Front Cover
 Part 1: Porto Rican archaeolog...
 Part 2: A large archaeological...
 Part 3: Porto Rican prehistory:...
 Part 4: Porto Rican prehistory:...
 Back Cover

Group Title: Scientific survey of Porto Rico and the Virgin Islands
Title: Scientific survey of Porto Rico and the Virgin Islands /
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00091487/00017
 Material Information
Title: Scientific survey of Porto Rico and the Virgin Islands /
Alternate Title: Scientific survey of Puerto Rico and the Virgin Islands
Physical Description: 19 v. : ill. ; 24 cm.
Language: English
Creator: New York Academy of Sciences
Jay I. Kislak Reference Collection (Library of Congress)
Publisher: The Academy,
The Academy
Place of Publication: New York N.Y
Publication Date: 1940-1952
Frequency: completely irregular
Subject: Scientific expeditions -- Periodicals   ( lcsh )
Natural history -- Periodicals -- Puerto Rico   ( lcsh )
Natural history -- Periodicals -- Virgin Islands of the United States   ( lcsh )
Natuurlijke historie   ( gtt )
Geologie   ( gtt )
Expedities   ( gtt )
Genre: bibliography   ( marcgt )
Spatial Coverage: Puerto Rico
United States Virgin Islands
Summary: Includes bibliographies.
Ownership: Provenance: Gift of Jay I. Kislak Foundation.
Statement of Responsibility: New York Academy of Sciences.
Dates or Sequential Designation: Vol. 1, pt. 1-
Dates or Sequential Designation: Ceased with vol. XIX, pt. 1.
General Note: Latest issue consulted: Vol. 18, pt. 4 (1952).
General Note: Kislak Ref. Collection: Vol. 18, pt. 2 (1941)-pt. 4 (1952).
 Record Information
Bibliographic ID: UF00091487
Volume ID: VID00017
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 01760019
lccn - 2002209050


This item has the following downloads:

VID00017 ( PDF )

Table of Contents
    Front Cover
        Page i
        Page ii
    Part 1: Porto Rican archaeology
        Page iii
        Page iv
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
        Page 175
        Page 176
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
        Page 183
        Page 184
        Page 185
        Page 186
        Page 187
        Page 188
        Page 189
        Page 190
        Page 191
        Page 192
        Page 193
        Page 194
        Page 195
        Page 196
        Page 197
        Page 198
        Page 199
        Page 200
        Page 200a
        Page 200b
        Page 201
        Page 202
        Page 202a
        Page 202b
        Page 203
        Page 204
        Page 204a
        Page 204b
        Page 205
        Page 206
        Page 206a
        Page 206b
        Page 207
        Page 208
    Part 2: A large archaeological site at Capa, Utuado, with notes on other Porto Rico sites visited in 1914-1915
        Page 208a
        Page 208b
        Page 209
        Page 210
        Page 211
        Page 212
        Page 213
        Page 214
        Page 214a
        Page 214b
        Page 215
        Page 216
        Page 216a
        Page 216b
        Page 217
        Page 218
        Page 218a
        Page 218b
        Page 219
        Page 220
        Page 220a
        Page 220b
        Page 221
        Page 222
        Page 222a
        Page 222b
        Page 223
        Page 224
        Page 224a
        Page 224b
        Page 225
        Page 226
        Page 226a
        Page 226b
        Page 227
        Page 228
        Page 229
        Page 230
        Page 231
        Page 232
        Page 233
        Page 234
        Page 235
        Page 236
        Page 237
        Page 238
        Page 239
        Page 240
        Page 241
        Page 242
        Page 242a
        Page 242b
        Page 243
        Page 244
        Page 245
        Page 246
        Page 247
        Page 248
        Page 249
        Page 250
        Page 251
        Page 252
        Page 252a
        Page 252b
        Page 253
        Page 254
        Page 255
        Page 256
        Page 257
        Page 258
        Page 259
        Page 260
        Page 261
        Page 262
        Page 263
        Page 264
        Page 265
        Page 266
        Page 267
        Page 268
        Page 269
        Page 270
        Page 271
        Page 272
        Page 273
        Page 274
        Page 275
        Page 276
        Page 277
        Page 278
        Page 279
        Page 280
        Page 281
        Page 282
        Page 283
        Page 284
        Page 285
        Page 286
        Page 287
        Page 288
        Page 289
        Page 290
        Page 291
        Page 292
        Page 293
        Page 294
        Page 295
        Page 296
        Page 297
        Page 298
        Page 298a
        Page 298b
        Page 299
        Page 300
        Page 300a
        Page 300b
        Page 301
        Page 302
        Page 303
        Page 304
    Part 3: Porto Rican prehistory: introduction; excavations in the West and North
        Page 305
        Page 306
        Page 307
        Page 308
        Page 309
        Page 310
        Page 311
        Page 312
        Page 313
        Page 314
        Page 315
        Page 316
        Page 317
        Page 318
        Page 319
        Page 320
        Page 321
        Page 322
        Page 323
        Page 324
        Page 325
        Page 326
        Page 327
        Page 328
        Page 329
        Page 330
        Page 331
        Page 332
        Page 333
        Page 334
        Page 335
        Page 336
        Page 337
        Page 338
        Page 339
        Page 340
        Page 341
        Page 342
        Page 343
        Page 344
        Page 345
        Page 346
        Page 347
        Page 348
        Page 349
        Page 350
        Page 351
        Page 352
        Page 353
        Page 354
        Page 355
        Page 356
        Page 357
        Page 358
        Page 359
        Page 360
        Page 361
        Page 362
        Page 363
        Page 364
        Page 365
        Page 366
        Page 367
        Page 368
        Page 369
        Page 370
        Page 371
        Page 372
        Page 373
        Page 374
        Page 375
        Page 376
        Page 377
        Page 378
        Page 379
        Page 380
        Page 381
        Page 382
        Page 383
        Page 384
        Page 385
        Page 386
        Page 387
        Page 388
        Page 389
        Page 390
        Page 391
        Page 392
        Page 393
        Page 394
        Page 395
        Page 396
        Page 397
        Page 398
        Page 399
        Page 400
        Page 401
        Page 402
        Page 403
        Page 404
        Page 405
        Page 406
        Page 407
        Page 408
        Page 409
        Page 410
        Page 411
        Page 412
        Page 413
        Page 414
        Page 415
        Page 416
        Page 417
        Page 418
        Page 419
        Page 420
        Page 421
        Page 422
        Page 423
        Page 424
        Page 425
        Page 426
        Page 427
        Page 428
        Page 429
        Page 430
        Page 431
        Page 432
        Page 433
        Page 434
        Page 435
        Page 436
        Page 437
        Page 438
        Page 439
        Page 440
        Page 441
        Page 442
        Page 443
        Page 444
        Page 445
        Page 446
        Page 447
        Page 448
        Page 449
        Page 450
        Page 451
        Page 452
        Page 453
        Page 454
        Page 455
        Page 456
        Page 457
        Page 458
        Page 459
        Page 460
        Page 460a
        Page 460b
        Page 460c
    Part 4: Porto Rican prehistory: Excavations in the Interior, South and East; chronological implications
        Page 461
        Page 462
        Page 463
        Page 464
        Page 465
        Page 466
        Page 467
        Page 468
        Page 469
        Page 470
        Page 471
        Page 472
        Page 473
        Page 474
        Page 475
        Page 476
        Page 477
        Page 478
        Page 479
        Page 480
        Page 481
        Page 482
        Page 483
        Page 484
        Page 485
        Page 486
        Page 487
        Page 488
        Page 489
        Page 490
        Page 491
        Page 492
        Page 493
        Page 494
        Page 495
        Page 496
        Page 497
        Page 498
        Page 499
        Page 500
        Page 501
        Page 502
        Page 503
        Page 504
        Page 505
        Page 506
        Page 507
        Page 508
        Page 509
        Page 510
        Page 511
        Page 512
        Page 513
        Page 514
        Page 515
        Page 516
        Page 517
        Page 518
        Page 519
        Page 520
        Page 521
        Page 522
        Page 523
        Page 524
        Page 525
        Page 526
        Page 527
        Page 528
        Page 529
        Page 530
        Page 531
        Page 532
        Page 533
        Page 534
        Page 535
        Page 536
        Page 537
        Page 538
        Page 539
        Page 540
        Page 541
        Page 542
        Page 543
        Page 544
        Page 545
        Page 546
        Page 547
        Page 548
        Page 549
        Page 550
        Page 551
        Page 552
        Page 553
        Page 554
        Page 555
        Page 556
        Page 557
        Page 558
        Page 559
        Page 560
        Page 561
        Page 562
        Page 563
        Page 564
        Page 565
        Page 566
        Page 567
        Page 568
        Page 569
        Page 570
        Page 571
        Page 572
        Page 573
        Page 574
        Page 575
        Page 576
        Page 577
        Page 578
        Page 579
        Page 580
        Page 581
        Page 582
    Back Cover
        Page 583
        Page 584
Full Text
, s-" r-






Porto Rican Archaeology

- =

PuHliBHEL, BI 1iI : A.C U.t Lm
I'. hlilH, hV 13. 1911)

<* :

jz *
%\ ^"

^L "

1",'I 4h I 6t l wl1,In 1









Porto Rican Archaeology

Froelich G. Raincy



Associate Editor (Porto Rico Publications)

This natural history survey of Porto Rico and the Virgin Islands, conducted by
The New York Academy of Sciences, was established in 1913. Continuous pub-
lication of the results of this survey is nuule possible through contributions from
the Department of A agriculture and Commerce of Porto Rico, and the Unwersity
of Porto Rico.



INTRODUCTION. ........... ......... .... ... 3
SEQUENCE OF CULTURES IN PORTO RICO ........... ........... 7
Excavations in Barrio Canas ..................... .............. .. 7
Excavation Number 1 . . ....... ............................. .. 8
Excavation N um ber 2 ........... .. .............................. 11
Sum m ary ........ ......... ....... ......... .. ... ............ 12
Collections from Barrio Canas................... ................ 14
Pottery of the Shell Level............ ........... ..... ........ 15
Associated Artifacts of the Shell Level. ..... ... ....... ....... 25
Variation within the Shell Level ... ................ ...... ....... 33
Pottery of the Crab Level. . .. .......... .. .......... 35
Associated Artifacts of the Crab Level ....... ..... ......... 54
Conclusions ............... . .................. ............ . 58
Excavations in Barrio Coto........ ....... .. .... ........ 62
Excavation N um ber 1 ............ ........... .... ............ 64
Excavation Number 2 ............. ..... .................. 66
Summary............. ........... ...... . .. 67
Collections from Barrio Coto....................... ... .. .... .... 69
Pottery of Shell Level Type........... .. .. . . 70
Pottery of Crab Level Type .................... ...... .. 71
Associated Artifacts......................... ..... ... .. . 72
C conclusions .......... ............ ....... ... .. ........ 74
Excavations in Barrio Monserrate ........ ....... ............... 75
Mound A ............ ................................... 76
M ound B ............ .......... ... .. .......... . .......... 78
Mound E ...................................... ... 82
Summary .......... .. ......................... . 83
Collections from Barrio Monserrate ............. ... .............. 83
Pottery of Shell Level Type in Mound A ........... ......... ...... 84
Associated Artifacts in Mound A ... ...................... ........ 87
Pottery of Shell Level Type in Mound B. ... ........... 88
Pottery of Crab Level Type in Mound B .. ... .. .. ...... 89
Associated Artifacts in Mound B ............ ............. ... . 91
Pottery of Crab Level Type in Mound E.. .... ...... .. . . ... 92
Associated Artifacts in M ound E . ......... ....................... 94
Conclusions .................. ...................... . 95
Additional Excavations in 1935 ........ ........ ..... . . . 97
Ball-courts in the Municipality of Orocovis . 98
Continued Excavation at Monserrate ............ ... .... 102
C conclusions ................. ............. .............. . 104
General Conclusions .. .............. . 105


PoRtRo Rlc'O ........ ... .. ........... I
Previous Excavations in Porlu Rico ................................... Ill
Santa Isabel ....... .......................... ........... 112
U tuado .......... ... .......... 113
M a n a ti . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 16
C abo R ojo .......... ................ ...... 117
Sum m ary .. ... ......... ............. ........... ......... 118
Santo D om ingo .................... . . . . . . .... 120
C ape M acao . .... .................................. . . . .. 120
San Pedro de M acoris ..................... ....................... 122
Hatt's Excavations in Eastern and Central Santo Domingo............ 124
Sam and B ay C aves ............................................... 125
Samand Peninsula, Village Sites ................................ : ... 126
M onte Cristi Province.............................. ............. 128
Andres Province .......................... ........... .......... 128
Sum m ary ........................................................ 130
H a iti . .. .. . .. .. . . . . .. . . . . . . . . . . . . .. . .. ... . 13 1
F ort Liberty B ay ............................................. 132
lie des Cabrits ................ .................................. 134
L a G onave ....................................................... 134
Summary......................................... ............. 134
C uba ........ ............ ........... ..... 135
Gran Tierra do M aya ............................................. 136
Jauco Region-Eastern Cuba ...................................... 137
Monte Cristo Region-Eastern Cuba .............................. 138
Gran Tierra Region-Eastern Cuba ................................. 138
Santiago Region-Southeastern Cuba . .. ......................... 139
Pinar del Rio Region-Western Cuba ............................... 140
Summary ......................... .......... ................. .. 140
Jamaica .. ......... ............................................ . 144
N orbrook .......................... ............................ 144
V ere . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . 14 5
R etreat ..................................... ............... 146
Jam aican C collections .............................................. 147
Baham a Islands................ ........................ ........... 149
Caicos Group ............................................... .... 149
Bahama Collections ........... ................................... 151
Survey of the Bahamas ..................... ...................... 151
Summary. ........ ............................................. 153
T he V irgin Islands .................................... ........... 154
De Booy's Excavations ............................................ 154
HIatt's Excavations ............................................... 158
Summary ............... .... .... ...... ...... ........ 161
St. K itts and N evis .... ......................... ....... ........... 162
Montserrat ...................... ............................... GG16


G uadeloupe ........ ................................................ 167
D om inica . ................................... . ........ 167
B arbados ......... ................................................ 168
St. Vincent ........................................... ........ .. 168
M ustique ................................................ .. ..... 170
C arriacou ............. ............................................ 170
Grenada .......... ....................................... 171
Trinidad ......... ....................................... 172
Erin Bay ......... ................. ...................... 172
C ape M ayaro .............. ..................................... 173
S um m ary .............. ........................................ 174
G general C conclusions .............. ................................. 175
BIBLIOGRAPHY ..................................................... 185
APPENDIX .. ............................................ .. 190
IN D E X ............ .............................................. 20 7

The present study is based primarily upon archaeological excava-
tions in Porto Rico carried out in the summer of 1934 under the direc-
tion of Peabody Museum, Yale University, and of the American
Museum of Natural History in New York with the cooperation of the
University of Porto Rico. These excavations were part of the scien-
tific survey of the island organized by the New York Academy of
Sciences. Investigation was continued in the summer of 1935 under
the same arrangement for a period of six weeks, and some additional
information obtained at that time has been included. During the
winter and spring of 1934 a general survey of archaeological sites and
collections was made in the Bahamas, Haiti, and Santo Domingo, and
some reference is made in this paper to the work on those islands.
In 1934 with the assistance of the University of Porto Rico a rapid
survey of archaeological sites in Porto Rico was made. Chancellor
Carlos E. Chardon assigned AMr. Gustavo Rodriguez to assist me in
the location of sites suitable for excavation, and with the cooperation
of the extension agents of the University of Porto Rico we were able
to cover a large area in a relatively short time. During the months of
April and May many promising sites were located both in the moun-
tainous interior region and along the coast. Since the primary object
was the excavation of sites with sufficient depth of deposit to show


S ,4 E I'. i247 20 O..s t I i'4 k

FIGURE 1. Map of Porto Rico. Sites excavated in 1934. t Sites excavated previous to 1934.


stratification if such existed, and since the most extensive sites were
found along the coast, excavations for the most part were limited to
the coastal region.
Three major sites were excavated during the months of June, July,
and August, 1934 (FIGURE 1). A large crew of laborers supplied by
the Porto Rican Emergency Relief Administration made it possible
to extend cuts through a large part of each site and in this manner to
determine the general composition. It was found that a single unit
of deposit in any site was of little significance in relation to the whole,
due to the nature of the deposition, and for this reason extensive ex-
cavation was of cardinal importance.
In July 1935 it was possible to carry out limited excavations in the
mountainous interior near Orocovis. Assisted by Arturo Morales
Carri6n of the University of Porto Rico and again with the aid of
Federal Relief Administration funds, 1 conducted the excavation of
four different sites near Orocovis attempting to correlate the interior
deposits with those on the coast. In August, 1935, we continued ex-
cavation for a period of two weeks at the Monserrate site which had
been partially excavated during the preceding summer.
Two periods of prehistoric occupation on the island were distin-
guishable in clearly stratified deposits of culture refuse found on the
north and on the south coasts. The artifacts removed from these
stratified deposits indicate that two distinct cultures are represented.
The first part of this paper is a description of the excavations and
collections which illustrate these cultures, while the second part is
concerned with the distribution throughout the West Indies of the
complex of traits characterizing each of them.
An extended archaeological monograph of Porto Rico will not be
included here. No attempt has been made to correlate the artifacts
removed from the excavations with the artifacts in the large and
numerous Porto Rican collections in museums and in the possession
of private collectors. Furthermore, a description of the numerous
sites known throughout the island is omitted as not pertinent to the
object of this paper. Descriptions and illustrations of individual
objects have been limited more than might seem advisable, but
emphasis has been placed on the description of common and char-
acteristic types of the material excavated. Obviously, a considerable
range of variation appears within a group of objects classified as a
type and this variation is indicated to some extent in the descriptions,
but a general classification of material is necessary in presenting a
concise description of several thousand specimens. Furthermore, the


generalized type becomes more significant than the individual object
when correlating the artifacts excavated with those found throughout
the West Indies.
The unpublished manuscript of Dr. S. K. Lothrop which locates
one hundred thirty-eight archaeological sites on the island, and the
work of many collectors who have amassed material from numerous
sites, open (lithe way to research in Porto Rico which should extend
over a period of years. The objective of the work inl 1934 was a
stratigraphie study of large deposits which might make it possible to
interpret the relation of various sites and the significance of the
thousands of specimens in I lhe collections. It is hoped that the tenta-
tive cultural chronology indicated by these excavations and described
in this paper may give direction and assistance in further and more
comprehensive research throughout the island.
I wish to express my gratitude to the many individuals and institu-
tions who have contributed to this research. Mr. Allison V. Armour
of New York made possible with his research yacht, the "Utowana,"
the general survey of the Bahamas and Haiti preceding the work in
Porto Rico, and has at all times given support and advice. To Dr.
('lark Wissler I am deeply indebted for the opportunity of under-
taking research in Porto Hico and for his advice throughout the course
of the work. To Dr. Cornelius Osgood I am also indebted for this
opportunity and for his criticism of this report. Dr. Carlos E. Char-
don, Chancellor of the University of Porto Hico, and the men of the
Agricultural Extension Service of the University, supplying transpor-
tation and knowledge of the country, made possible a survey which
might otherwise have taken months. General Bhlaton Winship,
(Governor of Porto Iico, and Colonel Francis Riggs, Insular Chief of
Police, evidenced a real interest in the work and solved many difficult
problems for me. Major Adolfo de Hostos, who has studied Porto
Rican prehistory for many years, contributed his knowledge of the
field and also proof-read this account. Dr. Montalvo Guenard of
Ponce, Porto Rico, located numerous sites and cooperated in the ex-
cavation of one which proved most productive. To such men as
Francisco Parra-Toro, Luis de Celis, Gustavo Rodriguez, Arturo
Morales Carri6n, D)ean .J. J. Osuna, acting Chancellor of the Univer-
sity, and the men of the Fajardo Sugar Company, I am deeply grate-
ful for the many iind varied ways in which they gave their assistance.
Finally, the accompanying pages were submitted to Yale University
in partial fulfillment of the requirements for the Degree of Doctor of
Philosophy, 1935.


Excavations in Barrio Canas
C('aas is a barrio of the municipality of Ponce. It lies in a fertile
plain between the mountains and the sea, on the southwest Porto
Rican coast which is now one of the most productive sugar cane
regions of the island. The southern coast is semi-arid, and the cane
fields are irrigated. The climate is always moderate and a tropical
xerophytic vegetation grows in areas which are not under constant
Colonia 1Miramar is a section of Barrio Canas within which lies a
field called "sitio de earacoles" or "place of shells." For many years
collectors of Porto Rican antiquities have taken objects of prehistoric
manufacture from this field as each year of cultivation uncovered
more of the refuse deposit. At the present time marine shells are
scattered over an area of several acres. In sections where the shells
are thickest low moutinds can be seen, suggesting that the refuse
was originally deposited in mounds of considerable size which have
been reduced and leveled off during the last four centuries of constant
Adjoining this field and extending along the banks of the Hio Canas,
is a small plot. of land which is at present overgrown with brush and
trees. It is probable that no great amount of cultivation has ever
taken place in this section as several mounds of considerable elevation
still remain. Four of these appear to form a unit since they are con-
nected by low grades or ramps. The largest mound has an elevation
of 2 meters and extends in a roughly oval shape over an area ap-
proximately 65 by 30 meters. The western end has been partially
cut, away by the river and falls sharply from the crest to the river
bed. Joining this at the opposite end is a roughly round mound
20 meters in diameter which rises abruptly to an elevation of 2.5
meters. On the east side of the latter is a still smaller mound with an
altitude slightly over one meter. The fourth mound lies on the river
bank south of the largest mound and connected with it by a low ridge.
This last mound approaches 25 meters in diameter and is 2 meters
Culture refuse made up essentially of marine shells and potsherds
extends over the whole area which includes the group of four mounds,
others in the wooded plot and the shell-strewn area in the cultivated
fields. No attempt was made to map the whole area of occupation
since this would require the removal of most of the trees and brush in
the wooded section, for which permission could not be obtained.


Excavation was confined primarily to the largest mound but test pits
were made in two of the others.
The extent of the largest mound, designated as A on the chart, was
such that its complete removal was impossible in the time allowed.
In order to obtain as much information as possible concerning its
composition as a whole, trenches were cut across both ends and these
were connected by a longitudinal trench running the full length of the
mound. A second longitudinal trench was begun but was not en-
tirely completed. A crew of twelve men was set to work at the
eastern end in what has been termed excavation No. 1, and at the
same time twelve more were started at the western end in the excava-
tion referred to as No. 2. Five weeks were spent in working the
Canas site.
All excavation in the Canas midden was carried out by working
down from the surface in the series of trenches shown on the chart
(FIGURE 2). These trenches were marked off in sections 2 meters
square and a team of two men worked in each section, one shoveling
while the other examined the soil thrown out of the cut. All refuse
was removed in levels 25 centimeters in thickness and the specimens
from each 25 centimeter level in each square were sorted, labeled with
the section and depth (i.e. A-I, 0-25 centimeters), and packed in a bag
as a separate unit.
Excavation No. 1 comprised two parallel trenches 28 meters in
length which cut across the northeastern end of the midden, and two
trenches extending 30 meters through the long axis of the midden in
the highest central section. An undug ramp 2 meters in width re-
mained between these longitudinal trenches and thus provided four
parallel faces or cross-sections which could be examined and charted
in order to determine any stratification of refuse should such exist.
The composition of refuse was largely the same throughout all
sections removed in excavation No. 1. Although primarily made up
of marine shells (Strombus, Tellina, Chama, Neritina, Cardium and
Donax), it contained a large quantity of sand, blackened earth, ash,
charcoal, and bone refuse. Near the surface, usually to a depth of
approximately 25 centimeters there was a greater amount of humus,
and shells were broken in small pieces, but below this most shells
were intact and the soil intermixed was loose and powdery. In
trenches No. I and No. 2 cut across the midden, no consistent strati-
fication of refuse appeared. In limited sections there were concentra-
tions of one or two types of marine shells, earth, or ash and charcoal,


FIGURE 2. Chart of Barrio Canas site.

but no well-defined strata and no distinct fire pits or hearths. In
longitudinal trenches F and H, however, the upper levels of the de-
posit contained primarily large marine shells (Strombus, Tellina, and
Chama) while the lower levels contained small marine shells (Neritina,
Cardium, and Donax). This lower level of small marine shells is


shown on FIGURE 3 (page 13). No sharp line dividing the deposit
could be seen, and the distinction is made simply on the preponder-
ance of large or small marine shells.
The only indications of house structures found in the deposit were
cylindrical cavities approximately the size of fence post holes which
may mark the location of disintegrated house posts. No alignment of
such cavities was found which might have indicated the size and
shape of house structures. One of the cavities, globular in shape and
75 centimeters in depth, may indicate a disintegrated storage basket.
Potsherds make up the bulk of the artifacts found in the excavation
and these were so numerous that several thousand were removed in
a single day. The great majority were fragments of coarse, undecor-
ated pottery, but a small percentage bore incised designs, modeled
ornamentation, or a red slip. Only large rim fragments and orna-
mented sherds were retained during the excavation. A few vessels
were found intact. Implements made from shell, stone, and bone
were found also but these were relatively rare compared to the num-
ber of pottery fragments. Collections of bone refuse from this de-
posit are made up largely of fish, bird, hutia and mianati bones. They
have not been studied and identified in detail.
The fifteen burials found in the four trenches of excavation No. I
appeared at depths varying from 75 centimeters to 1.75 meters in
several different sections. All skeletons were found lying in culture
refuse but none was associated with mortuary artifacts. The bones
were so badly disintegrated that few of them were removed intact
and no skull could be preserved for study. In several cases, however,
the original position of interment could be determined. The descrip-
tion of each burial is given in the appendix (page 190). Some of the
bones probably represent secondary burial of skeletons after the flesh
had been removed, while others are certainly primary burials of the
entire body. In most of these primary burials the bodies were interred
in a flexed position either on the left side, the right side, or upon the
back. At least one, however, was placed on the back in an extended
position, and some may have been deposited in a- sitting posture.
The four trenches of excavation No. 1 exposed an undisturbed refuse
deposit composed primarily of marine shells in which no significant
stratification could be seen. The composition of this refuse and the
artifacts which it contained are characteristic of numerous shell heaps
found along the southern coast of Porto Rico. Large quantities of
coarse, undecorated pottery and certain types of shell and polished
stone implements are typical. This northeastern part of the midden


at Canas is unusual only in its extent, depth, and undisturbed condi-
tion. The great majority of Porto Rican middens have been leveled,
to a large extent, by centuries of cultivation.

The southwestern end of mound A at Canas has been partially cut
away by the river Canas which is now simply a dry stream bed, and
there is a sharp decline from the crest of the mound to the bed of the
river. Excavation No. 2 was begun with a trench 28 meters long and
2 meters wide which cut directly across the midden at the edge of
this abrupt decline. Subsequently, two parallel trenches 2 and 3
(FIGURE 2) were also excavated but an undug ramp connecting with
that in excavation No. 1 was left intact until the last so that a longitudi-
nal cross-section extending the full length of the mound could be
charted. A single trench F running with the long axis of the midden
connected with excavation No. 1 while three more trenches, E, H, and
I, parallel to F, remain unfinished. As in excavation No. 1, all trenches
were divided in sections 2 meters square and all specimens were
segregated according to section and level.
In the first cut (trench A) in excavation No. 2 the deposit of large
marine shells, dry sandy soil, ash, and coarse pottery, like that found
throughout excavation No. 1, extended to a depth of only 40 to 90
centimeters. Below this we found a hard-packed yellow material
which appeared to be sterile sub-soil. As the excavation was con-
tinued, however, this sub-stratum proved to contain numerous pot-
sherds as well as occasional stone implements. Many of these pot-
sherds were decorated with brilliant red and white painted designs,
and practically all sherds found were hard, fine-grained, well-fired
ware representing what appeared to be an unusually advanced tech-
nique of pottery manufacture. The great majority of the sherds as
well as the occasional stone implements stood in sharp contrast to
anything found in the northeastern part of the midden and to the
sherds in numerous Porto Rican collections found in shell middens
throughout the island. Dr. M\ontalvo-Guenard, of Ponce, however,
had found a few sherds of this kind in a field near the Canas midden.
As excavation in the southwestern end of the midden continued
toward the center, the upper level of marine shells and coarse pottery
increased in depth and the yellow, hard-packed sub-stratum re-
mained, sharply segregated from the upper, loose, shifting, marine
shell deposit. The composition of this sub-stratum, which varied in
thickness from 40 to 60 centimeters was eventually recognized as a


mass of disintegrated land crab shells (Cardisoma guiahumi Latreille)
mixed with a yellow sandy loam. Ash and charcoal were also present
but in smaller quantities than in the marine shell level above. Bone
refuse was also common and like that in the upper level represented
principally the manati, hutia, turtle, fishes, and birds. As represented
on the chart (Flounj 3) this sub-stratum of land crab shells and
painted sherds extended from the bank of the river approximately 30
meters toward the center of the midden and then disappeared at a
point where the upper level of marine shell deposit reached a depth of
2 meters.
Of the nine burials found in excavation No. 2 (see appendix for de-
tailed description), seven appeared in the sub-stratum of land crab
shells. All of these were so far disintegrated that no skeleton could
be preserved for study. In only three cases was it possible to deter-
mine the position of interment, but in these cases it was clear that the
bodies had been placed in an extended position, two lying on the face
and one lying on the back. Although a comparison of burial types in
the marine shell level and the land crab level is hardly possible with so
few examples, it may be significant that all three identifiable in the
sub-stratum were in the extended or prone position while the great
majority of those in the upper marine shell level were in a flexed
position. Furthermore, no secondary burial of bones alone, quite
certainly a type of burial associated with the upper marine shell
level, was observed in the sub-stratum of land crab shells.

When excavation No. 2 was terminated after five weeks of excava-
tion at the Canas site, it was clear that two periods of occupation were
represented in the large midden by the two distinctly separate strata
of culture refuse. There was no sterile layer of soil between these two
strata which might represent a lapse of time between the two periods
of deposition, but this absence does not necessarily indicate a con-
tinuous occupation of the site as illustrated by the fact that the ac-
cumulation of a sterile layer of soil has not yet taken place on the
most recent deposit, now at least four hundred years old. The lower
stratum of yellow soil and disintegrated land crab shells appeared as a
relatively hard-packed mass always sharply separated from the loose
and shifting upper stratum of sand, ash, and marine shells. Approx-
imately the same types of bone refuse (manati, hutia, turtle, fishes
and birds) occurred in both strata and indicate that the people occupy-
ing the site during the two periods had certain foods in common, but


w.L.^ LAeR M.r -cr9

(P355 RE~rIN 'CR I



5 L LE-~- O L C I I -e .' '

S 2 aN N ^ o 7 s-A r OAN G -nO-3 9 0 99 .i* oi on 313 a-. 3 tM9 t aW d 39 3 0-7 S3 M 505 9 a0 3 4 a03 0

FIGURE 3. Cross-sections of Barrio Canas excavations. A, trench F, excavation 'No. 2. Showing opper marine shell deposit and lower land crab
shell deposit, at Canas. B, trench H, excavation No. 2. Showing two levels of deposition, at Canas C, flexed burial lying on back. In the last of
the marine shell deposit in trench It, excavation No. 1, Canas.


EP LE 7. F' FTs


the staple of the early inhabitants was certainly land crabs while
that of the later inhabitants was largely itmarine shellfish.
It would appear from the excavation completed that the late oc-
cupation extended over a longer period and a wider area than the
earlier. No refuse of the early period was found below mounds B and
C iii the test pits made, and in mound A it was confined to an area 30
meters in diameter along the river bank. The maximum( depth of
refuse from the early period was 60 centimeters while the late period
refuse in many sections reached a depth of 2 meters. Red and white
painted sherds, however, have been found along an irrigation ditch
inii a field about .5 kilometer from mound A, and it is probable that
they indicate another site occupied during the early period.

Collections from Barrio Canas
Since the site iii Canas was essentially a deposit of refuse, the
artifacts found were for the most part fragmentary. The great major-
ity were potsherds which appeared in large numbers scattered through
the food refuse. Complete vessels were very rare, although large
fragments were common. Implements made of shell and of stone
were comparatively numerous, but the majority of these also were
broken. Ornaments of shell and stone were very rare and were never
found with the burials encountered in the deposit. Implements and
carved figures of bone were limited to a few unique specimens.
Wheni the collections were arranged in the laboratory according to
their depth in the refuse deposit, it was clear that there was a sharp
distinction between the artifacts found in the early deposit of land
crab shells and those found in the late deposit of marine shells. During
the excavation of the site the two levels of occupation were recog-
nized when they were first encountered, and the material from each
level was segregated when packed and labeled, so that it has been
possible to treat the artifacts from each level as a unit. In the follow-
ing descriptions a classification of the material from each level is
presented. The classifications of pottery arc made on the same
general basis of vessel shapes, types of ware, and aspects of ornamen-
tation, but the distinctions between the pottery of each level are such
that the systems of classification do not conform in many details.
The term Shell Level is used ini the following sections to designate
the upper or recent deposit of mnarimne shells, and the term Crab
Level is used to designate the lower or older deposit of land crab
The summary which follows the descriptions of Shell Level and


Crab Level artifacts will serve to point out the differences and the
similarities where such exist.

The 2915 sherds which make up the collection of pottery fragments
from the shell heal) proper at Canas arc actually a small proportion
of the total number removed during the excavation, since all small
and undecorated fragments were discarded. The great bulk of the
pottery found was a crude unpainted ware, homogeneous in color and
composition. Only large rim sherds, lugs, handles, and the relatively
rare fragments of this crude ware which were ornamented with in-
cised or modeled decorations, were retained for analysis. A small
number of the sherds in the deposit were of a red slipped ware and
because of their rarity the majority of such sherds were also retained.
Thus on the basis of color the sherds in the collection fall into two
major groups which have been termed Crude Ware and Red Ware.
CRUDE WARE.-Fragments of cooking pots or service wares indi-
cate that the bulk of the pottery used was neither slipped nor painted,
and was as a whole crude and poorly made. The large part, or 2662
of the 2915 sherds in the collection, are of this type. The ware is
dominantly brown but varies in shade from a reddish to a dark fire-
blackened brown. It is sand-tempered but thick, coarse, and poorly
fired. Surfaces are generally rubbed smooth but are never polished.
Vessels of crude ware are as a whole irregularly formed and have the
appearance of being hastily or carelessly made. The majority of the
sherds are fire-blackened in part and suggest that crude ware vessels
were used for cooking. Pottery of this type was occasionally orna-
mented with incised patterns, or with modeled figures, either in
relief on vessel walls or in the round on rims and handles.
RED WARE.-Only 253 fragments of Red Ware were taken from the
deposit and these represent practically all that were found. The dis-
cards during the excavation were Crude Ware fragments, so that no
exact percentages of Red and Crude Wares can be given. However,
it is certain that Red Ware made up less than 5% of the pottery
Red Ware is distinguished by a clay slip which varies from a pink
to a light red. The slip is present either on the inside or the outside,
or on both sides. Surfaces of Red Ware fragments are generally rub-
bed to a dull smoothness but never to a bright polish. Fragments of
Red Ware vessels indicate that pottery of this kind was more care-
fully made than Crude Ware. This is shown in the regularity of


outline, and the finish of surfaces and modeled ornamentation. The
same types of incised and modeled ornamentation, however, appear
on both Crude and Red Ware. As far as can be determined vessel
shapes are also the same. Red Ware vessels were used as cooking
pots since many fragments show the fire-blackening, but this is less
common than on Crude Ware fragments.
The only real distinctions between Red Ware and Crude Ware
pottery are the red slip and the more careful finish. Therefore, the
color of ware and the presence or absence of slip have been omitted
from the following classification of Shell Level pottery.
THE BASIS FOR CLASSIFICATION.-Elements of ornamentation are
as a whole comparatively rare on pottery of the Shell Level. Painted
decoration is found on only six sherds. Incised decoration is present on
only 35 sherds out of the 2915 retained, which is in turn only a small
part of the total number in the deposit. Relief decoration on the walls
of vessels is likewise very rare since there are only 31 examples.
Modeled figures on rims and handles are more common but still rare,
with only 158 examples. Such small percentages of ornamented
sherds are obviously not adequate for a classification of sherds on the
basis of ornamentation alone. A classification of rim types is of little
significance since the great majority are simple vertical or incurving
Complete vessels are very rare. Only 10 vessels were found com-
plete, or nearly so. However, the majority of sherds retained are
large fragments of the characteristic boat-shaped bowls which are
easily identified by the handles and lugs. Round bowls and dish forms
can in turn be distinguished from the boat shapes. Thus a classifica-
tion of the pottery on the basis of shapes is possible, although subject
to some degree of error. This classification on the basis of shapes is
supplemented by another classification of the elements of ornamenta-
tion, confined to incised designs and modeled forms.
FRAGMENTS (FIGURE 4).-Shape A: The most common type of vessel
found in the Shell Level is a deep boat-shaped cooking pot with broad
vertical loop handles. These pots are generally about twice as long
as they are wide. They have the form of a boat with a high bow and
stern. The handles are formed at each end of the vessel, and have
the appearance of an extension of the rim which has been folded back
down the wall of the vessel to form a flat loop. All pots of this type
have a small, flat, round base. The rims are vertical and form a line
or ridge in joining the rounded body about the circumference.






..... ... :Y;.....(1


------ ---- --~----------------




FIGou i 4. hell I evel vessel shapes.


Three pots of this type are practically complete. The smallest is 16
centimeters in longest diameter and 7 centimeters high. The largest
is 34 centimeters in longest diameter and 19 centimeters high. The
424 handles taken from the Shell Level can definitely be identified as
fragments of shape A, and 892 large rim sherds can be associated with
this type of deep boat-shaped pot due to the characteristic curvature
of the rim.
The majority of fragments of this type are crude undecorated ware,
but there are a few examples of Red Ware, and also some fragments
ornamented with rude modeled forms which will be described in a
following section.
Shape A-1: A variant of shape A is a pot of the same general shape
and size which is distinguished by vertical ridge lugs in place of broad
loop handles. Fragments of shape A-1 are much less common than
those of shape A, with only 42 identifiable sherds. The vertical lugs
of shape A-1 are formed in the same position as the loop handles of
shape A at each end of the pot. The lugs rise gradually from the
walls of the vessels to form a ridge 1 to 3 centimeters high. Ordinarily
the lug extends over the lip of the vessel, and in rare cases terminates
in a round flat knob. In a few examples the lug is perforated to form a
true handle. Vessels of this shape have no modeled ornamentation.
Shape B: There are 47 fragments which can be identified as shallow
boat-shaped bowls or dishes with rectangular lugs. Two nearly com-
plete vessels of this type make it possible to determine the shape
quite definitely. The bowls are elongate, and may be described as
boat-shaped although the ends do not rise sharply as they do on the
pots of shape A. As a whole the vessels are much smaller and gener-
ally not over 22 centimeters in greatest diameter. As far as can be
determined, the bases are round, rims are curved, and no ridge
appears about the circumference. The lugs as a rule rise from the lip
at the same angle as the vessel wall. Generally the corners of the lug
are slightly drawn up, producing a somewhat concave upper edge.
Shape B-1: Seven fragments indicate a shape which may best be
termed a variant of shape B. It is determined by roughly semi-lunar
lugs on shallow oval, or boat-shaped, bowls of the same form as shape
B. The semi-lunar lugs take the place of rectangular lugs on vessels
of the same general shape and size. The semi-lunar lugs are rarely
perforated to form a handle for suspension. Shallow bowls or dishes
of shapes B and B-1 are apparently not ornamented in any way. They
are as a rule made of Crude Ware, but there are a few examples of
Red Wanre.


Shape C: Boat-shaped bowls with modeled head lugs are represented
by 47 fragments. Many of these fragments are large enough to
indicate the prevalent form, which is that of a shallow oval or boat-
shaped vessel of the general form and size of shape B. Modeled
heads are evidently formed at each end of the vessel as an extension of
the rim. Vessels of this type may be either Red Ware or Crude Ware.
As a whole, they appear to be more carefully made than shape A, and
less often show use as cooking bowls, being probably too small and
shallow to be of use on the fire. The various types of modeled heads
will be described under the heading, "Modeled head lugs on rims."
Shape D: Round bowls and dishes are suggested by a number of rim
sherds which are not large enough to indicate the complete vessel
shape. Some 543 sherds have been grouped in this general classifica-
tion as representing vessels with a round rim in contrast to the custom-
ary oval rim of the boat-shaped vessels.
A large number of these sherds grouped as shape D probably are
fragments of a wide-mouthed shallow bowl or dish. Rim sherds of
this type often show an enlarged lip or edge which turns sharply in
toward the interior of the vessel. Obviously such a loose classification
as round bowls and dishes is very inadequate, but it does have some
meaning in that it sets off a group of vessels which are round as
opposed to the great majority which are of the characteristic boat-
Shape E: Thick clay platters or griddles, so common in the West
Indies as a whole, are abundant in the Shell Level deposit. No com-
plete griddles were found, but fragments indicate circular, disc-shaped
platters of a great variety of sizes. These are composed of a very
coarse, poorly-fired clay, and are as a rule very fragile. In thickness
they vary from 1 to 4 centimeters at the rim and are always much
thicker at the rim than at the center. A noticeable characteris-
tic of the majority of these griddles is that the upper side is rubbed or
smoothed while the lower side is very coarse and rough. Samples of
griddle rims were taken from each level of each section during the
excavation, but no attempt was made to retain all fragments.
There are a very few examples of griddles which are ornamented
with deep, rough parallel line incisions about the rim. A simple
design or pattern is formed, and the incisions are restricted in all
cases to a border around the edge of the griddle.
Miscellaneous rare shapes.-There are a few sherds which can be
recognized as fragments of bowls representing either unique or com-
paratively rare shapes.


Shape f: Vessels with spouts arc indicated by 3 rim fragments
with tubular spouts about 4 centimeters long. In all three cases it
can be determined that the spouts were attached approximately 1
centimeter below the lip and at right angles to the wall of the vessel.
The rim fragments suggest shallow round bowls but the complete
shape cannot be determined.
Shape g: One complete bowl is a unique specimen as no fragments
of similar vessels were found. It is a shallow bowl somewhat more
round than boat-shaped. It has a flat base and a semi-spherical body
terminating in a vertical rim. On one side, at the rim, is a loop handle
and at the opposite side is a hole 2 centimeters in diameter which was
apparently used for pouring. The pot is unusually crude and irregu-
larly made.
Shape h: One complete and one partially complete bowl represent
a rare shape. They are small round bowls which round out from a
flat base to the equator and then rise vertically to the lip, thus com-
bining a vertical rim with a semi-spherical body and a flat base.
Shape i: A single vessel which is practically complete rests on a
flat round base and rises gradually to the equator, at which point the
walls turn sharply in to form a restricted mouth. No fragments sug-
gesting this shape were found.
Shape j: Bowls of a shape characterized by a peculiar indentation
of the rim can be recognized from 8 sherds. A complete bowl of
this kind was found in Barrio Monserrate. It appears as if, when still
soft, the vessel had been picked up with a pair of tongs which left a
small groove, indentation, or fold at the lip on each side.
Shape k: Vessels with restricted necks are represented by 4 sheids.
Very little concerning the complete shape of the vessels can be deter-
mined beyond the fact that jar forms with collars or necks are present
but rare.
Shape 1: Bowls partitioned off into two compartments can be
determined from 3 examples. One was found practically complete
and is a shallow boat-shaped bowl with rectangular lugs. A partition
is modeled through the center extending from one end to the other.
Much of the partition has been broken out and it is not possible to
tell whether it extended the full depth of the bowl.
SHERDS (PLATE 1).-Modeled head lugs on rims: Modeled heads, or
"adornos," are the most outstanding aspect of ornamentation in all
the Shell Level material. There are 68 examples of this type of
ornament applied to rims. They appear most commonly on sherds


which can be identified as belonging to shallow boat-shaped bowls,
but there are a few sherds which are fragments of round bowl forms.
Invariably they are applied to the peaks, or raised ends, of the boat-
shaped vessel. No complete pots with modeled head lugs were
found, but it is presumed that in most cases lugs were applied at each
end. On the basis of the figure represented, these rim adornos have
been divided into 6 classes. All represent some kind of zoomorphic
Class a. The so-called bat-heads are the most common identifiable
type. They are characterized by bulging eyes shaped like doughnuts,
protruding snouts with nasal perforations, and crests or head-dresses
formed by deep grooves. Often a modeled form below or around the
chin is present, which may represent folded arms or wings. The crest
and head-dress are at times modified to forms which may represent
ears. The features as a whole are made up of knobs and deep grooves.
Adornos of this class are found in most of Porto Rico and have become
known as one of the most typical forms of ornamentation on Porto
Rican pottery. There were 21 examples of this class found in the
Shell Level. Of these, 16 face the inside of the vessel and 5 are placed
so that they face along the rim, 2 toward the left and 3 toward the
right. They are found on Crude Ware as well as Red Ware vessels.
Class b. A group of 10 of these modeled figures can be identified as
human representations. Human-heads are smaller and much more
simple than the bat-heads. They are characterized by a prominent
and well-formed nose, and by eyes and mouth formed by a slit or per-
foration. No other features are represented. In simplicity and
crudeness they contrast sharply with the bat-heads. All of the 10
examples face the interior of the vessel. In most cases these human-
head lugs are flatter than the bat-heads which are always in complete
Class c. Three heads look remarkably cat-like. They have been
set aside as a separate class because they are identical and markedly
distinct from other heads. The eyes are sunk deep in the face and
outlined by rings of clay. The nose protrudes in a snout. Sharp-
pointed ears stand straight up on the head. It is possible that these
heads are a variant of the bat-head class, but the appearance as a
whole is quite distinct. The three face the interior of the vessel.
Class d. Four heads can be recognized as bird figures. Each one is
to some degree distinct, but a common trait is a well-formed beak.
One example represents the whole body, while the other 3 are forms of
the head alone. All face the interior of the vessel.


Class e. Pelican-heads are clearly represented by 2 specimens.
Both examples are applied to Red Ware shallow boat-shaped bowls,
and are identical except for size. In each case, two pelican-heads are
joined facing in opposite directions along the rim of the vessel. The
long thick bills are brought down against the neck in the position
which the pelican so commonly assumes. Eyes are well formed on
each side of the head.
There are 27 modeled heads of zoomorphic beings which have been
left unclassified. Each one is unique and cannot readily be associated
with any known creature. Three of these may represent the turtle,
but have no distinct traits in common. Two others are probably
meant to represent humans, but are distinct from the usual human-
heads. Bulging, doughnut-shaped eyes, and eyes formed by slits and
punctures, are present on many examples of the unclassified group.
There are 13 which face the inside of the vessel, 12 which face the
outside, and 2 which look along the rim.
Modeled forms on loop handles: A distinct aspect of ornamentation is
found on 76 broad loop handles which are clearly fragments of the
deep boat-shaped bowls, shape A. This ornamentation is carried out
by the addition of modeled forms on the top of the handles. Appar-
ently the figures were formed by drawing up the clay while the handles
were being made, or in many cases by applying strips or knobs on the
finished handle and modeling them as desired. Such ornamentation
is regularly found on the Crude Ware cooking pots, but also in rare
cases on Red Ware. Decoration of this kind can most conveniently be
divided into two classes on the basis of the type of figure.
Class a. Geometric figures are the most common, numbering 57
examples. Spirals, scrolls, ridges, loops and car-like appendages
predominate. The ear-like motives were evidently formed by drawing
up the edges of the handle. In many cases the ears are impressed with
lateral or transverse grooves, or pinched together to form a groove.
Knobs of clay applied to the handle are impressed with crosses and
grooves, or drawn out to form a straight or curved ridge. Rarely an
S-shaped figure appears in high relief lying over the top of the handle.
No single type of geometric figure predominates and the majority of
examples are unique. Scrolls, ridges, and loops in the center of the
top of the handle may be combined with car-like elements at the edges.
Class b. Zoomorphic heads modeled on loop handles are less com-
mon than geometric figures. There are only 19 examples present.
All such heads are of the same type, and are unusually small, crude,
and simple. They were applied at the middle of the top of the handle,


and formed by simply impressing a small knob with grooves and
punctures to form the eyes and mouth of the head. In some cases a
prominent nose has been pinched out between the punctures represent-
ing the eyes. The majority of these heads are not over one centimeter
in height. Most of them probably represent bat-heads and human-
Geometric figures are at times combined with the zoomorphic
heads. Of these, ear-like appendages at the edges of the handle are
the most common, but curved ridges also appear on the same handle.
Modeled figures in relief on vessel walls: Ornamentation of vessels
by the application of figures in low relief was not common, since only
31 sherds carrying such ornamentation were found in all the Shell
Level. Nevertheless, on pottery which is rarely decorated in any
way, this limited number of sherds forms a distinct class. So far as
can be determined, relief decoration was generally applied to round
bowls, and always near the rim. All figures of this type were crudely
executed. They appear on both Crude Ware and Red Ware sherds,
but no complete vessel ornamented by relief figures was found.
Figures which commonly appear in relief can be grouped in four classes:
Class a. Figures which can best be described as sigmoid scrolls.
These appear in low relief, roughly parallel to the rim and near the lip.
They arc composed of a series of joined curves, a section of which
resembles a capital S lying on its back. The number of curves is not
determined since all examples of this motive are found on sherds, but
it is clear that the scroll did not extend in all cases entirely around the
Class b. Figures which are closed spiral scrolls. In no case does the
spiral make more than two turns about the inside end. The whole
figure is rarely more than 4 centimeters in diameter. Spiral scrolls
are in low relief, and also on the outside wall of vessels near the rim.
Class c. Figures which may be described as worm-like. They form
a half circle near the rim of the vessels, with one end of the half circle
always beginning at the lip. The worm-like appearance is emphasized
by a series of transverse grooves impressed over the length of each
figure. In many cases two such figures are found together with the
convex sides facing.
Class d. More elaborate relief decorations are indicated by several
sherds which carry portions of a zoomorphic figure in low relief.
Heads and arms or feet and legs appear together. There are no cases
of a complete body. It is very probable that the feet and legs were
applied on the wall of the vessel at one side and the head and arms on


the other side, which is not unusual in the so-called Arawak pottery
of the Greater Antilles. Examples of this class are so rare and frag-
mentary that no conclusion can be reached as to the prevalent type of
figure represented, but one specimen is the familiar bat-head so com-
mon to Porto Rican pottery. Bat-heads have been described among
the modeled head lugs on rims.
Incised decoration: Sherds ornamented with incised lines are
surprisingly rare in the Canas site. Decoration of this kind has long
been associated with West Indian ceramics, but only 35 sherds were
found bearing incisions. As far as can be determined from the sherds,
incised ornamentation was generally applied to round bowls and not
to boat-shaped vessels, but the fragments are small and the complete
forms remain uncertain. Designs may be grouped in three classes.
Class a. Horizontal parallel lines in a panel near the rim, often
terminating in a pit or puncture, form the most common design pat-
tern. Incisions are rude and often irregular, and were applied before
the firing. In some cases the parallel lines curve in to connect, thus
forming a long rectangular figure with rounded ends. The puncture
or pit, which appears to have been applied with a sharp awl, is a
characteristic element of this design. A series of parallel lines were
often used in this type of pattern, but apparently they were confined
to a space within 3 or 4 centimeters of the lip of the vessel. Class a
designs are generally found on the outside of the rim, but a few ex-
amples are present which indicate that the incisions were applied
within the rim of what must have been an open bowl.
Class b. Parallel lines in curvilinear patterns on the outside of the
body are less common than the horizontal rim incisions. Sherds
bearing patterns of this kind are so rare and fragmentary that no
adequate characterizing description can be given.
Class c. A series of vertical parallel lines forming a band about
the rim of vessels is a distinct pattern. This type of incised design is
even more crude than those mentioned above. It generally appears
on bowl forms of Crude Ware. Apparently the band of incised lines
does not extend around the rim of the vessel. This pattern is often
associated with the worm-like relief figures.
Painted decoration: Ornamentation with paint is limited to 6 sherds
which are all fragments of shallow bowls or dishes. A red paint,
which appears to be the same as the slip applied to Red Ware vessels,
has been added to the lip of these bowls to form a border, and simple
curvilinear designs have been applied to the interior. None of the
fragments are large enough to give an idea of the complete design but


all the sherds of this group bear simple red lines which are segments of
curvilinear elements. These painted designs are applied to the usual
Crude Ware vessels. The 6 sherds of shallow bowls or dishes orna-
mented with a simple red line pattern are significant in the Canas site
since a large number of the sherds removed from the middens at
Monserrate are of vessels of the same shape ornamented in the same

Artifacts of shell and stone were numerous in the Shell Level de-
posit though much more rare than potsherds (PLATE 2). Tools of
shell and stone make up the bulk of associated artifacts, and of these,
shell tools predominate. Of a small group of ornaments, about half
are shell and half stone. Bone artifacts are limited to a few awls or
needles, carved figures, spatulas and problematic objects.
SHELL IMPLEMENTs.-Chisels: The most prevalent shell tool re-
sembles a chisel. The tool is in every case made from the lip of a
Cassis tuberosa, and the row of knobs or ridges which are present on
the shell remain on the body of the chisel, as if intended to aid in
gripping the implement. It is dagger-shaped, and slightly curved
with the natural curve of the shell. The length varies from 10 to 15
centimeters, and the width from 2 to 4 centimeters. The thickness
depends largely on how well the tool has been ground down. Many of
them are very rough, and a flange which is a remnant of the body of
the shell remains on the side. Many of the chisels are only roughly
pointed and have a comparatively blunt end. Others have been
rubbed down to a sharp curved bit or to a polished point. The poll is
usually left rough but in some cases is smoothed off to form a blunt or
rounded butt. Only 1 specimen has been ground until all traces of
the shell surface have disappeared. There were 32 examples of this
type of tool found.
Hoes: Tools roughly shaped like the sole of a shoe can probably
best be classed as hoes. Cut from the Stromnbus goliath, they have a
slight curve of the shell. They are cut only around the edges and the
flat surfaces retain the undulations of the natural shell. They are
never polished or rubbed, and the edges remain jagged or rough. The
working end is rounded like the end of the sole of a shoe and is never
ground down to form a sharp bit, but remains the thickness of the
shell. No evidence of hafting is present. In size they vary from 15 to
18 centimeters in length, and 6 to 8 centimeters in width. The thick-
ness is generally about Y to 1 centimeter.


Celts or axes: Celts are very similar in form to the hoes, except that
the working end has in all cases been ground down to a fairly sharp
curved bit. Many of them have the shoe-sole shape, and they are
invariably made from the Strombus goliath shell as are the hoes. The
surfaces are unworked but the edges are occasionally ground smooth.
A few examples have a roughly pointed poll while others are to some
extent rectangular. All have a curved bit. Lengths are 11 to 14
centimeters, widths from 4 to 8 centimeters; the thickness of the
body is generally from 1 to 1 centimeter. The celts are less common
than hoes, as there are only eight examples.
Unclassified shell tools: Forty-four worked implements roughly the
size of celts and chisels have been left unclassified as their use is un-
known and they do not fall into distinct types. Some of them are
roughly semi-lunar and about 14 centimeters long. Like the celts,
they are cut from Strombus shells and left unpolished. There are no
cutting edges or points. Two specimens are dagger-shaped with
something resembling a hilt, but the point is blunt. Of these one is
small and quite well finished, the other is large and rough. One ex-
ample has the outline of the three-pointed ceremonial objects, but is
flat like the celts or hoes. Another is a long, slender, blade-like
implement with well-finished edges, but it also is blunt. Of the 44
examples, no two are identical. Some may be tools in the process of
manufacture, but others are evidently finished implements of some
sort. It may be that some were used as net measures, and yet none
are notched. They have not been termed ceremonial as is often cus-
tomary, because none is carefully made or finished.
Shell discs: Of the 8 disc-shaped artifacts, 5 are rough, irregular, and
unpolished. The convolutions of the shell still appear on the surfaces
and the edges remain jagged or rough. One has a small perforation
near the edge. The diameters are 6, 8, 9, 9, and 91/ centimeters. The
3 remaining have been partly polished on both surfaces and the edges
have been ground smooth. One is perforated at the center, while
another has a perforation at the center and also a series of perforations
circling the rim. These rim perforations are spaced regularly ap-
proximately 1 centimeter apart. The use of these discs remains un-
certain. The discs with holes at the center cannot be spindle whorls
as the holes are much too small for a spindle. They seem to be rather
large and rough for ornaments, 812 centimeters in diameter. The
unfinished discs may be counters or weights.
Shell ornaments: Only 4 specimens can quite definitely be termed
ornaments. One is a small, flat, rectangular piece, 4 by 3 centimeters,


which is perforated at the center and engraved with what is probably
a conventionalized frog figure. Legs and feet are clearly shown drawn
up in a flexed position, but no head is represented. Another specimen
is a simple, flat, oval ring, 4 centimeters in diameter, with grooves cut
on one side at each end. A third is a small and very thin disc 3 centi-
meters in diameter and less than }J. millimeter in thickness with a
tiny hole at the center. It may have been sewn on some garment.
The fourth is an I-shaped object 6 centimeters long, 2 centimeters wide,
and 3 millimeters thick. It is carefully made and well polished, but
it is not perforated or engraved.
A cylindrical shell object, 6 centimeters long, may represent a shell
bead in the process of construction. A curious triangular object, 6
centimeters high and 112 centimeters thick, is notched at the top and
engraved with a series of parallel transverse lines across the bottom.
Its use is problematic, but because of its finish and engraving it has
been grouped with ornaments.
Shell "zemi": A small three-pointed object resembling the three-
pointed stone zemis of the fourth class as established by Dr. Fewkes
was found at a depth of 50 centimeters. It has exactly the same
shape as the stone objects of this type found at Barrio Coto and Barrio
Monserrate. It forms an equilateral triangle 6 centimeters long at the
base, 3 centimeters high, and 2 centimeters thick. The base is flat and
there are no carvings or engravings.
STONE IMPLEMENTS.-Petaloid celts: The polished triangular, or so-
called petaloid celts, common to all the West Indies, are present
throughout the whole Shell Level at Canas. There were 21 examples
found which could be identified as petaloid shapes although the
majority are fragmentary. \lost of them are made of a very dense
green stone, but a few were made of black stone. All but one are
highly polished and evenly made with a poll that is generally blunt-
pointed, and a sharp and semi-circular bit. The more or less complete
examples vary in length from 6 to 12 centimeters, and from 4 to 8
centimeters in width at the cutting edge. Petaloid celts in the Shell
Level become very significant, since they stand in sharp contrast to
the rectangular adze-like celts found in the Crab Level below.
Stone beads: The term bead may be poorly chosen for a group of
stone objects which are unique and characteristic of the Shell Level.
They are tubular, well-polished stone objects which have the shape of
pork sausages. They vary in length from 5 to 9 centimeters and in
diameter from 2 to 3 centimeters. Most of the examples have been
drilled longitudinally, while some were evidently discarded in the


process of drilling, since the holes remain unfinished. All were made
from a black and white mottled granite. Many of them seem unusu-
ally large and heavy for use as beads, but the perforations indicate
suspension of some sort. There were 8 of these objects found.
Stone hammers and rubbing stones form a group of objects varying
in shape and form, which were found in the culture deposit and show
some use. Some have a chipped surface at the butt showing use as
striking implements, while others have a smoothed surface caused by
rubbing or polishing. A few stones of this group have the manoo"
shape, but show no signs of having been artificially formed.
Pitted and grooved stones are exceptionally rare. Two examples
are water-worn pebbles which are flat and oval. They are pitted on
each side and roughly grooved at both edges. Another example is
grooved but not pitted. All three are probably hammer stones, as
the ends are chipped from striking.
Carved stone figures: Small zoomorphic figures carved in stone are
represented by 3 specimens, all of which are of the same type. The
figures are executed in three-quarters round and have a flat back or
base. Heads and faces which are well formed probably represent those
of men, animals, or mythical beings. The arms, which are not clearly
shown, seem to be drawn back as if tied behind the back. Legs and
feet are present on one specimen, and are drawn up in the flexed
position. A common trait of all three is a prominent penis. All are
perforated at the position of the shoulders. They were evidently tied
against some flat surface, possibly the forehead of the wearer, or sus-
pended. These little stone figures are not uncommon in many private
collections in Porto Rico, and are remarkably uniform in such char-
acteristics as the flat back, bound and flexed position, prominent
penis, and double holes for suspension at the shoulders. Evidently
the position was symbolic in some way, and the figures were more
than ornamental. Two of the figures found are 3 centimeters long and
1Y2 centimeters wide. The other is 5 by 2/2 centimeters.
A very small cylindrical object cut with three encircling grooves may
have been a bead or pendant. It is 2 centimeters long and 5 milli-
meters in diameter.
A cylindrical object with slightly concave sides is ornamented by
two encircling grooves and four raised oval knobs, which in turn are
engraved with a cross punctured at the intersection. Holes were made
at each end slanting through the edge. It is 6% centimeters long and
1Y centimeters in diameter, and is probably a symbolic ceremonial


Stone "zemi": Only one three-pointed stone was found at Canas.
It is the type referred to as the fourth class, and is unornamented in
any way. Much smaller than the large, carved, three-pointed stone
zemis so common in archaeological collections from Porto Rico, it
measures 7 centimeters in length at the base and 6 centimeters in
height at the central point.
Toy "dujo": A very small stone object has the form of the wooden
"dujo" or stool with its four legs and curved seat. It is roughly
formed from coral limestone, and is probably the utilization of an
accidental shape. It is 16 centimeters in length and stands 9 centi-
meters high. The four legs are roughly formed, and it is difficult to
tell how much has been worked and how much is natural.
Flint chips: Chips or flakes of flint were found throughout the
deposit but were not abundant. None of them shows secondary
chipping. It is possible that many of these flakes were utilized but no
manufactured implements of flint were found.
Miscellaneous objects of stone: One implement found in the shell
deposit has the shape of the so-called "Carib-stones" or "eared axes"
which are numerous in the Lesser Antilles. It resembles the blade of a
broad paddle and tapers down at the poll to a slender neck with knobs
on the poll which may be termed ears. It is 16 centimeters in length,
9 centimeters in greatest width, and 4 centimeters thick at the center.
There is no sharp edge or bit, and the edges have a rough, pecked ap-
pearance. It has been ground down but is not polished.
A fragment of what was evidently a boat-shaped stone bowl was
found at the bottom of the Shell Level. The fragment is a triangular
piece including a section of the rim. It was evidently pecked out of a
coarse granular stone, and resembles an Eskimo stone lamp rather
than a well-finished, thin-walled bowl. It is rough and unpolished.
A roughly rectangular stone 14 by 10 centimeters, found at the
bottom of the Shell Level, is cut with transverse grooves. There are
three grooves on one side and two on the other. The grooves are
straight and parallel, about 1 centimeter wide and Y2 centimeter deep.
Such a stone may have been used to sharpen pointed implements, or
possibly as an arrow shaft straightener.
BONE IMPLEMENTS.-Worked manati ribs: One manati rib found
has been ground down to form a sharp, curved, pick-like implement
20 centimeters in length and 212 centimeters wide at the butt. This
was the only complete and well-worked implement of manati rib
found in the deposit. The other ribs which show any kind of working
are only roughly grooved. Many of these have been broken at the


groove, suggesting that the grooves were cut in order to break the ribs
in sections. Possibly these fragments represent the discards, the
section of the bone cut away having been used for some finished imple-
ment. Excepting the pick, no such finished implements were found.
Turtle shell spatulas: One object which has been termed a spatula is
shaped like a narrow paddle blade, rounded at one end and squared
off at the other. The edges are regularly cut and smooth, and the
surfaces show some rubbing. It is 16 centimeters long, 6 centimeters
wide, and 5 millimeters thick. It is slightly curved with the natural
curve of the shell. The other object termed a spatula is fragmentary.
The piece is rectangular, and three sides are well cut and ground
smooth. The longest edge is beveled, forming a sharp bit. The sharp
edge or bit is 10 centimeters long. The rounded edges are 8 centi-
meters long. The fourth side is the rough broken edge. The thickness
is 5 millimeters.
Bone tubes: Three tubes are ahnost identical in size, 8 centimeters
long, and 8 millimeters in diameter. They were probably cut from
bird bones and then ground smooth to some extent. The ends show
marks of cutting. As a whole they are not well finished nor polished.
A fourth tube is very small (212 centimeters long, 5 millimeters in
diameter) and might be termed a bead. It is well polished. Grooves
encircle the ends.
Bone awls or needles: The triggers from trigger fish would undoubted-
ly make very serviceable awls and needles. A great many of these
were found in the deposit, and a few of them show that they have
been refashioned for such a use. Worked examples have been ground
down and resharpened, and the hole which is present in the hinge or
butt has been enlarged as if to take a thread. Since the triggers are
normally smooth and sharp, no retouching is necessary, and it is
probable many more were used than are represented by the few
definitely worked examples.
Awls cut from large bones are rare. Only 3 examples were found.
One is smooth and sharp, and is 17 centimeters long. The other two
are fragmentary and apparently less well finished.
Fish vertebrae counters: Fish vertebrae which have been cut down
to form a spool-shaped or checker-like object were found in all sites
excavated, but only one specimen was found at Canas. This is well
finished and the central hole has been enlarged, forming a ring-like
object. It is 2 centimeters in diameter.
Figures carved from bone: Probably the most remarkable single
object found at Canas is the representation of a human being carved


in low relief on what is evidently a human long bone. The complete
surface is utilized. Thus, the figure in relief extends completely around
the bone, with the head at one end and the feet at the other. It is in
a seated position and the hands rest on the knees. Most of the face
is missing. A female figure is probably intended, as breasts are
clearly shown. The most peculiar feature is a greatly exaggerated
navel, which is represented by a large perforated disc. The perfora-
tion extends through to the hollow section of the bone. A carved
figure found at Barrio Coto portrayed the same peculiarly exaggerated
navel, which fact suggests a religious symbolism in this feature. The
whole carved figure is 1012 centimeters high, and the bone on which
it is carved is 14 centimeters long and 21Y centimeters in diameter.
The figure is carefully executed and the bone is remarkably well
polished. Holes for suspension appear at the back of the neck and at
the top of the head. The specimen was found at a depth of 2 meters
and at the bottom of the refuse deposit.
Another carved bone object is fragmentary and unidentifiable.
A solid cylindrical bone, which is undoubtedly manati, is topped by a
curious winged figure. One side of this figure has two large, round
humps, while the other side contains a figure in relief which probably
represents a lizard. This figure is indistinct and apparently has no
head. The bone has been broken away at both ends, and all the
carved figure is not present. The fragment is 13 centimeters long and
4 centimeters in diameter. The carved figure at the end is 4 centi-
meters long. Possibly this is a fragment of a "swallow-stick."
The third carved bone object represents the head of a duck. It has
been cut from manati bone and is 12 centimeters long and 4 centi-
meters in diameter.
CLAY OBJECTS.-Clay discs: Discs which were evidently made from
potsherds were found in all three major sites. Only 2 were taken from
the deposit at Canas, 1 of Red Ware and 1 of Crude Ware. Perforated
discs were common at Barrio Monserrate and suggest spindle whorls,
but none of this type was found at Canas.
Clay ear plug: A small clay ring 18 millimeters in diameter and 8
millimeters wide was found in the Shell Level material. The ring is
raised at the edges, forming a broad groove around its circumference.
A ring of the same type was found at Barrio Monserrate lying beside
the head of a skeleton, which suggests that such rings were worn as
ear plugs.
of Shell Level pottery types is given in order to summarize the de-


scriptions which have been given, and to indicate the relative num-
bers of sherds representing each type.

Crude Ware. ..................................... (No. of sherds) 2662
Red W are . . . ....... ............................... (N o. of sherds) 253
Total No. of sherds 2915
Classification of shapes based upon identifiable fragments.
Shape A. Boat-shaped with loop handles ............... No. identified 424
Probable fragments 892
Shape A,. Boat-shaped with ridge lugs ............ . .. No. identified 42
Shape B. Boat-shaped with rectangular lugs....... ... No. identified 47
Shape Bi. Boat-shaped with semi-lunar lugs.............. No. identified 7
Shape C. Boat-shaped with modeled head lugs........... No. identified 47
Shape D. Indefinite round bowl sherds............. Possible fragments 543
Shape E ........................................... No. identified 159

Rare shapes.
Shape f. Vessels with spouts ..................... . No. identified
Shape g. Vessel with hole for pouring ............. ..... .No. identified
Shape h. Round bowls with vertical rim. ............... No. identified
Shape i. Round bowl with restricted mouth............ No. identified
Shape j. Bowls with indented rim ..................... .No. identified
Shape k. Vessels with restricted necks ......... ....... .No. identified
Shape 1. Bowls with partition ..................... . No. identified

Classification of the aspects of ornamentation based on sherds.
A. Modeled head lugs on rims ............................ (No.)
Class a ....................... ..................... 21
C lass b ......... ...................... ... ......... 10
Class c ........................ ......... ............ 3
C lass d .... ....... ....................... ........... 4
C lass e .. . .. . . .. .. .. . .. ... . .. . .. . .. .. .. .. 2
U classified .......................................... 27
B. Modeled forms on loop handles......................... (No.)
C lass a...................... ............ ........ . 57
C lass b...... ............ .......................... 19
C. Modeled figures in relief on vessel walls ................... (No.)
Class a ....................... ...................... 8
C lass b .. . . . .. . .. . . .. . ... . .. .. . .. ... .. .. 3
Class c............. ....................... ....... 14
C lass d .......................... ......... .......... 6
D Incised decoration......................... ........... (N o.)
Class a ........................ .................... 20
Class b ........................ .......... ....... .. 12
Class c ........................ ........ ............. 3
E Painted decoration......................... .......... .. (N o.)

Total number of ornamented sherds









Complex of traits distinguishing the Shell Level material.
1. Deposition in refuse heaps of marine shells.
2. Crude pottery unpainted and unslipped forming bulk of deposit.
3. Red Slipped Ware (rare).
4. Deep boat-shaped bowls with loop handles.
5. Shallow boat-shaped bowls with modeled head lugs.
6. Shallow boat-shaped bowls with rectangular lugs.
7. Double bowls.
8. Bowls with spouts.
9. All vessels with vertical or incurving rims.
10. Bat-head adornos on rims.
11. Crude human-head adornos on rims.
12. Modeled decoration on vessel walls.
13. Modeled decoration on broad loop handles.
14. Parallel incised line patterns terminating in pits or punctures.
15. Petaloid stone celts.
16. Shell chisels.
17. Shell celts.
18. Shell hoes.
19. Shell discs.
20. Carved shell ornaments or amulets.
21. Tubular stone beads.
22. Carved stone figures.
23. Three-pointed objects of shell and stone.

During the excavation of the Canas site it was remarked that the
modeled head lugs or adornos appeared more frequently in the first
25 centimeter cut or at the top of the deposit. These clay heads are
one of the most striking characteristics of all pottery collections in
Porto Rico. With the descriptions of pottery which have just been
given, it is seen that certain types of modeled heads are characteristic
of the Shell Level, while other types are characteristic of the Crab
Level and these types have been distinguished in the classification.
Excavation No. 1 was a series of cuts through a deposit which was
limited to Shell Level refuse. The maximum depth of this deposit
at the center of the midden was slightly over 2 meters. Since a
variation of pottery within the Shell Level was suggested by the
noticeable predominance of modeled head lugs in the first 25 centi-
meter cut, the collections from Canas, excavation No. 1, have been
used for a stratigraphic study of Shell Level material. This excludes
material from excavation No. 2 where the Shell Level overlying the
Crab Level did not average over 1 meter in depth.
Only those elements which were numerous enough to be meaningful
in such a study were used. The following charts show the result of


this stratigraphic analysis. Certain limitations of this method are
seen in the fact that the amount of deposit and the consequent num-
ber of specimens removed from the upper levels is much greater than
that from the lower levels. Only in a very limited area in the midden
(lid the deposit reach a depth of two meters. Another factor to be
considered is the probable shifting or sliding of the refuse during the
process of deposit. The refuse in excavation No. 1 was primarily
composed of large marine shells and ashes with very little soil. During
deposition this refuse was apparently heaped up and would undoubt-
edly slide and shift about during the period of occupation. With
these conditions in mind, charts showing distribution of sherds
illustrating various types of pottery must be taken with some reserva-


0 25 50 75 100 125 150 175
Depth in cms. to to to to to to to to
25 50 75 100 125 150 175 200
Cubic meters removed ...... 37 37 27 27 19 15 15 7
Shape A. Loop handles .... 24 62 41 50 36 25 25 5
Shape A-1. Ridge lugs..... 4 10 2 2 1 1
Shape B. Rectangular lugs. 2 17 2 4 4 2 10 2
Shape C. Modeled head lugs 7 9 6 5 1
Crude Ware............... 360 680 459 226 172 114 96 40
Red Ware................. 30 48 40 37 12 7 2
Modeled heads on rims ..... 21 12 8 5 1
Modeled forms on loop han-
dles .................... 6 6 12 9 5
Modeled forms on vessel walls 5 9 3 1 1
Incised designs ............. 17 11 2 1
Bat-head adornos ........ .. 6 4 2 1
Human-head adornos ....... 2 4 3 3 1

The preceding chart indicates that modeled and incised elements of
ornamentation are practically limited to the first meter of deposit.
Considering the fact that the great bulk of the material was found in
the first meter and that modeled and incised elements on sherds are
comparatively rare as a whole, the condition is not definitely indica-
tive of a change in the material culture represented. The presence of
21 modeled heads on vessel rims in the first 25 centimeter level, with
a sudden decrease in the next four levels, quite clearly shows, how-
ever, that this type of ornamentation was made use of more commonly


during the last phase of occupation of the site. It will be seen from
the chart that incised ornamentation also became much more common
in the last phase. The dominant vessel shapes and the types of ware
appear in approximately the same proportion throughout the deposit.
The sudden increase of modeled heads and incised designs near the
surface of the deposit may represent an impetus from other regions
at a later period and the significance of this condition may become
clear when the material from Canas is correlated with that from other
The pottery of the lower or Crab Level cannot adequately be
classified in the same manner as the pottery of the upper or Shell
Level. Classification on the basis of vessel shapes becomes much
more difficult in dealing with the Crab Level material because the
boat-shaped vessels typical of the Shell Level, which are easily deter-
mined by large sherds, are here absent, and the great variety of new
shapes suggested by the sherds cannot definitely be determined.
Furthermore, the presence of a large number of painted sherds makes
it necessary to place the emphasis on ornamentation and on the use of
Since there are 15 vessels which are complete, or sufficiently so
to determine the original shape, they have been treated separately,
and reference is made at all times to the sherds in order to indicate as
far as possible the prevalence of various types. The sherds collected
(859) do not represent the entire number of sherds in the deposit, since
only large rim fragments, painted sherds, and lugs and handles were
retained for analysis. Obviously this makes the percentage of painted
and unpainted sherds relatively meaningless, but with this in mind it
will be seen, when the figures are given, that painted pottery forms a
rather small proportion of the whole. Lugs and handles fall readily
into distinct types, and although they cannot be used to determine
vessel shapes as they could with Shell Level material and its dominant
boat shapes, they can be used as characterizing elements. Thus the
Crab Level material is classified in three ways: first, with reference to
pottery shapes on the basis of complete or nearly complete pots;
second, with reference to color of ware and painted and incised orna-
mentation; and third, with reference to types of handles and lugs.
As far as possible, the techniques of painting and the various types of
lugs and handles are correlated with the pottery shapes.












FIGUHE 5. Crab Level vessel shapes.


bowl type with a flat base and a wide flaring rim is represented by one
complete and one partially complete vessel. Bowls of this type rise
from a small, flat, circular base at a wide angle, then turn abruptly
upward, and gradually flare out to a wide, open mouth. A character-
istic of shape A bowls is a line or ridge about the circumference of the
body where the angle of the wall is changed. The complete bowl is
9 centimeters high and 25 centimeters in diameter at the lip. The
line about the circumference is 3 centimeters above the base. The
partially complete bowl is 8 centimeters high and 32 centimeters in
diameter at the lip. The line about the circumference is 8 centi-
meters above the base. The lip of the larger bowl is slightly thickened,
while that of the smaller is not.
A characteristic of shape A is a small, perforated knob which is
applied at the line or ridge circling the wall. Twenty-two sherds
with this perforated knob are apparently fragments of this type of
bowl. The knob is lacking on the partially complete example, but it
is possible that it was broken away with the missing part.
The complete specimen is of unpainted ware, while the incomplete
is decorated with red and white paint.
Shape B: Three bowls which are partially complete represent a type
which is not unlike shape A, but which is distinguished by the absence
of the encircling ridge or line. In other words, the bowls rise with a
slight bulge from a flat, circular base, and then gradually flare out to
a wide, open mouth. There is no sharp change in the angle of the wall.
All 3 examples of shape B vary to some degree. The first is 10 centi-
meters high and 29 centimeters in diameter at the lip, which is notice-
ably thicker than the rest of the vessel wall. The bulge from the
base is slight and there is little variation in the angle of the wall
from base to lip. The second is 12 centimeters high and 21 centimeters
in diameter at the lip. The base has a greater diameter than the first
example and the bulge from the base is more pronounced. The lip is
not thickened. The third is 10 centimeters high and 13 centimeters
in diameter at the lip which is very slightly thickened. The bulge at
the base is intermediate between the first and second examples.
All 3 bowls of shape B are ornamented on the outer wall with
designs in red and white paint.
Fragments of shapes A and B, for the most part, cannot be dis-
tinguished from one another. Bowls with a wide flaring rim which are
probably one shape or the other, are indicated, however, by 168 large
sherds. Many smaller sherds may be fragments of vessels of this
shape since they indicate a flaring rim, but their definite allocation


to shapes A and B would be very uncertain. The 168 sherds show that
shapes A and B fall into the unpainted ware, red painted ware, and
red and white painted ware groups.
Shape C: Round bowls with vertical handles at the rim on each side
form a type which may be termed cooking pots. Two examples of
this shape are complete, and a third is recognizable, though fragment-
ary. All 3 examples are of plain unpainted ware and fire-black-
ened on the outer surface. They are thin walled and well formed,
although less well finished than vessels of shape A and B. The bowls
rise with a slight bulge and at a broad angle from a flat, circular base
and then continue upward to form a vertical rim. The line formed by
the change in the angle of the wall is not a sharp ridge as it is on
vessels of shape A, but is rounded and more gradual. D-shaped
handles of the second class, which will be described in detail later, are
applied at the rim on opposite sides. They are vertical and do not
extend over the lip. One complete pot is 20 centimeters in diameter
at the lip and 10 centimeters high. The vertical rim is 4 centimeters
wide. The other pot is 22 centimeters in diameter, 9 centimeters high,
and has a vertical rim 5 centimeters wide.
Apparently bowls of this type are much less common than shapes
A and B as only 14 sherds can quite definitely be associated with this
shape. The 34 D-shaped handles of the second class may in part be
fragments of such bowls, but many of them indicate larger and heavier
vessels which undoubtedly vary in shape. All sherds recognized as
fragments of this shape are unpainted and fire-blackened.
Shape D: A large, round bowl with D-shaped vertical handles of the
first class represents a distinct type of cooking pot. With the excep-
tion of the handles, it is not unlike shape A, since it rises from a flat
base to a sharp line about the circumference, then turns upward and
gradually flares out to a wide mouth. It is unpainted and fire-black-
ened like the vessels of shape C. The handles are applied at the rim
but do not extend over the lip. It is 35 centimeters in diameter at the
lip and 16 centimeters high. The encircling line or ridge is 6 centi-
meters above the base.
Sherds which can be associated with shape D are rare. Some of
the 23 D-shaped handles of the first class present in the collection may
be from vessels of this kind, but only 3 can quite definitely be al-
located by means of the curvature of the rim and the fire-blackening.
Shape E: Large, thick-walled jars with restricted necks are repre-
sented by one complete vessel and nine fragments. The complete jar
flares out from a flat base to a height of 9 centimeters, then turns up-


ward and inward for 13 centimeters to the neck, where it again flares
outward 6 centimeters to the lip. The greatest diameter is at the
equator which is 35 centimeters. The height over all is 24 centimeters.
D-shaped vertical handles are applied at opposite sides and extend
between the neck and the equator. This jar was found at the bottom
of the Crab Level and was filled with crab claws.
The 9 fragments associated with shape E because they represent
large heavy jars with loop handles and restricted necks indicate cer-
tain variants of the shape described, but none are large enough to
suggest the complete shape. The complete jar and the sherds are all
plain undecorated ware.
Shape F: Round bowl forms with high, broad, annular bases are
represented by one partially complete vessel, and 16 recognizable
fragments which are complete or fragmentary bases. The original
form of the bowls cannot be determined from the vessel which is
partially complete since the upper part of the bowl is gone, and the
type is based primarily on the presence of the bases. These annular
bases resemble an inverted bowl with straight, flaring sides, varying
in height from 2 to 6 centimers and in diameter from 9 to 19 centi-
meters. The vessels with annular bases, as indicated by the partially
complete example, are apparently chalice-like with a broad open
bowl above the wide flaring base. The bowl forms a sharp angle with
the base since both base and bowl flare sharply outward from the
point at which they join. Vessels of this type were probably made as
a unit, since the bottom of the bowl and the top of the hollow base arc
one, and no bowl fragments were found from which the base had been
broken. On the other hand, several complete bases retain the jagged
edges of the bowl which has been broken away.
The 8 fragments of shape F and the partially complete example
are red and white painted, 7 are plain unpainted ware, and 1 is red
painted ware.
Shape G: An oval bowl with a rounded body and a sharp right-angle
rim or lip is a single example of a distinct shape. It rests on a flat oval
base and rounds gradually outward and upward to the edge, where it
suddenly turns sharply out to form a horizontal right-angle rim 1Y1
centimeters wide. Its length is 26 centimeters and its width 20 centi-
meters. The height is 7 centimeters. Part of a rectangular lug
formed by an extension of the right-angle rim remains at one end,
while the rim at the other end is broken away. It is probable that
rectangular lugs were affixed at each end. This bowl to some extent
suggests the boat shapes of the Shell Level material, but it lacks the
high bow and stern and the sloping rim.


Shape G is evidently a rare form as only 2 fragments are from oval-
shaped bowls and these lack the right-angle rim. Rim fragments of
the right-angle type number only 12 and are not large enough to
show the oval shape of the bowl. Rectangular lugs are rare, with only
9 examples, but these may be fragments of shape G. The complete
bowl is of red-painted ware.
Shape H: One vessel which is partially complete is a shallow, round
bowl which rests on a broad, flat base and rounds gradually outward
and upward to a height of 6 centimeters. It is 21 centimeters in di-
ameter. The lip is slightly thicker than the wall and is stippled with
rows of indentations which are bounded by an incised line. There
are no sherds which can definitely be associated with this shape, al-
though there are many sherds from a rounded bowl form. The single
example is of red-painted ware.
Shape I: Another single example of a distinct shape is a partially
complete bowl which is distinguished by the application of a collar
about the circumference where the vertical rim joins the body of the
vessel. It is a round bowl which rests on a flat base, flares sharply
outward and upward, and then turns straight up forming a vertical
rim 3 centimeters wide. The diameter is 22 centimeters and the
height 8 centimeters. The collar was applied as a triangular fillet of
clay which was broken away from the bowl during excavation. Raised
knobs appear on the collar at two points, but a good part of the collar
is missing and more knobs of this kind were probably originally present.
None of the sherds in the collection can be recognized as fragments of
this shape, although it is possible that many of them may be, since the
rim alone is not an unusual type. The single example is Red Painted
Shape J: Another unusual shape is represented by a single complete
vessel. It is a small discus-shaped bowl with a small restricted mouth.
The wall rises very gradually from a flat base to the equator, where it
turns sharply in and rises gradually to a small mouth. There is a
restriction about the vessel half way between the equator and the
mouth. The diameter at the equator is 16 centimeters, while the di-
ameter at the mouth, which is the same as that of the base, is 7 centi-
meters. The height is 612 centimeters. The wall of the vessel above
the restriction is ornamented with incised lines inlaid with white paint,
thus forming a panel or border of painted and incised decoration about
the mouth.
There are no large fragments which can definitely be allocated to
this shape, but 9 small sherds of those ornamented with incised and


inlaid designs show the sharp angle of the wall and the restriction, and
may well be fragments of this type.
Shape K: The flat platters or griddles so common in the West Indies
are present in the Crab Level as well as in the Shell Level. No com-
plete examples were found, but 40 rim fragments indicate the usual
form, a round, flat, thick platter generally thicker at the edge than at
the center. No griddles with rude incisions about the rim were found
in the Crab Level. A unique specimen is a griddle fragment which was
treated with a well-polished red slip or paint on both sides. The rest
are coarse rough clay generally rubbed smooth on the upper side only.
Shape L: Some form of vessel with legs is indicated by 1 sherd.
This sherd is a hollow pottery leg which contains rattling pellets. The
leg is 4 centimeters long and 2 centimeters in diameter. It is per-
forated on one side and at the bottom with small round holes through
which the clay pellets were probably inserted. The only other ex-
amples of objects containing rattling clay pellets are 2 modeled clay
heads, or "adornos," which will be described in a following section.
Crab Level stands in sharp contrast to pottery of the Shell Level
because of the quality of the ware.* Crab Level pottery is, as a whole,

A few sherds of Crab Level type as well as some from the Shell Level have been examined
by Anna 0 Shepard of the Laboratory of Anthropology. Her descriptions and comments
referring to Crab Level types are as follows:
"Two of the specimens examined contained numerous calcareous tests. One of these
(F-11 1.75) has a homogeneous, extremely fine textured, calcareous paste. Mineral inclu-
sions, principally quartz and feldspar, are fine and sparse. The organic remains are mostly
fragmentary but they appear to have been principally foraminifera, a species of Rotalia
being abundant and Tetularia less common. There are also larger fragments from the
shell of some bivalve.
"The surface of F-11 1.75 (thin section 1419) is highly polished but when a section through
the vessel wall is examined with a binocular microscope, magnification of 48 diameters, a
distinct surface coat cannot be detected. The difference in color between surface and paste
is only such as might result from difference in texture. The flaking of the surface noticeable
on the interior just below the rim is unusual, however, for a polished paste and the thin
section shows an exceedingly thin layer of a material similar to the paste in color when viewed
in transmitted light but more brightly birefringent. The thin section was cut on a diagonal
through the vessel wall instead of cross sectioning, therefore the thickness as measured from
the thin section is greater than the actual thickness of the coat, This measurement, 0.016
mm., gives some idea of the extreme thinness of the surface film.
"D-3 1.00 (thin section 1420) (White paint on red) is sherd tempered. The fragments of
sherd temper have a fine textured, dark red-brown paste. Mineral grains, principally quartz
and feldspar are of the same grading and as abundant as the sherd fragments.
"D-3 1.00 (thin section 1421) (White paint on red) contains numerous angular to sub-
angular grains of quartz, feldspar and green hornblende. A few irregular dark brown,
nearly opaque fragments with inclusions of quartz may be potsherd but are too rare in
occurrence to have been intentionally introduced as temper.
"As there are no two sherds in the lot with identical paste, the fact that the two white on
red sherds differ from the others does not necessarily indicate foreign origin. The use of
sherd temper does represent, however, a distinct ceramic technique, whereas the mineralogi-


thin, well fired, polished or rubbed to a smooth surface, and is com-
posed of a fine-grained clay. Pottery of the Shell Level is coarse,
thick, granular, and in many cases tends to crumble. Pottery of the
Crab Level is hard, brittle and well knit. When struck it rings, in
some cases like tile, in others like porcelain. Unquestionably a much
more advanced technique in pottery manufacture and firing is shown
in all forms of the Crab Level pottery. On the basis of color the ware
falls into three groups: Plain Brown Ware, Red Painted Ware, and
Red and White Painted Ware.
Plain Brown Ware: This type of ware forms the largest percentage
of all sherds removed from the deposit. It might be considered as the
service ware of everyday use. The color varies from a light reddish
brown to a dark smoke-blackened brown. Of the 859 sherds retained
for analysis 492 are of the plain unpainted brown variety, but these
figures do not indicate the true proportion, since all sherds of the red
and white painted type were conserved while only large rim fragments,
lugs, and handles of the plain were taken. Thus plain ware forms a
larger proportion of the pottery remains than is indicated in the
Plain Ware is generally slipped with a very fine brown clay or, when
all clay used in the construction was of the very fine kind, the surface
was polished without the addition of the slip. Thus, the sherds which
have relatively the same surface appearance actually represent two
distinct techniques of manufacture; pottery composed of a coarse
clay was treated with a thick slip of fine clay and polished, or pottery
composed throughout of fine clay was polished without the addition
of a slip. The firing process may change the composition of the sur-
face of a vessel so that it appears to be slipped when actually no such
technique was employed, but many sherds in the collection definitely
indicate a slip since this has begun to disintegrate, leaving part of the
sherd slipped and part unslipped. Broken sherds often show a black

cal differences in the paste of unslipped and red slipped sherds may result merely from
minor local variations in materials. The percentage occurrence of the various pastes could
be determined hy examining a large number of sherds which would show whether the paste
of the white on red sherds occurs in any types recognized as indigenous and also whether the
white on red ware is uniform in paste.
"Although the tempering material in these sherds is of considerable interest, the style of
decoration of the new ware is so distinctive it would seem to offer the most simple and con-
venient means of studying chronological and geographic distribution and thus determining
whether this ware is intrusive, introduced, or of local origin.
"The technique of decoration of the white on red ware seems unusual. The white paint
was in part painted directly on the highly polished surface of the paste and on one of the
sherds examined only lines and smaller areas were applied on the red, which therefore is
hardly a slip in the ordinary sense of the word. If entire vessels or larger fragments are
available for examination a study of the layout of the design should be of particular interest."


interior and brown surfaces, but more commonly the composition is
of a brown or bay-colored clay which varies in shade.
All pottery shapes which have been distinguished, with the excep-
tion of the discus-shaped bowl, are represented in Plain Ware, but
wide, flaring-rimmed bowls of shapes A and B are unquestionably
the most numerous.
Class a. Incised designs applied before firing are found on only 7
sherds of the Plain Ware. In every case they are found on the inside
of what arc apparently shallow bowl forms. There are 6 of the sherds
which show a simple pattern of concentric circles in most cases en-
graved about the flat bottom of the bowl. The seventh has a design
composed of a series of parallel straight lines. The incisions are all
deep, smooth and regular. As will be seen, incised designs appear on
Painted Ware as well as on Plain Ware, but are as a whole very un-
Class b. Incised designs applied after firing are found on 11 sherds
of the Plain Ware. Designs of this kind are all of the same type and
appear in every case in a panel on the inside of the rim of bowls. The
deep incised parallel lines marking the panel enclose a solid field of
fine cross-hatchings which have the appearance of a small meshed net.
Panels of cross-hatching may be double or arranged in long rectangles,
but always are restricted to a rim border. The technique of incising
after the pottery has been fired is peculiar to the Crab Level and is
limited to cross-hatch designs at the rim.
Red Painted Ware: Pottery which has been treated with red paint
is of the same general composition as Plain Brown Ware. That is,
it is simply slipped and polished Brown Ware to which red paint has
been applied. In no case has red paint been applied to both the inside
and the outside of the same vessel. It is clearly a paint and not a slip
since the majority of sherds show both the brown slip and the super-
imposed red paint. Painted surfaces are generally a bright red which
is burnished until it shines. It contrasts clearly with the Red Ware
of the Shell Level, which is a dull pinkish red applied as a clay slip.
Red Painted sherds have been divided into three groups; those
showing paint on the rim or lip only, those showing paint on one side
entire, and those which indicate a simple pattern in red paint outlined
by incisions. Obviously such a classification is subject to error since
a sherd classed as painted on the rim only may be a fragment of a
vessel which was ornamented with a painted design. However, since
only large rim fragments as a rule were retained, and since complete
vessels of the various types were found, the classification has some
point and will suffice as a basis of description.


Class a. Sherds indicating paint on the rim or lip only are numerous,
and are predominantly fragments of bowls with a thick rim, or a rim
which is at right angles to the wall. On sherds of this kind the paint
appears only in a narrow band along the lip. Thus the edge of the
sherd alone is red and the rest of the surface is the polished brown.
There are a few sherds of wide, flaring-rimmed bowls which show paint
applied in a wide band along the inside on the curve of the flaring rim.
Many red-painted sherds suggest that they are fragments of shape G,
the oval bowl with right-angle rim, but no definite association can be
made. Of the 165 sherds of Red Painted Ware, 68 show paint on the
rim only.
Class 1)b. Sherds painted on only one side have been grouped together
as representing vessels which were treated with red paint on the inside
or the outside. As has been observed above, the classification is sub-
ject to error since a fragment does not determine the treatment of the
whole, but numerous large sherds which extend from lip to base of
vessels indicate that complete bowls were ornamented in this manner.
Sherds painted on one side, numbering 78, are slightly more numerous
than those painted on rim alone. The majority appear to be frag-
ments of flaring-rimmed bowls of shapes A and B, while a few are
bowl forms with an incurving rim, and others are of indeterminate
shapes. A distinctive feature is that red paint is in no case applied to
both the inside and outside surfaces of the same vessel.
Class c. Sherds bearing simple designs in red paint outlined by
incisions are, apparently, fragments of shallow bowl shapes, and the
designs are in all cases on the interior of the vessel. The most com-
mon design of this kind is a red disc which was applied at the bottom
of the bowl (ImGURE 7-2). The flat bottom has been outlined by a deep
incised circle which is filled in with red paint. There is one V-shaped
figure on the inner wall outlined by incisions and filled in with red.
Another figure resembles the letter G. Two of these figures appear
back to back on a fragment, and are painted and incised (FIGURE 7-1).
The other designs of this type are fragmentary and the elements are
not clear, but in all cases painting and incising appear together, and
circles or curvilinear designs are paramount. Sherds of this type are
the least common of the Red Painted Ware group, numbering 19.
Red and White Painted Ware: The most striking feature of Crab
Level pottery is the use of red and white paint in the formation of
numerous and complicated designs. Some 7 bowls which are nearly
complete and 202 sherds, many of which are large fragments of bowls,
comprise the collection of Red and White Painted Ware. From these


it can be seen that certain characteristic traits hold for all pottery
ornamented in this manner. Designs are in all cases restricted to the
outer surface, while the inner surface remains the polished brown
color common to all pottery of the Crab Level. In many cases the
painted design is limited to an upper or lower section of the outer
surface, and it is clear that this type of ware is again the Plain Brown
Ware with its slip and polish to which the red and white painted de-
signs have been added. All designs are executed in white paint on a
red background, that is, in all cases white paint is applied over red
paint, and the red is applied over a polished or slipped brown surface.
Another characteristic of this type of ware is the presence of nega-
tive designs. On all sherds which are large enough to carry a recogniz-
able design, part or all of the design elements are in the negative red
outlined by white paint. In some cases design elements are in white
paint and therefore positive, as the white is always the last to be ap-
plied, but they are in turn always associated with elements in red which
naturally are in the negative. It is difficult to determine from small
sherds and fragmentary designs whether the elements in their complete
form are negative or positive, but it is clear from large sherds that
negative designs in red predominate (FIGURi 6).
A thick bright red paint has been used which has been polished to a
glossy surface. It rarely shows disintegration in any way and remains
surprisingly vivid. The white paint on many sherds is thick, polished,
and very slightly disintegrated, but on others it has cracked away and
partly disappeared so that spots of red paint show through the white.
The Red and White Painted Ware has been divided into four
groups on the basis of the technique of ornamentation: (a) sherds
with two color designs, (b) those with two color designs with white
inlaid in an incised pattern, (c) those with three color designs, a small
group in which the polished brown surface has been left to form an
element in the design, and (d) an aberrant group of four sherds on
which new colors are used to form a three color design.
Class a. (PLATE 3) Sherds ornamented with two color designs far
outnumber the other three classes since 161 of the 202 sherds are of
this type. The designs are all geometric and are composed of scrolls,
circles, dots, and various other geometric figures as well as straight
and curving lines or bands. A general characteristic which can be
determined from the complete bowls and the large sherds is the use of
decorative panels or distinct fields of ornamentation. In no case does
a single pattern or design clement extend completely around the
vessel. These panels are apparently rectangular sections which en-



4 a
F1,rIIE 6. Crab Le~vel redL andI wh1ile~ minted desigim~


close a single design unit. A common design unit of this kind is a
spool-shaped figure in negative red in a field of white which is in turn
outlined by a rectangle formed by a negative red line (FIGURE 6-4).
Another unit of this type is a circle with four tangential scrolls in
white on a red field again forming a rectangular panel (FIGURE 6-1).
Since the possibility of distinguishing complete patterns or design
units on a collection of sherds is very limited and subject to so much
error, no attempt will be made to base the description on designs as a
whole. After the passing remarks concerning the use of the panel, it
seems best to break the designs up into common elements and describe
them on that basis.
Linear elements are present on all sherds. If the sherds were grouped
according to the dominant design element the majority would fall
into a "linear group," which is probably quite natural in dealing with
a large number of small sherds. Certainly straight and parallel lines
go to make up a large part of every pattern. Broad parallel bands
and lines are commonly found circling the edge of rim fragments.
Straight horizontal and vertical lines in red aind white regularly
separate the design panels.
Curvilinear elements are present on the 44 sherds which are the
large fragments. Practically all large sherds carry a curvilinear
element in some form. There are four main curvilinear forms: scrolls,
circles, dots, and elliptical figures. They appear in negative red or in
white. Thirteen scrolls appear on the sherds and fourteen on the
partially complete vessels. Many arc apparently tangent to circles,
and not uncommonly they are connected, forming a double scroll
unit. Circles are present on 12 sherds and it is probable that many
of the curved lines on broken sherds are segments of circles. Circles
appear regularly in negative red, and at times enclose another figure
or a dot. Dots are distinguished clearly on only 7 sherds and are
usually enclosed by circles or lines forming part of a design. Elliptical
figures in red and white are distinct on only 4 sherds, but again
curved lines on fragments probably indicate figures of this kind
A spool-shaped figure in negative red is the most common and
distinctive single element on all Red and White Painted pottery.
It is usually the central figure in a panel, and may be horizontal or
vertical. It is often enclosed by a circle or rectangle. There are 34
examples of this spool shape present on the sherds, and in all cases it
is a distinct element created purposely and not the result of a design


A series of crosses formed by white lines on a red field appear on the
fragment on an annular base. This element is unique in the Canas
material, but appears again on an annular base found at Monscrrate
(FIGURE 6-7).
Rectangles and squares of solid color or in outline in red and white
are common, and as has been mentioned, generally form the bound-
aries of design panels.
Conventionalized zoomorphic faces composed of geometric elements
are present on 3 large sherds (FIGURE 6-3). Faces of this kind are
represented in a triangular panel. Eyes appear as red dots or as red
rings enclosing a horizontal spool-shaped figure representing the pupil.
The nose of each face is a vertical spool-shaped element, and the
mouth is a long rectangular element in which teeth are represented by
two rows of white dots. Thus both negative red and positive white
elements make up the figure. These are the only examples of realistic
representation of any kind, and are entirely made up of geometric
Class b. Two color designs with white inlaid in an incised pattern
are present on 23 sherds and on 1 complete pot (shape J discus-
shaped, small-mouthed bowl). Pottery of this kind has been incised
with deep regular grooves which have been filled with white paint,
forming a pattern on a red background (FIGURE 7-3, 4). In a few cases
a triangular or rectangular area in the design confined by incised lines
has been entirely filled in with white paint, but ordinarily the white is
limited to the grooves. Spiral and scroll designs are apparently the
most common. Four tight spiral figures remain partially intact on
the sherds. Two of these have as many as six turns forming the spiral
or spring-like figure. One sherd carries a design formed by two inter-
locking scrolls. The complete pot is decorated with a panel about the
rim which is composed of a flat scroll design extending completely
around the mouth. A section of this design looks like a series of
horizontal parallel lines, but as a unit it is seen as a complicated scroll
figure. It is probable that many of the sherds which contain a series
of parallel lines arc fragments of such a design.
Most of the sherds of this group are small and the designs cannot be
distinguished, but it is clear that straight and curving lines, spirals
and scrolls make up the majority of the patterns.
Class c. Three color designs in red, white, and brown slip appear on
only 14 small sherds (FIGURE 7-5). A variant of the customary tech-
nique is seen in the use of the basic slip color to form an element in the
design. The sherds are so small that no complete designs can be




FIGURE 7. Crab Level painted sherds. 1. Red Painted Ware, class c. Red painted
design outlined by incisions. G. figures. 2. Red Painted Ware, class c. Red painted design
outlined by incisions. Red disc figure. 3. Red and White Painted Ware, class b. White
paint inlaid in an incised pattern. Locked scroll figures. 4. Red and White Painted Ware,
class b. White inlaid in an incised pattern. Spiral figures. 5. Red and White Painted
Ware, class c. Three color design with the brown slip exposed to form the third color ele-
ment. Spool figure and scroll enclosing a dot. 6. Red and White Painted Ware, class d.
Miscellaneous sherds. Salmon-colored paint used with red and white to form an eye figure.


determined, and even the elements of design are fragmentary. It is
clear, however, that designs as a whole were laid in red paint over the
brown polished slip and then outlined by lines or bands of white paint.
Both straight and curved lines are present. One sherd of this type
hears the spool-shaped figure in negative brown slip outlined by a
field of white paint. On the same sherd appears a scroll figure en-
closing a dot, both in negative brown slip in a field of white. The rest
of the design is in red and white. A wavy line in white paint which is
the only example of such an element on any of the sherds appears on
one of these three color fragments.
Class d. Four sherds form an aberrant group which might be in-
cluded with the three-color-design group except that new and rare
colors are employed. On 2 of these sherds black paint is used with
red and white to make up the design. Apparently red was applied to
the slip, black over the red, and white over the black, all three colors
being exposed in certain areas to make up the pattern. Circles and
spool-shaped figures are present. The other 2 sherds carry red, white,
and salmon colors on a brown slip, but only the red, white, and salmon
form the design. A peculiar figure is executed on these 2 sherds. It
resembles a long, slanted eye in salmon color on a red field, the salmon
and red being separated by a white line. The pupil of the eye is red
outlined by white on the salmon-colored field (FIGURE 7-6).
Summary of Red and White Painted Ware: A synthesis of the char-
acteristics of red and white painted ware may be given as follows:
geometric designs in red and white are applied at all times to the outer
surface of vessels; the painted designs are in all cases applied over a
brown polished slip and may cover part or all of the outer surface;
white is at all times applied over the red; designs are in negative red
or positive white, or both; designs commonly appear in units segre-
gated in panels; the elements of ornamentation are spirals, scrolls
(double and interlocked), circles, dots, straight, and curving lines;
the most outstanding single element or unit is a spool-shaped figure in
negative red; in the incised patterns included with this group, white
is inlaid in the incisions; three-color designs are present with the basic
slip color forming the third element.
So far as can be determined, there were examples of pottery decor-
ated with red and white paint of each shape distinguished, but sherds
indicate that the majority of red and white painted vessels were wide
flaring-mouthed bowls similar to shapes A and B. Bowls with handles
are represented by only 5 sherds. Bowls with annular bases treated
with red and white paint are indicated by 7 fragments.


4).-Pottery fragments which are easily distinguishable, and which
form a definite class of objects, are the handles and lugs broken from
various vessels the shapes of which, as a rule, cannot adequately be
determined. As was observed above, handles and lugs of the Crab
Level are fragments of many undetermined vessels, and unlike handles
and lugs of the Shell Level, which regularly can be identified as frag-
ments of definite boat-shaped bowls, they are not used to designate
recognizable shapes. Fragments of this type have been arranged in
five groups: (a) D-shaped handles, (b) rectangular lugs, (c) semi-
lunar lugs, (d) perforated knobs on vessel walls, and (e) modeled head
lugs. When possible these fragments are correlated with known
pottery shapes.
D-shaped handles: All true handles of the Crab Level pottery are
characterized by the outline which forms a D against the vessel wall.
They are all broad bands or ribbons of clay attached to the wall, and
do not extend over the lip. The D-shape and the application to the
wall, which does not allow for an extension over the lip, sharply
distinguish all handles of the Crab Level from those of the Shell
Level which regularly form a flat loop extending above the rim.
Crab Level handles are further distinguished by being thicker and
heavier. The edges are rounded off and smoothed down. The
handles fall into four sub-classes.
In the first class are grouped the simple D-shapes which are vertical
on the vessels and not distinguished by added modeled forms. Han-
dles of this class were applied on the wall extending from the lip down-
ward, or less commonly on the body of the vessel below the rim. The
second class of D handles is distinguished from the first by the pres-
ence of a low round knob at the top of the handle. This knob, or
button, which is found on a large number of D handles at Canas as
well as at all the other sites, is a distinct trait of Crab Level pottery.
The purpose of this knob has not been determined. It can hardly be
termed a decorative element and yet it apparently has no serviceable
function. Of the 108 complete and fragmentary handles in the Crab
Level deposit at Canas, 68 are complete enough to be recognizable
without question, and of these 34 are of the second class with the
knob or button. The third class are D handles which are applied
vertically from the rim down and are topped by a semi-lunar, lug-
like appendage. In other words, a semi-lunar flange or ridge is added
to the handle where it joins the rim of the vessel. Only 8 examples
of this class were found. The fourth class is limited to 4 examples.
These are D-shaped but are applied horizontally near the rim.


A correlation of these four classes of handles with definite bowl
types is not possible. From complete specimens we know that
handles of the second class are found on shape C, and handles of the
first class are found on shapes D and E. A good many of the sepa-
rate handles suggest these shapes, but it is also true that other forms
are indicated. It is certain that all D-shaped handles were applied
to round vessels of some sort and never to oval or boat-shaped
Rectangular lugs: Lugs formed by an extension of the rim to produce
a rectangular flange are present in the Crab Level as they are in the
Shell Level. The smallest of these is 5 centimeters long and 2 centi-
meters high. The largest is 9 centimeters long and 3 centimeters high.
In no case does the lug form a sharp right angle with the rim, but
always joins it in a more or less gradual curve. In some cases the
corners of the lug are drawn up so that the upper edge is concave.
Only 10 of these rectangular lugs were found, of which 7 are Red
Painted Ware, with the paint limited to the rim and lug. A rectangu-
lar lug is attached to the complete bowl, shape G, which is oval in
shape. Several of the separate lugs suggest oval forms of some kind,
while others are evidently fragments of round bowl forms.
Semi-lunar lugs: There are 6 sherds which are fragments of shallow
round bowl forms with semi-lunar-shaped lugs. Four of these lugs are
perforated, apparently for the suspension of the bowls. Lugs of this
type actually vary little from rectangular lugs since they are drawn
out from the lip in the same manner and simply have a convex upper
edge instead of a straight or slightly concave edge.
Perforated knobs on vessel walls: Like the curious knobs on D-shaped
handles, perforated knobs on the vessel walls are a distinctive trait of
Crab Level pottery. The low round buttons or knobs are the same as
those on the handles except that they are transversely perforated at
the base. A knob of this kind is found on one complete vessel, shape
A, which is a deep bowl with a wide flaring rim and a line or ridge
about the equator where the wall turns abruptly upward before
flaring out to the lip. The knob was applied to this line or ridge.
All the 24 sherds which bear this type of perforated knob are ap-
parently fragments of Shape A bowls, and the knob in all cases was
applied on or near the encircling ridge. It is probable that vessels of
this kind were suspended by means of the perforated knob, but it is
curious that on the complete vessel only one knob is present, which
suggests that vessels were hung up by means of this perforation when
not in use.


Modeled head lugs: "Adornos," or modeled figures, on pottery of
the Crab Level are represented by 23 fragments of vessels. All but
two are attached to the rim. One of the 2 exceptions is modeled on
the vessel wall about 2 centimeters below the lip and the other ap-
pears on a D-shaped handle. There are 6 attached to the rim where
the rim is joined by a handle but the rest are not associated with
handles. As far as can be determined, the 21 adornos attached to the
rim face the interior of the vessel.
None of the modeled heads found in the Crab Level resemble any
of the heads found in the Shell Level. Although Crab Level heads are
zoomorphic representations of some sort, none resembles the familiar
bat-heads and only two might be termed human representations. All
Crab Level heads are more carefully formed and polished. There are
8 ornamented with red and white paint, 7 are painted red, and 8 are of
Brown Polished Ware. As a whole, Crab Level heads are distinguished
from Shell Level heads by the technique of the formation of the
features. On the former, features are more carefully modeled and
emphasized by deep incised lines. On the latter, features such as
eyes, mouth, and nose were apparently often applied to the figure
separately, and are rough and irregular.
A distinctive trait of the majority of the Crab Level heads is the
bulging or protruding perforated eye which is emphasized by a deep
incised line about the base. Knobs, bumps, or buttons regularly
appear on all heads as features or added adornments. All these
buttons are rubbed smooth. Many represent mouths which, like the
eyes, are emphasized by encircling incised lines. The modeled heads
have been divided into four groups, three of which represent distine-
live types. The fourth group is composed of various unique forms
which do not readily fall into any classification.
Type a. Semi-spherical heads. The features face the interior of the
vessel and the back of the head has been pushed in to form a cup-
shaped hollow. Hollow semini-spherical heads of this kind, although
limited to 5 specimens, indicate a unique technique of modeling. One
of these heads probably represents a human being with perforated
ears, a well-formed nose, and eyes and mouth shaped like a coffee bean.
The other 4 are indistinguishable, but all have the characteristic
bulging eyes.
Type b. A flat semi-lunar form upon which is modeled in relief a
long straight nose flanked by two protruding eyes. On one of these,
the nose-like ridge has been drawn up to form a loop. Heads of this
kind might be described as semi-lunar lugs upon which nose ridge


and eyes have been added. One head of this type was found on a
large fragment of a round shallow bowl red-painted about the rim.
The others are fragments of undetermined vessel shapes and are of
Brown Polished Ware.
Type c. Hollow spherical heads which contain rattling pellets. Only
2 such heads were found. Both are perforated at the back which
suggests that the pellets were inserted through this hole after the
formation of the head. Both these heads are white painted, and both
have the characteristic bulging eyes and mouth. One appears to be a
combination bird-human being with the nose developing into a crest
over the top of the head.
Of the various heads which do not fall into the three groups de-
scribed above, 3 probably represent the manati. The only features
are bulging eyes and a slit-like mouth, but the shape is that of a
manati head with its blunt snout. Another quite clearly represents a
turtle's head and neck. A fifth head suggests a crab with great pro-
truding eyes, but it is further complicated by a row of triangular
incisions alternating red and white. Another unique specimen of this
group probably represents a human head with perforated ears, well-
formed nose, puffed cheeks and coffee-bean eyes and mouth. Of this
group, five are red and white painted, three are red painted, and four
are Plain Brown Ware. The trait common to all is the bulging per-
forated eye outlined by deep incision.
A correlation of modeled head lugs with definite bowl shapes is
impossible. It can be determined, however, that the heads were at-
tached to some form of round open bowl, and to bowls with D-shaped
handles. No complete vessels with modeled head lugs were found.

SHELL ARTIFACTS.-Celts, hoes and chisels which are abundant in
the Shell Level are entirely lacking in the Crab Level. Shell artifacts
of any kind are noticeably rare, and of those found, the majority are
individual. A group of eight objects has been set aside as problematic,
and although each has a distinct shape and size, all have a common
origin in the Cassis tuberosa shell lip. They are elongate in form and
well rubbed or polished. Rarely is any remnant of the shell surface
left. They are not pointed or sharpened in any way, and none has a
flat, smooth surface. None of the shapes suggests a use or function,
but all are unquestionably manufactured by hand. It is possible that
they represent tools in the process of manufacture, but if so, one should
expect to find at least fragments of the finished tools. As a whole


these problematic objects are distinct from the problematic objects
of the Shell Level in that they are much smaller and more thoroughly
rubbed. They vary in length from 5 to 10 centimeters and in width
from I to 3 centimeters.
A shell cup found in the deposit is a Cassis tuberosa shell, the interior
coils of which have been cut away. The vessel is not well finished since
the cut edges remain rough and jagged and the original surface of the
shell remains. The lip of the shell forms the lip of the cup, and the
point of the shell forms a sort of rough spout.
A spoon made of a Cypraea (Cowry) shell is another unique object.
The shell has been cut away in such a manner as to leave an imple-
ment with well-polished edges which resembles the head of a deep
spoon with a square tip. The object is 5 centimeters wide and 6 centi-
meters long. Spoon shapes of this kind are found in all the other sites
associated with Crab Level material, and although the total number
is small it seems likely that this shell spoon is characteristic of the
Crab Level.
Two of the olivia shell rattles, common to the whole region, were
found in the Crab Level at Canas. They are made by cutting the
shell in two transversely and then rubbing through a hole for perfora-
tion on the side of the shell at the uncut end.
Two flat polished shell objects probably are ornaments or amulets
in the process of manufacture. One is an oval shape 8 centimeters in
length with well-polished surfaces and well-rounded edges. The other
is rectangular and also well polished. It is 3 by 5 centimeters.
A shell cylinder 5 centimeters long and 1 centimeter in diameter
appears to be a bead in the making, since a longitudinal perforation
has been begun at one end.
A curious shell object found in the Crab Level looks very much like
a cleat for fastening a line. The object rests on a flat rectangular base
and has prongs at each end which extend up and outward. It is
horizontally perforated at the base of each prong, and apparently was
bound against a flat surface. The object is 6 centimeters long on the
base and the prongs extend upward 2 centimeters. It is very well
polished and regularly formed. An object of the same kind was found
at Barrio Coto.
STONE IMPLEMENTS.-No petaloid stone celts were found in the
Crab Level at Canas. The typical stone implement in this level is a
rectangular adze-like celt with one flat surface and one convex surface.
There were nine celts of this type obtained. In cross-section it would
appear as a half moon. The bit is sharp and semi-circular. The celts


are made with remarkable precision, and the sharp edges of the flat
side are as straight as if made on a plane table. Contrasting with the
petaloid celts of the Shell Level, these rectangular adze-like celts
become one of the most distinctive characteristics of the Crab Level.
Two of the celts found are variants of this type with two flat sur-
faces instead of one. On these the bit is bevelled and straight rather
than semi-circular.
Most of these rectangular celts have been broken so that the exact
shape of the poll is questionable, but the complete examples are either
rounded or straight on the butt. No marks of hafting can be seen.
One of the celts is of a fine-grained, green stone, while all the rest arc
of a black fine-grained stone. All are remarkably well polished, and
the regularity of their manufacture places them among the finest ex-
amples of polished stone implements found in the West Indies.
There are 4 fragments of celts which lack the flat surface. These
4 fragments arc of chert which is patinated so that the objects appear
to be covered with a white paint. The complete shapes cannot be
determined, but they are distinct from the rectangular celts, lacking
the flat surface. The bits are curvilinear.
Fragments of polished stone, which were found, may be tools in
process of manufacture or else rubbing and polishing stones. They are
broken so that the original form is questionable. One has a roughly
formed, gouge-like bit.
One discoidal stone has evidently been worked down for some
purpose. Numerous elongate stones with blunt points may represent
some form of tool but none shows marks of use and it is possible that
they are only water-worn pebbles which have been used for rubbing
and polishing.
A few fragments of flint in the form of chips or flakes were found,
but there are no signs of secondary chipping, and if the flint flakes
were utilized it could have been only as rude scrapers.
BONE IMPLEMENTS.-The only object of bone identified as a tool
is a fragment of a manati rib which has been worked to a blunt point.
The object is probably part of one of the manati rib picks. There are
four more manati rib segments which have been rubbed down and
charred in the fire. It is possible that they are also sections of picks.
A thick, heavy piece of turtle shell has, roughly, the form of a large,
stemmed spear head. It has obviously been rubbed down for some
purpose but its use is questionable. It is 14 centimeters long and 6
centimeters wide at the shoulders. A small, rectangular piece of
turtle shell also shows rubbing and polishing. The edges are ground


smooth and regular, and one surface is polished. It has evidently
been broken away from a larger section.
summarizes the Crab Level pottery types and the traits which par-
ticularly characterize the Crab Level material.

Classification of shapes based upon complete or partially complete vessels.
Shape A. Round, open bowl with wide flaring rim and perforated knob at ridge
about circumference.
Shape B. Round, open bowl with wide flaring rim.
Shape C. Round, flat-bottomed cooking bowl, wide vertical rim and D-shaped
Shape D. Round, flat-bottomed cooking bowl with flaring rim and D-shaped
Shape E. Large, heavy jar with restricted neck and D-shaped handle.
Shape F. Round bowl with high annular base.
Shape G. Oval bowl with right-angle rim and rectangular lugs.
Shape H. Round, shallow bowl with walls curving in at rim.
Shape I. Round bowl with vertical rim and collar.
Shape J. Discus-shaped, small-mouthed bowl.
Shape K. Griddles.
Shape L. Unidentified vessel with hollow legs containing rattling clay pellets.
Classification of the aspects of painted and incised ornamentation based primarily
upon sherds.
Plain Brown Ware..... . ......... ............. (No. of sherds) 492
a. Incised designs................... ............... 7
b. Incised designs applied after firing .................... 11
Red Painted Ware ....................... . ... . (No. of sherds) 165
a. On rim or lip only.................... .......... 68
b. On one side............ .. ....... ... ........... 78
c. In simple pattern outlined by incisions ............. .. 19
Red and White Painted Ware .................. . . (No. of sherds) 202
a. Two color designs............. ................. 161
b. Two color designs with white inlaid in an incised pattern. 23
c. Three color designs........... ................... 14
d. M miscellaneous ......... ........ ................ 4

Total Number of sherds................. .............. 859
Classification of handles and lugs based upon sherds.
D-shaped handles.......... .................... (No. of sherds) 108
1st class....................... ....... ............. 23
2nd class........ ................. .............. 34
3rd class.... ............................. ....... 8
4th class ............... ............................... 3
Rectangular lugs.. . .................. .......... (No. of sherds) 10
Semi-lunar lugs........................... ............ (No. of sherds) 6
Perforated knobs on vessel walls........................... 24
M odeled head lugs..... ............ . .. ............. 23


Type a-hollow semi-spherical ...................... 5
Type b- semi-lunnar........................... .. .. 3
Type c-hollow, spherical, with pellets .... ............ 2

Complex of Traits Characterizing the Crab Level Material
1. Deposition in yellow soil and disintegrated land crab shells below the normal
shell heap.
2. An advanced technique in pottery manufacture with the bulk of the pottery
slipped and polished, fine-grained, hard, thin, and regularly formed.
3. Red and white painted designs over a brown slip.
4. White paint at all times applied over red.
5. Negative designs.
6. White paint inlaid in an incised pattern.
7. Simple red painted designs outlined by incisions.
8. Incised cross-hatch designs applied after firing.
9. Round bowls with D-shaped handles.
10. Jars with restricted necks.
11. Oval bowls with rectangular lugs.
12. The great majority of vessels with a flaring rim.
13. Round bowls with high annular bases.
14. Zoomorphic "adornos" on rims and handles with bulging features empha-
sized by deep incised lines. "Adornos" commonly painted red and white.
15. Concave zoomorphic "adornos" on rims facing the interior of vessels.
16. Hollow zoomorphic "adornos" containing a rattling clay pellet.
17. Perforated knobs on vessel walls.
18. Low round knobs or buttons on D-shaped handles.
19. Rectangular adze-like stone celts with one side flat and the other convex.
20. Celts of chert with a white patination.


The preceding descriptions of Shell Level and Crab Level artifacts
have been based on the material removed from each of two strata of
deposition distinguished by different types of food refuse. A com-
parison of the artifacts deposited during the two periods of occupation
represented by these strata is presented in the following tabulation.

Types of Crude Ware without slip, polish, Polished and slipped Brown Ware
Ware or paint
Thick, coarse-grained (predom- Hard, thin, fine-grained (predom-
inating) inating)
Red Slipped Ware, pink to red Red Painted Ware
(rare) Paint applied over polished and
slipped Brown Ware (abundant)
Red and White Painted Ware
Painted designs applied over pol-
ished and slipped Brown Ware




of orna-


Boat-shaped bowls (predominat-
Some round bowl forms

Vessel rims vertical or incurving

Flat loop handles extending over
rim on boat-shaped vessels

Rectangular lugs on rim of boat-
shaped vessels
Semi-lunar lugs on rim of boat-
shaped vessels
Ridge lugs on boat-shaped vessels
Modeled head lugs on round and
boat-shaped vessels
Modeled head lugs on vessel rims
Rough zoomorphic figures princi-
pally representing hat and hu-
man heads in the round

Modeled figures on loop handles
Geometric figures in relief and
small, crude bat and human-
Modeled figures on vessel walls
Geometric and zoomorphic
Incised decoration on the exterior
of bowls in panels near rim
Parallel lines terminating in punc-
Painted decoration
Six sherds with crude curvilinear
figures in red on Crude Ware

Problematic Implements
Three-pointed "zemi"

Round bowl forms (predominat-
Oval shapes rare, no boat-shapes
Annular bases
Vessel rims principally wide-flaring
but occasionally straight and
D-shaped handles on walls of
round bowls never extending
over rim
Four distinct types
Rectangular lugs on rim of oval
and round bowls
Semi-lunar lugs on round bowls

Modeled head lugs on round bowls
Perforated knobs on vessel walls
Modeled head lugs on vessel rims
Polished, painted, well formed
Unknown zoomorphic beings
Semi-lunar, semi-spherical, and
spherical-hollow with pellets
Modeled figures on D-shaped
Large, painted
Like those figures recorded above

Incised curvilinear designs on the
inside of shallow bowls (rare)
Cross-hatch incised designs ap-
plied after firing
Red painted designs outlined by
Bowls with red painted lip
Bowls red painted on one side
Red and White Painted Ware
Two color designs
Two color designs with white in-
laid in incisions
Three color designs

Problematic Implements
One spoon
One cup
Two rattles
One cleat-like object


Stone Petaloid celts Rectangular adze-like celts
Objects Tubular beads White patinated celts of chert,
Hammers and rubbing stones shape uncertain
Small carved zoomorphic figures Rubbing stones
Three-pointed "zemi"
Bone Worked manati ribs Worked manati ribs
Small tubes
Awls and needles
Carved figures
All pottery found in the Crab Level deposit is easily distinguished
from that found in the Shell Level because of the type of ware, and
indicates a much more advanced technique of modeling and firing.
The only identifiable shape common to both levels is the clay griddle,
and this is found throughout the West Indies and northern South
America. Rectangular and semi-lunar lugs appear on sherds from
both levels, but so far as can be determined were applied to vessels of
different shape. Modeled head lugs were found in both levels but
distinct types were limited to each level, and there is a marked differ-
ence between all such figures found in the Shell Level and those found
in the Crab Level. Incised ornamentation is found on sherds from
both levels but the types of incision are distinct, and a new technique
of incising appears in the Crab Level which has been termed incising
after firing. Painted designs are limited to the Crab Level pottery,
except for crude curvilinear designs painted on six sherds found in the
Shell Level. Shell chisels, celts, hoes, discs and ornaments were found
only in the Shell Level. Petaloid stone celts, tubular beads, carved
figures and zemis were limited to the Shell Level, while rectangular
adze-like stone celts, and white patinated chert celts were found only
in the Crab Level.
Such distinctions between the artifacts found in the two strata of
occupation unquestionably indicate that the site was occupied during
the late period by a people with a material culture distinct from that of
those who settled there in the early period. The conditions of the
deposit and the distinctive traits are such that an immigration of an
alien people may be inferred, and quite definitely no cultural fusion
took place, since traits characterizing the early culture were not taken
over by the late. The fact that traits were not adopted suggests that
some time elapsed between the two periods of occupation. If an
immigrating group had come in contact with an early population
either in a friendly or a hostile relation, it is probable that some ele-
ments, at least, of the early ceramic complex would have been taken


over as units or combined with the ceramic complex of the new group.
The fact that types of ware, vessel shapes excepting the griddle, and
elements of ornamentation are all distinctive in each stratum pre-
cludes the possibility of adoption or combination. Furthermore, the
abrupt change in the type of food refuse substantiates the inference
that some time elapsed before the arrival of the immigrating group,
since direct contact would undoubtedly result in a carry-over of the
food complex, at least for a certain time, before giving way to a
gradual change.
An analysis of the collections has led to the following conclusions;
Canas was occupied at an early period by a group of people with a
material culture which, judging from the pottery remains, was sur-
prisingly advanced in comparison to the material culture represented
by remains found throughout the Greater Antilles. These people
disappeared, and were followed after some lapse of time by an immi-
grating group with a ceramic complex less well developed but with a
technique of stone carving more advanced, if it is justifiable to judge
from a few examples of small carved stone figures. The complex of
traits characterizing this late culture is in detail distinct from that of
the early culture, but a basic similarity exists, which suggests a com-
mon place of origin or a common source of diffusion.
Evidence of a common prototype is indicated in such general
similarities as: the application of modeled zoomorphic figures on the
rims of vessels; clay griddles or burens; rectangular and semi-lunar
lugs; the relation between oval and boat-shaped vessels; polished
stone implements and the absence of chipped flint tools; and the
relation between flat loop and D-shaped handles. These basic ele-
ments are found throughout the West Indies and in northern South
America. Future work may lead to the discovery of the diffusion
center for the two cultures represented in the Canas deposit, since it
is obvious that the late culture did not develop directly from the early
The presence of remarkably well made pottery decorated with
designs in red and white, in the early period deposit, was the most
striking element encountered. It is so markedly distinct from the
pottery in collections from Porto Rico that its discovery came as a
complete surprise. Stray sherds have been found in southwestern
Porto Rico from time to time, but have remained a puzzle since the
composition did not in any way correspond with the general ceramic
development represented by the more or less crude forms of pottery
found scattered over a large part of the island. Its deposition in a


sub-stratum explains its rarity in collections. As a description of the
following excavations will show, this painted pottery is not restricted
to southwestern Porto Rico and in all cases appears in stratification as
a deposit of an early period. It becomes the easily recognized criterion
of the early culture.

Excavations in Barrio Coto
Coto is a barrio of the municipality of Isabela which lies between the
city of Isabela and the mouth of the Guajataca river on the northwest
coast of Porto Rico. This section of the island is a plateau-like area,
dotted with low rolling hills, which drops suddenly to the shore of the
ocean in sheer rocky cliffs. Rainfall is abundant and the land is fertile
though rough and rocky in many sections. At the present time most
of the barrio is under cultivation. It is divided into small plots, and
farmed by individual families who raise various food products for sale
and for their own consumption. Large cane plantations are rare in
this region due to the rolling terrain.
Four kilometers west of the mouth of the Guajataca river and two
kilometers from the shore is a valley surrounded by low hills. The
floor of this valley is very fertile and is now almost entirely under
cultivation. Near the center of the valley is a low hill approximately
one-half kilometer in diameter, over the surface of which are scattered
numerous shells, potsherds, and broken implements. The shells are
primarily terrestrial gastropods although various marine shells also
are found. Collectors of relics have gathered numerous prehistoric
artifacts from the fields which extend over a large part of this low hill,
and have made various minor excavations on the west side which at
present is pasture land.
No mounds appear on the site and there are no pits, earth works, or
other indications of dwelling sites.
Excavation of the site was undertaken by cutting a series of trenches
through the deposit on the western side of the hill (FIGURE 8). All
trenches were divided into sections four by four meters and each
section was given a key letter and number. All material was removed
in 25 centimeter levels, and the artifacts from each section and level
were packed in bags as a unit. Unfortunately the deposit had been
much disturbed throughout. The numerous pits dug by collectors
spread across the side of the hill, and many had been filled in and
grown over so that the disturbance could not be detected until cut by
the trenches. Furthermore, the slope of the hill was such that con-
siderable erosioni had taken ph.ce.


-/ *' /'

W vlterI hole '
:/ .. '

Di- '*'y
b~ed '
jiad m

r i ;

:*' :'''

.*. / t : i


Scale 1-400
interval 750ma.

Extent of/.refuse

Top of rise

X /4,




FaunsB S. Chart of Barrio Coto site.


The first trench was forty meters in length and extended from the
base of the hill to the comparatively level summit. It was divided
into ten sections which are referred to as A-i to A-10, with section A-1
at the base and A-10 at the top. Excavation was continued well into
sterile sub-soil at the bottom of the refuse deposit throughout the
length of the trench. Refuse was found to a depth of 25 centimeters
in section A-1 at the base of the hill. Proceeding up the slope it
increased gradually to 75 centimeters at the crest from which it de-
creased gradually to 65 centimeters in A-10 on the level summit. A
cross-section of this deposit is shown on the accompanying chart
The deposit removed from this trench was primarily composed of
blackened sandy soil through which were scattered numerous ter-
restrial gastropod shells, a few marine shells of various species, some
land crab shells, charcoal and ashes, hutia, manati, fish and bird bones,
potsherds, and implements of shell and stone. The shell content of
this deposit was relatively very low when compared to that removed
in Barrio Canas. In some limited sections only scattered bits of shell
appeared. No stratification of refuse could be determined in any
section, although toward the bottom of the deposit the soil was less
blackened with charcoal and organic decay. Also artifacts appeared
less commonly in the lowest 25 centimeters of deposit.
Burials were encountered in sections A-6 at 50 centimeters, A-3 at
50 centimeters, A-2 at 50 centimeters, A-4 at 75 centimeters, A-7 at
25 centimeters, A-6 at 50 centimeters, A-2 at 75 centimeters, A-7 at
50 centimeters, and A-7 at 75 centimeters. Complete clay vessels
were found with three of these skeletons, associated so that they were
obviously buried with the bodies. One of the skeletons was that of a
child and the other two were those of adults. Detailed description of
each of these burials appears in the appendix. Three skeletons were
found lying in sterile sub-soil below the refuse deposit, and the others
lay in the refuse in all levels from the top to the bottom.
A second trench was cut parallel to the first and joining it along its
southern edge. This trench was divided into sections corresponding
to those in the first trench and the key letters and numbers B-2, B-3,
B-4, B-5 and 13-6 were used to designate these sections which corre-
sponded with A-2 to A-6 in the first trench.
The deposit of refuse removed from this second trench (trench B)
varied in depth between 25 and 75 centimeters. The greatest depth
was found near the crest of the hill and the least depth near the base.

Terrestrial shells
Shell Level artifacts
Primary flexed burials.

65 ms .


Fevw terrestri al, sme crab shells
Crab Level Artifacts
Primary Flexed buril&s.

Terrestrial shells
Shell Level artifacts
Primary flexea burials.


Scale Icm.2m.

FIGURE 9. Cross-sections of Blarrio Coto excavations.


The composition of the refuse was much the same as that already
described. Terrestrial gastropod shells were scattered through black-
ened soil. Some marine shells and land crab shells were also found.
Potsherds were numerous in the upper levels but gradually decreased
in number toward the bottom. Shell and stone implements were
found in all levels. Sections of the deposit in trench B contained more
wood ash, charcoal, shells, and general culture refuse than others,
but no well defined fire pits or distinct hearths were found. No
stratification of refuse appeared.
Burials were struck in sections B-6 at 25 centimeters, B-5 at 50
centimeters, B-6 at 50 centimeters, B-4 at 50 centimeters, B-2 at 50
centimeters, B-6 at 75 centimeters, and B-3 at 50 centimeters. One of
these burials was that of a baby, the bones of which were contained
in a small shallow bowl. This apparently was a secondary burial of
bones since the bowl was too small to contain the entire body of a
child. Details of each burial are given in the appendix.
A third trench (trench C) was cut paralleling trench B and extending
the same length, or 20 meters. The deposit in this trench varied in
depth between 25 and 50 centimeters. It was of the same general
composition as that in the first two trenches and contained numerous
potsherds and implements and a few burials. The material was
removed and recorded in the same manner as that from the other
A fourth trench (trench D) was next run paralleling trench C through
its entire length of 20 meters. The refuse deposit was noticeably
decreasing in the trench, varying between 20 and 40 centimeters. The
number of artifacts perceptibly decreased and burials were much less
These excavations which were proceeding southward along the west
side of the hill were finally interrupted by a road running diagonally
across the top of the hill.

Owing to certain restrictions imposed by the owner of this site it
was necessary to leave a small section unexcavated between the first
series of trenches and the second series. This second series (excavation
No. 2) was begun with a trench 24 meters in length running in a
northeasterly direction along the base of the hill, 8 meters north of
section A-1 in excavation No. 1. Five trenches 4 meters wide were
dug paralleling this first trench and proceeding toward the summit of
the hill. Each trench was divided into six sections and each section


given a key letter and number. All refuse was removed in 25 centi-
meters levels and the artifacts from each level in each section were
packed as a unit.
The first three trenches of excavation No. 2 exposed a deposit
similar to that found throughout excavation No. 1. The depth varied
between 25 and 75 centimeters with the greatest depth in those sec-
tions near the crest of the hill. Terrestrial gastropods made up the
bulk of the food refuse but scattered marine shells, land crab shells,
and bones of manatis, hutia, fish and birds were also found. Fire pits
or hearths were not clearly defined although charcoal and ashes were
abundant throughout. Sherds and broken implements were numerous
though less common toward the bottom of the deposit. No clay
vessels or other artifacts were found with any of these. There was no
evidence of stratification in the deposit.
The next two trenches (trenches 4 and 5), which extended along the
crest of the hill, cut through a deposit of refuse which reached a depth
of one meter. This depth of deposit was confined to sections A-4,
B-4, C-4, A-5, B-5, and C-5, or an area of 96 square meters. Toward
the bottom of the deposit in these sections the soil became yellow, and
terrestrial and marine shells less common. A few red and white
painted sherds were found, and all the pottery removed from this
yellowish soil near the bottom of the deposit was like that found in
the Crab Level at Canas. No distinct line of separation between an
upper and a lower level could be seen when trench walls were squared
off and charted, but a change in the nature of the refuse was unmis-
takable between 20 and 30 centimeters from the bottom.
Burials were numerous in trenches 4 and 5 at all levels. In sections
D-4, E-4, D-5 and E-5, five burials were found within a radius of 3
meters lying from 50 to 75 centimeters in depth. Two well preserved
skeletons were found in the last traces of refuse at 1 meter. The
details are given in the appendix.
Trench 6, the last to be excavated, was not entirely completed.
The depth of deposit in this trench, which was cut across the summit
of the hill, was definitely decreasing. The refuse was like that through-
out most of the deposit and contained numerous artifacts.

The refuse deposit exposed by the series of trenches in excavation
No. 1 and in excavation No. 2, for the most part varied between 25
and 75 centimeters in depth. It was composed of blackened sandy
soil, ashes, charcoal, terrestrial gastropod shells, scattered marine


shells, a few land crab shells, and bones of hutias, inanatis, birds and
fishes. Shells of various terrestrial gastropods made up the major
part of the food refuse. Potsherds, and artifacts made of shell and
stone, were numerous throughout the deposit. As the following
descriptions of the collections will indicate, the great majority of all
artifacts found in the deposit were similar to those found in the
Shell Level at Canas, but Canas Crab Level types were also found in
limited numbers.
In a limited area in excavation No. 2 the deposit reached a depth of
1 meter, and in this section there was a noticeable change in the com-
position of the refuse near the bottom. The deepest 20 to 30 centi-
meters of deposit was yellow and contained only scattered terrestrial
and marine shells. Land crab shells were found but in no great num-
bers. Ashes and charcoal were less abundant and artifacts were more
rare than in the upper 75 centimeters of refuse. The artifacts found
in this lower level were almost entirely of Crab Level type. There was
no distinct line separating an upper from a lower level of occupation,
and no complete segregation of distinct types of food refuse as in the
Canas site. A noticeable change in the types of artifacts indicates,
however, a condition very similar to that in Barrio Canas where two
periods of occupation were observed.
There were 60 burials encountered during the excavations. They
were found scattered throughout the refuse deposit in all sections at
all levels and also in clear sub-soil below the refuse. No particular
orientation of the skeletons prevailed since they were found lying in
all directions. Many were so entirely disintegrated that the position
of interment could not be determined, while others were fairly well
preserved, and could be photographed in position and removed for
Of the intact skeletons 17 were found to lie on the left side with the
legs flexed. The amount of flexing varied considerably from a slight
bend of the knees to a tight flex in which the knees reached to within a
few inches of the forehead. Many skeletons lay as if the body had
been tightly bound when interred. There were 18 skeletons which
lay on the right side in the same flexed position.
Also there were 14 skeletons found which lay on the back with the
legs drawn up and flexed tight against the body. With this type of
burial the skull was generally found forced forward or raised above the
level of the spinal column. One skeleton lay on its face with the legs
flexed beneath the body. Another burial was evidently a secondary
burial of bones with the flesh removed since the bones were found
stacked in a heap, one fragment of the skull lying over another.


A single urn burial encountered was that of a baby, the bones of
which filled a small round howl. It was clearly a secondary burial
of the bones after the flesh had been removed.
Clay vessels were found associated with 3 skeletons as mortuary
offerings. In a few cases artifacts such as worked bones, single beads,
and roughly worked or fragmentary stone implements were found in
close proximity to skeletons, but, since all lay in a matrix of culture
refuse, it could not definitely be determined that they were buried with
the body.
The details of each burial are given in the appendix.

Collections from Barrio Coto
Complete vessels were rare in the deposit in Barrio Coto and the
five that were found intact were associated with burials. Large
fragments, however, were common, and throughout the excavation
only these large fragments and sherds which carried handles, lugs, or
ornamental elements were retained. It may be estimated that over
half of the potsherds removed were discarded since they lacked any
kind of ornamentation and were too small to indicate the vessel shape.
As at Barrio Canas, the great majority of sherds were unpainted.
A study of the collections has made it clear that the pottery of Coto
to a large extent duplicates that found at Canas. The great majority
of pottery types, the characteristics of vessel shapes, composition of
wares, and elements of ornamentation identified in the Canas collec-
tions appear again in the Coto material. As described in the section
concerning excavations, however, clear stratification of refuse deposit
with the definite segregation of certain pottery types in each level of
occupation was not found in the Coto deposit. The sub-stratum found
in sections A-4, B-4, C-4, A-5, B-5, and C-5 closely resembled the sub-
stratum or Crab Level at Barrio Canas, and the pottery found in this
sub-stratum duplicates that found in the Canas Crab Level, but all
the rest of the Coto deposit contained pottery which correlates with
the pottery from both Crab Level and Shell Level at Canas, with the
latter predominating.
The pottery and associated artifacts from Coto are described by
means of the following charts which list the Canas Shell Level and
Crab Level types found at Coto. New types found at Coto are de-
scribed in sections following the charts. The collection has been
treated as a single unit since the sub-stratum was very limited and
not clearly distinguished from the rest of the deposit. The distribu-
tion of the various types in relation to depth in the deposit and the
stratigraphic significance are discussed in a following section.


The following chart lists those pottery types identified with the
Shell Level at Canas which are found in the collections from Coto.
Depth in the deposit is indicated in 25 centimeter levels. The mnum-
hers refer to identifiable sherds.

Depth in cmis.

75 Total
to No. of
100 sherds

C rude W are .. . ........ ...............
Red W are . ......... .................
Shape A .......... ..............
Shape A- . . ......... ................
Shape B .......... .............
Shape C ................. .......
Shape D ............................ ..

Shape E . . .. ...................

Shape F. ..... ...............
Shape J.......................
Shape L .. .. ... ... ...........
Modeled head lugs, class A...............
Modeled head lugs, class B...............
Modeled head lugs, class C.............
Modeled head lugs, class D .............
Modeled head lugs, class E.......... .
Modeled head lugs, unclassified...........
Modeled forms on handles, class A .......
Modeled forms on handles, class B........
Modeled figures on vessel walls, class A..
Modeled figures on vessel walls, class B..
Modeled figures on vessel walls, class C.
Modeled figures on vessel walls, class D...
Incised decoration, class A...............
Incised decoration, class B ..............

349 324 134 9 816
37 49 14 100
132 161 71 5 369
15 12 2 29
7 20 13 1 41
19 10 8 3 40
Round bowl sherds present, but not
definitely identifiable
Griddles present, but not distinguished
from Crab Level type
3 2 1 6
4 4 8
1 1
16 2 1 3 22
17 8 3 28

Canas Shell Level pottery types not found in the Coto material are
the following: vessel shapes Bi, g, h, i; incised decoration, class c; and
red painted decoration of the Shell Level type which is the application
of simple curvilinear designs on the inside of Crude Ware bowls. Only
six sherds of this kind were found at Canas.
NEW TYPES FOUND AT COTo.-The rounded bottom of a vessel with
four legs is a unique specimen in the Coto collections. The fragment
has been broken away from a vessel the shape of which is unidentifi-


able. It was supported on four short solid legs which have been par-
tially broken away and remain about 1 centimeter in length and 1
centimeter in diameter. Four sherds which are undoubtedly pottery
legs were also found and may be fragments of other vessels of this
kind. The legs and the partially complete vessels are of the Crude
Ware typical of the Canas Shell Level and may be associated with the
Shell Level complex.

The following chart lists those pottery types identified with the
Crab Level at Canas which are found in the Coto collections. Depth
in the deposit is indicated in centimeters. The numbers refer to
identifiable sherds.

0 25 50 75 Total
Depth in ems. to to to to No. of
25 50 75 100 sherds
Shapes A and B (fragments of each type not
distinguishable).... ................. 1 3 6 8 18
Shape F ........... ..... ... ....... 1 7 2 10
Shape G .................... .......... 1 2 3
Shape K ..... ................ ... Griddles present, but not distinguished
from Shell Level type
Plain Brown Ware ................. 21 41 57 39 158
Red Painted Ware, class A ............... 1 6 7 4 18
Red Painted Ware, class B .. . . ........ 1
Red Painted Ware, class C............... 1 1 2
Red and White Painted Ware, class A..... 1 6 7 4 18
D-shaped handles, 1st class .. ........... 8 15 19 17 59
D-shaped handles, 2nd class .............. 4 8 12 6 30
D-shaped handles, 3rd class ......... .... 1 1
Rectangular lugs.. .. . . . .. 6 6 16
Perforated knobs....................... 2 2 4
Modeled head lugs, type B............... 1 2 1 1 5
Modeled head lugs, unclassified........... 3 1 4

Canas Crab Level types absent from the Coto collections-No
example of shapes C, D, E, H, I, J, and L could definitely be deter-
mined from sherds at Coto. Since these types are based on complete
or partially complete vessels found at Canas, and complete bowls
were very rare at Coto, the absence is not surprising. As explained in
the discussion of Canas material, Crab Level shapes are difficult to
determine by means of sherds, while Shell Level shapes are easily
distinguished because of the predominating boat-shape.


Red and White Painted Ware sherds of classes b, c, and d are not
found. These types, which are rare at Canas, are defined as: (b), two
color designs with white inlaid in an incised pattern; (c) three color
designs; and (d) miscellaneous Painted Ware, a group of four sherds
found at Canas with black and salmon-colored designs. Red and
White Painted Ware found at Coto is comparatively rare as a whole
and the sherds of this type which were found are for the most part
small fragments. No complete design patterns were found and con-
sequently a comparison of designs on sherds from the two sites is not
possible. Linear and curvilinear elements are found on all sherds.
It is clear even from fragmentary design elements that many are in
negative red as they were at Canas. White is in all cases applied over
red and both red and white over a polished or slipped brown surface.
The red and white paints used at Coto appear less well preserved on
the sherds, and lack the gloss or burnish of the paints on Canas sherds.
Apparently the technique of painting and firing was less advanced at
Coto, or else the materials at hand were not so durable. It is also
possible that the natural conditions at Coto were less adapted to the
preservation of the paint.
Modeled clay heads of type a, or the semi-spherical heads impressed
from the back, and of type e, the hollow heads with clay pellet rattles,
are not represented in the Coto collections. These heads were rare at
NEW TYPES FOUND AT COTO.-An oval bowl with a sharp right-
angle rim or lip resembles the Canas Crab Level shape C, except that
it lacks the rectangular lugs. It is of Red Painted Ware, class A.
Numerous sherds found at Coto may be fragments of such oval bowls
but cannot be classed definitely as such.
A modeled head lug of Crab Level Red Painted Ware, found in the
deposit, is an unusual type. It is in the form of some zoomorphic
head with a sharp snout and two pronged appendages on the head
resembling horns. Another unique modeled head clearly represents a
Shell, stone, and bone implements found in the Coto deposit to a
large extent duplicate those found at Canas. The chart on page 73
lists these implements under the type headings described in the
section concerning the Canas collections. It will be seen that, with the
exception of 4, all the implements listed are of the Canas Shell Level
type. Two of these exceptions are shell spoons which are the same as
the one found in the Canas Crab Level, while the other 2 are rectangu-


lar adze-like celts. These celts are of exactly the same type as the
celts limited to the Crab Level at Canas. One was found in the sub-
stratum at Coto and 1 was found in the surface 25 centimeters level.
Three unusual objects found at Coto (PLATE 5) require individual
description since objects of this kind were not found at Canas.
One of these is a bone piece 11, centimeters long and 2 centimeters
broad which has been carved into a shape resembling a canoe paddle
with a T-shaped handle and a broad pointed blade. This was found
at a depth between 75-100 centimeters near a flexed burial.
Another bone artifact is composed of two polished bone tubes
which were found glued together in a V-shape. A carved shell ring
bound the tubes together at the apex of the V. Carved bone rings
were applied about the opposite ends of the tubes. The tubes were 11
centimeters long and 1 centimeter in diameter. The object was found
in the surface 25 centimeters level. Tubes, used for snuffing pipta-
denia among the Ouitoto, examples of which are reproduced from
Crevaux by W. E. Roth (1924: pl. 52), are not unlike the object found
at Coto (see PLATE 5, FIGURE 3).
The third unusual object is a small carved figure probably represent-
ing a human being. It was found in excavation No. 1, A-4, at 25-50


0 25 50 75
Depth in cms. to to to to Total
25 50 75 100 Number
Shell chisels .................. .......... 2 2 1 5
Shell celts................... ......... .. 2 1 3
Problematic shell implements............. 9 3 3 15
Shell discs ............................. 1 1
Shell discs, perforated ...... .......... . 1 1
Shell rattles.................. .......... 2 2 2 1 7
Shell spoons (Crab Level type?) .......... 2 2
Petaloid stone celts .......... .......... 23 9 20 52
Rubbing stones......................... 6 1 7
Tubular stone beads...... ............. 2 2 4
Rectangular adze-like celt................. 1 1 2
3-pointed stones ........................ 2 2
Carved stone figures..................... 1 1 2
Eared Carib stones...................... 1 1
B one aw ls.............................. 2 2
Worked manati ribs. .................... 1 1
W worked tortoise shell . . . .... ..... 1 1 2


centimeters. It is 6 centimeters high and 3 centimeters wide and
carved in the half-round from some unidentified material with the
consistency of wood or burned bone. It is jet black and may be a
carved coroso seed. The eyes and teeth are inlaid with shell. A crest
on the top of the head resembles a three-pointed zemi. A transverse
perforation at the back of the head may indicate that it was suspended.
A curious feature is an exaggerated navel represented by a perforated
disc in relief. The same kind of navel appeared on a carved bone
figure found al Canas.
The charts listing the pottery types characteristic of the Shell Level
and the Crab Level at Canas which appear in the Coto collections,
indicate that the percentage of Crab Level types increases with the
depth of the deposit, and that the percentage of Shell Level types
decreases with the depth. This is more clearly shown in the following
simplified chart.

Depth in cms. 0-25 25-50 50-75 75-100
Sherds classed as Shell Level type........... .386 373 148 18
Sherds classed as Crab Level type............ 48 85 102 64

The limited sub-stratum contained Crab Level pottery types only,
and may be considered as duplicating the sub-stratum found in Barrio
Canas. This correlates with the early period of occupation and the
early culture distinguished at Canas. Crab Level pottery types were
found, however, throughout all levels of the deposit associated with
Shell Level types which represent the late culture at Canas.
The first interpretation suggested by this condition of the deposit
is that a cultural fusion took place in Barrio Coto. That is, that the
late arrivals either united with or took over the material culture of the
early group and deposited the bulk of the refuse found. A significant
fact, however, makes this interpretation unlikely. Crab Level pot-
tery traits are not found in combination with Shell Level traits in the
composition of a single vessel. For example, in no case is a decorative
element characteristic of the Shell Level found on a polished or slipped
Brown Ware, Red Painted Ware, or Red and White Painted Ware
vessel. Likewise, no Crab Level ornamental technique is found on a
vessel of Shell Level shape or ware. If a fusion of material cultures
had taken place, some combinations of ceramic traits would undoubt-
edly appear.


As described in the sections concerning excavations, cuts were made
along the side of a hill with a pronounced incline, and considerable
erosion apparently had taken place. Furthermore, numerous pits had
been dug and refilled by various collectors. The entire deposit did not
reach a depth of over 1 meter and for the most part was less than 75
centimeters in depth, so that these disturbances would affect sections
of the deposit from the surface to the bottom. A significant condition
appears in the fact that sherds classed as Crab Level types regularly
appeared in groups and were not scattered at random through the
upper levels of the refuse. The general disturbance of the site, and the
appearance of Crab Level sherds in groups in the upper levels, would
suggest that the intermixture of Crab Level and Shell Level pottery
in the deposit was due to this disturbance, and that they were origi-
nally deposited in stratified levels as at Barrio Canas.
The nature of the deposition at Coto remains something of a prob-
lem. It is clear, however, that the two periods of occupation and the
two cultures found at Canas are again represented at Barrio Coto.
The great majority of the traits characterizing each level at Canas
reappeared at Coto, which undoubtedly indicates a close cultural
connection between the two regions of the island during both periods
of occupation.
Excavations in Barrio Monserrate
The site excavated in Barrio Monserrate is situated on Embarcaderos
Point which is on the northeast coast of Porto Rico just west of the
town of Luquillo. The point is a low flat sandy area which is now
planted in cocoanut palms. A barrier reef extends along the shore and
encloses a shallow bay where shell fish of many varieties are abundant.
Conchs are found along the reef and the deep-sea fishing outside the
reef is excellent.
The refuse of a prehistoric settlement is found scattered over an ir-
regular area approximately 300 meters in length by 200 meters in
width which extends along the shore of the bay and a lagoon that joins
the bay at this point. Potsherds are strewn over the entire area, and
many artifacts made of shell and stone have been exposed by the
burrowings of numerous land crabs. Five low mounds can be dis-
tinguished. They rise a meter, or a meter and a half, above the level
of the surrounding area. One of these mounds had been partly cut
away by the surf, and culture refuse was exposed to a depth of one
During a month of excavation at this site, trenches were cut through
three of these mounds, mounds A, B, and E as shown on the chart


(FIGURE 10). The method of excavation was the same as that carried
out at the other sites. All trenches were divided into sections, in this
case 4 by 4 meters, and the refuse removed in 25 centimeters levels.
The artifacts from each level of each section were separated and
packed as a unit.
A trench 32 meters long and 4 meters wide was cut the length of the
mound with the western edge of the trench passing approximately
through the center. A deposit of refuse was found to a depth of 1.50
meters near the center of the mound at its highest point. At the edge
of the mound the deposit decreased to approximately 25 centimeters.
The last traces of refuse lay in wet sand slightly above sea level.
The general composition of the deposit was similar to that found in
the Shell Level at Barrio Canas. Marine shells, principally Strombus,
Tellina and Chama, were numerous and made up the bulk of the food
refuse. They were scattered in varying quantities through blackened
sandy soil. Ashes, charcoal, and bones of manatis, hutias, fishes and
birds were found intermixed with the shells and the blackened sand.
Artifacts were more numerous in this deposit than in any removed up
to this time. At the center of the mound in sections A-3, A-4, A-5 and
A-6 some variation in the composition could be determined. The
first 50 to 60 centimeters contained a greater percentage of shells than
the lower levels, and also more ash, charcoal, bones, and artifacts. In
the next 40 to 50 centimeters shells were less abundant and the number
of artifacts decreased perceptibly. The last 40 centimeters of deposit
contained few shells and artifacts, and was generally gray in color with
small quantities of ash and charcoal mixed with yellow sand. Below
this was clear sand at sea level. Sections dug to a depth of 1.60 to
1.70 meters rapidly filled with water.
No burials were found in the first 50 centimeters of deposit, which
contained the greatest number of shells and artifacts.
The next 50 centimeters contained several primary flexed burials
and a limited number of burial urns containing the bones of very small
In the 50 centimeters at the bottom, flexed primary burials were
very numerous, often lying so close together that one skeleton could
be distinguished only with difficulty from another.
Below all culture refuse in clear sand fragmentary skeletons were
found lying partly in the water which filled the trenches and marked
the sea level.
A second trench was cut through mound A parallel to and joining



Soale I-1500

,<---Extent of refuse

Mound A


Mound C

E -

S .. Mound B

Shoal water

Many shell f4

Embarcaderos Point


Fiouuit 10. Chart of Monserrate situ.




the first trench along the western edge. These two trenches which
made an excavation 32 meters long and 8 meters wide removed all the
central and highest section of the mound.
The deposit removed during the excavation of the second trench
was similar to that removed from the first. In sections B-3, B-4, B-5
and B-6, which were the area containing the greatest deposit, the same
variation in composition was found. The surface 50 to 60 centimeters
contained a great number of marine shells, black sand, bones, ashes,
and numerous artifacts, but no burials. The next 50 centimeters
contained less food refuse and artifacts, but several primary flexed
burials, and child burial urns. The lowest 40 to 50 centimeters of
deposit was filled with primary flexed burials, but still less food refuse
was found and artifacts were comparatively rare. In section B-5
from a depth of 75 centimeters to 1.50 meters the deposit was so
filled with skeletons that some sections appeared to contain a massed
burial. Below all refuse in clear sand and half submerged in water
were several skeletons, some of which were fairly well preserved.
iThe chart (FIGURE 11) shows a cross-section of mound A through
the long axis at the center. No distinct stratification of refuse appeared,
and the levels indicated on the chart were not sharply segregated in
the mound. These levels are simply distinguished on the basis of
more or less refuse content of the same general type, and by the ab-
sence or presence of different types of burial. The composition of the
mound clearly indicates that during the last phase of occupation no
interments were made with the refuse. During an earlier period
child urn burial was not uncommon and this type of interment was
practiced along with primary flexed burial in the refuse. Apparently
when the site was first occupied primary flexed burial prevailed and
urns were not used for the interment of small children.
Descriptions of each burial found in mound A are given in the
A house and garden plot occupied a large part of mound 1 so that
it was necessary to limit excavation to the northern end. A trench 4
meters wide was begun on the northeastern side and run in a south-
westerly direction through and a few meters beyond the highest rise.
The trench was 20 meters long and was divided into five sections,
numbered A-1, A-2, A-3, A-4, and A-5. A rich deposit of marine shells
and artifacts was found to a depth of 20 centimeters in section A-l,
and this deposit gradually increased to 75 centimeters in sections A-4
and A-5 at tihe highest rise. This deposit was very similar to that

Thick marine shell deposit
lany Shell Level arti f acts; nu b rials

Less shell deposit But some 51hll Level artifacts
Primary Flexed burials.

Child urn burials. (x )

Still less refuse
Primary Turials.

Sterile sana.
Primary buri &ls


Scale 1Cn-.Im


found in the surface 50 centimeters of mound A, being made up of
marine shells, some land crab shells, black sandy soil, ashes, charcoal,
bones, and artifacts.
Below this compact deposit of shells and artifacts in section A-I a
few artifacts and scattered marine shells were found in yellowish to
gray sand extending on down to the water level and below. Sterile
sub-soil was not reached in section A-1 since water rapidly filled the
trench below 1 meter. In section A-2 gray sand and scattered shells
and artifacts were found to a depth of 75 centimeters and below this
was clear sterile sand. In section A-3 the compact mass of shells and
artifacts reached a depth varying between 50 and 75 centimeters, and
below this appeared gray to yellow sand through which were scattered
land crab shells. Sterile sand was reached at 1.50 meters. In section
A-4 a sub-stratum of deposit was distinguished below the mass of
marine shells. This was composed of gray soil and land crab shells
with scattered potsherds which reached a depth of 1.75 meters.
Section A-5 exposed a stratified deposit the same as that found in
Barrio Canas. A 75 centimeters deposit of marine shells, blackened
sand, various food refuse and artifacts lay above a well-defined stratum
of yellowish sand filled with masses of disintegrated land crab shells,
some ash and charcoal, and artifacts, among which were numerous
red and white painted sherds. The sub-stratum reached a depth of
1.50 meters and lay on sterile sand. The line between these refuse
levels was indistinct in sections A-3 and A-4, but was clearly defined
in section A-5.
A second trench (FIGURE 12) was cut through mound B joining the
first trench along its southern edge and extending the same length.
The deposit exposed was essentially the same as that found in the first
trench with some variation in the depth of deposit. The sub-stratum
of land crab shells appeared in section B-3 and was distinct in section
B-5. The maximum depth of deposit was 1.75 meters.
Two refuse levels, or periods of occupation during which different
kinds of food refuse were deposited, could readily be distinguished near
the center of mound B. It is unfortunate that excavations could not
be continued southward through the highest and most extensive
section where the house and garden were situated, since the sub-
stratum appeared to be increasing in depth in that direction. With
the excavations completed, the extent of the early deposit of land
crab remains was not determined.
Burials were rare in mound B trenches. The three that were found
are recorded in the appendix. No burials were found in the land crab

Thick marine shell deposit
Shell level artifacts.

Few marine shells
Shell level artifacts.

Sterile sas-d
w at'e level e

Land crab deposit.
Crab level- artifacts.

75 cma.

50 "

Scale I cm-Im.

Few marine shells.
Very fevw artifacts.

Land crab deposit
Crab level artifacts.

B-5 B B-4 B 3 B- -I

Sterile san,) 50 "

Scale Icm- Im.
FioGuni 12. Cross-sections of mound B and mound E, Mionserrate.


The third mound to be excavated lay along the edge of the water,
and a section at the north side had been cut away by the surf. A
trench 20 meters long and 4 meters wide was run along this northern
side and excavated to the depth of the water level. This trench ex-
tended through the highest elevation but did not pass through the
entire length of the mound.
Refuse was found in limited sections to a depth of 1 meter and below
this there was a level of clear wet sand which extended below the high
tide level. The composition of the refuse in mound E was consider-
ably different from that found in mounds A and B. The surface 25
to 35 centimeters removed contained scattered marine shells, some
ash and charcoal, gray sand and a few scattered potsherds. Below this
and extending down to sterile sand, marine shells were still more rare,
land crab shells appeared in increasing numbers to the bottom of the
deposit, the sand was yellow in general appearance, ashes and charcoal
were scarce, and potsherds were more numerous although not so
plentiful as in the upper levels of mounds A and B.
In clear sand and partly below sea level a deposit of conch shells
was found. No artifacts, bones, ashes, or other culture refuse were
found with them. The conch shells were complete, and it is probable
that they were a natural deposition since they lay at the water's edge.
A second trench was excavated along the southern edge of the first
and extending the same length. The refuse reached a depth of 1
meter, and was similar to that removed from the first trench with
some variation. The upper 20 to 35 centimeters contained less marine
shells and more land crab shells. Sherds were very scarce and frag-
mentary. Below this a few scattered marine shells were found but
land crab shells became abundant and in some sections the bulk of
the deposit was like that found in the sub-stratum of mound B.
As a whole the deposit exposed in mound E more closely resembled
that found at the bottom of mound B than the deposit found in the
upper levels of mound B, and the bulk of mound A. Crab shell was
the most common type of food refuse throughout, but marine shells
were numerous near the surface and scattered through all levels.
The number of burials found in mound E was very limited. They
are recorded in the appendix. There was one unusual burial, however,
encountered in the first trench, which requires special comment in
this section. This was an adult skeleton which lay on the left side
with the legs flexed, at a depth of 50 centimeters. Covering the skull
was part of a large open dish, over the right shoulder was a second


bowl right side up containing a third bowl, and in the bend of the knees
was a fourth bowl bottom side up. The burial was the only one found
in mound E with which artifacts were associated. The vessels were
all of Shell Level type. As a description of the material from mound
E will show in the following section, all other pottery found in mound
E was of Crab Level type. This fact undoubtedly indicates that the
burial was intrusive and not made during the deposition of the mound.
The two periods of occupation discovered in Barrio Canas and
again in Barrio Coto appear for a third time in Barrio Monserrate.
Distinctly stratified deposits were found only in a limited section of
mound B but were sufficient to indicate that during the earliest period
of occupation land crab shells made up the bulk of the food refuse.
Again, the extent of the early occupation could not be determined.
The composition of mound E was somewhat confusing as was that of
the midden in Barrio Coto, but apparently it was deposited primarily
during the early period. During the late period, or that in which
marine shells were deposited, occupation extended over the entire site.
Further excavation at the Monserrate site would be profitable since
no cuts were made in mounds C and D and a large part of mound B.
It is one of the most extensive prehistoric dwelling sites in Porto Rico,
and was undoubtedly occupied over a long period, and very possibly
from time to time since the first occupation of the island, as the condi-
tions for fishing and also for agriculture are excellent.
Collections from Barrio Monserrate
There was some variation in the kind of refuse deposit and in the
types of artifacts removed from each of the three mounds excavated.
However, the majority of the artifacts found duplicate those removed
from the Barrio Canas and Barrio Coto sites, and in the following
description charts are employed to indicate these duplications at
The collections from each mound are treated separately. Culture
refuse extends over the entire area and no segregation of the mounds
appears on the surface, but the excavations made it clear that the
occupation of all mounds was not contemporaneous.
Potsherds found at Monserrate, as at the other sites, made up the
great majority of all artifacts present. It may be estimated that less
than half of these sherds were retained in the collections since all but
large rim fragments, handles, lugs, and ornamented sherds were dis-
carded throughout the excavation. Large collections of sherds and


associated artifacts were sent to the University of Porto Rico and are
not recorded in the following descriptions. Complete or partially com-
plete vessels were comparatively numerous for a refuse site, and these
have made it possible to determine the prevailing shapes. All pottery
types which are distinct from those encountered in the other sites are
described separately.
Objects of shell, stone, and bone in the collections represent all
that were found in the deposit, since all such objects recognized as
artifacts were retained. They are, for the most part, similar to those
found at the other sites. The exceptions are described in some detail.
The following chart lists those pottery types identified with the
Shell Level in Canas which are found in the collections from mound A
at Monserrate. Depth in the deposit is indicated in 25 centimeter
levels. The numbers refer to identifiable sherds or rarely to complete

Depth in cms.

Crude W are ..........................
Red Slipped W are . . . . . . . . ...
Shape A .............................
Shape A-1 ....... ......... .........
Shape B ......... ........ ........
Shape B -1 ............... ..........
Shape C ........................
Shape D . . ......... ................

Shap e E .............................
Shape f .......... .............
Shape h ........................
Shape i ............. .................
Shape j .......... ..........
Shape k ........ ...............
Shape I ......... ........... ....
Modeled head lugs on rims, class b......
Modeled head lugs on rims, class d ......
Modeled head lugs on rims, unclassified..
Modeled figures on vessel walls, class a. .
Modeled figures on vessel walls, class b. .
Modeled figures on vessel walls, class c. .
Modeled figures on vessel walls, class d. .
Incised decoration, class a..............
Incised decoration, class b .............
Incised decoration, class c..............
Painted decoration ................

0 25 50 75 100
to to to to to
25 50 75 100 125 s
112 138 60 36 10
23 32 9 1
10 25 13 13
5 5 3 1
6 5 2 6 3
2 1 1
7 8 1 1
RMund bowl sherds present, but
definitely identifiable
10 10 4
1 1

NTo. of

3 4
1 1
8 14
1 11
26 12 169


Shell Level pottery types which do not appear in the Monserrate
mound A collections are listed as follows:
Miscellaneous rare vessel, shape g.
Modeled figures on loop handles, class a.
Modeled figures on loop handles, class b.
Modeled head lugs on rims, class a.
Modeled head lugs on rims, class c.
The absence of modeled figures on broad loop handles of boat-
shaped vessels is significant in the Monserrate site, since this is a
relatively common element of ornamentation, primarily in the late
phase of the Shell Level deposit, at Canas and also at Coto.
The absence of class a, modeled head lugs on rims, is even more
significant, since this type is based on the bat-heads so common in
Porto Rican collections and found in large numbers in the last phase
of the Shell Level deposit, at Barrio Canas. Since the other pottery
types conform to those found at Canas, the suggestion is made that
the mound A deposit correlates with an earlier phase of the Shell
Level deposit at Canas.
Another distinctive variation in the mound A collections is the
large number of sherds bearing crude red painted designs. Only 6
sherds of this type were found at Canas and none appeared at Coto.
One complete and two partially complete vessels with this type of
painted ornamentation were found in mound A. Sherds were abund-
ant. Paint was applied in a hand about the rim or lip of vessels and in
broad, irregular curvilinear patterns on the interior of what were
apparently shallow open bowls or dishes. The bowls and sherds are
of Crude Ware regularly associated with the Shell Level deposits,
and the paint varies from a pink to red resembling the slip on the
Red Slipped Ware. The complete vessel is a shallow boat-shaped
bowl with rectangular lugs (shape B). A double scroll figure is
painted on the interior.
Seven large burial urns found in mound A were intact or complete
enough so that the shapes could be ascertained (FIGURE 13). One is a
large boat-shaped bowl with semi-lunar lugs which resemble shape
B-1, except that it is much larger than bowls of this shape found in
mound A or in the other sites. The others are simple, round, open
bowl forms similar to those smaller vessels which have been grouped
as indefinite round bowls, shape D. All the urns are of thick, coarse,
crude ware, and undecorated in any way.
A singular object, found at a depth of 50 centimeters in refuse, is a
miniature boat-shaped vessel with four short, solid legs. It is 8 centi-


Cover (Inverted)

Ixrnrie 13. Shapes of child burial urns from AMons'rrate.


meters long and stands 4 centimeters high. The modeling is unusually
crude and the surfaces are rough. It has the appearance of having
been hurriedly pressed out of a small lump of clay with no attention to
symmetry or finish. Irregular lumps and ridges remain on the sur-
face. The whole has been slipped with a thick pink clay.
In the last traces of refuse at the bottom of mound A, three sherds
of Crab Level type were found. One is an annular base designating
the Crab Level shape F. Another is the rim of a shallow round bowl
with a red painted lip; and the third is a D-shaped handle sherd of the
first class. The three sherds are distinguished from the Shell Level
types on the basis of ware and shape.

The mound A deposit produced numerous implements of shell and
stone. Bone objects were very rare. Practically all material of this
kind duplicates that found in the Shell Level at Canas. None of the
rectangular adze-like celts or shell spoons found in the Canas Crab
Level were present.
A stone pipe bowl taken from the first 25 centimeter level is the
only object of this kind found in all excavations made. It is approxi-
mately the same shape and size as a modern European tobacco pipe
bowl. The stem has been broken away. It is made of a soft granular
trap or limestone and is ornamented by rude cross-hatched incisions
about the entire bowl. The interior is obviously smoke-blackened.
Since the pipe was found within a few centimeters of the surface it
may be of modern origin.
A bird figure in clay found at a depth of 50 centimeters closely
resembles the carved 3-pointed stone zemis so common in Porto Rican
collections. The head has been broken away but the wings and tail
are well represented. The figure is crouched on an elongated base
and rises to a blunt point to form, in outline, a rough triangle. The
complete object is 10 centimeters long and 6 centimeters high.
Perforated clay discs appear in the Monserrate excavations, but are
absent at Canas and Coto. Six of these were found in mound A.
They are evidently made of potsherds and vary in diameter from 3 to
6 centimeters. The perforations are in all cases at the center and vary
slightly about 1 centimeter in diameter. The size of the discs and the
size of the perforations suggest that they were used as spindle whorls.


0 25 50 75 100
Depth in ceis. to to to to to Total
25 50 75 100 125 No.
Shell chisels ........................... 2 6 2 10
Shell celts................. ......... .. 4 2 3 1 10
Shell hoes............................. 7 2 1 1 11
Problematic shell implements............ 4 6 4 3 3 20
Shell discs............................ 1 2 3
Shell discs perforated ................... 1 1 1 1 -1
Olivia shell rattles ................... . 1
Petaloid stone celts.................... 24 20 9 7 6 66
Rubbing stones........................ 4 5 1 10
Tubular stone heads.......... ....... 1 1 2
Bone awls................ ............ 2 2
Worked manati ribs................. . 1 2 6
W worked tortoise shell ................. I 1

The following chart lists those pottery types identified with the
Shell Level in Canas which are found in the collections from mound B
at Monserrate. Depth in the deposit is indicated in centimeters.
The numbers refer to identifiable sherds.

Depth in cms.

Crude Ware .... ............. .......
lied W are..................... .......
Shape A ..... ...................
Shape A-1 .. .... . ........
Shape B .. ........ .. ..... ..........
Shape B -1 .. ...........................
Shape C ......... ..................
Shape D. ........................

Shape E .........................
Shape i ............... ............
Shape k ........................
Modeled head lugs on rim, class b.........
Modeled head lugs on rim, unclassified...
Modeled figures on vessel wall, class 1).....
Modeled figures on vessel wall, class c.....
Modeled figures on vessel wall, class d.....
Incised decoration, class a............ .
Incised decoration, class b . ...........
Incised decoration, class c................
Painted decoration .... ................

0 25 50 75
to to to to Total
25 50 75 100 No.
13 26 27 I; 72
4 1 1 6
6 12 17 2 37
15 3 18
3 4 4 1 12
1 1
3 2 5
Round bowl sherds present, but not
definitely identifiable
7 4 1 12
1 1
1 1

2 22 12 3 39


Shell Level types which do not appear in the mound B collections
are listed as follows:

Miscellaneous rare shapes f, g, h, j, and i.
Modeled head lugs on rims, classes a, c, d, and c.
Modeled figures on loop handles, classes a and b.
Modeled figures on vessel walls, class a.

It will be noted that the types absent from mound B are those
which are absent from mound A. Most significant of these are the
bat-heads and the modeled figures on loop handles.
Sherds with red painted decoration are also numerous in mound B
collections, and are of the same type as the red painted sherds in
mound A.
A distinctly new type of modeled zoomorphic figure was found on a
few Crude Ware sherds. These figures are modeled in the round and
applied to the rim of round bowl sherds. Heads and forelimbs are
represented in such a manner that the creature appears to be crawling
over the rim into the vessel. In general composition the figures
resembled the usual modeled "adornos" found on Shell Level pottery,
but the added forelimbs and the position on the rim distinguish them
from other Shell Level types.

The following chart lists those pottery types identified with the
Crab Level in Canas which are found in the collections from mound
B at Monserrate. The depth in the deposit is indicated in centimeters.
The numbers refer to identifiable sherds.

0 25 50 75 100 125 150
Depth in ems. to to to to to to to Totals
25 50 75 100 125 150 175
Shape A or B .................... 1 3 3 2 9
Shape F ......................... 1 1 1 3
Plain Brown Ware ................ 4 16 32 18 3 2 75
Plain Brown Ware, class b .......... 1 2 4 7
Red Painted Ware, class a ......... 3 7 3 2 1 16
Red and White Painted Ware, class a 2 6 6 2 1 17
D-shaped handles, 1st class ........ 1 3 6 10
D-shaped handles, 2nd class ........ 5 7 1 13
Rectangular lugs.................. 3 2 2 7
Perforated knobs ................. 3 1 1 5
Modeled head lugs, type a ......... 2 1 3


Crab Level types which do not appear in the mound B collections
are as follows:

Shapes C, D, E, G, H, I, J, and L.
Plain Brown Ware, class a.
Red Painted Ware, classes b and c.
Red and White Painted Ware, classes b, c, and d.
D-shaped handles, 3rd and 4th classes.
Semi-lunar lugs.
Modeled head nlugs, types b and c.

Since the classification of Crab Level shapes was based on a few
complete or partially complete vessels found at Canas, and since com-
plete vessels found in mound B were very rare, a correlation of shapes
is not very significant. As mentioned before, the determination of
vessel shapes from sherds of Crab Level type is difficult and subject
to much error. The significant fact remains, however, that boat-
shapes are absent and that rim sherds, for the most part, indicate
round vessels with wide flaring rims.
The types of ware and the elements of ornamentation which are
missing in mound B were rare in the Canas deposit. Red and White
Painted Ware, which is the most striking element of Crab Level
pottery, is not abundant in mound B collections, but enough sherds
are present to show a marked similarity with those found at Canas.
White is in all cases applied over red and both colors are applied over
a polished or slipped Brown Ware. Negative designs in red and
positive designs in white make up geometric patterns. Panels of
decoration, a series of crosses in white paint, and the spool-shaped
figure appear in the mound B sherds, which definitely correlate with
similar design elements found on sherds in the Canas collections. The
paint is less vivid than that found at Canas and in many cases is
partly disintegrated. In this respect the painted sherds resemble
those found at Coto.
A unique type of vessel is represented by a single fragment found at
a depth of 75-100 centimeters. The fragment is a hollow effigy head
of a human being, the neck of which apparently formed the neck of a
jar. A prominent nose and brow ridges are modeled in relief. The
eyes are represented by two round punctures and the mouth by a
long straight slit. Prominent rectangular and engraved ears are
appended. A hole 2 centimeters in diameter passes through one ear
to the interior of the effigy and has the appearance of a spout, or hole,
for pouring. The object is 10 centimeters from ear tip to car tip and
is 5 centimeters deep. The piece may be the upper section of an


effigy water jar similar to those found in Santo Domingo, but the
features are on the top surface and not on the side. Strangely enough,
the object suggests a clay lamp with the perforated ear forming a wick
aperture, but there is no precedent for this in 1\Test Indian collections.
It is of hard, thin, slipped Brown Ware and was associated with Crab
Level pottery.
A comparison of the charts on page 88 and on page 89 will indicate
that Shell Level types dominate in the upper 75 centimeters of the
mound, while Crab Level types dominate in the lower 75 centimeters.
During the excavations clear stratification of refuse was found in the
central sections of the deposit only, and it was observed that in these
sections Shell Level types were limited entirely to the upper level, and
Crab Level types entirely to the lower level. This complete segrega-
tion does not appear on the charts since all the pottery from the
mound has been presented as a unit and the arbitrary levels of the
chart do not conform to the levels of deposition. Furthermore, near
the edges of the mound there was an obvious confusion of refuse
levels and an intermixture of Shell Level and Crab Level pottery
Mound B, like mound A, produced shell, and stone objects which to
a large extent duplicate those found in the Shell Level at Canas and at
Coto. All but 6 of these (5 shell discs, 1 shell hoe) were found in the
upper 75 centimeters of deposit with Shell Level pottery.
A shell spoon and 2 white patinated chert celts were found in the
lower level, or sub-stratumn, with Crab Level pottery. Similar objects
were found in the Crab Level at Canas. Rubbing stones and worked
manati ribs found in the lowest level in mound B were found in both
Shell and Crab Level at Canas.
There were 9 unusual petaloid stone celts found in the lowest 75
centimeters. These have the petaloid shape characteristic of all Shell
Level celts but they remain unpolished. The facets of a flaking process
of manufacture can be seen on the surfaces, and no section has been
rubbed or polished. The implements are regularly made and sym-
metrically shaped. The curved bits are relatively sharp. In general
appearance they resemble a well-flaked flint celt.
A unique object found at a depth of 50 centimeters resembles the
artifacts generally referred to as pottery stamps. It is made of baked
clay and in outline is midway between a bell and a cone-shape. On
the broad end is an engraved pattern composed of straight lines and
punctures. It is 6 centimeters in length and 5 centimeters in diameter


at the flaring base. Since none of the pottery found is ornamented
with a stamped design, in all probability the object is not a pottery
stamp, but it may have been used for some other stamping process.
It is remarkable that no other object of this kind was found in all the
excavations undertaken.
The following chart indicates the distribution of shell, stone, and
bone objects in the deposit of mound B at Monserrate.

0 25 50 75 100 125
Depth in ems. to to to to to to Total
25 50 75 100 125 150 No.

Shell chisels .......... ...............
Shell celts............. ..... ..........
Shell hoes.................. ...........
Problematic shell implements...... .....
Shell discs ...................... ....
Shell discs perforated ...................
Shell rattles ...........................
3-pointed stones.......................
R ubbing stones........................
Petaloid stone celts-polished . .
Petaloid stone celts-flaked .............
White patinated chert celts. Crab Level
ty p e . . . . . .. .. .. . .. ... . ..
Shell spoons. Crab Level type ..........
Worked manati ribs .................. .

2 1
2 1
11 1

9 15

2 3 1

1 4
1 4
1 1 4
2 3 2 8

2 2 5
4 4 1 9

1 1 2
1 1
2 1 9

The chart on page 93 lists those pottery types identified with the
Crab Level at Canas which are found in the collections from mound E
at Monserrate. Depth in the deposit is indicated in centimeters. The
numbers refer to identifiable sherds and in some cases to complete or
partially complete vessels.
Crab Level types which do not appear in the mound E collections
are as follows:
Vessel shapes C, D, G, H, I and J1.
Plain Brown Ware, class a.
Red Painted Ware. classes b and d.
Red and White Painted Ware, classes b, c, and d.
D-shaped handles, 3rd and 4th classes.
Perforated knobs on vessel walls.
Modeled head lugs, types a and c.

The vessel shapes and types of ware which are absent, were rare in
the Camas collections. D-shaped handles of the 3rd and 4th classes


0 25 50 75 100 125
Depth in cms. to to to to to to Total
25 50 75 100 125 150 No.
Shape A or B .........-- ............ 4 5 3 3 15
Shape A .............................. 1 1
Shape B............... .......... 2 2 4 8
Shape E ............. ........ 1 I
Shape F .............................. 1 1 3 2 7
Shape K (Griddles-not distinguished from
Shell Level type) .................... 1 1 1 3
Plain Brown Ware ........ ............ 9 24 30 17 14 5 99
Plain Brown Ware, class b. . . . . 1 1
Red Painted Ware, class a. .............. 2 19 15 7 8 51
Red Painted Ware, class c .............. 3 3
led and White Painted Ware, class a. .. 1 1 4 1 7
D-shaped handles, 1st class ............. 5 10 13 13 3 4 48
D-shaped handles, 2nd class ....... ..... 1 1 2 1 5
Rectangular lugs ............ ..... .... 4 5 3 2 14
Sem i-lunar lugs........................ 1 1
Modeled head lugs, class h. ............. 2 2

were also rare, but perforated knobs on vessel walls were common.
Their absence in mound E collections may be significant. Half the
material removed from mound E was sent to the University of Porto
Rico and has not been studied so that the correlations are not exact.
Three unusual modeled head lugs were found at a depth of 75
centimeters in mound E. All of them clearly represent turtle heads,
and fore-flippers are modeled on the vessels at each side of the heads.
The figures, with eye punctures and mouth slits, are applied about,
one inch below the lip on the outside of round bowl sherds with red
painted rims. The sherds are Crab Level Red Painted Ware, class a.
The fragments apparently represent bowl shapes A or B.
Another unique object is a three-legged disc-shaped vessel. A
slightly concave disc 10 centimeters in diameter rests on three solid
legs 2 centimeters high. It resembles a miniature griddle on legs.
The center of the top surface of the disc has been burned black and
has partly crumbled. It has the appearance of a crucible on which a
hot flame has been concentrated. The object is light brown in color
antd resembles the usual Plain Brown Ware of the Crab Level.
A miniature boat-shaped vessel 6 centimeters in length and 2
centimeters high was found at a depth of 1 meter. The rim of the
vessel rises to a peak at each end and one of these peaks is perforated.
It is irregularly made and poorly fired. This tiny boat-shaped vessel
is peculiarly significant in a mound which contained pottery of Crab


Level type only, since boat-shapes normally are not associated with
Crab Level pottery.
A sherd ini the form of a vessel spoutl was found at a depth of 1.25
meters. It is 3 centimeters long and 1 centimeter in diameter, and is
slipped and polished Brown Ware. The object quite definitely repre-
sents a vessel spout since a small segment of the wall remains, but
nothing of the complete shape or position on the wall can be deter-
With the exception of the 4 bowls of Shell Level type, found
associated with a burial, which have been referred to in the section
concerning excavations as intrusive, and the miniature boat-shaped
vessel, all pottery found in mound E is of Crab Level type. Some of
the types established at Canas are absent and there is some variation
within the types recorded but, as a whole, the pottery correlates very
closely with that in the type site. A comparison of the charts indicat-
ing the Crab Level types in mound B and in mound E will show an
even closer similarity, and the assumption has been made that the
deposition of mound E and of the lower levels of mound B was contem-
Numerous objects of shell and stone found in mound E introduce
something of a problem. The 2 shell chisels, 8 shell discs, and 9
polished stone celts are types of artifacts which have been correlated
with the Shell Level deposit in Canas, Coto, and in mounds A and B.
The pottery of mound E is all of Crab Level type.
The 5 shell spoons are the same as the one found in the Crab Level
at Canas and two found in the Crab Level of mound B. The other
shell and stone objects and the worked manati ribs are not distinctive
of either level.
This association of a few shell and stone objects of Shell Level type
with Crab Level pottery in a mound which was composed of an inter-
mixture of marine shells and land crab shells, is difficult to interpret
in relation to the deposits at Canas and in mound B. If the deposit
of mound E was contemporaneous with the early deposit in mound
B, which is definitely indicated by the pottery types, then some Shell
Level implements may have been in use during this early period. It
is not entirely impossible, however, that these implements were
intrusive from a late occupation. Refuse of the late occupation was
found scattered over the entire site and the refuse in the top 25 centi-
meters level was obviously the same as that in the entire deposit of
mound A. This refuse was primarily marine shells, but in the upper


25 centimeters of mound E only a few scattered artifacts were found.
Some intrusion into the lower levels of mound E would be possible as
the discovery of a burial associated with Shell Level pottery has shown.
This may be, however, the interpretation of the nature of the deposi-
tion in mound E remains an embarrassing problem.
There are 21 rough-flaked petaloid stone celts which were found in
mound E, and are similar to implements of this kind found in mound
B. They have the shape of the Shell Level polished petaloid celts but
the surfaces remain with the facets of a flaking process. The number
found precludes the possibility that these implements were the com-
mon petaloid celts in the process of manufacture and indicates that
they were in use during the early period of occupation. Rough-
flaked celts of this kind were found only in the lower or Crab Level of
mound B and in mound E. No objects of the kind were found in
mound A or in the deposits of Barrio Canas and Barrio Coto.
The following chart indicates the distribution of objects of shell,
stone, and bone in the deposit of mound E, at Monserrate.

0 25 50 75 100
Depth in cms. to to to to to Total
25 50 75 100 125 No.
Shell chisels ........................... 2 2
Problematic shell implements ............ 11 3 1 1 16
Shell spatula .......................... 1 1
Shell discs............................. 1 5 6
Shell discs perforated ................... 1 1 2
Shell rattles .......................... 4 2 6
Shell spoons ........................ ... 2 1 1 1 5
Petaloid stone celts polished............. 2 5 2 9
Petaloid stone celts rough-flaked ......... 9 4 5 3 21
Rubbing stones ...................... ..1 1
Worked manati ribs .......... .......... 2 2 2 6

Excavations in Barrio Monserrate were limited to three low mounds
of the five which appeared in an extensive area strewn with the refuse
of prehistoric occupation. These mounds have been referred to as
A, B, and E.
Mound A was composed of culture refuse which was similar to that
found in the Shell Level at Canas. Marine shells made up the bulk of
the food refuse and the artifacts were of Shell Level type. Some
variation in the deposit could be determined. The upper levels con-


trained a great mass of food refuse and large quantities of potsherds
and associated artifacts, but no burials. Below this was an indistinct
stratum containing markedly less culture refuse but numerous pri-
mary flexed burials and several child urn burials. The last 40 to 50
centimeters of deposit contained still less refuse and a large number of
primary flexed burials often lying so closely associated that one
skeleton could be distinguished only with difficulty from another.
There were no urn burials in this lowest stratum. No marked dis-
tinctions in types of artifacts could be determined in the levels and
all pottery and associated artifacts found throughout, to a large
extent, duplicated those found in the Shell Level at Canas.
Mound B definitely indicated two periods of occupation. In the
central sections of the mound, a distinct stratum of land crab shells
was found below the stratum of marine shells. The lower stratum con-
tained pottery which was entirely of Crab Level type as established at
Canas, and the upper stratum contained pottery which was of Shell
Level type. Burials were unusually rare in mound B.
Mound E, as a whole, contained much less culture refuse than the
other two mounds. The surface 20 to 25 centimeters were composed
of humus interspersed with marine shells and scattered land crab
shells. Below this, marine shells appeared in small numbers and land
crab shells were numerous, but were not found in masses as at Canas
and in mound B. Only a few scattered sherds were found in the first
20 to 25 centimeters of deposit, and below this, sherds and associated
artifacts were more numerous but relatively scarce. Practically all
pottery found was of Crab Level type. Some of the associated arti-
facts were of Crab and some were of Shell Level type. Only a very few
burials were found.
All Shell Level types established at Barrio Canas were not found at
Mlonserrate. Most significant of these were the bat-head "adornos"
and the modeled figures on broad loop handles. These were absent
from mound A as well as mound B. Pottery with crude curvilinear
designs in red paint was abundant in the marine shell deposits of both
mound A and mound B. Only six sherds of this type of ware were
found in the Shell Level at Canas. With these exceptions and also
with the exceptions of a few rare pieces, the marine shell deposit at
Monserrate duplicated the marine shell deposits at Canas.
Crab Level types of pottery found in the sub-stratum of mound B
were the same as those found throughout mound E. A few of the
Crab Level types established at Canas were absent. Most significant
of these were Red and White Painted Ware, classes b and c, and the

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs