time. 1 Analyst, Earl V. Shannon, Un!ted States National for anaty8l!t examined optically and pronounced homogeneous by E. S. Larsen, of the United States Geological Survey.

PAGE 339

IGNEOUS ROCKS. 299 The kaoliniza ti on of the plagioclase is noticeable on weathered expo sures of the quartz diorite, particularly in porphyritic phases of the rock, where the phenocrysts are attacked first so as to give the rock a spotted appearance. AGE OF INTRUSION. The intrusion of quartz diorite probably occurred between the end of Lower Cretaceous and the beginning of middle Eocene time. Although the quartz diorite is not definitely known to intrude the Lower Creta ceous argillite series, the presence of large and small quartz veins cutting these argillites between Cerca-la-Source and Lamielle and the meta morphism of the argillites indicate that the quartz diorite is younger. In the Dominican Republic, along the edge of the pre-Tertiary basal com plex south of Sabanet.a and between Sabaneta and Restauraci6n, sericitic argillites that probably are of the same age as the argillites of the Repu b lic of Haiti are intruded by quartz diorite and dikes of granitic dioritic .. porphyry.1 Although the argillite south of Sabaneta is somewhat more metamorphosed than most of the Cretaceous ( ?) rocks in the Republic of Haiti, it resembles, according to the Dominican report: sheared slaty srgillites south of Restauraci6n that are without doubt a continuation of the band of Cretaceous ( ?) rocks extending southeastward from Lamielle. The relation of the quartz diorite to the late Upper Cretaceous lime stones found only in the region around Grande-Riviere du Nord is not known a relation that if determined would place the age of the intrusio:g within narrow limits. The quartz diorite porphyries in the Montagnes de Jean Rabel are over lain unconformably by the upper Eocene limestone, and pebbles of a porphyry presumably related to the quartz diorite in origin are found in the basal upper Eocene conglomerate on Morne du Cap. Dacites much altered by the intrusion are unconformably overlain by relatively 11nal tered Plaisance limestone (middle Eocene) at the northeast base of Mont Puilboreau. No effects of contact metamorphism have been found in any of the Tertiary limestones. Tbe quartz diorite is therefore probably of late Cretaceous age, as before Eocene time at least the outer shell of the batholith had been exposed by erosion near what is now the Montagnes de Jean Rabel. D.ACITE PORPHYRY OF 'rHE MONTAGNES NOIRES. D1smmuTION AND STRUCTURAL REIAATIONS. A dacite porphyry is exposed southwest of St.-Michel, in the intermon tane valleys of the northwestern part of the Montagnes Noires. This rock constitutes the greater part of the igneous rocks in the valley in Savane 1 A geological reconnaissance ot the Dominican Republic : Survey Mem., vol. 1, pp. 52, 53, 56, 1921. 2 Op. cit., pp. 52-53. Dominican Republic Geol.

PAGE 340

300 GEOLOGY OF THE REPUBLIC OF HAITI. la Cidra back of the first limestone range southwest of St.-Michel and part of the rocks flooring the second savanna, in which stands the village of Paul. These savannas are separated by limestone ranges. In Savane la Cidra the dacite porphyry is in contact with hornblende andesites and sedimentary rocks of supposed lower Cretaceous age. The Cretaceous ( ?) rocks are mildly contact metamorphosed near the porphyry, as shown by a development of epidote, chlorite, quartz, and in places cubes of pyrite. The andesites have been altered to greenish-gray rocks containing abundant secondary epidote, quartz, chlorite, and calcite, and in some specimens the plagioclase is altered to albite and sericite. In the dacite porphyry itself there is some recrystallization of the hornblende, sericitization of the plagioclase, and formation of a manganese epidote from the ferromagnesian minerals. At some places farther southwest in the savanna the dacite porphyry shows a well-developed flow structure and evidence of brecciation. There is a lack of definite structural evidence to demonstrate either the intrusive or extrusive origin of the porphyry, although its texture and the accompanying metamorphism indicate that part of the porphyry exposed belongs to minor intrusive bodies. As the Cretaceous ( ?) beds presumably were the surface formation in early Tertiary or late Cretaceous times, the intrusive bodies probably were rather close to the surface, which may acco11nt for the texture and widespread resorption of the hornblende in the porphyry. The porphyry may, however, have broken through the comparatively thin cover of Cretaceous ( ?) deposits in some places. PETROORAPHY. The dacite porphyry is a conspicuous purplish-gray to brown rock with large crystals of plagioclase, prisms of hornblende, grains of quartz, an
PAGE 341

REPUBLIC OF IT.1\lTI GEOLOGICAL SCR\'BY PLATE x .. r .. 1. JlTTO'I 'O)fTC'non R.\PIT OF QT":\R'"fZ DlORI':rE FRO:\I :\IOR:\TE 80'C' 1 '11 ()l"' I.iES !EI{( I), i>l<1gi(>c lil c, r>artl)" scricitizE>cl: Q. qt1 <11tz; II, h o rnbl'n<1 e ; C, cblorite. nico l s, X 10. ... Q Q R. OF DA CITE PORPIIYRY SA "AXE LA ClDR.\, NOIRES. Q, rtz; P, plagiocl11se: B, biotite: 1\-f, n1ngnetite; Pel, 1)iecl111ontite. Or
PAGE 342



PAGE 343

IGNEOUS ROCKS. 301 The mica is a deep-brown pleochroic variety, with high birefringence and unusually high refraction, and possibly contains some manganese (p = 1. 7 2V medium to small, a= straw yellow, y =deep brown). It is usually resorbed around the borders of flakes and is partly altered to a manganese epidote, piedmontite (a=yellow, y=deep rose, y/\001 cleavage= 32 ). The mica is in flakes 3 or 4 millimeters in diameter and in amounts su. bordinate to those of the other dark constituents. The groundmass consists of mi croli tes of oligoclase in slender and stout sec tions, and of roicrofelsitic intergrowths of a sodic plagioclase with quartz and probably orth9clase. Magnetite is scattered in the groundmass in small octahedral grains together with some small scales, probably of hematite. The apatite is in stout yellowish pleochroic prisms. Some sericite is developed along cracks in the plagioclase and some scattered piedmontite a .nd calcite occurs jn the groundmass. CORRELATION AND AGE OF THE PORPHYRY. The porphyry is in some features unlike the other igneous rocks in the Republic and can not be correlated definitely on its petrographic character. As it appears to have intrusive relations both to the Cretaceous rocks and to the hornblende andesites, and as it is older than the upper Eocene limestone, it is probably of the same or nearly the same age as the quartz diorite in the Massif du Nord. It may be a chilled intrusive phase of the quartz diorite batholith that was intruded into the rocks of the northern part of the Republic in Cretaceous time. The nearest known outcrops of quartz diorite are those south of Limbe and near Bahon, although intrusive bodies of considerable magnitude unquestionably under lie the metamorphosed volcanic rocks in Section Las Lomas, about 12 kilometers northeast of Savane La Cidra. The variation in the mineral composition of the dacite porphyry from the normal contact porphyries of the quartz diorite might be accounted for by the conditions of oxidation and of considerably reduced pressure and temperature in the Cretaceous ( ?) rocks near the surface. The con cretionary masses of iron and manganese oxides found on its surface are exactly similar, both in texture and in the relative proportions of iron and manganese, to those formed by the weathering of quartz diorite debris near Acul Samedi. LATER QUARTZ DIORITE GROUP OF THE MONTAGNES DE TERRE-NEUVE DISTRIBUTION AND STRUCTURAL RELATIONS. Rocks of the later quartz diorite group, so far as known, are confined to a narrow zone, about 11 kilometers in length, in the Montagnes de Terre-N euve. They consist of a series of small elongated stocks and a few accompanying dikes and pegmatitic veins. At some places irregular

PAGE 344

302 GEOLOGY OF THE REPUBLIC OF HAITI. or laccolith-like bodies appear to have spread out laterally from areas near the apex of the main intrusive masses. These rocks were intruded, probably during Miocene time, through the older volcanic rocks and into the upper Eocene limestone. The exposed rocks of this group are predominantly chilled contact porphyries, although there are a few small bodies and dikes of coarser grained quartz diorite and granodiorite. Their distribution and structural relations and the associated ore deposits are considered more fully in the description of the Terre-N euve district. (See pp. 425-459 and II.) PETROGBAPHY AND CHEMICAL COMPOSITION. Quartz diorite. The quartz diorite phase of the intrusive at Hilaire, in the Comm11ne of Terre-Neuve, is a gray rock of medi11m-grained texture in which the dark minerals are usually subordinate in amount, al though hornblende may be abundant locally. It is similar in appearance to the Cretaceous quartz diorite shown in Plate XIX, A, except that it is of somewhat finer grain. Prisms of plagioclase and hornblende, grains of quartz, and in some rocks flakes of biotite can be recognized mega scopicall y. In thin section the rock has a somewhat porphyritic texture, containing euhedral to subhedral prisms of plagioclase and subhedral prisms of hornblende in a matrix of smaller grains of quartz and some orthocla.se of later crystallization. Much of the hornblende may also be interstitial with the quartz, depending on its relative abundance. The plagioclase is in zoned subhedral prisms, generally with albite and Carlsbad twinning, from 2 to 4 millimeters in length. Extinction angles on combined twins and on the 010 zone show a composition ranging from andesine (Ab&1 Ans) to oligoclase ( Ab70 An80) Rims of even more sodic oligoclase may be present on the more zonal crystals, so that the average plagioclase is probably near Ab65 An85 The plagioclase for1ns from 45 to 50 per cent of the rock volume. Quartz is in formless interstitial grains between the prisms of plagioclase and hornblende. It is somewhat fractured and has a slight wavy extinction. The amo11nt of quartz is rather high, forming 20 to 30 per cent of the rock. Orthoclase, in general very subordinate, is fo11nd interstitial like the quartz but may forrn irregular plates plastered on the outside of prisms of plagioclase. Orthoclase generally appears to form less than 5 per cent of the rock in this facies of the intrusive but is sufficient in amount to differentiate the rock from the Cretaceous quartz diorites, most of which contain no orthoclase. Hornblende, which at some places forms 15 to 20 per cent of the rock, is pleochroic from yellow-green to olive-green, and some has a shreddy or uralitic appearance. The largest prisms of hornblende indent only the borders of the plagioclase, as they began to crystallize somewhat later than the calcic plagioclase. The hornblende is quite idiomorphic toward quartz. Diopside or is found only in subordinate amounts in some parts of

PAGE 345

IGNEOUS ROCKS. 303 the rock, chiefly in rather small prisms and granules interstitial to the plagioclase. A small amount of brown biotite was noted in one section, associated with somewhat larger amounts of orthoclase, so that the rock apparently approaches a granodiorite in composition. The biotite was altering to chlorite. Magnetite, titanite, and apatite are the principal accessory minerals. Titanite is especially abundant in some rocks, owing in part probably to decomposition of an earlier titanium:bearing horn blende. Chlorite, calcite, and quartz are secondary minerals. Fibrous or uralitic-appearing hornblende may have been for1ned by the recrystalliza tion of a more basic hornblende and is usually accompanied by grains of ti tani te and magnetite. Granodiorite. Rocks of this type were fo11nd only in Meme Valley as dikes or small intrusive bodies cutting the fine-grained porphyries. '!,hey are gray or slightly greenish rocks of medium grain, speckled with flakes of dark-brown biotite and grains of pyroxene or hornblende. (See Plate XX, B.) Plagioclase feldspar and quartz can be recognized. Close inspec tion shows that in some specimens the biotite occurs in poikilitic flakes enclosing other minerals. In a thin section of a dike rock from a locality just south of the R-Oche Glisse workings (Plate XXII, A) the plagioclase is in euhedral to sub hedral prisms between which are irregular grains of quartz, orthoclase, and biotite. The prisms of plagioclase are from 0.2 to 2 millimeters in length but average about 1 millimeter. The plagioclase is zonal, ranging from andesine, and in some crystals from sodic labradorite (Ah.cs Ancscs), to oligoclase, and probably has an average composition of sodic andesine or oligoclase ( Ab76 An25 to Ab70 An80). Orthoclase occurs in irregular interstitial grains, some of them as much as 1 or 2 millimeters in diameter, generally untwinned. Many of the larger grains are poikilitic and inclose a few small prisms of plagioclase. The quartz is in irregular grains, some showing undulatory extinction. The brown biotite is in entirely irregular flakes which are interstitial to or inclose minerals of earlier crystalliza tion, such as plagioclase, pyroxene, and iron ores. The flakes range from small specks up to pieces 2 or 3 millimeters in larger diameter. The pyroxene, probably diopside or some nonal11minous variety, is colorless to pale greenish in section and forms rather stout prismatic grains with ragged outline that very commonly inclose numerous grains of iron ore and small prisms of plagioclase. The angle Z /\ c is about 42. A light green hornblende is intergrow11 with or has replaced some crystals of diopside, forming an imperfect network and giving the crystals a mottled appearance. Small grains of massive hornblende also occur which are not associated with the diopside. Magnetite, the commonest accessory, occurs in irregular or octahedral grains. 'l,i tani te, epidote, and apatite are other minor accessories. A chemical analysis of this rock and averages of some analyses of grano diorite and quartz diorite are given in the following table:

PAGE 346

304 GEOLOGY OF THE REPUBLIC OF HAITI. Chemical analysis of granodiorite from Meme Valley, average analyses of grano diorite and quartz diorite, and limits of variation of granodiorite. 1 2 s Si0.1 59.37 65.10 59.47 59-68.5 A120a 13 .23 15.82 16.52 14-17 FE!20a 8.45 1 .64 2.63 1.5-2.25 FeO 4.25 2.66 4.11 1.50-4.5 MgO 3.76 2.17 8.75 1.0-2.5 Cao 5.63 4.66 6.24 S-6.5 Na20 3.69 8.82 2.98 2.5-4.5 K20 2.62 2.29 1.93 1 8.5 H20 + 1.08 1.09 1.39 H20-0.06 e e e e e e I e e e e 4 I I e I e I I e e I I I e I e I I I I Ti02 2.14 0.54 0.64 P:.iOtS 0.22 0.16 0.26 MnO 0.17 99.67 100.00 100.00 1. Granodiorite, Meme Valley, Republic of Haiti. H. S. Washington, analyst. 2. Average of 12 analyses of granodiorite. Daly R. A., Igneous rocks and their origin, p. 25, 1914. 3. Average of 20 analyses of quartz diorite. Daly, R. A., op. cit. p. 26, 1914. 4. Limits of variation of granodiorite. Lindgren, Waldemar, Granodiorite and other intermediate rocks: Am. Jour. Sci., 4th ser., vol. 9, p. 272, 1900. An approximate mode, determined by a combination of measurement by the Rosiwal method and by chemical calculation, and the norm are shown in the following table : Approximate mineral composition and norm of M eme granodiorite. Approximate mineral composition. Quart;z. .............................. 17 9 Ortboclase . . . . . . . . . . . . . . . 8. 9 Andesine (Abro All3() ) ................... 40. 7 Bioti'te ................................... 12. 7 Diopside . . . . . . . . . . . . . . . . 7 Hornblende . . . . . . . . . . . . . . . 4 Magnetite . . . . . . . . . . . . . . . . 3.? Ilmenite ................................. 2.1 Apatite . . . . . . . . . . . . . . . . . O 4 Titanite . . . . . . . . . . . . . . . . . 1 Norm. Quartz .................................. 13 .18 Ortboclase .............................. 15. 68 Albite .................................. 31.34 Anorthi te ............................... 11 48 Diopside ................................ 12 .18 Hypersthene . . . . . . . . . . . . . . 5. 87 Ilmenite ................................ 4.(J"[ Magnetite ............................... 4 99 Apatite . . . . . . . . . . . . . . . . O. 51 The rock is dacose (11.4'' .2''.''4). The rock is lower in silica and higher in iron and magnesium oxides than the average analysis of the granodiorites, the amounts of these oxides clearly corresponding more closely to those in the average quartz diorite. The alumina is low for rocks in the diorite or granodiorite fami lies, but in this respect the rock is similar to the Cretaceous quartz diorite from Mome Madeleine. (See p. 292.) The titani11m is rather high. The chemical calculations indicate that ferric iron probably replaces the alu mina t.o some extent in the biotite and hornblende, and the pyroxene

PAGE 347

REPUBLIC OF 1-IAITI GEOLOG I C A L S UR\. E Y Q. PLATE X XII 4 PFIOTO:.\II CROGRAPII OF GRAKODIOilITN FR0:\1 ::.\IE'ME VALLEY, 0 orthoclas c ; r, OF TERRE-NEU\'E. plngioc.:lase; Q, quartz ; B biotite ; D, cli o p i c l e ; C1oss ec1 nic o l s X 3 0 111agneti t e B PHOTOMICROGRAPH OF GLASSY PYROXENE ANDESITE WITH PERI,ITIC TEXTU R E FROM SOUTH SLOPE 01'"' iVIORNE D U:.\IU R \ ILLE, COl\1i\ilUNE OF '.rERRE-NEUVE P, plagioclnse; A at1gite ; H, hyperstbene; M, magnetite. Or
PAGE 348



PAGE 349

IGNEOUS ROCKS. 305 presumably is a variety of diopside. The relative amounts of the calcium, sodium, and potassi11m oxides correspond to those of Daly's average granodiorite, and the ratio of the orthoclase to the plagioclase in the mode brings the rock well within the limits of the ratio that is common in rocks of the granodiorite family. The low percentage of orthoclase and the high percentage of ferromagnesian minerals indicate, however, that the_ rock approaches a diori te in composition. Dacite porphyries. Contact porphyries are the most widespread type of this group. Although they differ in appearance, owing to alteration and variations in texture, they have the same general character under the microscope. The porphyries are light gray to dark gray or greenish gray, in places stained yellowish or brownish by weathering, and commonly much jointed. The plagioclase phenocrysts, prisms of hornblende, and grains of quartz can easily be recognized with the unaided eye or with a hand lens, unless alteration has been so intense as to have recrystallized the hornblende or altered the plagioclase. In thin section the plagioclase phenocrysts are in more or less euhedral zonal prisms, some of them fragmental, and from 1 to 3 millimeters in length. They are generally andesine, ranging in composition from Abao An40 to Absrs An35, although phenocrysts as calcic as labradorite (Ab4o Anso to AbGo An6o) are found in some specimens. These phenocrysts constitute from 30 to 50 per cent of the rock volume. The quartz is in rounded bipyramidal crystals, some of them fractured, the largest 1 milli meter in diameter. Quartz phenocrysts are generally rather sparingly present, for1ning from a small percentage to about 10 per cent of the rock, most of the quartz being in the groundmass. The hornblende is olive green or blue-green and may be partly resorbed to magnetite or recrystal lized to a shreddy uralitic variety. The hornblende alters to chlorite, iron oxides, and calcite. A colorless augite is sparingly present in some sections, generally surrounded by a granular aggregate of albite, chlorite, titanite, and other secondary minerals. The groundmass in most speci mens is a fine-grained intergrowth of sodic andesine or oligoclase with qua .rtz and probably some potash feldspar. Magnetite and apatite are accessory minerals. The groundmass of some of the specimens contains abundant microlites of a uralitic hornblende, apparently of secondary origin. Sericite, chlorite, calcite, and quartz are secondary minerals. Titanite is generally present in traces and may be abundant where altera tion has decomposed some of the f erromagnesian minerals. A more com plete description of the alteration of these rocks is given in the descrip tion of the ore deposits of the Terre-Neuve district (pp. 439-440). Dike rocks. A small number of dikes cut the older andesitic lavas out side of the main intrusive zone in the Terre-Neuve region. A large prominent dike of this type at Rocher, southeast of Terre-Neuve village, is the country rock of some of the copper veins. The rock is a gray porphyry 20

PAGE 350

306 GEOLOGY OF THE REPUBLIC OF HAITI. containing white plagioclase phenocrysts 2 to 3 millimeters in length, some hornblende prisms, and a few grains of quartz in a dark-gray aphanitic groundmass. Except for a few flakes of biotite partly altered to chlorite the rock is much like the contact porphyries. The plagioclase phenocrysts are sodic andesine with borders of oligoclase. The ground mass consists of oligoclase, some quartz, and potash feldspar. Aplites and pegmatites. Aplitic and pegmatitic facies of the intrusive are not abundant. Pegmatite veins consisting of untwinned orthoclase plagioclase, quartz, and biotite and some pyrite and chalcopyrite were seen in Meme Valley. An aplitic rock of rather unusual type, consistjng largely of albite, a little quartz, and spherulitic or radial growths of mus covite, was found near Jeanty Ravine. Some chlorite, titanite, and mag netite are present in the rock. RELATIONS OF THE VARIOUS TYPES. The dacite porphyries make up the larger part of the intrusive masses and generally were the first rocks intruded. They represent chilled magma that first came into contact with the relatively cold volcanic rocks and limestones. The porphyries in Meme Valley have been cut by dikes and irregular masses of the granodiorite. The relation of the quartz diorite at Hilaire to the porphyries is not known. The intense alteration of the porphyries at some places and the general absence of this alteration in the coarser-grained rocks indicate that the coarser-grained rocks are intrusive into the chilled porphyries. Veins of pegmatite and aplite cut both the dacite porphyries and the granodiorite. The relative age of the surrounding dikes, such as the one at Rocher, is not known. This dike does not have chilled borders and may be one of the later intrusives. Al though no analyses of the altered porphyries are available they probably are more similar in composition to the quartz diorite at Hilaire than to the fresh granodiorite at Meme, which is a later differentiate, probably richer in potash and soda. METAMORPHIC ROCKS ENTIRELY OR IN P.ART OF IGNEOUS ORIGIN. DISTRIBUTION AND GENERAL FE.A.TURES. Metamorphic rocks largely of igneous origin are found at a number of places in the northern region. The most notable localities where these rocks were fo11nd during the reconnaissance are on the North Plain at Morne Beckly and west of Le Trou, near Grande-Riviere du Nord, west of Limbe, on Limbe Mountain, and in Section Las Lomas, northeast of St. ltiiichel. All the metamorphic igneous rocks at these localities appear to have been derived by intense dynamic and hydrothermal contact meta morphism from the earlier basalts and associated pyroxenites and perido tites;probably during the folding and batholithic intrusions of Cretaceous

PAGE 351

IGNEOUS ROCKS 307 time. The more intense phases of this metamorphism have resulted in the formation of chloritic, amphibolitic, talcose, and serpentinous rocks and in the production of lo cal zones of schists The following table gives a brief classification of the more notable types of metamorphic and igneous ro cks in the northern region. 'l,he more notable types of metamorphjc rocks are described in the following text. Classification of igneous and metamorphic rocks in the northern part of the Republic of Haiti. Original state of rocks. Present state of rocks. Age or relations. 1. Sediments. Impure quartz sandstone ( ?) Garnetiferous quartz-mica schist. Limestones . . . . . . M etamorphic limestone schists Probably Paleozoic. and marble. 2. Basaltic lavas and tuffs ...... Tuft s and calcareous tufts ( ?) .. 3. Basic and ultrabasic intru s1ves. Fine-grained or '' greenstones '' ; daloidal basalts. porphyritic and amyg-Uralitized and albitized basalts; chloritized basalts ; chlorite schists ; some amph ibolites ( ?) Chlorite schists; chloritic calc schists Diabases and gabbroa........ Uralitized and albitized dia base s ; rnetadiabases ; epido sites ; hornblende schists. Pyroxenites ( ?) . Amphib olites and hornblende schists. Peridotites Partially serpentinized perido tites ; talc schists; serpen tines.a. 4. Andesites and dacites ........ 5. Sediments. Clays tones, siltstones, sand stones, limestones. Largely nonmarine but partly ma rine. Andesites and dacites. Lo cally albitized and sericitized and otherwise slightly altered. Chloritic or sericitic argillites ; slaty shale s and sandstones; calcareous sand-.tones; crystal line limestones and marble. Locally schistose. Early or (Triassic middle Mesozoic ( ?) and Jurassic). Associated with and probably intrusive into the basaltic lavas and tuff s. Early or middle Mesozoic. Jurassic ( ?) possibly in part early Cretaceous. Early or middle Mesozoic. 6. Quartz diorite and associated Quartz diorite, etc., locally Late Cretaceous. rocks. sheared and fractured. 7. Basaltic and andesitic diabase, essexite. tized. lavas Zeoli-Practically same, except for local alteration. Late Cretaceous or early Eo cene. Exact age unknown. a. Large areas of serpentine were not found in the Republic of Haiti but are found in the Dominican Republic. (See p. 287.)

PAGE 352

308 GEOLOGY OF THE REPUBLIO OF HAITI. AM:PHIBOLITES AND HORNBLENDE SCHISTS. Rocks generally without schistose structure but in some places showing it and consisting predominantly of hornblende are found southwest of Limbe and on the North Plain at Morne Beckly and east of Le Trou. Practically all gradations from incipient amphibolitization of basaltic rocks to rocks that have been fully recr31stallized to amphiboles are found. Generally some chlorite, talc, or serpentine is fo11nd in the amphibolites, and iron ores are invariably present. Dark-gray amphibolites, which on weathered surfaces are conspicuously porphyritic, are found east of Le Trou along the road between Cap Ha!tien and Ouanaminthe. These rocks were originally coarsely porph.Y. ritic, containing many large crystals of diopside or augite from 1 to 5 millimeters in diameter. On freshly broken surfaces the rocks are dark green and the porphyritic texture is not so apparent. In thin section some of the rocks show remnants of unaltered pyroxene or less commonly of unaltered olivine. One rock was completely converted to pale amphibole with some chlorite and iron oxides, though the original porphyritic tex ture could be easily recognized. In another rock from the same locality and of similar appearance the phenocrysts, which were largely diopside, were altered to a pale scarcely pleochroic amphibole with some flaky ser pentine. Along fractures in some of the phenocrysi;s green isotropic grains, probably of pleonaste, have been deposited. A grayish-green hornblende schist from Morne Beckly is greasy to the touch and is spotted with dark-green patches of chlorite. Microscopic examination shows that it was derived by metamorphism from a coarse porphyritic basalt or pyroxenite probably similar to rocks east of Le Trou. The groundmass of the rock now consists largely of a very pale amphibole with considerable serpentine or chlori te and iron ores. Along seams i11 the rock the opaque iron ores and in some places reddish-brown trans parent grains, probably of pictotite or chromite: have been deposited. The phenocrysts, which resemble in outline large crystals of pyroxene and olivine ( ?) are completely altered to serpentine and interlacing needles of actinolitic hornblende, with some chlorite. The ultrabasic rocks from which these serpentinous amphibolites were derived probably are of the sa. me age as the basic rocks from which the nickel-bearing serpentines of the Dominican Republic and of Cuba were derived. It is noteworthy that no areas of serpentine rocks were fo11nd in the Republic of Haiti. Amphibole is a major constituent and serpentine is a minor constituent of most of the altered rocks. An augite peridotite found in Section Las Lomas showed only incipient serpentinization (see p. 288). Some amphibolites and hornblende schists at Morne Beckly are com posed of greenish-black to gray hornblende in thin fragments that are pleochroic from colorless or yellowish to blue-green (a=l.62.003,,8= 1.635.003, y=l.645.003). The hornblende is arranged in parallel,

PAGE 353

IGNEOUS ROCKS. 309 radial, or confused aggregates and in some rocks in b11ndles of acicular crystals 5 to 6 centimeters in length. Between the dark crystals of horn blende or along shear planes some yellowish or brownish asbestiform acti nolite is found in places. The complete recrystallization of these rocks obscures their texture and origin. Probably, however, they represent only the final stage in the amphibolitization of rocks such as those just described. TAI,Q SCHIST. A relatively pure talc schist derived through metamorphism from an enstatite-bearing rock was found on Morne Beckly. About 15 per cent of the rock consists of remnants of a colorless orthorhombic pyroxene, and the rest is made up of talc and iron oxides. The rock appears to have been brecciated or sheared. Talc and some stringers of iron ores replace the primary constituents, starting from the fracture lines. Enstatite is the only primary mineral preserved in the rock. CHLORITIO SCHISTS. Chlorite schists occur as country rock associated with the metalliferous veins in Section Cormiers, west of Grande-Riviere du Nord, and in Sec tion Las Lomas, northeast of St. Michel. These light-greenish schists are spotted with dark flakes of chlorite and consist predominantly of calcite and chlinochore, with some iron ores and pyrite. Quartz, chalcopyrite, or titanite may be present. The schists may have once been predominantly chlorite and clearly show replacement of the chlinochore by later calcite. Although the exact origin of these calcitic chlorite schists is \1nknown, they may have been formed by chloritization of basaltic rocks along zones of hydrotherinal activity, and the resulting chlorite rocks, during sub sequent periods of mineralization, were recrystallized and partly replaced by calcite and quartz carrying sulphides. Almost complete chloritization of amygdaloidal basalts was observed in lavas collected at habitation La Selle, in Section Cormiers north of the locality where the chlorite schists were fo11nd. This type of schist, so far as known at present, is confined to mineralized shear zones. A different type of chloritic schist is fo11nd on Morne Beckly. This is a reddish-brown distinctly schistose rock spotted with deep-green flakes of chlorite. In thin section the rock is seen to be of fragmental origin and is probably a metamorphosed tu:ff. The fragments are angular and of varying size and are contained in a clouded base, consisting largely of calcite with scattered flakes of chlorite and some recrystallized iron oxides. The fragments appear to be largely of igneous origin and are altered to pale amphibole, tale, or chlorite. Some fragments consist of calcite and may possibly be fragments of limestones.

PAGE 354

310 GEOLOGY OF THE REPUBLIC OF HAITI. SUMMARY OF THE RELATIONS OF THE EXTRUSIVE AND INTRUSIVE SERIES. Briefly summarized, the igneous rocks of the northern region comprise a n older complex series of volcanic rocks and a younger group of batho lithic rocks that have intruded the volcanic rocks. The old basement on which the volcanic series rests has largely been engulfed in the quartz diorite batholjth in the central part of the Massif du Nord, although patches of it are probably preserved on the North Plain and on Tortue Island north of the main axis of intrusion. The structural relations within the volcanic series are only imperfectly lrnown. In general, there appears to be an older basaltic series and a younger andes itic series, but it is doubtful if a complete section including both series is to be fo11nd at all places. The volcanic rocks contain some old intrusive igneous bodies of basic or ultrabasic composition that are associated with and presumably intrude the oldest basal tic rocks A blanket of Cretaceous sedimentary rocks overlaps the surface of the younger volcanic rocks. In middle or late Cretaceous time, when these sediments had attained considerable thickness, all the rocks were intensely folded and were intruded by the batholith of quartz diorite. Among the latest lavas of late Cretaceous or early Eocene age are some basaltic and andesitic rocks, in part of alkaline composition. Their rela tions to the quartz diorite are unknown. Later folding, possibly in Miocene t i me, was accompanied by minor stocks of quartz diorite and granodiorite that intruded the older volcanic rocks and the Tertiary limestones in the Montagnes de Terre-N euve. The sections given in Figure 18 are somewhat generalized, but they show diagrammatically the relations of some of the igneous and sedimentary rocks. The minor intrusions south of Limbe (in section 0) lie along the crest of an anticlinal arch, the core of which consists of the older basaltic rocks, and the south flank and probably also the north flank are composed of the younger andesitic lavas. Figures 27 and 28, on pages 442-444, show in sections the structural relations of the post-Eocene intrusives in the Montagnes de Terre-Neuve. Presumably the series of rocks ranging from the early basaltic lavas in Jurassic ( ?) or earlier Mesozoic time through the andesitic and dacitic lavas to the batholitbic intrusions of quartz diorite of late Cretaceous age comprise a single cycle of igneous activity. The relation of the later Miocene ( ?) quartz cliorite group to the earlier cycle of activity is not apparent. The similarity of the chemical analyses of the Cretaceous quartz diorite and the post-upper Eocene granodiorite is very marked and surely indicates that these rocks were derived from a similar parent magma, though probably during separate cycles of igneous activity. (See pp. 292 and 304.) The later basaltic and alkaline andesitic rocks of the vicinity of Ennery and of the Montagnes de Terre-N euve appear to have been erupted along

PAGE 355

Metz-es 1000 iaMer s Metres 1000 'Q Ba tto urt.l s v 'a t_)Q v @ l@ -, _,, V x J( x-IC JC T JC IC IC X )CIC IC JC .I( K "l( llC >c IC)( I( J( I()( .!_ "1 I l; \ i<; I XX Kl( IC K >c IC X >C >t IC l( )( l( >c IC J( )( l( )( X IC J( J( )()(I( I( XI(>()( JC J(X k IC' J( .J 1.-, 'Yt !,-x JCICK>cJCK)CIC >c X I( ')( l( I( )( )( )( )( JC )t 0 Qal .__....., __ ......,.._. __ __.. ___ ....., ___ -4:Kilometres N q, N Qe.1 . mc . u:,m;a;t!i'ma!t&.i:.. =t -v 1:> Qal. II( JC )f )( )( )( ' . L X x X X X X X M X X I I o I 1'."-'1 ,._,_, , / :\...-' M >< X X X X X X X X >< I -\'I / ,' ,,-X X X X X X )( >c '1 .+i +-t:"'f .;."t:t, : + "'.', 1 1" ..... :-' ... .., ,. -.-\ .. .... ., a.,...,., ".-t-, ,_., I ,...,.. v ,.. y \I ,._-' t. I ., ,_,, '!:II t-,,'i ... -\ ? I r \I"' I" I ., >c : X X ,.: . ; / \ / .; \ ; L "\ .-,,"y )C )( "f' )C >t )( K )( )C / I / "" ;. :J\ ' :\ \ \, v,. I'\ \ 1 x '-=,c(! X )( >< X X )C >c I I I ;" ..-I ,, \ ""'"' J< X er >t X >t x X >C : !. : :.><" ... \ I I -\ ..,. \ \ \ ... I JI L >< 4 ? X X >< >t "' >< >c .A,..;:.:, I I / ,. -\ .- .... ..--..-.. --.-....... i l : ... ,:: :.: : .: l ::: = :. .. .. . "........... .. .. . . . . . . . : ... :t ... .. , ..... . . .. ..... ... .- ..... . ... ..... ,. . .......... ::. .. .. :: ,\,.; ....... ,. ........ ........ .,,,;\;oXl) o o o 0 o I Alluvions 1 r dq )t?""'".r.. v ....... &,. ',,. ; 'f' y . r Tertiaire Mesozoiqu.e .1_ -.... \ r. V I \ , +"+" +'.+:+ K )C llC K .IC / \T I --1..l I .. ''-;'1V.' 1 os 1 r-r-4 es-r-i ...,y. 1 : ... + JC .J( 1 ... J1't 1 1 .. .-,r-:=-. +.. . ..... II( JC )( ...... \' : :.-: CaJ.caire Ca.J.ca:ire CeJc.oire Diorit.e .,.. .. ne Pla15a;nce, nli:J!' , ouCz.eta..ee et. roche.s Eocene .L.,;;;: qua:r;t-z.ifere 1nferi.ear I superieur super1et1r mayen l.nconnu oumoyen ... 1 FIGURE 18. Generalized sections showing the relations of the igneous and sedimentary rocks in the Massif du Nord. A, Section from Cerca-la-Source to Ouanamlnthe. B, Section from Cerca-Cabajal to the vicinity ot Morne Beckly near Limonade. a, Section from the d'Ennery between Ennery and Poteau through Plaisance Valley and LimM to the north coast ...... 0 z 0 t"tl pj 0 a m

PAGE 356

312 GEOLOGY OF THE B.EPUELIC OF HAITI. a late Cretaceous ( ?) structural trough south of the main axis of the Cre taceous batholith. They probably are relatively minor eruptions belonging to the later stages of the Mesozoic cycle of activity. CENTRAL REGION. GENERAL FEATURES AND DISTRIBUTION OF IGNEOUS ROCKS. In the central part of the Republic there were at least three periods of igneous eruptions Cretaceous or Eocene, middle or upper Oligocene, and Miocene. Basaltic lavas and tuffs representing the earliest of these periods of eruption were definitely recognized at only two localities, one south west of Las Cahobas and the other in the central part of the Cha1ne des Mateux. Nephelite basalts of Oligocene age were found northeast of Thomazeau, and basalts of the same composition near Saut d'Eau may be of the same age or possibly younger. Miocene basaltic rocks were found in the Miocene sedimentary rocks just north of the Cul-de-Sac Plain and north of l' Arcahaie. So far as known, all the igneous rocks are extrusive lavas or tuffs"'1and no indications of intrusive activity were recognized in any part of the central region. Fragments of quartz and of yellowish iron-rich epidote in some of the Miocene sedimentary beds just south of St.-Marc probably were derived from the central parts of the Montagnes Noires, where these minerals are found in and associated with the intrusive rocks. PRE-TERTIARY OR LOWER EOCENE BASALTIC ROCKS. Basaltic lavas of pre-Tertiary or lower Eocene age are exposed about 4 kilometers southwest of Las Cahobas, where the road to Mirebalais crosses the gap in the Montagnes Noires. The mo11ntains are anticlinal, and the ba-aalts underlie the upper Eocene limestone near the center of the gap. (See Fig. 19.) Basaltic volcanic rocks crop out in small areas south and east of Couyau, near the crest of the anticlinal arch of the Chaille des Mateux. They prob ably lie at the base of the upper Eocene limestone or underlie it. Most of the outcrops are deeply weathered and have a reddish or yellowish iron-stained soil. At one exposure a basaltic tuff composed largely of altered basaltic glass and containing small Foramjnifera is interbedded in the upper Eocene limestone. This material may represent reworked tuff in the basal parts of the upper Eocene limestones. Exposures of the basal parts of the Eocene limestone and any underlying or interbedded volcanic rocks may be found in the unexplored deep valley north of Couyau that drains westward to Mont Rouis. PETROGRAPHY. Hypersthene basalts. Two specimens collected southwest of Las Caho bas are hypersthene basalts.

PAGE 357

s.s.o MoRNE DuPRE PurNE OU -Tos Niveaade .. la.Mer .... . ..... ..L. 7 Too i 7 dou.blee s MoRNE MacHEL N.N.E :! ,.g laNer F1ounE 19. Section across the Montagnes du Trou d' Eau and the Montagnes Noires from Maneville to Las Cahobas, showing the relations of the middle Eocene basalts in the Montagnes Noires and the supposed relations of the nephelite basalts to the middle and upper Oligocene limestones in the Montagnes du Trou d'Eau. Qal, Quaternary alluvium; Tm. Miocene; Tos, upper Oligocene limestone; Tbn, bedded lavas, agglomerates and tutfs, principally nephelite basalts; Tom, middle Oligocene limestone; Tes, upper Eocene limestone; b, early Tertiary or Mesozoic basaltic lavas . ..... Q z ltj 0 00 t:d 0 a 00 Clo)

PAGE 358

314 GEOLOGY OF THE REPUBLIC OF HAITI. The rock has a sheet-like jointing. One of the specimens is a dark-gray to black highly vesicular lava, with vesicles 3 to 4 millimeters in diameter. Some of the vesicles are lined with altered glass or chloritic substances and opal. In thin section the lava is seen to contain prisms about 1 milli meter in length of a n e arly colorless orthorhombic pyroxene. The fine groundmass shows a flow structure and consists of thin prisms of labradori te in a base of brown partly altered glass. The hypersthene phenocrysts may be in clusters of as many as six or seven crystals. Small hypersthene microlites are present in the glassy base. There are a few scattered prisms of plagioclase about 0.2 millimeter in length, and a few greenish to brownish grains of augite of about the same size. Fine specks of magnetite are scattered in the groundmass. The second specimen is a greenish-gray rock of aphanitic texture. The hypersthene phenocrysts are altered to greenish serpentine and a mineral of low refraction ( n= 1.47), poss j bly some form of silica. There are a few grains of unaltered augite. The groundmass of this rock is similar in texture to the other lava except for a smaller proportion of glassy base. Tuff. No lavas suitable for petrographic study were collected in the Cha.lne des Mateux. The tuffaceous rock east of Couyau interbedded in the upper Eocene limestone is a light-brownish rock resembling a much weathered porph3rry. It consists of fragments of a brown fine-vesjcular altered glass or palagonite in a matrix of calcareous material. The original shapes of the glass fragments have been destroyed, either by alteration or transportation. NEPHELITB BAS.ALTS. DISTRIBUTION AND STRUCTURAL RELATIONS. Near Saut d'Eaw.. Southwest of Saut d'Eau (Ville Bonheur) the nepbelite basalts rest on an eroded surface of limestone of middle Oligo cene age. Along the trail from Mirebalais to Saut d'Eau, 3 k i lometers southeast of Saut d'Eau, the basalts rest on soft, white marly beds that are considered a marly facies of Oligocene or Eocene limestone, which is also fo11nd on the hill west of Saut d'Eau. The relations of the basalts to the Miocene beds of the Artibonite Valley are not known. The basalt floors the <;>pen grass-covered savanna southwest of Saut d'Eau called Savane Madame Michel. Between this savanna and the Savane Madame Michaud, to the southwest, the trail to Fond-des-Orangers crosses low hills, where the underlying middle Oligocene limestone crops out. Northwest of the trail, just beyond the divide on Savane Maaame Michaud, a more or less conical dissected hill rises about 200 meters above the savanna. The southeast slopes consist of bedded volcanic material and some vesicular flows. The bedded material is in thin, even beds, which in some places dip 20 to 30 southwestward. It consists of small fragments, some of which are partly ro11nded, and a few fragments of vesicular lava larger than a man's head. The general character of the material and

PAGE 359

IGNEOUS ROCKS. 315 bedding indicate that it is partly reworked volcanic tuff and agglomerates, although their source was not d i scovered. (See Pl. XVIII, B, p. 280.) They probably overlie the flows to the east and south. The basaltic lavas extend southward into the valley called Fond-des-Orangers, where the underlying limestone is of uppe r Eocene age. Near Thomazeau. Nephelite basalts of the same character as those found at Saut d'Eau are exposed at the foot of the mountains nortl1east of Thomazeau and north of Maneville. On the trail from Thomazeau to Cormillon, at the foot of the mountains, conglomeratic beds near the base of the overlying upper Oligocene limestone contain pebbles of basalt and of a limestone, presumably of upper Eocene age. All the igneous pebbles in this conglomerate consist of altered nephelite basalt containing abun ... dant crysta ,ls of olivine and generally are well rounded, as are the few pebbles of limestone. The igneous pebbles may be reworked tuffaceous and agglomeratic material. About 4 kilometers northeast of Thomazeau there are exposures of material that appears to consist of reworked ag glomerates and tu:ffs, loosely cemented and dipping as much as 15 to 18. Along the trail from Cornillon to Marche Canard, on the north slope of the mountains, about 1 or 2 kilometers beyond their crest, weathered basalt underlies the upper Oligocene limestone. This basalt pres11mably is the same as that exposed farther southwest, near Thomazeau, although no specimens were examined petrographically. (See Fig. 19, p. 313.) PETROGRAPHY AND CHEMICAL COMPOSITION. The massive basalts of this series are dark-gray rocks when unaltered, although some are brownish. Extremely vesicular or scoriaceous lavas may be brown, coated with yellowish alteration products. The weathered surfaces of the gray rocks are bleached to a rusty brown, and vesicular lavas may be deeply pitted. The lavas from Saut d'Eau and Maneville are of exactly the same appearance and mineral composition, and in thin sections they are practically indistinguishable. N ephelite basalt. A typical unaltered rock from a locality about 2 kilometers northeast of Maneville is dark gray and not noticeably porphyritic but contains small phenocrysts, the largest about 1 millimeter in length, of glassy olivine and of augite in a dark-gray dense groundmass. In thin section the texture is porphyritic with phenocrysts of olivine and augite in a fine-grained holocrystalline groundmass of augite, nephelite, and zeolites. (See Pl. XXIII, B.) The unaltered olivine pheno crysts are rounded to subhedral crystals of characteristic outline or are straight cross-fractured prisms, partly in clusters. A yellowish to brown ish-yellow augite is less common as phenocrysts. It usually occurs in long, thin, cross-fractured prisms, which make up about 5 to 10 per cent of the rock. The augite is zonal, with extinction on 010 (..L /3), ranging from 42 to 52 on the outside, pleochroic (a=l.'l'O, y==l.72+ ). The augite

PAGE 360

316 GEOLOGY OF THE REPUBLIC OF HAITI. in the gro11ndmass is in small prisms and microlites. Titaniferous magnetite and probably ilmenite are the principal accessory minerals and form irregular grains and clustered aggregates. Both apatite in small prisms and some melilite are scattered in the gro11ndmass in minor quanti ties. The nephelite forms irregular grains except near borders of the miarolitic cavities, where it forms 8tout hexagonal pris ms that are sharply idiomorphic against the surro11nding later crystallized zeolites. (See Pl. XXIII, B.) The nephelite is perfectly clear and unaltered. The zeolites scattered in the groundmass and :filling the cavities in the rock could not be definitely identified. As the nephelite is clear and unaltered in conta c t with the zeolites, some of the zeolites may be of primary origin. Small amo11nts of calcite and analcite, both probably of secondary origin, are present in the rock, generally in cavities with the zeolites. The mineral composition of the rock could be only roughly estimated because of the texture of the gro11ndmass and is about as follows: Olivine, 15 per cent; augite, 50 p e r cent; nephelite, 15 per cent; iron ores, melilite, and apatite, 10 to 12 per cent; zeolites, 8 to 10 per c ent. A chemical analysis of this ro c k, an analysis of a similar rock from Grenada, British Wes t Indies, and average analyses of nephelite and melilite-nephelite basalts for comparison are given in the following table: Analysis of nephelite basalt from Mane ville, an analy s is of a similar rock from Grenada, and ave rage analyses of nephe lite basalts. 1 2 8 Si02 38. 6 4 42. 83 39.87 37.56 Al20a 11.14 10.9'2 13.68 10.08 5.35 4.33 6.71 6.82 FeO 6.31 8.82 6.43 6.94 MgO 13.04 14.02 10.46 16.32 OaO 14.40 18.20 12.36 18.82 Na20 3.48 3.24 3.85 8.11 K20 1.90 .64 1.87 1.53 H20 + 8.01 1.80 2.22 2.52 H 20.29 Ti02 2.85 .05 1.5 0 2.66 P205 71 .89 .9, MnO .14 .12 .21 .06 100.21 100.86 1. Nephelite bas a lt, n ear Maneville Republic of Haiti. H. S. Wa shington, analyst. 2. ''Olivine basalt,'' Grenada, British West Indies. J. B. Harriso n, analyst. Rocks of Grenada, p. 10, 1896. Given by H. S. Washington in Chemi cal a n alyses of igneo u s rocks, U. S. Geol. Survey Prof. Paper 99, p. 709, 1917. 8. Average of 26 analyses of nephelite basalt. Daly, R. A., Igneous rocks and their origin, p. 88, 1914. Average of 6 analyses of melilite-nephelite basalt. Daly, R A., op. cit., p. 38.

PAGE 361

JtE 1' R L 1 C ()J;' HA I J I <;EI >I,< > G I l \ I 8 l R \' ) 4. l'II C ) l ():\I I< OF BARAI,T r l 'IIE :\IASRII1 LA SEf,f,F., XOl{' l 'If f)I l{l,? IJ; :Jll.: .. l'. D, J>s i1i\i11<' : .\, tit:1niff'rot1s D.llf?i1P: X. nephPlitP: :\I. tita11if11..;; 111:1gn<'tit< : Z Zl'<)li tc-s light, X 90.

PAGE 362



PAGE 363

IGNEOUS ROCKS 317 The norm of the Maneville nephelite basalt, calculated according to the quantitative classification, is as follows: N arm of M aneville nephelite basalt. Anorthite .............................. 9.48 Calcium orthosilicate .............. 7. 75 Leucite . . . . . . . . . . . . . . . . 8. 72 Magrietite ................................ 7. 78 N ephelite ............................... 15 70 Ilmenite ................................. 5. 40 Diopside ................................ 25 .10 Apatite ................................... 1. 71 0 Ii vine ................................. 15. 80 The rock is uvaldose ((III) IV. 2.3.2''.2.) In chemical composition the rock falls between Daly's nephelite and melilite-nephelite basalts, except that it is higher in calcium and slightly higher in titanium. The iron oxides are somewhat lower. The gradation toward melilite-nephelite basalt is also indicated by the presence of some calcium orthosilicate in the norm. Some of the basalts at this locality are notably higher in augite and iron ores than the rock analyzed. In a specimen collected about half a kilometer northeast of Maneville the groundmass is composed largely of augite prisms and microlites together with a smaller percentage of interstitial nephelite and zeolites. Besides the accessory minerals it con tains small brownish cloudy grains of an isotropic feldspathoid, probably haiiynite or some member of the sodalite group, and flakes of reddish brown biotite. The olivine in this rock is partly altered to serpentine and iron ores. Calcite accompanies the zeolites, both in the groundmass and in the cavities, and these minerals are probably to some extent secondary, although probably of late magmatic or deuteric origin: The augite prob ably comprises about 60-65 per cent Hauynite-nephelite basalt. A specimen collected about 2 kilometers northeast of Thomazeau contains 3 to 5 per cent of haiiynite as an acces sory mineral. In thin section it shows hexagonal cross sections about 0.2 millimeter in diameter, generally clouded with brown or gray dusty inclusions, some of them in sectors or symmetrically arranged. Some sec tions are elongated, forming hexagonal prisms, which are probably due to dodecahedral twinning. Reddish-brown biotite is also present in small flakes. Z eolitized nephelite basalt. A brownish-gray rock, collected about 4 kilometers south e ast of Saut d'Eau, is a nephelite basalt containing zeo lites. It contains scattered vesicles and amygdules and is speckled with brownish-yellow grains of olivine altering to serpentine and iron oxides. In thin section the olivine is seen to occur in euhedral to subhedral crys tals, some of which are marked by embayments of the gro11ndmass. The augite phenocrysts are brownish green and have the characteristic zonal and ''hourglass'' structure. The augite is partly in very long prisms whose ratio of length to width is more than 10 to 1. This rock differs from some of the basalts in the development of the nephelite, which may be in relatively large crystals that poikilitically inclose the augite and acces sories of the gro11ndmass, and in the presence of considerable analcite,

PAGE 364

318 GEOLOGY OF THE REPUBLIC OF HAITI. which forms part of the base and replaces the nephelite. The mineral composition of the rock is approximately as follows: Olivine, 10 per cent; augite, 60 per cent; nephelite, 7 ( ?) per cent; iron ores, 5 per cent; analcite, 10 ( ?) per cent; zeolites, 8 ( ?) per cent. Apatite is an acces sory mineral. M elilite-nephelite basalt. A large vesicular bomb or fragment from the bedded volcanic debris northwest of Savane Madame Michaud is a melilite-nephelite basalt. It differs from the other nephelite basalts only in the presence of a few small yellowish crystals of melilite, some of which have characteristic biconcave sections. RELATIONS AND ORIGIN OF 1HE LAV AS. In six sections studied from both localities and in specimens of con glomeratic volcanic debris that included fragments of lava, none of the basalts contained feldspar. The lavas from Thomazeau, from near Maneville, and from Saut d'Eau are remarkably siinilar in petrographic character. Accordingly, the lavas at these different localities must either have been derived from the same parent source of magma or the magmas must have had a similar genetic history. As the lavas overlie the middle Oligocene and Eocene limestones the magmas in reaching the surface must have passed through at least a thousand meters of limestone. The uniformity of the lavas and the failure to find more siliceous types of igneous rocks associated with them indicates that the magmas were fully formed before they reached the chambers or conduits of the volcanoes from which they w ere erupted. A more ex tended study might, however, reveal the presence of other types of lavas. AGE OF THE ERUPTIONS. As the basaltic rocks at Saut d' Eau and northeast of Thomazeau are of exactly the same petrographic character and cont.a.in similar associated beds of reworked volcanic debris the eruptions in the two localities prob ably occurred at the same time, although this has not been proved by studies of the structure. The basalts at Saut d'Eau are of post-middle Oligocene age, and those near Thomazeau are overlain by upper Oligo cene limestone. If the rocks at the two localities are of the same age the eruptions occurred in late middle or early upper Oligocene time. No evidence was fol1nd during the field work to indicate that the eruptive rocks were ever continuous between the two regions. In the San Juan Valley and the valley of Rio Yaque del Sur, in tl1e Dominican Republic, nearly flat-lying beds of limburgite capping mesas and upland areas of gravel are considered Pleiatocene.1 The fact that these limburgites are probably of similar composition to the nephelite basalts of the Republic of Haiti indicates that eruptions of lava of 1 A geological reconnaissance of the Dominican Republic : Dominican Rep. Geol. Survey Mem., vol. 1, p. 203, 1921.

PAGE 365

IGNEOUS ROOKS. 319 ultrabasic composition may have occurred intermittently over a con siderable part of later Tertiary time. The lavas at Saut d'Eau may therefore be younger than those near Maneville, but their relation to the Mio cene beds of the Artibonite Valley is not known. MIOCENE ( ?) BASALTIC ROCKS. GENERAL FEATURES AND STRUCTURAL RELATIONS. / Basaltic lavas and waterworn debris derived from them are interbedded in Miocene sediments on the so11th slope of the Cbaine des Mateux. About 4 or 5 kilometers north of I' Arcahaie, on the trail to Couyau, basalt is exposed in a ridge 50 to 60 meters in width and appears to be interbedded in the Mioc ene serie s. (See Fig. 5 and pp. 213-214.) Another exposure, which probably represents the sa .me bed, was seen along the same trail, about 2 or 3 kilometers farther north. Along the trail between Saut d'Eau and the Cul-de-Sac Plain, about 3 kilometers southeast of Source Morissel, conglomerate that consists probably of reworked basaltic agglomerates and tuffs is interbedded in the Miocene series. PETROGRAPHY. A fragment or cobble of dark-gi;ay amygdaloidal lava in the Miocene sediments south of Saut d'Eau contains amygdules lined with chlorite and calcite. In thin section this rock consists of abundant needles of labradorite and thin prisms and grains of augite in a somewhat altered base that contains chlorite. Greenish augite also forms phenocrysts from 0.2 to 1 millimeter in diameter, partly in clusters of several grains. The thin prisms of augite, especially in the gro11ndmass, have a brownish color and are zonal, giving an extinction Z /\ c ranging from 46 to 52. Magnetite is an accessory mineral in small grains. Some aggregates of a serpentinous or chloritic mineral may be after olivine, although their shapes are not distinctive. Chlorite occurs interstitially in the altered base and also lines some amygdules. Zeolites and calcite are also present in the amygdules. The augite, which forms 40 to 50 per cent of the base, is somewhat similar in character and development to the augite found in the nephelite basalts near Saut d'Eau and Maneville This basalt is further1nore similar in many respects to the amygdaloidal basalts of the later basaltic eruptions of the northern region. (See p. 284.) AGE AND REL.A.TIO NS OF THE LAV AS. Conclusive evidence that the volcanic debris in the Miocene beds near Source Morissel was derived from lavas of Miocene age was not obtained. Some of the eruptions, however, probably occurred in Miocene time, as is indica,ted by the bed of solid basalt in the Miocene north of l' Arcahaie. (See Fig. 5, p. 128.)

PAGE 366

320 GEOLOGY OF THE REPUBLIC OF HAITI. SOUTHERN REGION. GENERAL FEATURES. The igneous rocks of the southern part of the Republic are more uniform than those of the northern and central regions. So far as known, all are of extrusive origin or are parts of minor intrusive bodies associated with eruptions of lava. With the exception of small areas of andesitic rocks near Baraderes the lavas are predominantly basalts. 'fhe largest area of these basaltic rocks is in the de la Selle, where they are ex posed on the crest of the major anticlinal arch. Smaller areas are exposed farther west, where the overlying Tertiary limestones have been removed by erosion. With the possible exception of a minor occurrence of basalt that may be of Miocene age, a Cretaceous period of igneous activity is the only one that has been recognized in the southern region. The small zone of ande sitic rocks near Baraderes may not belong to the main period of basaltic eruptions, but as the andesites also underlie the upper Eocene limestone, the two periods of eruption probably were separated by only a relatively short interval. LATE CRETACEOUS BASAI,TIC ROCKS. DISTRIBUTION AND STRUCTURAL REI,ATIONS. Basaltic rocks, probably of late Cretaceous age, constitute most of the basement on which the Tertiary limestones were deposited in the southern part of the Republic. Their surface distribution is entirely dependen t on later erosion. At many places the basalts are exposed only in small patches beneath the cover of Tertiary sediments. A n11mber of the larger exposures are shown on the geologic map (Pl. I). Descriptions of the distribution and structural relations in the larger areas examined during the reconnaissance are given below. Massif de la Selle. In the Massif de la Selle the overlying upper Eocene limestone has been removed over an area from 500 to 600 square kilometers in extent. The major structure of the area is that of an eroded anticlinal arch, although this structure is modified by more or less exten sive faulting. The northern boundary of the area, near Furey, which is marked by a conspicuous scarp of limestone, appears to be a high-angle thrust fault. (See Fig. 20, A, p. 322.) At the southern bo11ndary, along the Grande Riviere de Jacmel, the contact is also marked by a fault, upturned basal beds of Eocene limestone being in contact with massive basalt. (See Fig. 20, B.) The faulting at this locality appears to be normal. 'fhe basal beds of the upper Eocene limestone are clearly exposed along the Grande Riviere de Jacmel below the boundary between the basalt and the limestone. Near Furey the upper part of the series is in contact with the basalt. The southeastern boundary of the area was not seen, except

PAGE 367

\ IGNEOUS ROCKS 321 from a distance, but it appears to be irregular, owing to high limestone ranges that jut into the area of basalt. At the southern boundary of the Leogane Plain, near the Riviere des Citronniers, basaltic agglomerate and breccia pass under the alluvial deposits. Farther east, along the Riviere Momance, the basaltic breccia and limestone seem to be in fault contact. A s1nall area of basalt is exposed on the south coast, on the west side of the small stream at Guillaumone. This basalt clearly underlies the upper Eocene limestone. An exposure on the slope leading down to the coast shows that the limestone was deposited on an irregular surface of the basalt. A larger area of basalt lies back of the plain at Cayes de J acmel. Impure tuffaceous limestones are interbedded with the basalts south of Etang J Bossier. There is no basis for the supposition that Etang Bossier is a era ter lake. Near the base of the basalts in the Massif de la Selle are interbedded impure tuffaceous limestones and shaly rocks of marine origin, giving evidence of contemporaneous volcanic activity during their deposition. (See p. 95.) Patches of older sheared metamorphic limestones are en gulfed in the basalts at several localities. (See p. 92.) Both the basalts and their associated tuffs and limestones are intricately folded. Except for thin local beds or lenses the basalts are remarkably free from interbedded tu:ffs or other pyroclastic debris, although in the valley of the Riviere des Citronniers there are exposures of coarse volcanic breccias and tuffs, w.hich appear to be of considerable extent and thickness. Whether these rocks are at the top or the bottom of the eruptive series was not definitely determined, although probably are near the top. The generalized sect i ons in Figure 20, A and B, show the relations of the basalts and the associated limestones in the Massif de la Selle. Etang de Mira.,qoane. Basalts are exposed in the depression occupied by the Etang de Miragoane. Along the road at the north side of the lake the basalts are overlain unconformably by basal beds of the upper Eocene limestone, consisting of fine conglomerates or sandstones and dark slates. The south side of the basin was not explored, but it consists of a straig:Qt and rather steep mountain front, the lower slopes of which may possibly be of volcanic rocks but the crest is composed of limestone. This mo11ntain wall may be flank of an eroded anticline, the crest of which passes through the center of the depression occupied by the Etang de Miragoane Vicinity of Asile Valley. Basaltic rocks associated in part with lime stones, probably of upper Cretaceous age, are exposed along the Grande Riviere des Nippes and the Riviere Serpent. The structural relations of these limestones to the basalts indicate that they were deposited at the time of the earlier eruptions. Pillow structure in basalt exposed north of the Grande-Riviere des Nippes on the trail to Anse-a-Veau shows that some of the :flows were 21

PAGE 368

Mveau, detaNsr '"a1 .s s .. .... Q.el b b :: I Io I .... . 1-DIC l;)Kc b Kc Te:i L'ASILETm ..... ... "Lil ......... __ _... Tes .,, .,, MOAN& MARDI Gw i .. .... I 0 ... b .,."' ,, bt ''""' ,,;1 ..... Tes r ;:;! ..... -'7 14'.: 9 ::::;::>.".C; """' s1 ., -., v ., ., Jlt "' v .,, ., MORN D' ENF'ER .. .,, ., .,, w .. ., .,, .,, .,, ...... = b '---., .,, .., s v ./ v ., ., .., .., ., y 1000-f Tes .. -b .. Tes .., ., ., ,,, .., ,'"r ,... / ',, 1'V,;t" ,,, / ... ,__ 17/ MoRNE N TRAN ... ., .., N v .,, '?' / ? \ .-M,veo.a "'t ., ., '-'"" \_ ...... dtl/,a.M'6r ., v ., v ., / $ 0 -Q.11at,exna:ire Tertieire Mesozoique JD ta t 10 atat , v v . : ... ... \ .... .... .......... .,: .. ,: ...... .. :: . . .. ., ..... ....... . . . . . .... ... .. ....... ......... .. ..,._. ,J,;0 ; 1 ; ; .. f , ; i ; ---r ., Q, v J --J;,,ffui:1 Couches Celca:ire et-Celtaires E::=Tm- :res ... .... ... Alluvions 0 loarne et eocene agglomerate dam calcsire; mioc.enes et tuf5 de C:ta.cA Cretad infneur Pliocene C1&.00 .. F1ounm 20. Generalized sections across the Southern Peninsula showing the relations of the basaltic lavas and the sedimentary rocks. A, Se ction from Aquln through the Asile Valley to Anse-A-Veau. B, Section from Jacmel to the Momance, southeast of the Uogane Plain. 0, Section from Cayes de Jacmel to Morne ".rranchant. Q tzj 0 s 0 t.!J =::d tJ;J Q td a 0 t:Q t-4

PAGE 369

IGNEOUS ROCKS. 323 submarine and included fossiliferous calcareous mud similar to that which forms the more massive limestones. (See Plate VIII, B, and p. 96.) At other places the lavas have overflowed or buried more indurated cal careous deposits. The contacts at some places are parallel to the bedding planes of the limestones. Figure 20, C, is a generalized section which shows diagrammatically the relations of the upper Cretaceous ( ?) sedi mentary rocks to the basalts. Vicinity of Aquin and St. Louis du Sud. Basalts crop out in some parts of the Aquin Plain and along its borders, where they are overlain by upper Eocene limestone. A few low hills between Aquin and Vieux Bourg consist largely of basalt, but their sides are strewn with limestone float, which probably comes from a cap of upper Eocene limestone. A good exposure in a road cut about 8 or 9 kilometers west of Aquin shows basalt overlain by brownish and greenish shales and impure limestone. Farther west along the coast, near St. Louis du Sud, the basalts form the lower parts of the mountain ranges, which are capped with massive lime stones, and also form many of the coastal hills. The lower alluvial valleys of some of the rivers, such as that of the Riviere du Mesle, are underlain by basalts, although the bed rock is concealed in places by river deposits. East of St. Louis du Sud many of the basalts show pillow structure and weather into ellipsoidal or round boulders, which strew the ground. Basa .It underlying the basal beds of the upper Eocene limestone is exposed on the trail from Cavaillon to St.-Louis, near the crest of the hill 6 or 7 kilo meters southeast of Cavaillon. Western part of peninsula. North of Port-a-Piment the basaltic rocks occupy an east-west troughlike valley that separates the limestone moun tains near the coast from the outskirts of the main range of the Massif de la Hotte to the north. From Les Anglais to Tiburon the low basalt ranges skirt the coast and a higher limestone range, which adjoins them to the north. Between Tiburon and Anse d' Ainhault basalt is the pre dominating rock of the foothills and low ridges along the coast, although here and there spurs of limestone extend to the coast. Limestone crops out on the crests of the higher ranges to the east. Sources Ohaudes. The interior valley at the Sources Chaudes is in basaltic rocks, which are apparently overlain along the northern border by basal shaly beds of the upper Eocene limestone. The hot springs issue from fissures in the basalt along the south side of the valley. PETROGRAPHY AND CHEMICAL COMPOSITION. Basalts. The normal basalts are dark-gray to black rocks, less com monly brownish gray or brown because of alteration. In texture they gen erally are aphanitic, but contain a few phenocrysts of plagioclase, augite, or olivine. Some of the rocks, particularly those in the central part of the peninsula, are a mygdaloidal, and pillow structure is common everywhere. The amygdules generally are filled with chl oritic minerals, chalcedony,

PAGE 370

324 GEOLOGY OF THE R EPUBLIO OF HAITI. calcite, or zeolites, and a few contain pyrite. The blocks of the pillow lavas vary from subangular to smoothly ellipsoidal in outline and range in diamete r from 8 or 10 centimeters to a meter or more. The spaces between the blocks may be occupied by secondary minerals or, less com monly, by sediments (Pl. VIII, B). Some of the more altered purplish brown lavas are amygdular and have a banding or eutaxitic structure. The black lavas weather at the surface to a brownish gray or rusty brown, and have rusty-brown films along the joint cracks. The pillow lavas weat her into round cobbles or boulders that may strew the ground. They disintegrate by spalling off parallel to the surface. In thin section the dark-gray aphanitic lavas generally are porphyritic, containing phenocrysts of basic plagioclase and less commonly of diopside or augite in a groundmass of intergranular texture, consisting of granules of augite interstitial to a plexus of thin plagioclase laths. (See Pl. XXIII, A, p. 316.) Many lavas contain a few phenocrysts of olivine. The plagioclase phenocrysts are euhedral to subhedral prisms, generally from 0.2 to 2.0 millimeters in length, and comprise from 5 to 10 per cent of the rock volume. They differ in composition in different lavas but are either bytownite or labradorite when :unaltered and are somewhat zonal . The more calcic plagioclase may have centers as calcic as Ab25 An715, but most of the phenocrysts are calcic labradorite ( Ab.0 Anao to An6cs). The plagioclase may be partly in clusters of prisms. A few smaller pheno crysts of diopside or augite may be present, generally associated with the clusters of plagioclase. Olivine is sparingly present as phenocrysts in many of the lavas but rarely comprises more than a small proportion of the rock. It generally is partly or completely altered to serpentine or iddingsite. Olivine or its alteration products comprise as much as 10 or 15 per cent of the rocks in only one or two specimens examined. Such rocks are olivine basalts. The plagioclase of the groundmass is in thin subhedral to euhedral prisms, consisting mostly of labradorite (Ab.0 An60). They range in length from 0.05 to 0.5 millimeter. In many of the basalts the prisms are interlacing. In specimens containing a high percentage of augite or a partly glassy base, they may form only a very incomplete network. The plagioclase, including the pbenocrysts, forms 30 to 45 per cent of the rock. The pyroxene, which is the predominating constituent in most of the rocks, generally appears to be a variety of diopside but varies in different rocks (a.=1.675, ,B=l.685, y=l.70-1.705). It is nearly colorless or slightly greenish to brownish, and except for a few scattered phenocrysts in some of the rocks is in small granules ( 0.05 millimeter) interstitial to the plagioclase. The diopside rarely occurs in grains large enough to inclose prisms of plagioclase. It forms 40 to 60 per cent of the rock. Magnetite, probably titaniferous, occurs in small grains interstitial to the plagioclase or as larger grains of earlier crystallization. In some lavas

PAGE 371

IGNEOUS ROCKS. 325 the iron ore is certainly largely ilmenite The iron ores comprise from 5 to 10 or 12 per cent of the rock. A few grains of greenish to brownish ac cessory hornblende were noted in one rock. Some rocks are coarser grained, containing pyroxene grains as much as a millimeter or more in diameter inclosing the plagioclase and producing a subophitic texture. The texture is in some places intersertal, with glass or alteration products occupying the spaces between the diopside or augite. A brow11ish to greenish micaceous mineral, resembling the serpentine that replaces some of the olivine, fills pores in the groundmass of some lavas. A chemical analysis of a basalt from the Massif de la Selle, north of the Riviere Gosseline, and the average of a number of analyses of basalts for comparison are given in the following table: Analysis of basalt from the Southern Peninsula of Haiti, and average analysis of basalt. 1 2 Si02 48. 97 Al208 14. 90 Fe20a . . . . . . . . . . . . . . . . . . . 96 FeO . . . . . . . . . . . . . . . . . . . . 10.27 MgO . . . . . . . . . 7. 09 cao . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11. 72 N a 2 0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. 06 K20 . . . . . . . . . . . . . . . . . . . . . . 33 H20 + . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. 85 H.,O . . . . 28 Ti02 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. 05 P205 24 MnO . . . . . . . . . . .15 100.37 48.78 15.85 5.37 6.8 6.08 8.91 3.18 1.6.1 1.76 1.89 .47 .29 1. Basalt, Southern Peninsula of Haiti, Ma ssif de la Selle north of the Rivi ere Gosseline, Republic of Haiti. H. S. Washington, analyst. 2. Average of 161 analyses of basalt. Daly, R. A., Igneous rocks and their origin, p. 27, 191. The mineral composition of the rock analyzed, as measured approxi mately by the Rosiwal method, and the norm, as calculated according to the quantitative classification, are as follows: Approximate mineral composition and norm of basalt from the Southern Peninsula of Haiti .Approximate mineral composition. Labradorite .. 38. 6 Diopside ......... 60. 6 Olivine and serpentine ............ 4 1 Ilmenite and magnetite ............ 6. 7 Norm. Orth oclase . . . 1. 8' Albite ...................... 17 .40 Anorthite .................. 30. 4.4 Diopside ................. 21.51 Hypersthene .... 20. 60 Olivine .............. 1.09 Ilmeni"te ...... 3.88: Magnetite .............. 1.39 Apatite . . . . . . . . . . . . . . . . .57 The rock is auvergnose (111.6.4.6).

PAGE 372

326 GEOLOGY OF THE REPUBLIC OF HAITI. The basalt is notably high in lime and is rather high in titanium. The low ferric iron indicates that the iron ore is largely ilmenite;. Albitized or stnlitic basalts. Reddish-brown to gray lavas of amygda loidal texture, in which the f e ldspars are largely albite, apparently of secondary and late magmatic origin, are fo11nd in the valley of the Riviere des Citronniers. The amygdul e s have a maximum diameter of 5 or 6 centi meters. They are filled with c alcite, chlorite, or zeolites. In thin s e ction the rocks are porphyritic, containing phenocrysts of plagioclase and some of olivine in a groundmass of intergranular texture similar to that of the normal basalts. The plagioclase consist.a largely or entirely of albite. Their original texture and twinning lamallae are gen erally preserved, although in some rocks the plagioclase is r e crystallized to a granular aggregate of albite. The olivine once present in some of the ro c ks is altered to iron oxid e s or chloritic minerals. The pyroxenes are rathe r fre s h, although they may b e partly r e placed b y cl1lorit e Chlorite and zeoli tes replace the plagioclase in some of the rocks. The amygdules of the rocks consist of z e olites, c hlorite, and calcit e The zeolites do not app ear to corre spond in opti c al prop e rties to any d e s c ribed species. In some rocks zeoli tes ( ? ) are the principal minerals in the amygdules. They form paralle l or radial crystall i ne growths or hard fibrous growths r e s e mbling chalcedony. Some of the mine rals are stain e d pinkish in places from impurities. The optical properti e s of three minerals that are common in many of the lavas were determined as follows: 1. A colorl e ss to translucent or whitish mineral forms parallel or ra diating growths of prismatic plates. Fragments colorless. Optically+, 2V medium. Prismatic plate s with Z normal to perfect cleavage and plates. Y is parallel to the pris matic cleavage and elongation. Crush e d fragments t end to lie on a face that is normal to Z and that in convergent light shows an acute bisectrix. a.= 1.523 o.ooa P= 1.524 0.003 y = 1.536 0.003. 2. Translucent to white or bluish fibrous platy mineral. Hardness 5 to 6. Fragments may be clouded with threadlike inclusions para Ile! to cleavage. Opti cally+, 2V large. Pris matic plate s with Z normal to platy cleavage. Xis parallel to fibe rs or pris matic cl e avage. Crushed frag ments tend to lie on a face that is normal to Z and that shows a posi tive bisectrix. a.= 1.508 .003. f3=1.51 .003. y = 1.513 .003. a. Whitish to bluish translucent mineral in tough fibrous growths. Hardness 6. Parallel fibers with parallel extinction and negative elonga tion. Birefringence very low ( .001 ). n= 1.52 approximately. Diabase porphyries. Coarse-grained basaltic rocks containing large pbenocrysts of plagioclase and some of augite were found at several places in the Massif de la Selle south of Furey. The plagioclase phenocrysts,

PAGE 373

IGNEOUS ROOXS. consisting of labradorite and bytownite, range from a few millimeters to 1 centimeter in length and comprise 10 to 15 per cent of the rock. The phenocrysts are partly in starlike groups. The plagioclase phenocrysts in one of the specimens are considerably clouded with saussuritic aggre gates and are replaced along cracks by chlorite. The texture of the groundmass of these rocks is subophitic. The grains of augite are about a millimeter in diameter and only partly wrap around the plagioclase in the groundmass. Besides platy titaniferous magnetite or ilmenite a few flakes of partly altered biotite occur as an accessory mineral. Pyrite is a secon dary mineral in one specimen. Although the texture of these rocks is not typical of the coarse-grained diabases, they presumably are parts of minor intrusive bodies. Basic augite andesites. Rocks similar to the ordinary basalt in appear ance and texture, but in which the plagioclase is predominantly andesine ( Ab60 An,0), were found at a few places in the central part of the penin sula between Miragoane and Aquin. They contain no remnants of olivine and generally have some interstitial brown glassy base. The few specimens studied are considerably more altered than most of the basalts. Chlorite ( delessite), calcite, and zeolites are abundant alteration products, replacing plagioclase and obscuring the texture of some of the rocks. Calcite and heulandite are associated in veinlets and replace the plagioclase. The plagioclase seems to be the :first mineral to alter, the augite in some rocks remaining almost 11naltered when the plagioclase is nearly destroyed. ALTERATION. The only extensive alteration product.a in the basalts are chlorite, zeolites, calcite, and serpentine. The chlorite, calcite, and zeolites form amygdules or line vesicles and may fill the spaces or replace the basalts around the blocks in pillar lavas. In some places the rock is almost completely replaced by these minerals. The plagi oclaee is the :first mineral attacked in alterations of this type. The zeolites are analcite, heulandite, and several unidentified or undescribed species found in the albitized basalts along the Riviere des Citronniers. (See p. 326.) Heulandite is widespread and generally is intergrown with or accompanies calcite. In a few places pyrite accompanies analcite and calcite in amygdules. The presence of these alteration products only in the amygdules of some rocks or between the blocks of pillow lava indicates that they were formed during or shortly after the period of volcanic activity. They were presumably deposited from aqueous solutions expelled by the solidification of the lavas themselves or from similar solutions of extraneous origin and from hot springs associated with the volcanic activity. The complete alteration of some lavas probably was due to their position near volcanic centers or near conduits bearing such solutions. The alteration of olivine to serpentine or iddingsite appears to be of slightly different character. This alteration may have taken place in rock

PAGE 374

328 GEOLOGY OF THE REPUBLIC OF HAITI. that is otherwise quite fresh. It probably occurred during or soon after the solidification of the lavas and was assisted only by the presence of a small amount of expelled water. Local albitization and formation of zeolites and other minerals ( ?) was noted in the reddish amygdular lavas from the valley of the Riviere des Citronniers and in reddish amygdular lava from the Grande Riviere de Nippes. There appears to be little doubt as to the secondary origin of the a.lbite in some of these rocks, although the augite is unaltered The basaltic lavas weather to hematite-bearing soil. ORIGIN OF THE I .. AVAS. The uniformity of the basalts over the greater part of the southe1n region and the general lack of extensive pyroclastic deposits indicate that they were erupted principally through :fissur es. A few small dikes, however, cut the basaltic rocks. The large deposits of agglomerate and tuff along the north side of the Massif de la Selle south and east of the Leogane Plain may have been formed during a late period in the volcanic activity, as they are not near the base of the basalt series. These deposits must have been formed by eruptions from central vents or volcanoes. It may be of some significance that the amygdalo idal albitized or spilitic lavas are fo11nd in the vicinity of the fragmental deposits in the valley of Riviere des Citron niers. During periods of relative quiet between periods of explosive activity the molten lava in the volcanic necks or conduits would be under conditions more favorable for the concentration of the light constituents than during the relatively quick eruptions that produced the fissure type of rocks. The concentration in the upper part of the magma chambers of the volatile and alkal ine constituent.a would produce alkaline magmas yielding highly gaseous vesicular and amygdaloidal lavas and would cause the local albitization either of the lavas themselves or of the rocks adjacent to the volcanic vent. The association of the lavas with limestones north of the Asile Valley, the presence of pillow structure in many of them, and the interbedding of calcareous shales and limestones in the lavas of the Massif de la Selle all seem to indicate that the eruptions took place at or near sea level and were sometimes interrupted by the deposition of material derived from them on flood plains or in the sea. Some of the flows were l1ndoubtedly submarine, possibly the greater part of those that show pillow structure. Some of the submarine flows may not have pjllow structure. The centers from which the eruptions of the southern region took place are not known and in all probability were not confined to the present out lines of the peninsula, which were largely determined in late Tertiary time. The agglomerates and tufts in the northwestern part of the Massif de la Selle probably are close to later centers of explosive activity.

PAGE 375

IGNEOUS ROCKS. 329 AGE OF THE LAVAS. As the lavas in the Massif de la Selle at Etang Bossier are interbedded with limestones containing fragments of supposed upper Cretaceous mollusks, a part of the eruptive rocks at this locality is probably of late Cretaceous age. (See p. 95.) Pillow lavas east of the Asile Valley along the Grande Riviere de Nippes are also associated with limestones which are probably of the same age. (See Pl. VIII, B.) The lavas both in the Massif de la Selle and at other places in the Southern Peninsula contain interbedded marine deposits, indicating that the eruptions took place at or near sea level, probably for the most part during Cretaceous time. Between J acmel and Leogane the basalts have buried sheared limestones, which may be lower Cretaceous or older. The periods of greatest igneous activity were probably in upper and possibly late middle Cretaceous time. Probably the activity ceased during the folding that took place in most parts of the Republic in very late Cretaceous time. No evidence of contemporaneous activity was found in any part of the basal upper Eocene of the Southern Peninsula. ANDESITES. DISTRIBUTION, STRUCTURAL RELATIONS, AND AGE. Andesitic lavas were found only west of Baraderes, where they underlie the upper Eocene limestone unconformably, like the basalts east of the town. Their relations to these basalts are unknown, but they may be minor flows or intrusive bodies of the same age as the Cretaceous basaltic eruptions. So far as known they are not interb e dded with basalts. They extend from the vicinity of the Riviere Salee eastward to the Riviere du Baraderes. Small areas may extend along the west side of the Baie des Baraderes. PETBOGRAPHY. Hypersthene andesites. Gray or brown rocks oI fine-grained porphy ritic texture containing abundant plagioclase phenocrysts as much as 1 millimeter in length and scattered smaller prjsms of altered hypersthene make up part of the andesites. One specimen showed fine platy partings or sheeting planes about 4 millimeters apart, probably genetically related to the contraction of the ro c k after solidification. In thin section the sheeted lava consists of plagioclase phenocrysts, mostly in euhedral prisms from 0.2 to 1 millimeter in l e n g th, comprising about 50 per cent of the rock, in a fine grained semicrystalline base. The plagioclase phenocrysts range from labradori te ( Ab4o An80) in the more calcic ones to calcic andesine and are strongly zonal. They probably average a calcic andesine or sodic labradorite (Ab6o AnlSo) in composi tion. Bastite pseudomorphs after hypersthene are less abt1ndant as pheno crysts, a few of them a millim e ter in length. The groundmass con s ist s mainly of small plagioclase prisms and microlites, prisms and microlites

PAGE 376

' 330 GEOLOGY OF THE REPUBLIC OF HAITI. -of bastite, and some light-colored glassy base. A small amount of magne tite and nt1merous small apatite prisms are scattered in the groundmass. rrhe rock is notably low in dark constituents. Hornblende andesites. Light-gray to brownish-gray porphyritic rocks containing plagioclase and hornblende phenocrysts in a fine-grained to felsitic gro11ndmass form another variety of the andesites. Some of these andesites weather into round boulders that spall off in concentric zones. These boulders are stained on the outside with brown iron oxides. In a thin section of a rock collected west of the Riviere des Baraderes, the plagioclase phenocrysts form euhedral to subhedral laths, from 2 to 5 milljmeters in length, and comprise about 15 to 20 per cent of the rock. The more calcic ones have centers of labradorite (Ab40 An60), but most are andesine. Some are strongly zoned, ranging in composition from andesine ( Ab65 An4lS) to oligoclase (A bas AnsG). The hornblende pheno crysts, as much as 4 or 5 millimeters in length, generally are completely resorbed, leaving a skeleton of magnetite or a granular aggregate of magnetite, pyroxene, plagioclase, and calcite. There are two generations of plagioclase in the groundmass, the later generation of small laths showing a trachytic texture. The plagioclase is sodic andesine or oligoclase (Ab65 An35 to Ab70 An80) in composition. Magnetite and small browr1ish prisms of apatite are accessory minerals POST-EOCENE ( ?) BASALTIC ROOKS. Basaltic rocks underlying Pleistocene marine deposits crop out at the base of a prominent sea cliff about 5 kilometers northwest of Chardon nieres. A bed of gravel several meters thick, composed of cobbles of the 11nderlying basalt and of limestone, lies at the base of the Pleistocene marine deposit.a. The basalt itself contains large blocks of white limestone resembling Eocene limestone and blocks of brown metamorphic limestone .. The relations suggest that the basalt may overlie the upper Eocene lime stone exposed a few hundred meters to the southeast along the coast. As the Eocene age of the limestone fragments in the basalt has not been established the suggestion that the basalt is younger than Eocene is only tentative. No exposures of basalt known to be younger than Eocene have been recognized south of the Cul-de-Sac Plain. The Oligocene and Mio cene basalts in the Chaine des Mate11x and Montagnes du Trou d'Eau are entirely different from any basalts found in the Southern Peninsula. The basalt in the cliff near Chardonnierea is mottled from greenish gray to red or black by weathering. It is thickly veined with calcite, part of it in crystals 10 to 15 centimeters in length. The lava is an ordi nary basalt or an augite-rich andesite, as the plagioclase is labradorite and calcic andesine. The rock contains considerable augite, so that it appears to correspond in mineral composition to the Cretaceous basalts.

PAGE 377

TECTONICS. 331 TECTONICS. By WENDELL P. WOODRING. GENERAL FEA'rURES. The relation of the West Indies to North America is somewhat similar to the relation of the Alps to Europe and of the Hjmalayas and East Indies to Asia. The West Indian region is a young mountainous complex, and its major tectonic features are due to cr11mpling of the earth's crust during the Alpine period of folding. It lies between the rigid mass of the Venezuelan highlands and the coastal plain of the southeastern United States, which, although covered in part by the Tertiary seas, was virtually rigid during the folding. During late M e sozoic and Tertiary time the West Indian region was part of the equatorial geosyncline that apparently almost completely encircled the globe. The track of this geosyncline is outlined by a zone of Alpine folds. The tectonic trends of the Republic are shown in Plate XXIV. In the iegions of broad folds these lines represent the crests of anticlines and the troughs of synclines, but in regions of closely spaced folds they merely represent the prevailing strike of the rocks. Some of the tectonic lines show the direction of major lines of rupture. A striking feature of the tectonic lines, as shown by Plate XXIV, is their arrangement in arcs resembling the arcs of the geographic features, which the tectonic features have largely deter1nined. No attempt has been made to discover the significance of the arrangement of these arcs, but they are like the arcs of folded mo11ntains in other regions. Most of the arcs trend northwestward and are convex southward. In the North west Peninsula the west end of the arc representing the crest of the anti clinal arch formed at the end of Miocene time bends southwestward, thus paralleling the northeast end of the Bartlett Deep, which separates the islands of Haiti and Cuba and which at some places between them attains a depth of more than 1,000 fathoms. Some of the arcs in the Southern Peninsula are convex northward, and at the west end of the peninsula they seem to virgate, or branch out sheaf-like. In this part as in other parts of the Tertiary equatorial geosyncline, beds that are no older than Pliocene, even Quaternary beds, are folded. Here also, as in other regions, the beds were folded at different times. There were apparently three principal periods of folding, one at the end of Cretaceous time, one at the end of Eocene time, and one during and at the end of Miocene time. Probably each of the periods of folding con tinued into the succeeding geologic period, and in parts of the Republic the fold ing that began in Miocene time still continu es. A surprising result of the reconnaissance is the discovery that the tectonic features of

PAGE 378

332 GEOLOGY OF THE REPUBLIC OF HAITI. a large part of the Republic are due to folding and crumpling of the beds during the last period of folding that is, during Miocene and Pliocene time. TECTONIC HISTORY. The pre-Mesozoic and early Mesozoic tectonic history of Haiti is so obscure that it can not be deciphered from the information now available. The schistose limestones of Tortue Island and the float of similar intensely metamorphosed rocks seen on the North Plain and the Leogane Plain indicate a record of Paleozoic or early Mesozoic folding that is too fragmentary to read. The most extensive beds of known or supposed Cretaceous age are in the Massif du Nord. These beds are everywhere complexly folded. In the eastern part of the massif their strike is to the northwest, but in the western part it bends aro11nd toward the north, thus forming arcs that are convex to the south. Beds in the northwestern part of the Montagnes Noires and in the Southern Peninsula that are considered of the same age are similarly folded. This folding probably took place at the end of Cre taceous time, although its reEsults can not easily be distinguished from those of later folding. For this reason no age can be fixed for some of the tectonic lines shown on Plate XXIV. The folding was accompanied or followed by intrusions of relatively large batholiths and stocks of quartz diorite, which reach the surface only in the northern part of the Republic and probably were confined to that part. The eastern part of the Massif du Nord contains the largest exposed batbolith, and this same region was rigid after the intrusion. This rigidity presents a marked contrast to the mobility that prevails in the central part of the Republic, where thick deposits of younger marine sedimentary rocks have been considerably folded and crumpled. Beds of marine limestone of upper Eocene age are the most widespread surface rocks in the Republic, as is shown by the geologic map (Pl. I) and the descriptions on pages 106-138. These beds a re folded and crumpled, in some areas more complexly folded than the younger rocks, so that there may have been a period of folding at the end of Eocene time and in early Oligocene time. It is difficult to distinguis h the results of this period of folding from the results of the folding at the end of Miocene time. In the northwestern part of the Montagnes Noires, where there are no Miocene rocks, the trend of the folds is the same as in the southeastern part, where Miocene beds are involved in the folding. In several parts of the Southern Peninsula the Eocene beds have been folded since Miocene time. The large basalt area south of Port-au-Prince seems to be on the crest of a broad anticlinal arch, which probably dates back to the end of Eocene time. The thrust faulting along the north side of the area of basalt may be later. (See fig. 20, 0.) The wide distribution of folded upper Eocene beds gives the impression that the results of the folding that

PAGE 380

TECTONICS. 333 occurred at the end of Eocene time are extensive, but the impression is probably misleading. In the mobile central part of the Republic, where there are extensive areas of Miocene rocks, the Oligocene and Miocene beds have the same strike and dip as the Eocene beds. In the south w e stern part of the Northwest Peninsula, however, beds apparently of upper Eocene age strike north-northwestward, whereas the crest of the main anticlinal arch formed at the end of Miocene time trends west southwestward. Some of the folding in the Montagnes de Terre-Neuve and in the mountains in the c entral part of the peninsula (Montagnes du Nord-ou est) probably took place at the end of Eocene time. The folds in the upper Eocene beds along the northern and southern borders of the Massif du Nord are probably of the same age. The Eocene beds extending across the crest of the Massif du Nord near Dondon and Christophe's Citadelle were apparently deposited in a trol1gh diagonal to the trend of the Cretaceous rocks. Meag e r evidence indicates that the beds in this t rough were folded at the end of Eocene time along lines diagonal to the older trends. The re is no evidence that Eocene deposits formerly extended across the crest of the massif east or west of this trough. The folds in the northwest ern part of the Montagnes Noire s, as well as those in part of the interior of the Southern Peninsula, may also date back to the close of Eoc ene time. The evidence available indicates that there was no folding during Oligocene time or at its end, but the post Eocene folding may have continued into lower Oligo c en e time. Upper Oligo c ene deposits are appar ently conformable with middle Oligocene, and Mio c ene b e ds rest conf ormably on upper Oligocene. Considered in relation to the present morphology, the most important occurrence in the tectonic history of the Republic wa.s the folding and crumpling of the rocks during Miocene and Pliocene time. The tectonic features of the entire mobile central part of the Republic, as well as the outlines of the Northwest Peninsula and of parts of the Southern Peninsula are the results of this folding. The Central Plain, Artibonite Valley, and Cul-de-Sac Plain are deep synclinal troughs; the Northwest Peninsula, Montagnes Noires, Chaine des Mateux, and Gonave Isla11d are anti .. clinal arches. Miocene rocks are involved in all of these folds. The Miocene beds of the Northwest Peninsula seem to be arched in a broad anticline trending west-southwest. Miocene marl crops out under the cover of Quaternary limestone near the crest of the arch along Riviere du Mole. Evidence presented on page 159 indicates that the Miocene beds in the reentrants near the head of Riviere de Jean Rabel and Riviere Cadet on the north side of the p e ninsula are in a shallow synclinal trough diagonal to the general trend of the arch but parallel to the strike of the upper Eocene rocks farther west. The syncline may be a survival from an older structural feature in the upper Eocene rocks flanking the trough. The reentrant of Oligocene and Miocene rocks on the south side of the

PAGE 381

334 GEOLOGY OF THE BEPUBLIO OF HAITI. same gap and the similar reentrants heading into the gap separating the Montagnes du N ord-ouest from the Montagnes de Terre-N euve may be similar structural f e atures. Along the ea s t side of the Arbre Plain the strike of the Miocene beds parallels the Montagnes de Terre N euve. Some of the folding in the Montagnes de Terre N euve and in the Montagnes du Nord-ouest probably took place at the close of Miocene time. In the Montag nes de Terre-N e uve the folding was accompanied or followed by intrusions of quartz diorite and granodiorite. The upper Oligocene or Miocene limestone covering most of Tortue Island seems to be arched in a broad anticline The sections on Plate XXXVI show that the Miocene rocks of the Central Plain are folded in a deep syncline and that they are upturned a gains t the mountains along the borders of the plain. Near the foot of the Montagnes Noires in the northwestern part of the plain the lower part of the Thomonde formation is thrust northeastward over the Ma!ssade tongue (se e Pl. XXXVI, se ction B-B'), and there is probably another high angle thrust fault of greater displacem ent at the mountain front. The northeast limbs of the Thomonde and Chamouscadille anticlines are very steeply tilte d near the mountains, and in depth these folds probably pass in to thrust faults. The great scarp south of Cerca-la-Source, near the southern border of the Massif du N 01 d (see Pl. XXV, A), seems to be a fault scarp between upper Oligocene limestone and Cretaceous ( ?) argilli te, but the evidence available to prove the existence of this fa ult is stratigraphic only. Northwest of the gorge of Riviere Artibonite the Montagne s Noires constitute a compound anticline, but southeast of the gorge the y constitute a single anticlinal arch separated from the compound anticline by a syncline. Figure 12, page 207, and Figure 13, page 208, show that the Artibonite Valley is a deep synclinal trough resembling the Central Plain and containing s e condary anticlines. Exposures along the road from Mirebalais to Las Cahobas show that the Miocene rocks are crumpled near the foot of the Montagne s Noires. Plate XXVI, A, is a view of a small, peculiar asyrnmetrical anticline that extends across the narrow valley of Riviere Fer-a-Cheval at Savanette and, on the south side of the stre am, bends eastward up the valley. West of Savannette a low arch almost at right angles to the trend of the trough of the syncline brings upper Oligocene limestone to the surface. The Chaille des Mateux is perhaps the most perfect example of a large anticline in the Republic. Figure 5 page 128, a se ction across the range, shows that on the southwest flank Oligocene rocks are thrust south westward over Miocene rocks and that there probably is another thrust fault at the edge of the Arcahaie Plain. The secondary anticline on the northeast flank of the mountains near St.-Marc is shown in Figure page 207.

PAGE 382

REPUBLIC QF HAITI GEOLOGICAI-' SUR\ "EY ... -,. 4. SuPPOSED UL'I, SCARP SOUTII'\rEST OF CERCA-LA.-SOL'IlCE. 'be ri
PAGE 383



PAGE 384

TECTONICS. 335 \ The trough of the Cul-de-Sac Plain is perhaps the most remarkable surface feature of the Republic. It has been interpreted as a downfaulted block bounded by normal faults and has been cited as a subaerial form of trough of the sort that is generally submerged in the West Indies, such as the Bartlett Deep, Brownson Deep, and Anegada trough. The Cul-de Sac trough is clearly a syncline, a.a Miocene beds are upturned on the flanks of the mountains bo1dering it and similar beds have been penetrated by wells in the plain. Moreover, the trough is bounded on the south side by a zone of high-angle thrust faults that dip toward the mounta ins, and there is evidence on the north side of similar high-angle thrust faults. Figure 21 is a section drawn across the trough and part of the bordering mountains. (See also Fig. 15, p. 219.) The Eocene and Miocene rocks are more intricately crumpled than is shown in the figure. The main fault on the south side of the trough is well exposed on the Grande Riviere du Cul-de-Sac a short distance above Bassin General. Steep southward dips ... s.o .., o er 2; z 0 "< 0 II J Hetres l: t-tt :re 1000 0 N.E 0 I 'Z. 4 6 8 10 -----Kilometres Ha:12t.eurs doublees FIGURE 21.-Section across the Cul-de-Sac trough and adjoining mountains. Qal, Quaternary alluvium; Tm, Miocene; To, Oligocene limestone; Te, upper Eocene limestone. were found on the north limb of the overturned arch south of the main fault. Plate XXVI, B, shows minor imbricated high-angle thrust faults exposed on the Grande Riviere du Cul-de-Sac in a zone of crumpled upper Eocene rocks south of the main fault. Port-au-Prince Bay and the St.-Marc Canal are apparently the submerged prolongation of this trough, which thus trends northwestward in an arc convex southward, parallel to the arcs of the Cha1ne des Mateux, the Artibonite Valley, the Montagnes Noires, and the Central Plain. The tect-Onic features of several parts of the Southern Peninsula clearly show that the widespread upper Eocene limestones were folded during Miocene time. Marine Miocene beds in the interior lowland on the Grande Riviere de Jeremie are now separated from the sea by a range composed of upper Eocene limestone. As shown on page 227 and in Figure 7 (p. 137), these Miocene rocks are not folded to form a synclinal trough but dip persistently northward toward the mountains, indicating that the upper Eocene limestone has been thrust southward over the Miocene beds. Lignite-bearing Miocene rocks at Camp Perrin are thrust northward over coarse conglomerates deposited as deltas and alluvial fans near the lakes

PAGE 385

336 GEOLOGY OF THE REPUBLIC OF HAITI and swamps where the lignite-bearing beds were laid down. (See pp. 235236, and Fig. 17, p. 234.) These nonmarine beds were probably at one time continuous with the marine Miocene beds that crop out on the Cayes Plain. The range of upper Eocene limestone now separating them may also have been thrust northward. Although the tectonic features of the largest interior lowland, the Asile Valley, are not fully known, it seems prob able that a fault bo11nds the lowland on the south. Most of the nonmarine Miocene beds in this low Iand are but slightly folded. Gonave Island is a broad anticlinal arch. The trend of its crest paral lels the Chaine des Mateux and the folds in the northern part of the Massif de la Selle. Tilted Miocene beds cover the lower part of the flanks of the arch, and in the northwestern part of the island these beds extend across the crest. All the Miocene rocks involved in the folding seem to be of lower and middle Miocene age. The folding probably began in late Miocene time. The evidence already given (seep. 241) shows that the marine conglom erates and marls in the valley of Riviere Gauche near J acmel are of Plio cene age. These rocks are as strongly folded as any of the Miocene beds, so the folding must have continued into Pliocene time. The Pliocene beds strike northwestward and at all the localities where they were ex amined dip southwestward. The southern margin of the lowland was not examined, but if the southwestward dip continues it may be inferred that the upper Eocene limestone in the range bordering the lowland has been thrust northward. The same relation is suggested by the discovery that there are middle Oligocene rocks at sea level on the west side of J acmel Bay and upper Eocene rocks at a higher altitude not far to the west. Many high-angle thrust faults are associated with the folds, particularly those that were formed during the last period of folding. No exten sive overthrust sheets such as characterize Alpine folds in many other parts of the Tertiary equatorial geosyncline were discovered. Detailed work among the upper Eocene limestones, which consist of many different types of rock, may reveal overthrnst sheets. The remarkable normal faults along the Trois Rivieres trough are prob ably later than Miocene. The rocks in the trough are of middle and upper Oligocene age. Details of the fault along the west side of the trough south of Gros-Morne are given on pages 120-121, and the fault is show11 in the section forming Figure 27, A-A' (p. 442), and Figure 30 (p. 471). The fault plane seems to dip steeply eastward, and the throw is several hundred meters. Normal faults of this magnitude are 11nusual in the Republic. The fault along the east side of the trough was not closely examined. The scarp that along its southward prolongation rises abruptly above the Gona.lves Plain is shown in Plate XXV, B. Morne Grammont, an outlier of the Montagnes Noires in this region, resembles the main mo11ntain front, as it seems to be a block tilted northeastward and bounded on the west by a norinal fault.

PAGE 386

REPUBLIC OF HAITI GEOLOGICAL SURVEY PLATE :XXVI : l ASY)J:\IETTIIC .\T J OF AXD T.1T:\:IEST()XE i\T ' SA\ .A:'\ETTE, 0:-\ RIVIERE FERA -CHEV..:\ L. B IIIGH-ANGLE THRUST FAULT IN LI:\1ESTONE OF SUI)POSED "GPPER AGE ON GRANDE RI," IERE DU CUL-DE-SAC.

PAGE 387

1. ...

PAGE 388

TECTONICS. 337 In the mobile central part of the Republic and in the Northwest Penin sula the distribution of the Quaternary reef caps 1 is intimately related to the major folds produced during the Miocene folding. In these parts of the Republic the reef caps are confined almost entirely to the flanks and plunging crests of anticlines. This relation is shown by the reef caps on the crests and flanks of the northwestward-plunging anticlines north and south of St.-Marc but is most strikingly shown in the Northwest Peninsula, wher e the Quaternary reef caps have an altitude of 400 to 450 meters ttbove sea level on the crest of the anticline -formed during the Miocene fold i ng. The r eef caps are most numerous at the crest of the arch but de crease in number and altitude down on the flanks, away from the end of the peninsula. Gonave Island is the only major anticline striking into the sea on which reef caps were not fo11nd, but the extremities of the island were not examined and the caps may be there. So far as known all the reef caps are s ymm etric ally arched over the anticlines. This arching of the reef caps shows that the folding in places continued through Quaternary time. In the Northwest Peninsula, at least, it is probably still going on. The significance of the distribution of the reef caps in the southern Peninsula is not known BEARING OF TECTONICS ON GEOLOGIC HISTORY OF WEST INDIES. The folding at the close of Miocene time deter1nined the location of such geographic f eatures as the Central Plain, Artibonite Valley, and Cul-de-Sac Plain, whi c h if submerged would be similar on a small scale to the great submer g ed troughs of the West Indies. The Cul-de-Sac Plain would be a particularly striking trough if submerged, comparable to the Anegada trough and having about the same curvature. Accumulating evidence indicate s that the submerged troughs were at least deepened if not entirely formed at the close of Miocene and during Pliocene time! The Northwest Peninsula, which was the most mobile part of the Republic during Quaternary time, is adjacent to the northeast end of the Bartlett Deep, perhaps the most remarkable of the submerged troughs. The submerg e d troughs, whi c h have been rec ently d e scribed by Taber,* have been interpreted by Vaughan and Taber as downfaulted blocks bounded by normal faults. The subaerial troughs are deep synclines bounded in part by a zone of imbricate d h i gh-angle thrust faults. It has recently b e en suggested that it is more reasonable to believe that the sub merged troughs are similar deep synclines probably limited by high-angle 1 The Quaternary limestone s at many localities resemble material in the near-by living fringing reefs; at other localities they are merely coralllferous limestones. For con venience all these d e p o Rit s called r ee f caps. 1 See Vaughn n T. W., Geol. S oc. America Bull., vol. 29, pp. 625 627, 1918; U. S. Nat. Bull. 103, pp. 609 610 1919. 1 Taber, S., The grea t fault troughs of the Antilles: Jour. Geology, vol. 30, pp. 89-114, Pl. 1, text fig. 1, 1922. "See Woodring, W. P . T ec t o nic feature s of the R e public of Hnlti and their benring on the geologic history of the West Indie s (abstract) : Washington Acad. Sci. Jour. (Await ing publication.) 22

PAGE 389

338 GEOLOGY OF THE REPUBLIC OF HAITI. thrust faults. Possibly both the subaerial and the submerged troughs were deepened by vertical movements later than the folding, probably of the same age as the movements that produced the high-angle normal faults at the edge of the Troia Rivieres Valley trough. This narrow trough would be a striking feature if submerged, but such troughs bo11nded by normal faults are unusual in the Republic. EARTHQUAKES. BY WENDELL P. WOODRING. RECORDS AVAILABLE Earthquakes are fre quent in the Republic, as in other parts of the Tertiary equatorj al geosyncline During the history of the colony and of the Republi c disastrous earthquakes have at times almost or completely aestroyed Port-au-Prince, Cap-Ha!tien, and other cities. Different parts of the Republic have had distinct seismic histories, and an attempt is here made to correlate the seismic phenomena with the known tectonic features The information on which this attempt is bas e d was obtained from records and accounts published by Rev. J. Scherer, Direct eur de l'Obs e rva'toire du Seminair e -College St.-Martial, Port-au Prince. M. S c herer deserves the hjghest praise for his patience in collect ing records and for his striking accounts of the disastrous shocks and of their relation to the surface and to the geologic features. H i s publication s are as follows : Les grands tremblements d e terre dans l ile d'Ha.lti: Observatoire Meteoro logique du S e minaire-College St.-Martial Bull. semest., July-Dec., 1911, pp. 153-162, 1912. An Engli s h translation of this article was publjsh e d in the Bulletin of the Seismological Society of America, vol. 2, pp. 161-180, map, 1912. Les tremblements de Terre de l ile d'Hai'.ti dans l eurs rapports avec le relief du sol et les f osses maritim e s qui l'entot1rent: Observatoire Meteorologique du Seminaire-College St.-M a rtial Bull. semest., July-Dec., 1912, pp. 132-139, 1913. Catalogue chronologique des tremblements de terre ressentis dans l'ile d'Ha!ti de 1551 a 1900: Observatoire Meteorologique du Seminaire-College St.-Martial Bull. semest., July-D ec., 1913, pp. 147-151, 1914. This catalogue is based principally on the following two catalogues: Poey y Aguirre, Andres, Catalogue chronologique des tremblements de terre ressentis dans les Indies Occidentals de 1530 a 1858; suivi d'une bibliographie seismique concernant les travaux relatifs au tremblements de terre des Antilles: (Extrait de l'annuaire de la Societe Meteorologique de France, tome 5, p 75, seance du 12 Mai, 1857) 76 pp., Versailles, 1858; and Tippenhauer, L. Gentil, Liste der Erdbeben auf Haiti: Die Insel Haiti, pp. 170-175, Leipzig, 1893. Tremblements de terre observes en Haiti de l'annee 1901-1910: Observatoire Meteorologique du Seminaire-College St.-Martial Bull. ann., annee 1920, pp. 100-104, 1921. In addition to these accounts and catalogues, M. Scher e r bas published records of shocks in the Bulletins of the Observatoire. The Bulletins ap peared semjannually from the last half of 1909 to the end of 1916, and

PAGE 390

EARTHQUAKES 339 annually since 1917. The records are based on observations made by M. Scherer in Port-au-Prince and by correspondents at twenty-three othe r localiti e s, namely: Cap-Ha1tien, Borgne, Port-de-Paix, Bassin-Bleu, Gros-Morne, Mole St.-Nicolas, St.-Michel de l' Atala ye, Pilate, Gona!ves, Mirebalais, St.-Marc, Thomazeau, Gantier, Fond-Verrettes, Petionville, Fure y, P e tit-Goave, Anse-a-Veau, Jeremie, Tiburon, Cayes, Bainet and Jacmel. This list is ba sed on the list of stations published in the bulletin for 1921, whi c h is the latest bulletjn now available. At times records were kept at Grande-Riviere du Nord, Bahon, Dondon, Baye11x, I' Arcahaie, Miragoane, Dame-Marie, and Chardonnieres. B eginning with the bulletin of July-December, 1911, the records of a s e ismograph installe d at the Obs e rvatoire have been published. The seismograph is the Omori-Bosch horizontal p endulum type recording south east-northw est and sou t hwest-northeast mov ements magnified 40 times. This instrument is very use ful in registering distant shocks, but it is use l e ss for d e t ermining the direction and amplitude of local shocks. There is an immediate nee d in Haiti for instruments designed to register local sl1ocks. Such instruments should be installed first at Port-au-Prince and Cape-Ha1tien, and later nt other places, particularly Mole St.-Nicolas and Anse-aVea u. The analysis of the 8eismic phenomena is divided into two parts, as the records gathered from diff erent parts of the Republic by M. Scherer since 1909 are muc h more detailed than the records before 1909. Shocks that originated in the Dominican Republic or elsewhere are not considered. DISASTROUS EARTHQUAKES FROM 1551 TO 1908. During the early colonial period (1630-1750), after the French and English bu c can e ers had drive n out the first Spanish settlers, the principal towns were on Tortue Island, along the north coast, at Petit-Goave, near the pre s ent site of Leogane (founded in 1712), and at St.-Louis du Sud. During the late colonial period ( 1750-1803), Cap-Haitien was the largest cjty, but Port-au-Prince was already established and the present cities and towns g rew gradually. During the colonial period and the early period of the Republic only eart hquakes that affected the aaministration of the government or the m e ans of liv e lihood of the people were recorded. Even in such records only the d estruction in the larger cities would be noted. Masonry and brick buildings, whose destruction during earthquake shocks causes the greatest havoc, are most numerous in the cities, and many of the larger cities are built wholly or in part on alluvium, which is more susceptible to the effe c t of sho c ks than bed roc k or residua I soil, particularly where ground water is near the surface. For these reasons, as Scherer has pointed out, records of the effects produced in the cities and in the more thickly populated alluvial plains, where the destruction was greatest, are the only

PAGE 391

340 GEOLOGY OF THE REPUBLIC OF HAITI. records that have come down to us. Little is known of the intensity of even the most disastrous shocks in the hilly country districts. But as cities were widely distributed even in the late colonial period, the inten sity of disastrous earthquakes as indicated by their effects in the cities may be safely used in detennining their origin. Moreover, the results of a study of the records of the disastrous shocks from 1551 to 1908 agree closely with the results of a study of the more complete records from 1909 to 1922. From 1551 to 1908 there were seven disastrous earthquakes that appar ently originated in the Republic of Haiti or under the sea near its shores. Many less severe shocks were recorded, most of them in the large cities, particularly in Port-au-Prince, but the records are of no value in deter mining the places of their origin. On May 7, 1842, Cap-Ha1tien, Port-de-Paix, and Mole St.-Nicolas were completely destroyed by the most disastrous earthquake recorded along the north coast. Cap-Ha!tien was then and had been during the colonial period the most beautiful city in the Republic, and the accounts of the destruction there are most complete. It is estimated that 5,000 people were killed at the Cap out of a total population of 10,000. Most of the city stood on alluvium close to its contact with bed rock. The only record of a sea wave is that the sea dashed against the buildings along the quay. At Port-de-Paix not a single building remained standing a .fter the shock. The sea withdrew 200 feet and then returned, covering the city with more than 15 feet of water. Probably most of the city stood on alluvi11m. Mole St.-Nicolas was an important military post, but the shock ruined the warehouses, forts, church, and aqueducts. There seems to be no definite record of a sea wave here, although in his catalogue Scherer records a wave along the entire Atlantic coast. Probably the entire town, except the fortifications, was built on alluvium. The destruction was great in the area extending eastward along the North Plain to and beyond the Domini can border. At Gona1ves and St.-Marc the shock was much less severe, although several houses fell. In the Southern Peninsula the shock was slight. These records show that the shock was most severe along the north coast from Mole St.-Nicolas eastward to Cap-Haltien. The recording of the sea wave only at Port-de-Paix may indicate that Port-de-Paix is near the place of origin, or the records may be deficient. The Bartlett Deep, as limited by the 1,000 fathom line, begins north of the Northwest Penin sula of the Republic and plunges sharply southwestward, parallel to the similarly plunging anticlinal c1est of the Northwest Peninsula. It h8.b been suggested on page 337 that the Bartlett Deep and other submerged deeps of the West Indies are synclinal troughs bounded by zones of hi gh angle thrust faults. The earthquake of May 7, 1842, was probably due to almost vertical displacement of the sea bottom a long the fault zone at the south edge of the Bartlett Deep off the coast of the Northwest Peninsula

PAGE 392

EARTHQUAKES. 341 This interpretation has already been gi,en by Taber,1 who claims, however, that the Bartlett Deep is a downfaulted block, bounded by normal faults. There is no evidence to substantiate the claim made by both Scherer and Taber that a prolongation of the Hartlett Deep extends along the channel south of Tortue Island and thence southeastward across the Cibao Valley of the Dominican Republic. On September 23, 1887, another severe earthquake shook the same region. The destruction was greatest at Mole St.-Nicolas, where nearly all the houses were ruined. The sea withdrew a great distance and in return jng augmented the disaster. At Port-de-Paix the recently erected church was destroyed. Farther east and south the destruction was not so great, but the sea wave was recorded as far west as Jeremie and Anse d'Hainault. The intensity of the shock and the size of the great sea wa,re at Mole St.-Nicolas indicate that this earthquake was due to almost vertical dis placement along the south edge of the Bartlett Deep, probably farther southwest than the displacement that caused the earthquake of 1842, as Taber 1 has suggested. No disastrous shocks are recorded in the central part of the Republic between 1551 a .nd 1908. During the colonial period and the early years of the Republic the Central Plain and adjoining parts of the Massif du Nord belonged to the Spanish colony. During the same period (1551 to 1908) there were several severe earth quakes in the southern part of the Republic, most of them centering near Port-au-Prince. The earliest shock occurred on November 9, 1701, when houses on the Leogane Plain were destroyed. The road from Uogane to Petit-Goave sank into the sea at places, but this movement may have been due to slumping. The records are too meager to show the place of origin of this shock. Port-au-Prince had not been founded at that time and the neighboring regions were thinly populated. On November 21, 1751, Port-au-Prince suffered one of its greatest dis asters from an earthquake. The newly founded capital then consisted of about 100 buildings, most of which were constructed of ma. sonry. It is said that only one building was left standing after the shock and that it was destroyed by equally severe shocks on the following day. There were numerous aftershocks, and the people lived in tents until December 8. Probably all the buildings at that time, except the fortifications, stood on the low alluvia I ground near the shore. In the Cul-de-Sac Plain, which is almost wholly covered with alluvium, many plantation buildings were laid in ruins. The shocks were felt at Leogane, St.-Marc, Gona1ves, and the Cap, but apparently the destruction was confined to Port-au-Prince 1 Taber, Stephen, The great fault troughs of the Antilles: Jour. Geology, vol. 30, p. 102, 1922. Sin c e this account was written Prof. Taber has pt1bllsbed another paper describing the disastrous earthquakes of the Republic (The seismic belt In the Greater Antilles: Seismological Soc. America Bull., vol. 12, pp. 199-219, pl. 7, 1922). The con clusions regarding the origin of these shocks are the same as those given in the earlier paper. 1 Idem.

PAGE 393

342 GEOLOGY OF THE REPUBLIC OF HAITI. and the Cul-de-Sac Plain. The field work done during the reconnaissance revealed a zone of imbricated high-angle thrust faults along the south side of the Cul-de-Sac trough. Some of these faults are well exposed on the Grande-Riviere du Cul-de-Sac above Bassin General. Overturned folds, some of which are ruptured along high-angle thrust faults, were discovered closer to Port-all-Prince (see p. 130). The probable west ward prolongation of this zone of overturned folds and faults is concealed near Port-au-Prince by nonmarine conglomerates, probably of Pleistocene age. Most of the folding and pres11mably of the thrust faulting took place at the end of Miocene and during Pliocene time. The region was mobile even later, as the Cul-de-Sac trough was completely submerged in Quater nary time. The earthquake of 1751, which was so disastrous to Port-au Prince, was probably due to movements along one of the fractures in this fault zone. If the shocks were due to vertical or nearly vertical submarine movements, as the earthquakes of 1842 and 1887 were supposed to be, there would surely be some record of a sea wave, as the coast from the site of the city northward along the Cul-de-Sac Plain is very low. Only 19 years later, on Jt1ne 3, 1770, Port-au-Prince and the Cul-de Sac Plain suffered another disaster, caused by one of the most severe earth quakes that has occurred since the island was settled by Europeans. The region of greatest destruction extended from Croix-des-Bouquets westward through the Cul-de-Sac Plain to Port-au-Prince and thence westward a1ong the coast through Leogane and Petit-Goave to Miragoane. Two hun dred people were killed in the capital Leogane was destroyed and only one building remained standing in Petit-Goave. The houses between Petit-, Goave and Etang de Miragoane were thrown down in ruins, even those that had been built on other than alluvial ground. The shock was felt all over the colony. Southey 1 states that the sea rose a league and a half up into the island, but this exaggerated report is not confirmed in cont e mpo rary accounts. At Grand-Goave part of the foot of a hill called La Saline was submerged, and'' something similar happ e ned in part of I' Arcahaie.= '' This movement may have been due to slumping If there had been a great sea wave it would be recorded, for all the towns destroyed along the coast were so situated that they would have been at least partly inundated. In the absence of authentic reports of a sea wave it seems reasonable to believe that this disastrous shock also had its origin in the fault zone along the south side of the Cul-de-Sac trough. Its intensity was apparently greater than that of the shock of 1751 and the area of destruction was greater along the thickly populated coast west of Port-au-Prince. The coast from PetitGoave to Anse-a-Veau had a moderately severe shock on April 8, 1860. At Anse-a-Veau the people sought refuge in La 1 Southey, Captain Thomas, Chronological history ot the West Indies, vol. 2, p. 407, London, 1827. 2 Scherer, J., Les grands tremblements de terre dans l'tle d'Haiti: Observatolre du St.-Martial Bull. semest., July-Dec., 19ti, p. 161, 1912.

PAGE 394

EARTHQUAKES. 343 Haute Ville, the eastern part of the town, which stands on a Quaternary 1eef cap about 10 meters higher than La Basse Ville, the western part, which is built on alluvi11m. The sea withdrew and then broke on the shore with a crash. The shock did some damage as far west as Baraderes, but was only slight at Jeremie. It was pronounced on the south coast at Aquin and Les Cayes, and even at Port-Salut. The damage in Port-auPrince was only slight. The sea wave was not a prominent feature, as no wave is recorded at coast towns other tha. n Anse-a-Veau. Moreover, the s hock was much stronger on the south coast than on the north coast. Although the evidence is conflicting it seems that this shock originated in the Southern Peninsula. The tectonic features of the Asile Valley are not fully known, but the abrupt mo11ntain slope along the south side of the valley seems to be a fault s c arp, along which the principal movement took place after the close of Miocene time. (See Fig. 20, A, p. 322.) Con-10 I.I.I u z 0 0 FIGURE 22. Graph showing the number of earthquakes recorded at different stations from 1909 to 1922. Lower curve (solid) iepresents general earthquakes felt in almost the entire country. tjnue d movements along this supposed fault may account for the shock of 1860 and many later less serious shocks. Taber 1 believed that the shocks of 1701, 1751, 1770, and 1860 were due to movements of the sea bottom off the north coast of the Southern P eninsula along a supposed fault zone extending westward from the Cul-de-Sa c trough. The description on pages 397-398 clearly shows that the submerged prolongation of the Cul-de-Sac trough bends northwestward into the St.-Marc Canal. Although the off-shore slope along the western half of the Southern Peninsula is very steep t here is hardly enough evi d e nce to warrant the supposition of faulting, a .nd the seismic phenomena can be interpreted otherwise. EARTHQUAKES FROM 1909 TO 1922. Figure 22 graphic ally shows the number of shocks recorded by Scherer and his correspondents from 1909 to May 4, 1922, inclusive. This graph 1 Op. cit., pp. 94.95, 1922

PAGE 395

344 GEOLOGY OF THE REPUBLIC OF HAITI. can not be rigidly used to show the frequency of earthquakes in different parts oi the Republic, for the records of the observations on which it is based were furnished by as many different persons as there are localities. At some of the stations, Aquin for example, records 'vere not kept continu ously during the period covered. The completeness of the record at Portau-Prince, which includes a large number of shocks, is probably due in part to the unfailing energy of M. Scherer. The lower line of the graph, representing the number of general shocks, would be smoother in some parts if all the general shocks were accurately recorded during the entire period at the different stations. The shocks considered general were re corded at five or more stations, but some of them were general only in parts of the Republic. Some of the general shocks for example, that of the Porto Rico earthquake of October 11, 1918 were of distant origin and were felt in virtt1ally all the Republic. Clearly recognizable after shocks were not included in the total number of shocks. The location of the stations also affects the graph. Some of them are on alluvial ground, others are on bed rock or on residual soil. In view of the facts just stated the graph may not seem to be very use ful, yet it shows clearly the relative frequency of earthquakes in some parts of the Republic. The large n11mber of shocks recorded in the Northwest Peninsula, at Port-au-Prince, and at Anse-a-Veau indicate that they are the regions of highest seismicity. Moreover, this indication agrees with the results of a study of the disastrous shocks from 1551 to 1908. The n11merous shocks recorded at Mole St.-Nicolas, Gona1ves, Port-de Paix, and Cap-Ha1tien probably originated along the fault zone at the south side of the Bartlett Deep, as did the disastrous shocks of 1842 and 1887. Sea waves are not recorded for any of the shocks from 1908 to 1922, which may therefore have involved sma ll vertical displacement. The largest n11mber of shocks were recorded at Gonaives and Mole St.-Nicolas, which are closest to the place of supposed origin. Most of the shocks have an intensity of II to IV (Rossi-Forel scale), but a few are as high as V and VI. Some of the shocks for example, that of February 4, 1918 were relatively strong at Mole St.-Nicolas (IV-V) but were not recorded elsewhere. The earthquake of March 20, 1910, is typical of the minor shocks felt in the northern part of the Republic, particularly in the Northwest Peninsula. It was recorded at Mole St.-Nicolas, Gona1ves, Port-de-Paix, Bassin Bleu (on Les Trois Rivieres between Gros-Morna and Port-de-Paix), Cap-Haitien, Grande-Riviere du Nord, Dondon, and Bahon. A shock that occurred on August 21, 1911, was recorded at Mole St.-Nicolas (IV-V), Gona1ves (V), Gros-Morne (V-VI), Port-de-Paix (V), Pilate (V), Cap-Ha!tien, St.-Michel de l' Atalaye, Hinche, St.-Marc (IV), Port-au-Prince (III), Furey (III), Petit-Goave (II), Anse-a Veau (III-IV), and Cayes. It is remarkable that this shock was not recorded at Petionville, for at Furey, farther from the place of its origin,

PAGE 396

EAR".l'H Q UAKEB. 345 it had an intensity of III. Both stations are on bedrock or on residual soil close to bedrock. Scherer remarks that this is not the first time that a gen eral shock was not felt at Peti onville. The highest intensity in the Southern Peninsula was at Anse-a-Veau, and Les Cayes was the only locality on the south coast where the shock was felt. The earthquake of September 6-7, 1912, apparently had its origin in the interior of the Massif du Nord, as it was strongest at Plaisance (VII), Limbe (VII), Grande-Riviere du Nord (VII), and St.-Michel de l' Ata laye (VII). At Port-au-Prince the intensity was IV-V. The shock was felt at all the stations except Tiburon, Aquin, and Les Cayes. St.-Marc is strikingly free from frequent shocks, the only shocks re corded there being general shocks originating at some distant locality. The records at St.-Michel de l' Atala ye and Mirebalais show similar f ea tures, indicating that the central part of the Republic has fewer earth quakes than the northern and southern parts. The serious earthquake of October 6, 1911, seems to have originated along the southern border of the Massif du Nord, or of its prolongation the Cordillera Central of the Dominican Republic, as Scherer has suggested. The greatest intensity was at San Juan de la Maguana (IX-X) in the Dominican Republic. It was felt at Cerca-la-Source (IX), Hinche (VII-IX), Valliere (VI), Cap-Ha1tien (VI-VII), Mirebalais (VI-VII), St.-Marc (VI), Gona1ves (VI), Port-de-Paix (IV-V), Port-au-Prince (VI), and throughout the Republic, the intensity diminishing in all directions from Cerca-la-Source a .nd Sa. n Juan de la Maguana. The destruction was not very great, as the region where the intensity was highest is thinly populated and most of the houses are built of wood. The great scarp overlooking Cerca-la Source from the south may be a fault scarp. (Seep. 334 and Pl. XXV, A.) If it is a fault it was active at the close of Miocene time, and if it is still active it may account for this earthquake. At Port-au-Prince this shock was the strongest recorded since September 23, 1887, the time of one of the disastrous earthquakes in the northern part of the Republic. Seven aftershocks of intensity II and III were felt at Port-au-Prince. More earthquakes have been reported at Port-au-Prince that at any other station. Many of them are supposed to be due to movement of the rocks along the fault zone at the south side of he Cul-de-Sac Plain, such as produced the disastrous shocks of 1751 and 177'0. If this supposition is correct it is surprising that so few of the shocks were recorded at Gantier, Petionville, and Thomazeau. Gantier and Petionville are on bed rock or residual soil, but Thomazeau stands on alluvi11m. A prono11nced vertical movement was felt during many shocks that apparently originated near Port-au-Prince. On July 31, 1914, a shock lasting 40 to 50 seconds at Port-au-Prince had an intensity of V. All of the movement seemed to be vertical. Despite its intensity the shock was purely local and was not felt farther than Petionville, Furey, and Leogane. The earth-

PAGE 397

346 GEOLOGY OF THE REPUBLIC OF HAITI quake of July 26, 1917, was the most recent pronol1nced shock apparently originating near Port-au-Prince, where it had an intensity of VI. The intensity in general decreased away from Port-au-Prince, but it was un usua lly high at Cap-Ha1tien (VI) and Limonade (V-VI). Although the shock was felt throughout the Cul-de-Sac Plain and at Gan tier (V), it "ras not felt at Petionville and Furey. }IIost of the sl1ocks recorded at Petionville, Gantier, and Thomazeau were felt at Port-au-Prince and generally with greater intensity there. The most striking feature of the Southern Peninsula is the large numbe r of shocks of low intensity (II-III) recorded at Anse-a-Veau. Probably most of them originated along the south side of the Asile Valley, as suggested for the earthquake of April 8, 1860, but on this supposition more of them should be felt at Aquin and Les Cayes. The earthquake of August 3, 1910, seems to have been most severe at Jeremie (VII). It was felt over virtually the entire Republic and had an intensity of V at Port-au-Prince and Cap-Ha1tien. At many stations a strong vertical movement was noted. An observer on the south coast of Gonave I sland felt the shock, but states that the sea was as calm as a lake. The origin of this earthquake is not so clear as for some other shocks, although it is supposed that a fault, active since Miocene time, extends along the north edge of the interior lowland south of Jeremie. (See pp. 226 -22 7 and Fig. 7, p. 137.) The most recent prono11nced shock was on January 15, 1922. The intensity was greatest at Les Cayes (V-VI), where houses even of reinforced concrete were cracked. It was felt all along the south coast as far east as ,T acmel (IV), along the nortl1 coast of the Southern Peninsula from Jeremie (IV-V) to Port-au-Prince (IV-V), and as far north as Gona1ves (III) and St.-Michel de l' Atalaye (II). In i : he absence of adequate information it may be suggested that this earthq uake had its origin either along the supposed fault at the south edge of the Asile Valley or along the scarp b e tween the Port-Salut Peninsula and the Cayes Plain. This scarp has the appearanre of a fault scarp (See p. 135.) CONCLUSIONS REGARDING FREQUENCY OF SHOCKS IN DIFFERENT p ARTS OF THE REPUBLIC. An analysis of the disastrous shocks from 1551 to 1908 and of the more detailed records of shocks from 1909 to 1922 indicate that the N ortbwest Peninsula, the region near Port-au-Prince, and the region nea r Anse-a Veau are the areas where earthquakes are most frequent. The frequency of earthquakes in the Northwest Peninsula agrees with the evidence derived from its geologic features. It is the most mobile region in the Republic, as Quaternary reef caps there have an altitude of 400 to 450 meters above sea level and are more numerous than elsewhere. The elevation of these reef caps is clearly due to the emergence of the

PAGE 398

EARTHQUAKES. 347 southwestward plunging crest of the Northwest Peninsula anticline. This continued rapid emergence causes strains in the rocks that are relieved by fracturing along the submerged northwest flank of the anticline where it plunges into the remarkable depths of the Bartlett Deep, or by slipping along former fractures at the same place. The strains would be relieved principally by vertical or almost vertical fracturing or slipping. The vertical movements transmitted through the elastic rocks cause the vibra tions. If this explanation is correct the serious shocks should ca use sea waves. The observer at Mole St.-Nicolas, the station nearest to the sup posed place of origin, should record any indications of a sea wave. Only one earthquake (September 6-7, 1912) seems to have originated in the interior western part of the Massif du Nord. There is a fault of considerable displacement on the south slope of Mont Puilbo1eau near its crest (see Fig. 18, C, p. 311), but it probably was active at the end of Eocene time and there is no other evidence that it is still active. The ii1terior eastern part of the Massif du Nord, which is a great batholith of quartz diorite, is probably the most stable area in the Republic. This supposition could be verified by obtaining records at Valliere. The south ern border of the massif is relatively stable, although one shock of high intensity (October 6, 1911), originated near Cerca-la-Source. Thrust faults along the southwest edge of the Central Plain, act ive at the end of J\'Iiocene time, seem to be no longer active, but this region is thinly populated, and minor shocks originating there might not be recorded. No shocks seem to have originated near St.-Marc or in the Artibonite Valley. On both sides of St.-Marc there are plunging anticlines bearing Quaternary reef caps resembling in miniature those of the Northwest Peninsula. It might be supposed that the emergence of these reef caps would produce strains; if so, they have not caused rupturing of the rocks during the time since the region was colonized. The thrust faults along the southwest flank of the Chaine des Mateux near the Arcahaie and the Cul-de-Sac plains (see Fig. 5, p. 128 and Fig. 21, p .. 335) are evidently no longer active, as no shocks have originated near l' Arcahaie. If the interpretation of the numerous shocks at Port-au-Prince given on page 342 is correct the thrust faulra along the south edge of the Cul-de-Sac Plain have been active during historic time, and there is no indication that their activity has ceased.. Many of tl1e shocks seem to be due to vertical or almost vertical displacement indicating corresponding movements along the high-angle thrust faults. The cause of the strains thus relieved is not known but may be related to the emergence of the Cul-de-Sac Plain, which was submerged early in Quaternary time. It is hoped that M. Scherer can find a correspondent at l' Asile. If the unusually numerous shocks recorded at Anse-a-V eau are due to dis placement along the south side of the Asile Valley the shocks should be

PAGE 399

348 GEOLOGY OF THE REPUBLIC OF HAITI. of higher intensity at l' Asile. A correspondent at Barad e res could furnish records that would be valuable in fixing the limits of this highly seismic area. The rest of the Southern Peninsula seems to be fairly stable, al though one shock (August 3, 1910) seems to have originated near Jeremie, and the most recent widespread shock (January 15, 1922) had its origin near Les Cayes. The central basalt in the lvlassif de la Selle is limited on at least the north side by faults, which apparently are no longer active. M. Scherer should enjoin his correspondents at all stations on the coast to collect faithfully any information concerning sea waves, particu la1ly during shocks of high intensity. There are so many faults in Haiti that any shock originating on the land could plausibly be consid e red the result of movements along some fault, but it is difficult actually to prove that any fault is still active. The only faults that have been regarded as probable places of origin of earth quakes have been active since Mio c ene time. Surface indications of recent activity are usually soon obliterated, and even immediately after they are formed a careful search generally is required to reveal the m. No surf ace indications were seen during the reconnaissance. The only way actually to prove, for example, that the numerous shocks at Port-au-Prince are due to movements along the faults at the south edge of the Cul-de-Sac Plain would be to determine the region of highest intens ity afte r a serious shock and then search for signs of a recently active fault. PRECAUTIONS AGAINST DAMAGE. Earthquakes have been very frequent in Haiti, particularly in the Northwest Peninsula, at Port-au-Prince, and at Anse-a-Veau. At times they have been disastrous. Shocks may be expected to continue at fre quent inte rvals and some of them may devastate the most thickly popu lated parts of the Republic. No one can predict when the sho c ks w ill come or at how frequent intervals they will be disastrous. Portau-Prince was almost or completely destroyed twice within the period of 19 years from 1751 to 1770, but disastrous shocks are usually not so frequent. In such a region disaster is invited unless precautions are taken. The e arthquakes can not be warded off, but buildings and other structures can be so located and constructed that they will most successfully stand the sho c ks. The following precautions are taken from the account of Professors Reid and Taber, who investigated the Porto Rico earthquake of October 11, 1918.1 Methods of building constru c tion in Porto Rico are very similar to those used in the Republic. Shocks are more severe on alluvial soils, especially soils that are satu rated with gro11nd water, than on bedrock or on residual soils. Most of 1 Reid, Harry Fielding, and Taber, Stephen, The Porto Rico earthquake of 1918, with descriptions ot earlier earthquakes: 66th Congress, 1st Session, Hous e ot Representatives Document 269, 74 pp., 6 figs., 1919 (see pp. 69-74) ; The Porto Rico earthquakes of October-November, 1918: Selsmol. Soc. America Bull., vol. 4, pp. 95-127, pis. 7-14, 1919.

PAGE 400

EARTHQUAKES. 349 the large cities and towns in Haiti stand along the coast and are built wholly or in part on alluvium. The location of many city buildings on alluvium is therefore unavoidable. The possibility of damage is diminished by extending the foundations to bedrock, if possible. Virtually the same result is obtained by building on a thick reinforced concrete plate, which causes the building to move as a whole. Contacts of alluvi11m with bed rock, escarpments, and river banks are especially dangerous locations. Buildings and other structures should be either so elastic that they yield without breaking or so strong that the shocks can not seriously injure them. Wooden buildings are one type of elastic structures. They should be well braced and the joints should be strong enough to resist pulling apart. For some time after the earthquake of June 3, 1770, the colonial authorities permitted only wooden buildings to be built in Port au-Prince. The chief objection to the of wooden buildings in the tropics is that they rapidly deteriorate through rotting and the attacks of termites and other insects. The common type of houses in the co11ntry districts and in the city districts where the poorer classes live are virtually immune from damage. They consist of wooden frames, usually lashed together with fiber thongs, and the walls are made of wicker, which is sometimes covered with plaster or adobe. Steel-frame buildings, such as the market at Port-au-Prince, are also strong, elastic structures. Buildings constructed of rigid material, such as brick, building stone, and concrete, should be strong enough to resist injury. Many of the dwel lings and public buildings show an inferior type of masonry construction, consisting of stones of irregular size and shape embedded in a poor lime mortar. The soft Quaternary coralliferous limestone is frequently used The lime used in the mortar is often incompletely burned and imper fectly slacked, and the sand contains many impurities. Several such build ings were being built in Port-au-Prince during our stay there. Buildings of this kind would be the first to fall during a shock of high intensity, and their construction should be prohibited. Brick buildings will resist any but the strongest shocks if a good mortar is used and the bricks are properly crossed. Clean sand should be used for mortar in both brick and masonry work. The only known large de posits of quartz sand relatively free from impurities are on the North Plain. The binding power of lime mortar is improved by adding cement. Firstclass concrete made with clean sand and sufficient cement is fairly strong. Concrete strengthened with steel reinforcements is believed to be the strongest material available. Heavy roofs, parapet walls aro11nd roofs, and overhanging cornices should be avoided. Solid partition walls should be firmly tied to the outer walls. Floor and roof joists should pass through the supporting wall, or at least half a meter beyond the inner face of the wall. They should be entirely free to slide in their recesses or should be firmly attached to the walls with iron plates.

PAGE 401

350 GEOLOGY OF THE REPUBLIC OF HAITI. SUMMARY OF GEOLOGIC HISTORY. By WENDEI.L P. WOODRING. Little is known of the Paleozoic and early Mesozoic history of the Re public. The schistose limestones of Tortue Island, which are probably marine, are of Paleozoic or early Mesozoic age, but the distribution of land and sea when they were laid down and later events in early Mesozoic time are not known. The intensely metamorphosed schists found as float on the North Plain and the Leogane Plain are altered igneous and sedi mentary rocks at least as old as the schistose limestones. There may be relatively large areas of these ancient rocks in some of the unexplored rugged mountains, such as the Montagnes de la .Hotte. They were formerly extensive in the eastern part of the Massif du Nord, as younger rocks contain material derived from them. The known Mesozoic history begins with a long-continued period of vulcanism in the no1"thern part of the Republic. There were extensive flows of basaltic lavas, followed by flows of andesitic and dacitic lavas. At some places, basaltic flows followed the andesitic lavas. The flows came from non-explosive vents, as very little pyroclastic material ac companies them. These volcanic rocks probably are largely of Jurassic age. The earlier basaltic l avas are very much altered and at places have been converted into greenstones and amphibolites No trace was found ()f the Triassic sea that extended westward from the European }ilediter ranean to Cuba and Mexico, carrying with it a Mediterranean and later a north European fauna. The early volca nic rocks of the northern part of the Republic were tl1en attacked by weathering and erosion and the n1aterial derived from them was deposited along the flood plains of streams that carved wide, gently sloping valleys. Some of the material was laid down in shallow seas tl1at extended over an 11nknown part of the country. rrh ese nonmarine and marine rocks are the argillites and limestones of supposed Lower Cre taceous age. At the same time shallow seas covered a large part of the present Southern Peninsula and the limestones found at many localities were laid down on a basement of unknown rocks. In Upper Cretaceous time a shallow sea covered at least part of the northern part of the Republic. Its limits are not known, as the rocks of this age that were found probably are mere remnants. A typical Mediterranean fauna, consisting principally of Rudistid mollusks, flourished in this sea. At places the Upper Cretaceous limestone is a Rudistid reef, consisting of masses of valves, some of which are a meter or more in length. This fauna, apparently of late Upper Cretaceous age, has been found in St. Croix, Porto Rico, the Dominican Republic, Cuba, and Jamaica, but it seems to be most extensive in J amaica.1

PAGE 402

EARTHQUAKES. 351 While this sea covered parts of the northern part of the Republic there were :fissure eruptions in the southern part, covering low-lying lands with thick flows of basalt that is remarkably 11niform from the Massif de la Selle westward to the end of the Southern Peninsula. Material derived from these flows and some of the flows themselves were deposited in shallow marginal seas. Mesozoic time apparently was closed by a period of folding, but it is difficult to ascertain, in most places, what elements of the present structure are due to Mesozoic movement and what to later Tertiary movements. Intrusion s of batholiths and stocks of quartz diorite accompanied or fol lowed the folding in the northern part of the Republic, altering the old volcanic rocks near the contacts to amphibolites and to chloritic and talcose schists. These masses of quartz diori te are exposed by erosion fron1 the Dominican border westward to the Northwest P eninsula The largest exposed batholith is in the eastern part of the Massif du Nord, and this r egion was very rigid during later time. Early Eocene time was a period of erosion, and if there were then any highlands they were rapidly worn down, exposing in the northern part of the Republic the batholiths and stocks of quartz diorite. During middle Eocene time the sea covered part of the 1fassif du Nord and the Northwest Peninsula and in it the Plaisance lim estone, wl1ich carries a Mediter ranean fauna, was deposited. The trough in which this limestone was laid down probably was much larger than the present outcrop of the limestone, but its boundaries are not known. The extensive transgression of the sea in late Eocene time is one of tl1e outstanding features of the Tertiary geo lo gic history. Its shallow waters covered almost the entire Republic except the northeastern and probably the western parts of the Massif du Nord. The many different kinds of lim esto ne, which cover so large an area, were deposited in this sea. Foraminifera of Mediterranean aspect are the most common fossils in the rocks. At the end of Eocene time the sea withdrew and for a period there was apparentl}r folding, the results of which, however, are almost as obscure as those of the folding at the end of Cretaceous time. Early Oligocene time, it seems, like early Eocene time, was a period of widespread emergence and erosion The sea returned a gain in middle Oligo cene time, but its transgression was less extensive than in late Eocene time. A sha llow sea covered large areas, particularly in the centra l part of the Republic, which was very mobile during Tertiary time. The rocks of middle Oligocene age, like the Eocene rocks, are limestones, indjcating that the waters were clear The history of the events between middle and upper Oligocene is rather obscure. During this interval there were flows of 1 See recent articles by C. T. Trecbn1ann, The Cretaceous and T ertiary question in Jamaica: Geol. Mag., vol. 59, no. 699, pp. 422-431, 3 text figs., 1922 ; The Barrettia beds ot Jamaica : Geol. Mag., vol. 59, pp. 501-514, pis. 18-20, 1 text fig., 1922

PAGE 403

352 GEOLOGY OF THE REPUBLIC OF HAITI. nephelite basalt in the Montagnes du Trou d'Eau. Farther west, near Saut d'Eau, the volcanic activity may have been of Miocene age. The nephelite basalt and other basalts interbedded with Miocene sedimentary rocks on the southwest slope of the Chaille des Mate11x represent the last know11 period of volcanism. The late Oligocene sea covered a smalle r area than t he middle Oligocene, the largest areas being in the mobile central part of the Republic. The upper Oli gocene rocks are limestones. The most common fossils in both the middle and upper Oligocene lime stones are Foraminif era and corals of Mediterranean aspect. The end of Oligocene time was a period of extensive elevation except possibly in the central part of the Republic. Detrital debris derived from the highlands was carried into the Miocene sea as it advanced. Near old land masses, as in the Central Plain, the Miocene rocks consist almost entirely of detrital material, although clearer waters at intervals permitt.ed the establishment of coral reefs. The early Miocene trangression was the most extensive except that in the late Eocene. The sea covered all of the Northwest Peninsula except the central part, all of the central part of the Republic, and parts of the Southern Peninsula. This was the last extensive transgression and also the last appearance of a Mediterranean fa11na. It would be more correct to consider the Mediterranean Tertiary faunas as having a West Indian aspect. Many of the genera of mollusks found in the Miocene Tertiary faunas of the West Indies and the Medi terranean region are now living in the West Indies but are extinct in the Mediterranean Sea. The present Mediterranean fa11na is the result of invasions of northern genera in Quaternary time. The sea probably covered the Cul-de-Sac Plain and marginal parts of the Southern Penin sula in middle Miocene time. N onmarine deposits in the interior of the Southern Peninsula seem to be of the same age. Toward the end of Miocene time there was a period of folding that determined the tectonic features of a large part of the Republic. Intrusions of quartz diorite and granodiorite in upper Eocene beds may have accompanied or followed this folding in the Montagnes de Terre-Neuve. The mot1ntain ranges in the centra l part of the Republic coincide with anticlinal arches formed at the end of Miocene time. It is not knowIL whether the arches were elevated as mo11ntains at the same time. Miocene and even older rocks are completely removed from their crests, and in the Chaine des Mateux there is evidence of peneplanation. All this erosion may have taken place when the present mountains were at a lower altitude. The mountains in the northern part of the Republic and in the Southern Peninsula probably have been highlands since the end of Oligocene time, although they may have shared in the elevation after the M ioce ne folding. The normal faults bo11nding the Trois Rivieres trough seem to be later than the folding and may be genetically related to the elevation of the mountains that followed the folding.

PAGE 404

EARTHQUAKES. 353 During middle or late Miocene time the West Indian islands were apparently larger than they are now, and they may have been joined to South or Central America, thus permittjng the invasion of a mammalian fa11na of South American aspect, principally rodents and gro11nd sloths. These mammals are repres ented by Quaternary remains found in caves, such as those explored near St.-Michel de l' Atalaye. The folding and faulting begun toward the end of Miocene time probably outlined the island of Haiti as we now know it. So far as known the sea covered only small areas near J acmel and PetitGoave in Pliocene time. The corals and mollusks in the beds in these areas are very similar to living W est Indian species. The period of folding begun in Miocene time continued after these beds were deposited. At the beginning of Quaternary time the sea covered the western part of the Northwest Peninsula, the Cul-de-Sac trough (which divided the island into two parts), and oth e r smaller areas along the present coast. These regions have emerged since then. In the Northwest Peninsula and the mobile central part of the Republic, where the emergence apparently is still continuing, the marine Quaternary rocks, or reef caps, are at a higher altitude on the crests of the anticlines, indicating that folding is still continuing. If observations could be extended over a long period it mjght be possible to determine the rate of emergence in the Northwest Peninsula, where Quaternary reef caps have an altitude of 400 to 450 meters above sea level. In the San Juan Valley of the Dominican Republic basaltic lavas of Pliocene or Quaternary age rest on Pliocene gravels. No trace of these lavas was fo11nd in the Central Plain. Folding and mountain-making have been active in the Republic since late Tertiary time and therefore earthquakes are frequent. There is no reason to believe that their frequency will diminish 23 ..

PAGE 405

PART III. GEOMORPHOLOGY. By WENDELL P. WoonRING, JoHN S. BROWN, and WILBUR S. BURBANK. The term geomorphology as used in this report signifies a description of the surface features of the earth with regard to their origin. Each of the geographic provinces of the Republic of Haiti comprises an area in which the surface features have a certain uniformity, usually the result of a common geologic history. These provinces are listed on page 31 and their location is shown on Plate XXVII. They are here described in the order in which they are listed on page 31, beginning in the northern part of the Republic. Only the major surface features of each province are described. As it seems unnecessary to repeat a discussion of the major events in the geologic history of each province, the reader should keep constantly in mind the g e ologic history of Haiti as outlined on pages 350-353. The lith ology and the structure of the surface rocks, which are the most potent agencies in determining the shape of the features, are described briefly for each province. TORTUE ISLAND. GENERAL RELATIONS. Tortue Island is 37 kilometers long and has a maximum widt h of 7 kilometers. It is separated from the mainland by a channel that is 15 kilometers wide at the west end of the island and 9 kilometers wide at its east end but that narrows to 7 kilometers off St.-Louis du Nord. The maximum recorded depth of the channel is 1,267 meters ( 693 fathoms) off the west end of the island. In the narrowest part of the channel the maximum recorded depth is 777 meters ( 425 fathoms). The axis of the island trends N. 78 W., parallel to the trend of the main structural features of the Massif du Nord. LAND FEATURES. UPLAND FORMS. Tortue Island is a plateau modified by erosion and at the margins further modified by the results of recent emergence. The interior is a rolling plateau, the crest of which has a mean altitude of about 300 meters above sea level, but low, rounded knobs rise 325 meters above sea level. The plateau slopes gently seaward, both to the north and to the south On the crest of the plateau there are no well-defined drainage channels. Near the 354

PAGE 407

GEOMORPHOLOGY. 355 roast the slope is steep and is scarred by ravines that have slightly dis sected the plateau by headward erosion. 411 the coastal ravines are short except those at La Vallee, on the south coast, where several extend far back into the plateau, forming a huge, intricately dissected amphitl1eater bounded by limestone cliffs. SHORE FEATURES. The most striking shore f eature s are emerged coastal terraces. These terraces are widest and be s t preserved at the extremities of the island, where the seaward slope is more g e ntle than on the flanks. Plate XXVIII, A, a view of the east end of the island as seen from the south, shows clear 1 y the profile of two emerged terraces. Each of the terraces has a gentle seaward slope and each has an emerged sea cliff at its inner edge. The lower terrac e is tr11ncat e d at the shore line by a similar but steeper cliff. As this part of the i sland wa s seen only from a distance no measure ments were taken, but at other localities visited the inner edge of the lower terrace stands about 15 meters above sea level. The altitude of the upper terrace is not known, but the vertical interval between its outer edge and the outer edge of the lower terrace is greater than that between the lower terrace and the shore line. On the south coast the terraces have been obliterated by erosion except in sheltered regions. One kilometer west of Pojnte des Oiseaux there is a narrow remnant of the lower terrace 50 to 80 meters wide. At the outer edge there is a perpendicular sea cliff 8 to 10 meters high. The steep slope at the inner edge of the terrace merges into the coastal slope of the plateau and is not clearly disc e rnible as a sea cliff. The abrasion platform is com posed of schistose limestone, which is covered with a thin veneer of soft conglomeratic coralliferous limestone. The west end of the island and the north coast were seen only from a distance. Terrac e profiles are visible at the west end, and according to the description of Moreau de St.-Mery 1 terraces are conspicuous features at some localities on the north coast. Along almost its entire length the north coast is truncated by sea cliffs and is inaccessible. The south coast is not so precipitous. At La Vallee an alluvial apron extends we s tward along the coast, embracing a narrow mud flat fringed with mangrove thickets. East of La Vallee the coast is bor dered by sea cliffs that truncate remnants of the lower terrace. SUBLITTORAL FEATURE