Skip to main content

Full text of "Annual report of the Bureau of American Ethnology to the Secretary of the Smithsonian Institution"

See other formats



'^■> h^/- ^ 













Smithsonian Institution^ 
Bureau of American Ethnology^, 
Washington, D. C, August 5, 1913. 
Sir: I have the honor to submit herewith the Thirty- 
fourth Annual Report of the Bureau of J^ierican Eth- 
nology, for the fiscal year ended June 30, 1913. 

With ajDpreciation of your aid in the work under my 
charge, I am 

Very respectfully, yours, 

P. W. Hodge, 
Eth nologist-in-cJiarge. 
Dr. Charles D. Walcott, 

Secretary of the Smithsonian Institution. 


Systematic researches 


Publications " _ .. 




Library ,. _ 






Miscellaneous _ _ 




A jirebistoric island culture area of America 

— .*i5 





F. W. Hodge, Ethnologist-in-Cliarge 

The operations of the Bureau of American Ethnology 
during the fiscal year ended June 30, 1913, were conducted 
in accordance with the act of Congress approved August 
24, 1912, making appropriations for sundry civil expenses 
of the Government, which act contains the following item : 

Amei-ican etlinology : For continuing ethnological researches 
among the American Indians and the natives of Hawaii, inchiding 
the excavation and preservation of archseologic remains, nnder the 
direction of the Smithsonian Institution, inchiding salaries or com- 
'pensation of all necessary employees and the purchase of necessary 
books and periodicals, including payment in advance for subscriii- 
tions, $42,000. 


The systematic researches were conducted by the regular 
staff of the bureau, consisting of seven ethnologists, and 
by other specialists not directly connected ^vith the bu- 
reau. These operations may be siuinnarized as follows: 

Mr. F. W. Hodge, ethnologist-iu-charge, was occupied 
almost entirelj^ during the year with administrative affairs 
pertaining to the bureau's activities. He was able to de- 
vote some time to the preparation of the Bibliography of 
the Pueblo Indians, the writings relating to the subject 
covering so extended a period (from 1539 to date) and 
being so numerous that much remains to be done. He 
devoted attention also, as opportunity offered, to the re- 
vision of certain sections of the Handbook of American 



iBdians, but as it is the desire to revise this work com- 
pletely, with the aid of the entire staff of the bureau as 
well as of other specialists, little more than a beginning 
of the revision has been made. Mr. Hodge continued 
to represent the Smithsonian Institution at the meetings 
of the United States Board on Geographic Names, and 
the Bureau of American Ethnologj^ on the Smithsonian 
advisory committee on printing and publication. 

Dr. J. Walter Fewkes, ethnologist, spent the summer 
months and part of the autmim of 1912 in correcting the 
proofs of his monograph on Casa Grande and of his report 
on the Antiquities of the Upper Verde River and Walnut 
Creek Valleys, Arizona, both of which appear in the 
Twenty-eighth Amiual Report of the bureau, and in com- 
pleting the draft of a memoir devoted to the Symbolic De- 
signs on Hopi Pottery, which it is designed to publish with 
nmnerous illustrations. The remainder of the autumn 
was occui^ied by Dr. Fewkes in gathering material for an 
eventual memoir on the Culture History of the Aborigines 
of the Lesser Antilles, these data being derived chiefly 
from a study of the early literature of the subject and of 
the rich West Indian collections from the island of St. 
Vincent in the Heye Museum of New York City. Pre- 
paratory to the publication of the final results, Dr. 
Fewkes, with the generous permission of George G. Heye, 
Esq., selected with entire freedom the necessary objects 
for illustration, and before the close of the fiscal year 
about 200 di'awings of the archeological objects in this im- 
portant collection had been finished. 

In October, 1912, Dr. Fewkes sailed for the West Indies 
imder the joint auspices of the bureau and the Heye Mu- 
seum, the special object in view being the gathering of new 
archeological data through the excavation of village sites 
and refuse-heaps and the examination of local collections 
in the islands. Dr. Fewkes visited Trinidad, Barbados, 
St. Vincent, Ballieeaux, Grenada, Dominica, St. Kitts, 
Santa Cruz, and other islands, excavating shell-heaps in 
Trinidad and Ballieeaux, and making archeological studies 


ill other isles. The results of the im^estigations in Trini- 
dad jiroved to be especially important, owing to the light 
which they shed on the material culture of the former 
aborigines of the coast adjacent to South America. 

Extensive' excavations were made in a large shell heap, 
known as Chii5-chip Hill, on the shore of Erin Bay in 
the Cedros district. This midden is historic, for it was 
in Erin Bay that Columbus anchored on his third voyage, 
sending men ashore to till their casks at the spring or 
stream near this Indian mound. Chip-chip Hill is now 
covered with buildings to so great an extent that it was 
possible to conduct excavations only at its periphery; 
nevertheless the diggings yielded a rich and unique col- 
lection that well illustrates the culture of the natives 
of this part of Trinidad. The collection consists of sev- 
eral fine unbroken pottery vessels with painted decoration, 
and more than a hundred well-made effigy heads of clay, 
in addition to effigy jars and many broken decorated bowls. 
There were also obtained from the Erin Bay midden sev- 
eral stone hatchets characteristic of Trinidad and the ad- 
jacent coast of South America, a few shell and bone gor- 
gets, and other artifacts illustrating the activities of the 
former inliabitants. It is an interesting fact that as a 
whole the objects here found resemble those that have been 
taken from shell heaps on the Venezuela coast and from 
the Pomeroon district of British Guiana more closely than 
they resemble related siiecimens from the other islands 
of the Lesser Antilles. Several other middens were 
examined in Trinidad, the most representative of which 
is situated near San Jose, the old Spanish capital. Prom- 
ising shell heaps were discovered also at Mayaro Bay on 
the eastern coast. 

One of the most important results of the "West Indian 
field work by Dr. Fewkes was a determination of the 
geographical distribution of certain types of artifacts and 
'a comparison of the prehistoric culture areas in the so- 
called Carib Islands. Evidence of the existence of a 
sedentary cidture on these islands preceding that of the 


Carib was obtained, showing it to have distantly resembled 
that of Porto Rico; this culture, however, was not uni- 
form. Dr. Fewkes also found that there were a number 
of subcultures in these islands. In prehistoric time Trini- 
dad and Tobago, it was determined, were somewhat suni- 
lar culturally, just as they are similar geologically and 
biologically, to northern South America, In Dr. Fewkes 's 
opinion perhaps nowhere is the effect of environment on 
human culture better illustrated than in the chain of 
islands extending from Grenada to Guadeloupe, which 
were inhabited, when discovered, by Carib, some of whose 
descendants are still to be found in Dominica and St. Vin- 
cent. The earlier or pre-Carib people were cidturally dis- 
tinct from those of Trinidad in the south, St. Kitts in the 
north, and Barbados in the east. The stone implements 
of the area are characteristic and the prehistoric pottery 
can readily be distinguished from that of the islands be- 
yond the limits named. 

A large niunber of shell heajDS on St. Vincent were 
visited and studies made of localities in that island in 
which caches of stone implements have been found. Six 
groups of petroglyi:)hs were examined, even some of the 
best known of which have never been described. Special 
effort was made to obtain information respecting the 
origin of certain problematical objects of tufaceous stone 
in the Heye Museum, said to have been collected from be- 
neath the lava beds on the flank of the Souf riere. 

Dr. Fewkes visited the locality on the island of Balli- 
ceaux where the Carib of St. Vincent were settled after the 
Carib wars and before they were deported to Roatan, on 
the coast of Honduras. Extensive excavations were made 
at the site of their former settlement at Banana Bay, 
where there is now a midden overgrown with brush. Here 
much pottery, as well as several hiunan skeletons and some 
shells and animal bones, were found. 

The mixed-blood sui'vivors of the St. Vincent Carib who ■ 
once lived at Morne Rond, near the Souf riere, but who are 
now settled at Cami^den Park, near Kingstown, were 


visited. These still retain some of their old customs, as 
making cassava from the poisonous roots of the maniliot, 
and preserve a few words of their native tongue. A brief 
vocabulary was obtained, but Carib is no longer habitually 
Slacken in St. Vincent. 

The fertile island of St. Kitts and the neighboring Nevis 
were foimd to be particularly instructive archeologically. 
Both have several extensive middens and well-preserved 
pictographs, the fonner having jdelded many artifacts 
that illustrate the material culture of its pre-Carib 
inliabitants. Through the courtesy of Mr. Comiell his 
large collection, which adequately illustrates the culture of 
St. Kitts and Nevis, was placed at the disposal of Dr. 
Fewkes for the purj^ose of study, and he was i^ermitted to 
make drawings of the more tyj^ieal objects, one of the most 
instructive of which is a sculptured torso from Nevis. 

In Barbados Dr. Fewkes examined the midden at Indian 
River, on the west coast, from which site the imi:)ortant 
Taylor archeological collection was gathered. Several 
other middens were visited on the lee coast from Bridge- 
town to the northern end of the island, where a marly hill 
strewn with potsherds was observed. He also examined 
the so-called "Indian excavations" at Freshwater Bay 
and others at Indian Eiver, and visited several cave shel- 
ters on the island. The most noteworthy of these caves 
are situated at Mount Gilboa and in the Scotland district, 
St. Lucy Parish. To one of these, known as the " Indian 
Castle," described in 1750 by the Rev. Griffith Hughes, 
who claims to have found therein an idol and other un- 
doubted Indian objects. Dr. Fewkes devoted much atten- 
tion. The gulches so characteristic of Barbados were 
favorite resorts of the aborigines, and, judging by the 
artifacts, furnished cave shelters for them. Although 
uninhabited at the time of its discovery, there is evidence 
of a considerable i^rehistoric aboriginal i)opi;lation in Bar- 
bados, whose culture was influenced largely by the charac- 
ter of the material from which their artifacts were made, 
most of them being fashioned from shell instead of stone — 


a characteristic seemingly constituting this island a special 
culture area. 

A collection of stone implements, including celts, axes, 
and other objects, was gathered at Santa Cruz. Several 
local collections of archeological objects were examined, 
and the large midden at the mouth of Salt River was 
visited. The prehistoric objects obtained on this island 
and from St. Thomas resemble those from Porto Rico. 

Although the Carib inhabitants of the Lesser Antilles 
are no longer of pure blood, and their language is known 
to only a few persons in Dominica and St. Vincent, and to 
these but imperfectly, it was found that the Negroes, who 
form more than nirie-lenths of the insular i:)opulation, re- 
tain in modified form some traces of the material culture 
of the Indians. Cassava is the chief food of manj'^ of the 
people, and the method of its i^reparation has been little 
changed since aboriginal times. Cocoa is ground on a 
stone and made into cylindrical rolls in much the same 
manner as it was prepared by the Indians in early times. 
The basketry made in Dominica was found to be the same 
in style and materials as is described by the early mis- 
sionaries to the Carib ; while the Negroes of Nevis manu- 
facture pottery of the same form and ornament and biuni 
it in much the same way as that found in the middens of 
St. Kitts. In working their spells the obia men commonly 
sprinkle stone objects with the blood of a goat, and the 
common people regard petroglyphs as " jumbies," or 
bugaboos. A great niunber of folk tales of a mixed ab- 
original and Negro type are still recounted in the cabins 
of the lowly, where Carib names for animals, plants, and 
places are household words. 

On his return to Washington Dr. Fewkes undertook the 
preparation of a report on his archeological researches in 
the West Indies, and considerable progress therein had 
been made by the close of the fiscal year. 

Mr. James Mooney, ethnologist, was occupied during 
the greater part of the year with the investigation of In- 
dian pojiulation, which has engaged his attention for a con- 


siderable time. This research covers the whole period 
from the first occupancy of the country by wliite people 
to the present time, and includes the entire territory from 
the Rio Grande to the Arctic. To make possible system- 
atic treatment the area covered has been mapped into 
about 25 sections, each of which constitutes approximately 
a single geographical and historic unit for separate treat- 
ment, although lunnerous migrations and removals and 
the frequent formation of new combinations necessitate a 
constant overlapping of the work of the sections. Sev- 
eral of the eastern areas have been completed and more 
or less progress has been made with each of the others. 
More recently Mr. Mooney has concentrated attention on 
Alaska and western Canada, for the Arctic parts of which 
Mr. Villi jahnur Stefansson and Dr. Waldemar Jochelson 
have generously furnished new and valuable data. In this 
memoir the plan is to include chapters on notable epi- 
demics, vital statistics, and race admixture, and the work 
is intended to appear as a monograjDh on the subject. 

On June 18, 1913, Mr. Mooney proceeded to the Eastern 
Cherokee Indians in North Carolina to continue his in- 
vestigations of the medical and religious rituals of that 
tribe, commenced a number of years ago, as it was deemed 
wise to finish this part of his Cherokee studies as soon as 
practicable by reason of the changes that are so rapidly 
taking place among this people. Mr. Mooney was still in 
the field at the close of the fiscal year. 

Dr. John R. S wanton, ethnologist, continued, both in the 
field and at the office, his studies of the Indians formerly 
occupying the territory of the Southern States. He spent 
the month of November, 1912, with the Alabama and 
Koasati Indians in Polk County, Tex., where he recorded 
250 pages of texts in the dialects spoken by these two 
tribes, corrected several texts obtained on earlier expedi- 
tions, and added materially to his general ethnological in- 
formation regarding them. In December Dr. Swanton 
proceeded to Oklahoma, where he obtained about 50 pages 
of text in Hitchiti, a language now confined to a very few 


pei-sons among the Creek Indians, and collected a few 
notes regarding the Choctaw. 

Before his departure from Washington and after his 
return Dr. Swanton spent the greater part of the time in 
collecting information concerning the southern tribes from 
early Spanish, French, and English authorities. Con- 
siderable attention was also devoted to reading the i^roofs 
of the Rev. Cji'us Byington's Choctaw Dictionary, now in 
process of printing, in which labor he was efficiently aided 
by Mr. H. S. Halbert, of the Alabama State department of 
archives and history. Dr. Swanton also conmienced a 
general grammatical study of the languages of the Musk- 
hogean stock, particularly Alabama, Hitchiti, and Choc- 
taw, and in order to fm-ther this work he was subsequently 
engaged in making a preliminary stem catalogue of Creek 
from the material recorded by the late Dr. Gatschet, simi- 
lar to the catalogue already prepared for Hitchiti, Ala- 
bama, and Natchez. He began also the prej^aration of a 
card catalogue of words in Timucua, the ancient extinct 
language of Florida, taken from the grammar and cate- 
chisms of Father Pare j a. In May, Dr. Swanton visited 
New York in order to examine rare Timucua works in the 
Buckingham Smith collection of the New York Historical 
Society. Through the courtesy of this society and of the 
New York Public Library arrangements have been made 
for furnishing photostat coiiies of these rare and impor- 
tant books, and the reproductions were in prej^aration at 
the close of the fiscal year. 

In connection with the researches of Dr. Swanton it is 
gratifying to report that he was awarded last spring the 
second Loubat prize in recognition of his two publica- 
tions — " Tlingit Myths and Texts " and " Indian Tribes 
of the Lower Mississijjpi Valley and Adjacent Coast of 
the Gulf of Mexico ' ' — both issued by the bureau. 

Mrs. M. C. Stevenson, ethnologist, devoted her time to 
the conclusion of her researches among the Tewa Indians 
of New Mexico and to the preparation of a paper on that 
interesting and conservative peoi^le. A preliminary table 


of contents of the proposed memoir indicates that her 
studies of the customs and beliefs of the Tewa will be as 
comprehensive as the published results of her investiga- 
tions of the Sia and the Zuni tribe of the same State. As 
at present outlined, the work, which will soon be com- 
pleted, will contain six sections, dealing with the following- 
subjects, respectively: Philosophy, anthropic worship and 
ritual, zoic worship, social customs, material culture, and 

Dr. Truman Michelson, ethnologist, continued his 
studies among the Algonquian tribes. In the middle of 
July, 1912, he proceeded to the Fox Indians, at Tama, 
Iowa, from whom a large additional body of m\i;hological 
material was obtained ; this, in connection with the m}i:hs 
and legends in the fomi of texts gathered during the pre- 
vious season, approxiaiates 7,000 pages. "V^Hien the trans- 
lation of this material shall have been finished it will form 
one of the most exhaustive collections of m}i;hology of 
any Indian tribe. It is noteworthy that these myths and 
tales differ essentially in style from those gathered by the 
late Dr. William Jones (scarcely any of whose material 
has been duplicated by Dr. Michelson) — a fact that 
emphasizes the necessity of recording such material in the 
aboriginal tongue. It may be added that the myths and 
tales collected are also important in the light they shed 
on the dissemination of myths. Study of the social and 
ceremonial organization of the Fox Indians was likewise 
continued, and especially full notes were obtained on their 
Religion dance. Many of the songs of one of the drums 
were recorded on a dictaphone and several photographs 
of the native ball game were secured. 

Dr. Michelson next proceeded to Haskell Institute, the 
nonreservation Indian school at Lawrence, Kans., for the 
purpose of obtaining notes on Atsina (Cfros Ventre) and 
several other Algonqman languages, the results of which 
show definitely that Atsina shares with Arapaho all the 
deviations from normal Algonquian, and that Potawatomi 
is further removed from Ojibwa, Ottawa, and Algonkin 
than anv one of these is from the others. 


Dr. Michelson next visited the Muusee, in Kansas, but 
found that, unfortunately, little is now available in the 
way of information except as to their language, which is 
still spoken by about half a dozen indi-\dduals, though none 
employ it habitually. 

The Delawares of Oklahoma were next visited, Dr. 
Michelson finding that their aboriginal customs are still 
retained to a large extent. Extended observations were 
made on several dances, and, to a lesser extent, on the social 
organization. From a study of the Delaware language, 
together with the Munsee dialect of Kansas, it was ascer- 
tained, as had previously been surmised, that the Dela- 
ware language of the early Moravian missionary Zeis- 
berger represents no single dialect but a medley of several 

On his way to Washington Dr. Michelson stopped again 
at Tama to obtain additional notes on the Fox Indians; 
at the same time he succeeded in arranging for the acquire- 
ment of certain sacred packs for the National Museum. 
He also visited Chicago and New York for the purpose 
of making comparative observations on the material cul- 
ture of the Fox tribe, based on collections in the museums 
of those cities. 

On his arrival in Washington, at the close of December, 
Dr. Michelson undertook the translation and study of the 
Fox myths. The results indicate that very great firmness 
in the word unit in Algonquian is more apparent than 
real, and that the classification of stems must be revised. 
Dr. Michelsen also brought to conclusion his translation 
of the Kickapoo myths and tales, collected by the late Dr. 
Jones, to which were added notes on Kickapoo granunar 
and comparative notes on the myths and tales, the whole 
making somewhat more than 300 pages. 

Through correspondence Dr. Michelson succeeded in 
arranging for the acquirement of other sacred packs of the 
Fox Indians, which have been deposited in the National 
Museum. He also aided in furnishing information in 
answer to inquiries by various correspondents, and from 


time to time supplied data for incorporation in a new edi- 
tion of the Handbook of American Indians. 

Prom the investigations of the bureau it seemed that the 
Siouan and Muskhogean languages resembled each other 
morjohologieally. In view of these circumstances it was 
deemed desirable that the Catawba, one of the Siouan 
tongues, should be restudied, and accordingly, toward the 
close of May, 1913, Dr. Michelson proceeded to South Caro- 
lina, where the remnant of the Catawba tribe still reside. 
Unfortmiately, it was found that the language is all but 
extinct, not even half a dozen persons being able to recall 
l^hrases, although isolated words can still be had in goodly 
nmnber. Owing to this paucity of text material it is 
hardly likely that the granmiar of Catawba w^ll ever be 
completely elucidated, and as no comparative study with 
other Siouan dialects has yet been made, it is not prac- 
ticable at present to say with which Siouan groui^ the lan- 
guage is most closely associated. A considerable nmnber 
of native songs are still remembered by the surviving Ca- 
tawba, nearly all of which Dr. Michelson succeeded in re- 
cording by dictaphone. 

Mr. J. N. B. Hewitt, ethnologist, was occupied during 
the year in translating unedited Seneca texts of myths 
which were collected by himself in 1896 and at other times 
on the Cattaraugus Resei-vation in western New York and 
on the Grand River Reservation in Ontario, Canada. 
These mji:hs, legends, and tales number 13 in all. In ad- 
dition, Mr. Hewitt undertook the editing of two Seneca 
texts — " The Legend of S'hagowe''uot'ha', or The Spirit 
of the Tides," and " The Tale of Doa'daneggr' and 
Hotk\visdadege"''a' " — recorded by himself in the form of 
field notes in 1896 and aggregating 95 tyiDewi'itten pages. 
At the close of the fiscal year about one-third of this work 
was completed. To these texts interlinear translations are 
to be added for the purpose of aiding in the grammatie 
study of the Seneca tongue. 

Mr. Hewitt also devoted much time to the collection and 
preparation of data for answers to correspondents of the 

160658°— 34 ETH— 22 2 


bureau, especially with reference to the Iroquoian aud 
Algouquian tribes. 

Mr. Francis La Flesche, ethnologist, continued his in- 
vestigations of the ethnology of the Osage Indians, giving 
particular attention to their rituals and accompanj-ing 
songs. He was enabled to record on the dictaphone the 
songs and fragments of the rituals belonging to the 
Waxobe degree of the No^'ho^zhi^ga rites, of which, as 
noted in the last annual report, he has been making a spe- 
cial study. These rituals have been transcribed and, mth 
the 84 songs that have been transcribed in musical nota- 
tion by Miss Alice C. Fletcher, comprise 66 typewritten 

Mr. La Flesche has also been able to record the 
No^'zhi^zho", or Fasting degree, of the Puma and Black 
Bear gentes. These two organizations are closely re- 
lated; they now not only use in common the songs and 
rituals of the No^'ho^zhi^ga rites, but they even go to the 
extent of exchanging gentile personal names as full recog- 
nition of their relationship. The No°'zhi"zho'' degree em- 
ploys 12 rituals and numerous songs, of which latter 81 
have been recorded. These songs are divided into two 
great groups, first of which is known as " The Seven 
Songs," having 16 sets, and the second, " The Six Songs," 
having 17 sets. The Osage texts of these rituals and songs 
cover 207 pages, about three-fourths of which have been 
finally typewritten. The 81 songs have been transcribed 
in musical notation by Miss Fletcher, while the transla- 
tion of the rituals and the words of the songs is in progress. 

In the autmim of 1912 Mr. La Flesche was fortunate in 
securing in full the Ni'k'i degree of these intricate Osage 
rites. Hitherto he had been able to obtain only the begin- 
ning of this degree, but his informant was finally induced 
to recite it in its entirety, comprising 1,542 lines. The 
real title of this degree is Ni'k'i No^k'o", " The Hearing 
of the Words of the People." In it the genesis of the 
tribe is given in a story made up of myth, legend, and 
symbolism, the whole being clearly devised to keep the 


people ever mindful of the necessity of au orderly aud au- 
thoritative conduct of war. It goes to show that the prin- 
ciple of war was early recognized by the Osage as the 
surest means hj which not only tribal and individual life 
might be safeguarded against strange and hostile tribes, 
but also as the means by which the tranquil enjoAinent of 
game and other natural products of their environment 
might be won. It is to this coveted tranquillity that the 
closing lines of man}' of the rituals refer, invariably lik- 
ening it to a " serene day." This degree employs ritual 
almost entirely, there being only 10 songs. The native 
ritual comprises 57 typewritten pages, of which a large 
l^art has been translated. 

In the sj^ring of 1913 Mr. La Flesche obtained the Rush 
Mat Weaving degree of the Pimia and Black Bear gentes. 
Only the " Seven Songs " spoken of before, with various 
ceremonial forms, are employed in this degree, the " Six 
Songs " being entirely omitted. The distinguishing fea- 
tures are the ceremonial weaving of the rush mat for the 
sacred case in which were enshrined the bird and other 
sacred objects, the renewal of all the articles that make 
up the sacred bundle, and the ceremonial stitching of the 
ends of the case. In some respects this is one of the most 
extraordi}iary degrees of the Osage that INIr. La Flesche 
has yet observed, since in its performance there are used 
70 brass kettles, 70 red-handled knives, and 70 awls in 
making the various articles, all of which the votary is 
obliged to furnish, together with other expensive articles 
that constitute the fees of the initiator and other officiating 
No°'ho''zhi''ga, as also 70 pieces of choice jerked meat for 
distribution among the members attending the initiation. 
Three rituals not used in the other degrees are employed 
in this, namely, the Green Rush ritual, the Bark ritual, 
and the Stitching and Cutting ritual. There are 61 pages 
of Osage text, about half of which have been transcribed. 

Mr. La Flesche also obtained the rituals and songs of 
the Washabe Athi", " The Carrying of a Dark Object," 
with full description of the various processions and cere- 


inouial forms. This is a war ceremouy, which, although 
uot counted as a degree, is a rite to which the seven degrees 
lead. The name of this ceremouy is derived from the war 
insignia, which is the charcoal ceremonially prepared from 
certain sacred trees, and which symbolizes the black marks 
denoting the birds and animals used to typify strength, 
courage, and fleetness. Mr. La Flesche's Osage informant 
regards this as the final act of the seven degrees. The 
Osage text comprises 90 pages, nearly one-half of which 
has been transcribed, together with 36 songs, which have 
been transcribed by Miss Fletcher, and 7 diagrams. 

Mr. La Flesche was fortunate enough to procure the 
sacred bundle of the Deer gens and the reed- whistle bundle 
of the Wind gens ; the contents of the latter are of excep- 
tional interest. Mrs. Brogahige, one of the ceremonial 
weavers of the Osage, at considerable sacrifice to herself, 
presented Mr. La Flesche two sacred looms, one of which 
is used in weaving the buffalo-hair case, aud the other in 
weaving the rush case for the sacred bird. These packs, 
together with specimens of ceremonially made burden 
straps which Mr. La Flesche collected, have been placed 
in the National Museum. 

Dr. Franz Boas, honorary philologist, continued the 
preparation of the material for the Handbook of American 
Indian Languages. As stated in the last annual report, the 
manuscript of the grammar of the Chukchee language, to 
appear in part 2 of this handbook, was completed and in its 
final form was discussed with the author, Mr. Waldemar 
Bogoras, during the visit of Dr. Boas to Berlin in the sum- 
mer of 1912. The results of these discussions were em- 
bodied in the work, the manuscript was delivered, and the 
typesetting connnenced. At the same time Dr. Boas stud- 
ied the Koryak texts collected by Mr, Bogoras, published 
in accordance with the plan previously outlined, at the ex- 
pense of the American Ethnological Society, and the indis- 
pensable references were einbodied in the granmiatical 


The Coos grammar by Dr. Leo J. Fraelitenberg was 
completed, so far as the work of the editor, Dr. Boas, is 
concerned, the r»ag'e proofs having been finally revised. 

The manuscript for the Siuslaw grammar, also by Dr. 
Frachtenberg, was submitted and the editing considerably 
advanced. This will be comi^leted as soon as the entire 
series of Siuslaw texts are in print — a work that has been 
undertaken under Dr. Boas's editorship by Columbia 
University. All the collected texts are now in tyi^e, so 
that examples can be added to the manuscript of the 

Dr. Frachtenberg remained in Siletz, Oreg., throughout 
the year for the pur})ose of revising on the spot the mate- 
rials on the Oregon languages. He was engaged in collect- 
ing and arranging the Alsea material for part 2 of the 
Handbook of Languages, and in preparing for the discus- 
sion of his Molala linguistics. The rapid disappearance of 
the Calapooya may make it necessary, however, to com- 
plete the field work on the language of this people before 
closing the work on the other manuscripts, even though 
this procedure may entail delay in the printing of the 

Dr. Alexander F. Chamberlain, of Clark University, 
who has undertaken the preparation of a grannnar of the 
Kutenai language, expects to deliver his manuscript early 
in the new fiscal year. The printing of this sketch must 
necessarily be delayed until the text material is available 
in print. 

Miss Haessler contimied her jireparations for a careful 
revision of the Dakota Dictionarj^ by Riggs — a work made 
necessary by reason of the need of greater precision in 
phonetics and translation, as well as of a more systematic 
arrangement of the material. Miss Haessler exi^ected to 
complete all the preliminary work by the smmner of 1914, 
so that, should facilities be available, she would then be 
able to undertake the required field work. 

Miss Frances Densmore continued her studies in Indian 
music, devoting special attention to that of the Sioux, and 


during the year submitted three papers, comprising 252 
pages of manuscriijt, original phonographic records and 
musical transcrij^tion of 107 songs, and 23 original photo- 
grajDhic illustrations. Three subjects have been ex- 
haustively studied and a fourth is represented in such 
manner that the results may be regarded as ready for pub- 
lication. The three principal subjects are the sacred 
stones, dreams about animals, and the buffalo hunt. The 
fourth subject referred to relates to the warpath and is 
represented b}' about 20 songs, but it awaits further study 
of the military societies. A special group of songs con- 
sists of those which have been composed and sung by the 
Sioux in honor of Miss Densmore. 

A study of the music of the Mandan and Hidatsa at Fort 
Berthold, N. Dak., w^as made by Miss Densmore in the smn- 
mer of 1912, in cooperation with the Historical Society of 
the State of North Dakota. The results of this investiga- 
tion consist of a manuscript of about 50 pages, with tran- 
scriptions of 40 songs. 

Miss Densmore also read the proofs of Bulletin 53 
(Chippewa Music — II), which is now in press. 

Mr. W. H. Holmes, head curator of the department of 
anthropology of the United States National Museum, con- 
tinued the preparation of the Handbook of American 
Archagologj' for publication by the bureau, as far as the 
limited time available for the purpose permitted. Aside 
from the preparation of the text and illustrations for 
parts 1 and 2 of this handbook, Mr. Holmes made field ob- 
servations among the ancient mica mines in western North 
Carolina and among mounds and village sites in South 
Carolina and Georgia. He also visited a number of 
museums for the purpose of examining the collections of 
archaeological material, among them being the museums of 
Boston, Andover, New York City, Philadelphia, Colum- 
bus, Chicago, Milwaukee, Madison, Davenport, and St. 

Mr. D. I. Bushnell, jr., made good jirogress in the com- 
pilation of the Handbook of Aboriginal Remains East of 


the Mississippi, the manuscript material for which, record- 
ed on cards, now approximates 16,000 words. The collated 
material has been derived from (1) replies to circular let- 
ters addressed to county clerks in all of the States east of 
the Mississippi, (2) coimnunications from various so- 
cieties and individuals, and (3) publications pertaining to 
the subject of American antiqmties. It is gratifying to 
state that there are very few areas not covered by the ma- 
terial already in hand, and it is expected that through the 
systematic manner in which Mr. Bushnell is prosecuting 
the work the handbook will be as complete as it is prac- 
ticable to make it by the time it is ready for publication. 
The investigations conducted jointly in 1910 and 1911 
by the bureau and the School of American Archaeology 
have borne additional fruit. An extended memoir on the 
Ethnogeography of the Tewa Indians, by J. P. Harrington, 
was received and will appear as the "accompanying pa- 
per" of the Twenty-ninth Annual Eeport, now in press. 
Three bulletins, namely (No. 54), The Physiography of 
the Rio Grande Valley, New Mexico, in Relation to Pueblo 
Culture, by Edgar L. Hcwett, Junius Henderson, and 
W. W. Robbins ; (No. 55) The Ethnobotany of the Tewa 
Indians, by Barbara W. Freire-Marreco, AY. "W. Robbins. 
and J. P. Harrington; and (No. 56) The Ethnozoology of 
the Tewa Indians, by Junius Henderson and J. P. Har- 
rington, were also presented as a part of the results of the 
joint exi:)editions and are either published or in process of 
printing. Mr. Harrington also made progress in the prep- 
aration of his report on the Mohave Indians, and Miss 
Freire-Marreco is exj^ected to submit shortly an extended 
IDajDcr on the Yavaj^ai tribe. There remains to be men- 
tioned in this connection another memoir, namely, An In- 
troduction to the Study of the Maya Hieroglyi:)hs, by Syl- 
vanus G. Morley; while not a direct product of the joint 
work of the bureau and the school, this is in a measure an 
outgrowth of it. The manuscript, together with the accom- 
panying illustrations, has been submitted to the bureau, 
but is now temporarily in the author's hands for slight 


Since the publication of the Handbook of American In- 
dians, through which additional popular interest in our 
aborigines has been aroused, it has been the desire to make 
a beginning toward the preparation of a series of hand- 
books devoted to the Indians of the respective States. 
The opportunity was fortunately presented toward the 
close of the fiscal year, when the bureau was enabled to 
enlist the aid of Dr. A. L. Kroeber, of the University of 
California, who has kindly consented to midertake the 
preparation of the initial volume of the series, to be de- 
voted to the Indians of California. It is planned to pre- 
sent the material in each volume in as popular a form as 
l^racticable, in order that it may be made of the greatest 
use to schools, and it is hoped that the means may be soon 
available to make possible the extension of the series to 
other States. 

Under a small allotment from the bureau, Mr. James 
Murie continued his studies of Pawnee ceremonies. He 
devoted special attention to the medicine rites, and on 
June 13, 1913, submitted a description of the ritual jier- 
taining to the " Purification of the Buffalo Skull." 

The transcription of the manuscript French-Miami Dic- 
tionary in the John Carter Brown Library at Providence, 
R. I., to which attention has been directed in previous re- 
ports, was finished by Miss Margaret Bingham Stillwell, 
who submitted the last pages of the vocabulary (which 
niunber 1,120 in all) early in January, 1913. The bureau 
is under obligations to Mr. George Parker Winship, 
librarian of the John Carter Brown Library, for his gen- 
eroiis cooperation in placing this valued docmnent at the 
disposal of the bureau and to Miss Stilhyell for the effi- 
cient manner in which this difficult task was accomplished. 

In the latter part of the fiscal year Mr. Jacob P. Diinn, 
of Indianapolis, in whose hands the French-Miami Dic- 
tionary was placed for study, commenced the annotation 
of the transcription and the addition of English equiva- 
lents. This necessitated a journey to Oklahoma, where 
Mr. Dunn enlisted the services of a Miami Indian as an 


interpreter. The result of these studies consists of (a) 
the Prench-Miami-English Dictionary, from Ahhaiser to 
Cojeux; (6) The History of Genesis, Chapter I, being 
Peoria text with Miami-English translation; (e) English- 
Miami Dictionary, from Abandon to Aim; (d) "Wissa- 
katcakwa Stories, recorded in Peoria by the late Dr. 
Gatschet, for which Mr. Dunn has made an interlinear 

The compilation of the List of Works Relating to 
Hawaii was continued by Prof. Howard M. Ballon, of the 
College of Hawaii, who from time to time has submitted ad- 
ditional titles. The recording of the material by more than 
one person necessarily resulted in more or less incon- 
sistency in form ; consequently the manuscript, which con- 
sists of many thousands of cards, has been in need of edi- 
torial revision in order to insure unif oiinity. For this re- 
vision the bureau has been fortunate in enlisting the sei'v- 
ices of Mr. Felix Neumann, an experienced bibliographer, 
who is making progress in the work. 


The editorial work of the bureau has been conducted as 
usual by Mr. J. G. Gurley, editor. The following publi- 
cations were. issued dui-ing the year: 

Twenty-eightJi Annual Report, containing ''accom- 
panjdng papers" as follows: (1) Casa Grande, by Jesse 
Walter Fewkes; (2) Antiquities of the Upper Verde 
River and Walnut Creek Valleys, Arizona, by Jesse Wal- 
ter Fewkes; (3) Preliminary Report on the Linguistic 
Classification of Algonquian Tribes, by Truman Mich- 

Bulletin 30, Handhool' of American Indians North of 
Mexico, edited by Frederick Webb Hodge. By concurrent 
resolution of Congress, in August, 1912, a reprint of this 
bulletin was ordered in an edition of 6,500 copies, of which 
4,000 were for the use of the House of Representatives, 
2,000 for the use of the Senate, and 500 for the use of the 
bureau. This reprint, in which were incorporated such 


desirable alterations as could, be eouvenieutly made with- 
out affecting the pagination of the work, was issued in 
January, 1913. 

Bulletin 52, Early Man in South America, by Ales 
Hrdlicka, in collaboiation with William H. Holmes, 
Bailey Willis, Fred. Eugene Wright, and Clarence N. 

Bulletin 54, The Physiography of the Rio Grande Val- 
ley, New Mexico, in Relation to Puehlo Culture, by Edgar 
liee Hewett, Junius Henderson, and Wilfred William 

The work on the other publications during the year may 
be siunmarized as follows : 

Ttventy-ninth Annual Report (" accompanying paper," 
The Ethnogeography of the Tewa Indians, by John Pea- 
body Harrington) . Manuscript prepared for the printers 
and nearly half of the composition finished. 

Thirtieth Annual Report ("accompanying papers": 
(1) Animism and Folklore of the Guiana Indians, by Wal- 
ter E. Roth; (2) Tsimshian Mythology, by Franz Boas; 
(3) Ethnobotany of the Zuiii Indians, by Matilda Coxe 
Stevenson). Editing of the third paper and to a consid- 
erable extent that of the first paper completed. 

Bulletin 40, Handhook of American Indian Languages, 
by Franz Boas — Part 2. Work on the Coos section nearly 
finished and composition of the Chukchee section begun. 
Two sections (Takelma and Coos) are now " made up," 
aggregating 429 pages. 

Bulletin 46, A Dictionary of the Choctatv Language, by 
Cyrus Bjdngton, edited by John R. Swanton and H. S. 
Halbert. The editors have revised two galley proofs of 
the Choctaw-English section of this dictionary and have 
practically finished preparations for the printers of the 
English-Choctaw section. The first part of this bulletin 
is now in process of paging. 

Bulletin 53, Chip pe nut Music — II, by Prances Dens- 
more. Manuscript edited and the several proofs read, in- 
cluding proofs of 180 pieces of music. At the end of the 


year the bulletin was held in the Printiug Office awaiting 
I'eceipt of the necessaxy pajjer stock. 

Bulletm 55, Etlmohotany of the Tewa Indians, by Bar- 
bara Whitchurch Freire-Marreeo, Wilfred William Rob- 
bins, and John Pcabody Harringfon. Manuscript edited 
and the work in galley f omi at the close of the year. 

Bulletin 56, Ethnozoology of the Teiva Indians, by 
Junius Henderson and John Peabody Harrington. 
Manuscript edited and the work in page form at the close 
of the year. 

In accordance with the act of Congress approved Au- 
gust 23, 1912, the entire stock of publications of the bu- 
reau, with the" exception of a few coi)ies of each available 
work which have been retained at the Smithsonian Insti- 
tution for special purposes, was transferred to the Ctov- 
eiimient Printiug Office in October, 1912, for distribution 
from the office of the superintendent of docmnents on or- 
der from the bureau. It has been found that this plan of 
distribution is highly successful, and, of course, much less 
expensive to the bureau. 

The correspondence relating to publications, of which 
15,070 were distributed during the year, was conducted 
\uider the inmiediate supervision of Miss Helen Munroe, 
of the Smithsonian Institution. The distribution of the 
publications may be sunmiarized as follows : 

Series : Copies. 

Report volumes and separate papers 3,895 

Bulletins 11,040 

Contributions to North American Ethnology 15 

Introductions 7 

Miscellaneous publications 113 


The demand for the Handbook of American Indians 
(Bulletin 30) continues unabated, by reason of the wide 
scope of the work, its j^opular fonn of treatment, and its 
usefulness to schools. There is an increasing demand for 
publications relating to Indian arts and crafts, and to 
archeology. The activity in the establishment of organ- 


izations of Camp Fire Grirls tlirougiiout the country has 
resulted iii a flood of requests for information relative to 
, Indian customs names, etc. 


As in the past, the preparation of illustrations for use 
in connection with the publications of the bureau, as well 
as the making of photographic portraits of the members 
of visiting deputations of Indians, continued in the im- 
mediate charge of Mr. De Lancey Gill, illustrator, whose 
work during the year included the making of negatives 
of 113 visiting Indians and of 93 miscellaneous ethnologic 
subjects; he also developed 298 negatives exposed by mem- 
bers of the bureau in their field work, printed 975 photo- 
graj)hs for official publication, exchange, and presentation 
to Indians, and prepared 105 drawings for reproduction 
as illustrations for the iDublications of the bureau. 

The tribes or j)ueblos reijresented by Indians who visited 
Washington during the year are: Acoma, Apache, Chey- 
enne, Chippewa, Cochiti, Crow, Isleta, Kiowa, Osage, Pas- 
samaquoddy, Ponca, San Juan, Santa Clara, Shoshoni, 
Sioux, Taos, and Wichita. Among the more important 
Indians whose portraits were made may be mentioned 
Plenty Coups and Medicine Crow (Crow tribe). Big ]\Ian 
and Iron Bear (Brule Sioux), Hollow Horn Bear, Red 
Cloud, and Red Hawk (Teton Sioux), Daybwawaindung 
(Chippewa), and Two Moons (Cheyenne). Many re- 
quests are made by correspondents for prints from the 
large collection of negatives in possession of the bureau, 
but it has not been possible to supply these, owing to lack 
of means, although in many cases they are desired for 
educational purjjoses. The series of photographs of rep- 
resentative Indians, from 55 tribes, wdiicli was made dur- 
ing the last fiscal year for si)ecial exliibition at the New 
York Public Library, has been borrowed from the bureau 
by the Public Library Commission of Indiana for exhibi- 
tion in the public libraries throughout the State. In the 
work of the photographic laboratory Mr. Gill was assisted 
bv Mr. Walter J. Stenliouse. 



The library of the bureau coutiuued in immediate cliarge 
of Miss Ella Leary, librarian, assisted by Mrs. Ella 
Slaughter. During the year the accessions comprised 
562 volmnes (of which 129 were purchased) and 244 
pamphlets, bringing the total nmnber of volumes in the 
library . to 18,532, and the pamphlets to 12,744. The 
periodicals currently received by the bureau, of wliich 
there are several thousand imbound parts, number 629 ; of 
these all but 18 are obtained m exchange for the bureau's 
publications. Special attention was paid during the year 
to filling lacunae in the periodical series. 

The cataloguing kept pace with the new accessions, and 
some progress was made in cataloguing ethnologic and 
related articles in the earlier serials. A monthly bulletin 
for the use of the members of the bureau staff was com- 
piled and posted by the librarian, who also made a begin- 
ning in the preparation of a list of writings on the music 
of American Indians. 

As in the past, it was necessary to draw on the collec- 
tions of the Library of Congress, about 300 volmnes hav- 
ing been borrowed during the year. On the other hand, 
the library of the bureau is frequently consulted by officers 
of the departments of the Govermnent, as well as by 
students not connected \rith the Smithsonian Institution. 

While many volumes are still without binding, the con- 
dition of the library in this respect has greatly improved 
during the last few years ; 493 volumes were bound at the 
Government Printing Office during the year. 


The following collections were made by the bureau or 
by members of its staff during the fiscal year and trans- 
ferred to the National Museum : 

54311. Six photographs (unmounted) taken by A. J. Hortswill, San 
Jose, Mindoro, P. I., among the natives of Mindoro Ishand. Gift 
to the bureau by Munn & Co., New York. 


54465. Sacred pack of the Fox Indians of Iowa. Purchased for the 
bureau by Dr. Truman Michelson. 

64691. Five pieces of cotton painted with Assyrian subjects. Re- 
ceived by the bureau from an imknown source. 

54798. Three sacred looms and seven burden straps of the Osage In- 
dians. Collected by Francis La Flesche. 

54933. Three fragments of Indian pottery found at Eed Willow, 
Nebr., by Mrs. Ada Martin, by wliom they were presented. 

54934. Sacred bundle of the Fox Indians. Purchased througli Dr. 
Truman Michelson. 

54946. Two sacred bundles of the Osage Indians. Purchased by 

Francis La Flesche. 
55002. Sacred bundle of the Fox Indians. Purchased through Dr. 

Truman Michelson. 
55075. An Osage buffalo-hair rope (reata) and an Osage woven belt. 

Purchased through Francis La Flesche. 
55234. Two ethnological objects from the natives of British Guiana, 

presented to the bureau by Dr. AValter E. Roth, of Pomeroon 

River, British Guiana. 
55323. Set of five plum-seed gaming dice of the Omaha Indians and 

a bottle of seeds used by the same Indians as perfume. Pre- 
sented by Francis La Flesche. 
55420. Pair of Osage ceremonial moccasins and an Osage ceremonial 

" pipe." Presented by Francis La Flesche. 


As stated in pre^dous reports, the property of the bu- 
reau of greatest value consists of its library, manuscripts 
for reference or publication, and photographic negatives. 
A reasonable nmnber of cameras, dictagraphs, and other 
apparatus, chiefly for use in the field, as well as a limited 
stock of stationery and office supplies, necessary office 
furnitiTre, and equipment, are also in possession of the 
bureau. The sum of $893.21 was expended for office 
furniture (including fireproof filing cases) during the 
year, $452.57 for apparatus (including typewriters, cam- 
eras, dictagraphs, etc.), and $258.45 for books and pe- 

The manuscripts of the bureau, many of which are of 
extreme value, are deposited in metal cases in a small room 
in the north tower of the Smithsonian Building, which 


should be made as nearly fireproof as possible. Requests 
for a small appropriation to j^rotect the manuscripts 
against possible destruction have been made in the past, 
but unfortunately the means haA^e not been granted. The 
manuscripts, which have been in the inmiediate care of 
Mr. J. N. B. Hewitt, have increased from time to time 
during the year, chiefly by the temporary deposit of mate- 
rials preparatory to editing for publication. Mention 
may here be made, however, of the gift of some manuscript 
Chippewa letters from the Rev. Joseph A. Gilfillan, and 
the acquirement of a photostat copy of the Motul-Maya 
Dictionary^ made at the expense of the bureau from the 
original in the John Carter Brown Library, at Providence, 
R. I., as elsewhere noted. Mention may also be made of 
various vocabularies or parts of vocabularies, 23 items in 
all, which were restored to the bureau by Mrs. Louisa H. 
Gatschet. who found them among Dr. Gatschet's effects. 


Quarters. — Since the begiimiug of 1910 the offices of the 
bureau have occuijied nine rooms in the north tower of 
the Smithsonian Building, and a room (the office of the 
ethnologist in charge) on the north side of the third floor 
of the eastern wing, while the library has occupied the 
entire eastern gallery of the large exhibition hall on the 
first floor, and the photographic laboratory part of the 
gallery in the southeastern section of the old National 
Museiun building. While the natural lighting of the 
I'ooms in the north tower, by reason of the thickness of 
the walls and the narrowness of the windows, is inade- 
quate, and the distance from the library and the photo- 
graphic laboratory makes them not readily accessible, the 
office facilities are far better than when the bureau was 
housed in cramped rented quarters. Aside from the 
photographic laboratory and one room in the north tower, 
no part of the bureau's quarters is provided with rvmning 
water. It is presumed that after the rearrangement of 
the large exhibition hall in the Smithsonian Building and 


its adaptation to general libraiy purposes the facilities of 
the bureau library will be greatly improved. 

Office force. — The office force of the bureau has not been 
augmented, although the correspondence has greatly in- 
creased owing to the growing demand on the bureau for 
information respecting the Indians. The copying of the 
rough manuscripts, field notes, etc., prepared by members 
of the bureau, as well as the verification of quotations, 
bibliographic citations, and similar work of a minor edi- 
torial nature, necessitate the employment of temporary 
aid from time to time. Most of the answers to corre- 
spondents who desire information of a special character 
have been prepared by the ethnologist in charge, but every 
member of the bureau's scientific staff is frequently called 
on for the same purjiose to furnish information pertaining 
to his particular field of knowledge. 


It is difficult to extend the systematic researches of the 
bureau along new and necessary lines without an increase 
of appropriations. .When a special research is under- 
taken, several years are often required to finish it, conse- 
quently the prospective income of the bureau for a con- 
siderable period is required to carry out adequately the 
work in hand. Opportunities are often presented for con- 
ducting investigations in new fields which have to be neg- 
lected owing to lack of means. An increase in the appro- 
Ijriations of the bureau has been urged for several years, 
but unfortunately the estimates have not been met with 
additional funds. 

Respectfully submitted. 

F. W. Hodge, 
Ethnologist in charge. 

Dr. Charles D. Walcott, 

Secretary of the S))iithsonia)i Institution, 

Washington, J). C. 



160658°— 34 ETH— 22 3 







Introduction „ 49 

Historical considerations 51 

Preliistorlc cultural areas in the West Indies o3 

Trinidad , 62 

Erin Bay 65 

Chip-clilp shell-heap GO 

Pottery 67 

Handles of vessels .• 71 

Stone implements 74 

Bone objects 75 

Objects of wood 75 

Comparison of prehistoric objects from Trinidad with those from 

other islands 75 

Tobago 78 

Barbados 78 

Middens 81 

Caves 82 

Mount Gilboa caves 83 

Artificial excavations S3 

Indian Castle S3 

Indian excavations 85 

• Artifacts 86 

St. Vincent-Grenada area 8S 

Grenada 88 

Bequia 89 

Battowia 89 

Balliceaux 89 

St. Vincent 90 

Kitchen middens 91 

Pit'tographs 92 

Artifacts 93 

Stone implements 93 

Celts and axes 94 

Petaloids 94 

Axes and chisels 94 

Axes with caps i 100 

Grooved hammers and axes 101 

Asymmetrical axes 102 

Tools 104 

Implements of cresceutic form 107 

Eared axes 108 

Engraved axes 109 

Problematic stone objects 110 

Grinding implements and pestles 112 

Stone fetishes, amulets, and idols 113 

Enigmatical objects 115 



Prehistoric cultural areas in the West Indies — Continueti. 
St. ^'incent-Grenada area — Continued. 

St. Vincent — Continued. Page. 

Pottery US 

Pendants 122 

Shell objects 123 

Terra-cotta stamps 123 

Perforated disks 123 

Dominica 123 

Stone implements ]2.") 

Martinique 128 

Guadeloupe 128 

Guesde collection 129 

Axes with regular margins 138 

Axes with asymmetrical margins 13s 

Eared axes 13(» 

Engraved axes 144 

Perforated axes 147 

Anchor axes 148 

Incised and perforated stones 148 

Problematical stones 149 

Mortars 152 

Pestles, grinders, and hammers 154 

Conical stones 155 

St. Kitts 15S 

Middens 159 

Pictographs 160 

Altar stone IQO 

The Connell collection 160 

Grinders 162 

Shell objects 162 

Pottery 163 

Stone objects 165 

St. Croix 166 

Salt River midden 167 

Artifacts 168 

Porto Rico area 168 

Caves, shell-heaps, and ball courts ' 170 

Archeological specimens 171 

Petaloid celts 172 

Monolithic petaloid celt 17.3 

Engraved celts 174 

Human heads and figures 183 

Stone heads 184 

.Stone nodules ami masks 184 

Semicircular stones 187 

Stone collars 187 

Elbow stones 198 

Description of elbow stones 201 

Morphology and interpretation 207 

Ceremonial batons of stone 210 


Prehistoric cultural areas in the West Indies — C'ontiuueil. 
Porto Rico area — Continue<l. 

Arclieological specimens — Conliuueil. Page. 

Ttiree-pointed stones 211 

First type of three-pointed stones 211 

Second type of three-pointed stones 216 

Fourth type of three-pointed zeniis 217 

Three-pointed stone used for pestle 220 

Idols 221 

Bird stones 221 

Mortars and grinders . 221 

Pestles 226 

Ornaments 230 

Stone pendant 232 

Amulets 233 

Bone objects 234 

Shell objects ; 235 

Clay objects 235 

Clay cylinders 235 

Pottery 236 

Cuba 240 

Prehistoric culture of Cuba 242 

Historical . . 244 

Archeological objects 248 

Conclusions 252 

Isle of Pines , 256 

Jamaica 257 

Great Cayman 258 

Analysis of West Indian archeological data in its geographical distribu- 
tion 259 

Pottery 259 

Stone implements 262 

Ornaments 265 

Conclusions 266 

Authorities cited 269 

Index 273 



1. Map of the West Indies, showing distribution of objects in the 

Museum of the American Indian (Heye Foundation) in 1914 49 

2-3. Clay objects from Trinidad 70 

4. Clay handles from Trinidad 70 

as. Clay heads from Trinidad 70 

9. Stone implements from Grenada and St. Vincent 70 

10-12. Stone implements from St. Vincent 98 

13-15. Notched stone implements from St. A'incent 9S 

16-17. Stone implements from St. Vincent 100 

18. Fish tail and asymmetrical stone objects from St. Vincent 100 

19-20. Stone implements from St. Vincent TOO 

21-23. Asymmetrical stone implements from St. Vincent 100 

24. Spatulate stone implements from St. Vincent 106 

2.5. A-E, Miscellaneous stone implements from St. Vincent ; F, Pend- 
ant for necklace 106 

26. Miscellaneous stone implements from St. Vincent 106 

27-28. Stone implements from St. Vincent 106 

29. Crescentic stone implements from St. Vincent 106 

30. Eared stone implements from St. Vincent and Grenada 110 

31. Eared stone implements from St. Vincent 110 

32. Inscribed and eared stone implements from St. Vincent 110 

33. Problematic stone objects from St. Vincent 110 

34. Problematic stone implements from St. Vincent 110 

35. Pestles and other stone objects from St. Vincent 110 

36. Stone amulets and fetishes 116 

37. Problematic objects from St. Vincent 116 

38-61. Problematical objects, Fancy, St. Vincent 116 

62. Pottery, Carriacou 120 

63. Fragments of pottery, Carriacou 120 

64. Clay heads from pottery objects, Carriacou 120 

65-67. Clay heads from pottery objects, Carriacou 120 

68. Pottery objects. Carriacou 12ft 

69. Pendants of stone and shell, St. Vincent 120 

70. Objects of clay and stone 120 

71. Stone implements, St. Vincent. Guadeloupe 12<1 

72. Anchor-shaiied stone implements, Guadeloupe (Guesde collection. 

Berlin Museum) 148 

73. Incised and perforated stone.';, Guadeloupe (Guesde collection, 

Berlin Museum) 148 

74. Problematical stones, Guadeloupe (Guesde collection, Berlin 

Museum ) 14S 

75. A, pestle; B, stone ring; C, mortar, Guadeloupe (Guesde collection, 

Berlin Museum) 14S 

76. Problematical stone implements, Guadeloupe (Guesde collection, 

Berlin Museum! 1.54 




77. Problematical stones, Guadeloupe (Guesde collection, Berlin 

Museum) 1">4 

78-79. Pestles and problematical stones, Guadeloupe (Guesde collection, 

Berlin Museum) 1.56 

80. Problematical objects and pestles, Guadeloupe (Guesde collection, 

Berlin Museum) 1.56 

81. Pestles and otber stones, Gaudeloupe (Guesde collection, Berlin 

Museum) 15(3 

82. Stone and shell implements, St. Kitts (Connell collection) 162 

83. Pottery ring and decorated shell cylinder, St. Kitts (Connell col- 

lection) 162 

84. Various objects, St. Kitts (Connell collection) 162 

85. Pottery and stone objects, St. Kitts 162 

86. Shell objects, St. Kitts and Nevis (Connell collection) 166 

87. Stone objects, St. Kitts, Dominica, St. Vincent 166 

88. Petaloid celts (Berlin Museum) 178 

89. Engraved petaloid celt, Santo Domingo (Copenhagen Museum) 178 

90. Stone dirk 182 

91. Stone heads [A. Guesde collection ; B, C, Heye Museum) 182 

92. Stone heads (Trocadero Museum, Paris) 1S2 

93. Celt and stone balls with engraved faces, Hati, Cuba (Berlin 

Museum ) 182 

94. ^4. Stone nodule; B, C, stone masks; Haiti (A, Grosser collection, 

Berlin Museum ) 188 

95. Massive stone collar, Porto Rico 188 

96. Slender stone collars, Porto Rico 188 

97. A, B, C, Slender atone collars; D, massive stone collar, Porto Rico 

(Trocadero Museum, Paris) 188 

98. Elbow stones, Porto Rico 200 

99. Elbow stones. Lesser Antilles, Guadeloupe (Guesde collection, 

Berlin Museum) 200 

100. Unidentitied stone objects, Guadeloupe (Guesde collection, Berlin 

Museum) 212 

101-103. Tliree-pointed stones of the first type, Porto Rico (Berlin 

Museum) 212 

104. Three-pointed stones of the first type (Madrid Museum) 214 

105. Three-pointed stones of the first type (Trocadero Museum) 214 

106. Three-pointed stones of the first type, Porto Rico 216 

107. A, C, Three-pointed stones of the first type, Porto Rico ; B, Com- 

posite three-pdinted stone, first and second types, Porto Rico 216 

108. A, B, Three-pointed stones of the first type; C, fourth type; D, second 

type 216 

109. Three-pointed stone of the second type (Museum Santiago de los 

Caballeros, Santo Domingo) 216 

HO. Three-pointed .stones of the fourth type (Berlin Museum) 220 

111. A, B, Three-piiinted stones of the fourth type; C, fourth type with 

face on each end; D, E, third type, Santo Domingo 220 

112. A, Stone bird ; B, C, stone mortar, Porto Rico 220 

113. A, B, Stool (duho) and grinders; C, turtle-shaped mortar (B, Gue.sde 

collection, Berlin Museum) 220 

114. Pestles. A, St. Thomas; B, C, Haiti (Berlin Museum) 228 

115. A, B, C, Pestles, Santo Domingo; D, side view of C 228 



116. A, Unidentified stone object; B, tooth-sliaped object; C, D, E, F, 

amulets 232 

117. A, Stone amulet: B, C, D, E, P, shell and bone objects; G, pottery 

stamp, Santo Domingo 232 

118. Pottery, A. B, Porto Rico; C, Santo Domingo 238 

119. A, Broken pottery neck ; B, C, pottery, Santo Domingo 2.38 

120. Seated stone figure, Santo Domingo 238 


1. Notched ax, Trinidad 74 

2. Jadeite pendant, Trinidad 75 

3. Eared ax from Guadeloupe 108 

4. Pestle seen in profile -. 112 

5. Stone pestle with face 112 

6. Stone pestle with eyes and mouth 112 

7. Stone object in shape of pestle 113 

8. Head and handle of broken pestle 113 

9. Pestle-shaped problematical object 113 

10. Stone pendants 114 

11. Hollow clay cylinder with face in relief 121 

12. Ax with marginal notches 125 

13. Ear-.shaped blade 139 

14. Eared stone implement 140 

15. Stone implement with two beaks 141 

16. Eared stone implement 141 

17. Massive stone eared implement 142 

18. Perforated eared implement 143 

19. Stone implement with indented head 144 

20. Finely polished stone implement 145 

21. Incised ax from Guadeloupe 146 

22. Incised decorated stone implement (Guesde collection) 147 

23. Ax with bifurcated blade 148 

24. Problematical stone, Guadeloupe (Berlin collection) 151 

25. Lateral view of figure 24 152 

26. Unidentified stone object resembling a mortar 154 

27. Unidentified stone object resembling a mortar 1.54 

28. Outline of ax, shown from edge and side 172 

29. Broken monolithic celt 173 

30. Broken monolithic celt 174 

31. Stone idol 186 

32. Semicircular .stone with face, from front and side 187 

33. Schematic view of stone collar 188 

34. The Strube stone collar (Bremen Museum) 190 

35. Lateral view of " knob " of the Strube stone collar (Bremen Museum) . 191 

36. Dorsal view of " knob '' of the Strube stone collar (Bremen Museum )_ 191 

37. Decorated panel and panel border of the Strube stone collar (Bremen 

Museum) 192 

38. Panel of stone collar (Heye Museum) 192 

39. Panel of stone collar (Latimer collection) 193 

40. Panel of stone collar (Latimer collection) 194 

41. Panel of stone collar (Bremen Museum) 194 

42. Panel of stone collar (Latimer collection) 194 

43. Panel of stone collar (Latimer collection) 195 



44. Stone collar showing unique decorated panel border (Trooadero 

Museum) 19.5 

45. Decorated panel and panel border of stone collar (Latimer collection ) _ 196 

46. Underside of decorated panel of the Strube stone collar 196 

47. Elbow-stone in the United States National Museum 202 

48. Elbow-stone (Madrid Museum) 203 

49. Elbow-stone in the Latimer collection 20.5 

50. Three-pointed stone, first type 212 

51. Three-pointed stone with face on anterior end 219 

52. Three-pointed stone of fourth t.vpe 219 

53. Base of three-pointed stone of fourth t.vpe, showing longitudinal 

furrows 220 

54. Problematical stone recalling three-pointed idol, with superficial 

knobs on bod.v 220 

55. Chocolate grinder shaped like a seat or duho 224 

56. Problematical object shaped like a pestle 227 

57. Front and back views of head of an end of decayed pestle handle 228 

58. Stone pestle with head on end of handle 229 

.^0. Problematical stone implement, i)robably when in use la.shed to a 

wooden handle 229 

60. Various forms of stone beads, plain and decorated , 232 

61. Amulet in Vienna Museum 2.'?4 

62. Clay stamp or die with incised meander 236 

63. Bowl with inci.sed decoration 241 

64. Idol of coral rock from Cueva de Boruga, Baracoa, Cuba, (Santiago 

■65. Idol or pestle from Loma del Cayuco (Santiago Museum) 251 

66. Stone idol (University Museum, Havana) 252 

67. Petaloid celt (Santiago Museum) 253 

68. Fragments of pottery from Nipe Bay (U. S. National Museum) 2rp4 

69. Stone with face from Nipe Bay (U. S. National Museum) 255 



[Specimens with Arabic numbers, Catalogue Museum of the American Indian (Heye 
Collection I ; Roman numbers. Catalogue Berlin Museum.] 

Plate 9. A, Xo. IVC* 1755; B. No. IVC 1T6T ; C, No. 9636; D, No. 2/8308; P, 

No. 9708; F, No. 2/9151; G, No. 1/4412; H. No. 2/9848; /, No. 

Plate 10. .4, No. 2/7755; B, No. 9706; C, No. 2/7738; D, No. 2/8310; E, No. 

2/8330 ; F, No. 1/4398 ; G. No. 1/4411 ; H, No. 2/8753 ; /, No. 2/8294. 
Plate 11. A, No. 2/7761 ; B, No. 2/7761 ; C, No. 2/9154 ; D, No. 2/9151 ; E, No. 

2/9851; F. No. 2/9838; G, No. 1/4398; H. No. 1/4425; /, No. 

2/9064 ; ,/, No. 2/9844 ; K, No. 2/5342. 
Plate 12. A, No. 1/1924 ; B, No. 9702 ; C, No. 1/4385 ; D, No. 2/7771 ; E, No. 

2/8294 ; F, No. 3731 ; G, No. 2/7747 ; H, No. 3738 ; /, No. 2/7754 ; 

J, No. 9676; K, No. 1/1970; L, Xo. 1/1930; if. No. 1/1930; y. 

No. 2/5331. 
Plate 13. -4, 1/1931; B, No. 2/8742; C, X'o. 2/5331; D, No. 2/7771; E. No. 

2/7771 ; F, No. 1/1965 ; G, No. 1/1938 ; H, No. 2/5331. 
Plate 14. .4, No. 2/7771; B, No. 1/1971; C, No. 2/7771; /), No. 1/4393; F, No. 

1/4411 ; F, No. 2/4398 ; G, No. 1/4398 ; H. Xo. 1/4411. 
Plate 15. A, No. 9655; B, No. 2/9832; C, No. 2/7757; D. No. 1/4411; F, No. 

2/7771 ; F, No. 2/9044 ; G, No. 2/7711 ; H, No. 1/4411 ; /, No. 1/4411 ; 

./, Xo. 1/4411. 
Plate 16. A, No. 2/5331 ; B, No. 2/7771 ; C, No. 1/4411 ; D, No. 2/8303 ; F, No. 

2/8294; F, No. 2/9151; G, No. 1/1960; ff, No. IVC* 1023; /, No. 

2/3517; ./. No. 1/122. 
Plate 17. ,1, No. 2/9036; B, No. 3/2018; C, No. 9656; D. No. 1/1931; F, No. 

7696 ; F, No. 1/4411 ; G, No. 9147 ; H, No. 9641 ; /. No. 1/1974. 
Plate 18. A, No. 1/4411; B, No. 96&3; C, No. 2/7756; D, No. 1/4411; F, No. 

2/8742; F, Xo. 1/1964; G, No. 9634; if, No. 2/7749; /, No. 1/4411. 
Plate 19. A, No. 2/8294; B, No. 2/8742; C, No. 2/3524; D, No. 1/4399; F, No. 

2/9842 ; F, No. 96&8 ; G, No. 2/7777 ; H. No. 2/9144 ; /, No. 2/8633 ; 

J, No. 1/4411 ; K, No. 2/8294. 
Plate 20. A, No. 1/1960; B. No. 1/121; C, No. 1/120; D, No. 1/4411; F, No. 

1/4398; F, No. 1/4398; G, No. 2/7764; i7, No. 2/5331; /, No. 

2/9985 ; ./. 2/8651 ; K, No. 2/8294 ; L, No. 1/4411. 
Plate 21. A, No. 9&13; B, No. 2/8294; C, No. 1/4411; D, No. 9704; E, No. 

2/9832; F, No. 1/1885; G, No. 2/8308; H, No. 192. 
Plate 22. A, No. 9633; B, No. 9723; C, Xo. 1/4411; 7>, N-to. 1/1967; F, No. 

1/1969; F, No. 2/9849; G, No. 1/1886; H, No. 1/4400. 
Plate 23. A, No. 2/9148; B, No. 2/3.^21; C, No. 2/9851; f). No. 1/4411; F. No. 

1/1949; F. No. 1/1931; G. No. 1/1942; H. No. 2/7788. 
Plate 24. A, No. 1/1948; B, Xo. 1/1942; C, Xo. 2/91.52; D. No. 2/8311; F, No. 

2^8293 ; F, No. 2/7757 ; G, No. 2/8293 ; //. No. 2/8301 ; /. No. 2/8293. 
Pi^te 25. A, No. 2/8293; B, No. 2/9831; C, No. 2/7757; D, No. 2/.5331 ; F, No. 

2/8261; F, Xo. IVC6 2022. 



Plate 26. A, No. 2/8293; B, No. 2/8768; C. No. 2/8268; D. No. 9702; E, No. 

1/1924 ; F. Xo. 2/8742 ; G, No. 2/5351 ; H. No. 2/8293 ; /. No. 2/7761 ; 

J, No. 2/9827; A', No. 1/1924; L, No. 4402. 
Plate 27. A, No. 2/8293; B. No. 2/9831; C, No. 2/7761; D. No. 1/1924; £. No. 

1/4422 ; F, No. 2/8293 ; 0. No. 9702 ; H. No. 2/8293 ; 7, No. 1/1924. 
Plate 28. ,4. No. 2/1924; B, No. 1/1933; C, No. 2/8305; 7), No. 2/9151; E, No. 

2/9151 ; F, No. 2/9151 ; G, No. 2/5330 ; i/. No. 2/5330 ; /, No. 2/7767 ; 

.7, No. 2/7757 ; K. No. 1/4391 ; L, No. 2/4391. 
Plate 29. A. No. 2/7753; B, No. 2/7743; C, No. 9710; 7), No. 2/7742; E, No. 

9734 ; F, No. 1/1991 ; G, No. 1/4435 ; 77, No. IV6 280 ; 7, No. 2/9846 ; 

J, No. 2/7739; A', No. 1^4408. 
Plate 30. A. No. 2/4855 ; B. No. 2/9845 ; C. No. 2/3516 ; 7), No. 1/1962 ; E. No. 

2/9851 ; F, No. 2/7771 ; G. No. 2/8294 ; 77. No. 1/1957 ; 7, No. 2/7751. 
Plate 31. .1, No. 1/4411; B, No. 1/1958; C, No. 2/9151; 7», No. 1/4411; E, No. 

2/9151 ; F, No. 2/7757 ; G, No. 2/9151. 
Plate 32. .4, No. 1/1959; B, No. 6709; C. No. 2/3515; D, No. 1/4406; E, No. 

9677; F, No. 9713. 
Plate 33. A, No. 2/9981; B, No. 1/1994; C, No. 2/8309; D, No. 1/4424; E. No. 

1/1976 ; F. No. 2/8299 ; G, No. 1/4424 ; 77, No. 2/8299 ; 7, No. 1/1893 ; 

J. No. 2/8293 ; A', No. 2/5334 ; L, No. 1/1963 ; M. No. 2/9142. 
Plate 34. A. No. 2/8315; B. No. 1/8683; C. No. 2/4391; 7), No. 1/1924; F. No. 

2/7763 ; F, No. 2/9848 ; G, No. 1/4378 ; 77, No. 2/8295 ; 7, No. 2/9856 ; 

J. No. 2/8624; A", No. 2/8613. 
Plate 35. A. No. 2/9847 ; B. No. 1/4398 ; C, No. 2/9837 ; 7), No. 1/4376 ; F. No. 

3/2185 ; F, No. 3/2135 ; G, No. 2/3532 ; 77, No. 1/1999 ; 7, No. 1/132. 
Plate 36. A, No. 9720; B. No. 2/9140; C, No. 2/7732; 7), No. 2/8717; F, No. 

Plate 37. .4, No. 3/1981 ; B, No. 2/9957 ; C. No. 2/9843 ; 7), No. 1/4386 ; E. No. 

2/8308 ; F. No. 2/9851 ; G. No. 2/9851 ; 77. No. 2/9046 ; 7, No, 2/8634 ; 

.7. No. 2/8619 ; A', No. 2/9037 ; L, No. 2/9040; M, No. 2/9039 ; .V, No. 

Plate 38. .4, No. 2/7745; B, No. 2/8264; C, No. 2/8616; 7), No. 2/8626; E, No. 

2/8653 ; F. No. 2/8625. 
Plate 39. A, No. 2/9621; B. No. 2/8650; C, No. 2/9895. 
Plate 40. A, No. 2/9958; A (outline) represents G. Plate 46; B, No. 2/9952; 

C, No. 1/12(4. 
Plate 41. .4, No. 2/8658; B, No. 2/9990; C, No. 2/9872. 
Plate 42. A. No. 2/8242; J5. No. 2/9896; C, No. 2/7746; 7), No. 2/82.14; E, No. 

2/9963 ; F. No. 2/9876. 
Plate 43. A, No. 2/8644 ; B, No. 2/8247. 
Plate 44. .4. No. 2/9865 ; B. No. 2/9865 : f\ No. 2/8649 ; 7), No. 2/9892 ; F, No. 

2/8611 ; F, No. 2/9853. 
Plate 45. A, No. 2/8639 ; B, No. 2/9870 ; C, No. 2/9868 ; D, No. 2/2805 ; E, No. 

2/8675 ; F. No. 2/8^41. 
Plate 46. .1, No. 2/8666 ; B, No. 2/8255 : C, No. 2/8652 ; D. No. 2/9889 ; F, No. 

2/9899 : F, No. 2/9878 : G. No. 2/9851 ; 77, No. 2/65 ; 7. No. 1/129. 
Plate 47. A, No. 2/9869; B, No. 2/8622; C, No. 2/8648; D, No. 2/7736; E, No. 

Plate 48. A, No. 2/8631 ; 7J, No. 2/9945 ; C. No. 2/9875 : D, No. 9880. 
Plate 49. A, No. 2/8623; B (left). No. 2/9886; B (right. No. 2/866."i; C, No. 

Plate .50. A, No. 2/2843 ; B, No. 2/8660 ; C, No. 2/8647. 
Plate 51. A; No. 2/8656 ; B, No. 2/8241 ; G, No. 2/7737 ; D, No. 2/8675. 


Plate 52. A, No. 2/8630 ; B. No. 2/9S8.'5 ; C, No. 2/8252. 
Plate 53. A, No. 2/7751 : B. No. 2/9874 ; C, No. 2/7744. 
Plate 54. A, No. 2/2S43 ; B, No. 2/8246 ; C, No. 2/8^14. 
Plate 5.5. .4, No. 2/77.50; B, No. 2/7736; C, No. 1/146; D, No. 2/9867; E, No. 

Plate 56. A, No. 2/9987 ; B. No. 2/9947 ; C, No. 2/8635 ; D, No. 2/9884. 
Plate 57. A, No. 2/9967 ; B. No 2,^9984 ; C, No. 2/8627 ; Z>. No. 2/8627. 
Plate 58. A, No. 2/9872 ; B, No. 1/4407. 
Plate 59. A, No. 2/9990 ; B, No. 2/8658 ; C, No. 2/8677. 
Plate 60. A. No. 2/8668 ; B, No. 2/8262 ; C, No. 2/8616 ; D, No. 2/9874 ; E. No. 

Plate 61. .1. No. 2/9864; B, No. 2/8665; C, No. 2/8659; D, No. 2/9951; B, No. 

2/8674 ; F, No. 2/8653 ; G. No. 2/8654 ; Ji. No. 2/7750. 
Plate 62. .4, No. 2/7784; B. No. 1/4427; C. No. 1/4427; D. No. 1/4428. 
Plate 63. A, No. 2^7784 ; B, No. 1/4427 ; C, No. 8125. 
Plate 64. A, No. 2/8271 ; B, No. 8125. 
Plate 65. A, No. 1/4427 ; B. No. 2/57 ; G. No. 1/4427. 
Plate 66. A, No. 1/4425 ; B, No. 1/8692 ; C, No. 1/4427 ; D, No. 3509. 
Plate 67. A, No. 2/8271; B. No. 8125; C, No. 1/4427; D, No. 1/4427; J?, No. 

a/7777; F, No. 1/4427. 
Plate 68. .4. No. 1/4429 ; B, No. 2/8752 ; C. No. 1/4427. 
Plate 69. A, No. 2/7740 ; B, No. 2/8305 ; C. No. 2/8304 ; D. No. 2/7765 ; E, No. 

2/3806 ; F. No. 9701 ; G'. No. 2/8274 ; H. No. 2/7741 ; /. No. 2/9139 ; 

.7. No. 3/1970; K. No. 9718; L. No. 1/4382; M, No. 1/4390; A', No. 

1/4390; O, No. 1/4390. 
Plate 70. A. No. 2/3537; B, No. 2/8273; C. No. 2/3536; D, 2/7778"; E, No. 

2/9825; F, No. 2/3535; G, No. 2/7778; H, No. 2/8316. 
Plate 71. A. No. IVC 286 ;B, No. IVC* 293; C. No. IVC'' 1017; D, No. IVC 

Plate 72. A, No. IVC* 291 ; B, No. IVC* 2S9. 
Plate 73. A, No. l\Ct> 262; B. No. IVC" 261; C, No. IVC" 265. 
Plate 74. B, No. IVC* 264; C. No. IVC" 135. 
Plate 75. A, No. IVC 1.52; B. No. 2/9145; C. No. IVC" 134. 
Plate 76. A, B, No. IVC" 300; C, D, IVC" 299. 
Plate 77. A. No. IVC" 159; B, No. IVC" 253; C, No. IVC" 1164; D. No. IVC" 

Plate 78. .4, B, No. IVC" 105; C. No. IVC" 163; D, No. IVC" 165; E. No. IVC" 

166 ; F, No. IVC" 292. 
Plate 79. A. No. IVC" 178; B. No. IVC" 211; C, No. IVC" 408; 7), E. No. IVC" 

1135 ; F, No. IVC" 171. 
Plate 80. A, No. IVC" 170" ; B, IVC" 164 ; C. No. IVC" 692 ; D. No. IVC" 170" ; 

E, No. IVC" 153. 
Plate 81. A, No. IVC" 703; B, IVC" 174; C. IVC" 151; E, No. IVC" 169. 
Plate 83. B, No. IVC" 1754. 

Plate 88. A. IVC" 1781; B, IVC 84; C. IVC" 30. 
Plate 90. A, B, C. 3/6312. 
Plate 91. B, C, No. 3045. 

Plate 93. S, C, No. IVC" 1778 ; 7). F, No. IVC" 78. 
Plate 94. .4, No. IVC" 1777. 
Plate 98. .4, A', No. 2/1988; B, No. 1/4018. 
Plate 100. A. B, IVC" 1326; C. n. No. 3/3936. 
Plate 101. A, No. 3694; 7). No. IVC" 5.5. 
Plate 102. A, No. IVC" 34 ; B, No. IVC" 75. 



Plate 103. 

Plate 106. 
Plate 107. 
I'LATE 108. 
Plate 110. 
Plate 111. 
Plate 112. 
Plate 113. 
Plate 114. 
Plate 115. 
Plate 116. 
Plate 117. 

Plate 118. 
Plate 119. 
Figure 2. 
Figure 3. 
Figure 4. 
Figure 5. 
Figure 6. 
Figure 7. 
Figure 8. 
Figure 9. 
Figure 11. 
FiGirRE 12. 
Figure 21. 
Figure 22. 
Figure 23. 
Figure 24. 
Figure 25. 
Figure 26. 
Figure 27. 

A, No. JVC* 54 ; B. No. 63488 ; C, No. IVC» 45. 

B, No. 1/2300; C, No. 3695. 
A, No. IVC» 45. 

C, No. 3695 ; Z), No. 3/3930. 

A, No. IVC» 125 ; B, No. IVC 124. 

A, No. IVC* 124 ; B, No. 1/4392 ; C, No. 2/3.529 ; D, No. 3697. 

A, No. 5431 ; B, C, No. 9719. 

A, No. IVC 33 ; B. No. IVC 1335 ; C. No. IVC» 1780. 

A, No. IVC> 1776; B, No. IVC 12; C, No. IVC 13. 

A, B, No. 2/49 ; C, D. No. 1/123. 

A, No. 3826; B. No. 2/8722; C, No. 1/9718; D, No. 1/9711. 

4, No. IVC' 92 ; B, No. 1/9714 ; D, E, No. JVC 1140 ; F, No. 1/9715 ; 

G, No. 2/7773. 
A, No. 1/9710; B, No. 1/152; C, No. IVC» 2025. 
A, No. 2/60; B, No. 1/150; C, No. 1/151. 

No. 2/3726. 
No. IVC" 308. 
No. 1/129. 
No. 2/3533. 
No. 2/1993. 
No. 2/3534. 
No. 2/8725. 
No. 2/9829. 
, No. 3/.5910. 
. No. IVC* 2175. 
, No. IVC 296. 
, No. IVC* 296. 
. No. IVC" 36. 
. No. IVC" 123. 
. No. IVC 123. 

No. JVC" 1327. 

No. IVC* 1327. 

Figure 28. No. 3/3937. 

Figure 29. No. 1/140. 

Figure 30. No. 3/2791. 

Figure 32. No. 3693. 

Figure 40. No. 8030 (U. S. N. M.) 

Figure 42. No. 17080 (U. S. N. M.) 

Figure 43. No. 8029 (U. S. N. M.) 

Figure 45. No. 17082 (U. S. N. M.) 

Figure 50. No. 3695. 

Figure 51. No. 2/3.529. 

Figure 52. No. IVC 125. 

Figure 53. No. 1/4392. 

Figure 55. No. 3698. 

Figure 56. No. 2/8269. 

Figure 57. No. 3/1996. 

Figure 59. No. IVC 79. 

Figure 63. No. IVC» 1788. 



By J. Walti:r Fewkes 


In the year 1911 the author was invited by Mr. Ceorofe G. Heye, 
of New York, to examine liis collection of Indian antiquities with 
a view to publishinfi a report upon them. This collection, even 
then, was very extensive, and since that time has been greatly en- 
larged in number ' of specimens, so that in Octolier, 1914. it con- 
tained about 9,500 prehistoric objects from the West Indies. 
The localities in which these were found are indicated on plate 1 and 
the number of objects obtained from each island is shown in the 
following table : 








- 200 




St. Kitts ^ 




Saba ■ 








St. Thomas 


Balliceaux _ _ 


Santa Cruz 

Porto Rico 





St. Vincent 






Republic of Santo Domingo 


Santa Lucia . 








' Since this article was written tlie number of Antillean objects in the Heye collec- 
tion has greatly increased through the addition of the coUeetions made by De Booy, 
Harrington, and others. Those from Trinidad, Santo Domingo, and the Virgin Islands ' 
have already been described by the late Theodoor de Booy, and a report on the Cuban 
collection made by Harrington will shortly be published. 

160658°— 34 ETH— 22 4 49 



Crooked 2 

Rasgeil 2 

Eleuthera 2 

Royal 1 

New Providence 3 

Great Abaco 1 

Mores 2 

Little Abaco 2' 

East Caicos 25 

Grand Caicos 68 

North Caicos 104 

Providence 59 

West Caicos 20 

Great Inagiia 20 

Mariguana 6 

Plana Cayo i 

Acklin 2 

Some of these specimens have been collected on various expeditions 
sent by Mr. Heye to the islands, but the majority have been pur- 
chased for him by Rev. Thomas Huckerby from collectors in the 
Lesser Antilles. A large part of this report is taken up in a consid- 
eration of these specimens. 

The author also visited these islands to gather data in the field 
before writing this report,^" and spent the winter of 1912-13 iii the 
Lesser Antilles, visiting Trinidad, St. Vincent, Barbados, St. Kitts, 
and Santa Cruz, where he obtained important material. He like- 
wise made short visits to other islands, examining and making notes 
on specimens in various public and private collections on the islands, 
which are embodied in this memoir. 

In order to get all jiossible information bearing on the forms and 
uses of tliese objects the author, in the winter of 191^14. visited 
several European museums rich in West Indian objects, as those 
in Copenhagen, Denmark; Bremen and Berlin. Germany; Vienna 
and Prague, Austria. A considerable number of drawings were 
made on this trip, especially in the Berlin Museum, which is one of 
the richest in these objects on the European Continent. On a re- 
cent trip to Europe, Prof. 'Marshall H. Saville, of Columbia Uni- 
versity, kindly obtained for the author several photographs of rare 
West Indian antiquities in the museums of London, Paris, and 
Madrid. The West Indian collection in the Heye Museum has also 
been enlarged by specimens purchased by the author from Seiior 
Seiyo. of Arecibo, Porto Rico, and other local collectors. 

The description of artifacts in the Heye Museum is accompanied 
by short accounts of related objects in other museums from the same 
islands from which the material was obtained. For convenience in 
the consideration of the subject the geographical method is adopted, 
the West Indies being divided into areas, which are supposed to indi- 
cate culture centers. The aim has been not so much a description of 
specimens as a consideration of a highly developed insular culture 
peculiar to America as a whole preparatory to a comparison of it 
with that of the neighboring continent. 

'" Tbe work was done under cooperation of the Heye Musenm and the Bureau of 
American Ethnolog}'. 


Since the author began the preparation of this report many other 
archeologists have been led to enter the West Indian field.^ Several 
collectors have been sent by Mr. Heye to the islands, find large col- 
lections have been brought to his museum from the Bahamas, Cuba, 
Santo Domingo, and Trinidad. In addition to work by the Heye 
Museum, other institutions have begun work, especially in Porto 
Rico, where important results are being brought to light by exca- 
vations in ball courts, shell heaps, and caves. The New York 
Academy of iScience. in cooperation with the Insular government, 
made excavations in ball courts and shell-heaps of Porto Rico, under 
the supervision of Dr. F. Boas, in 1915. This wealth of new material 
sheds some light on many doubtful questions which pioneer students 
in Antillean archeology have been unable to answer, and will prob- 
ably, when published, antiquate some of the theories brought forward 
by the author in this article. For obvious reasons no adequate 
reference can here be made to details of unpublished material, but it 
is very gratifying to the author that his prediction, made over a decade 
ago, that the AVest Indian field will afford a rich harvest to arche- 
ologists provided with ample means for intensive study on any one 
of the chain of islands connecting South America with the south- 
eastern part of the United States, has been confirmed. Of all the 
islands superficially explored none still offer greater facilities for 
study than Santo Domingo and Porto Rico, the central points of the 
characteristic Antillean culture, where, no doubt, it originated. Much 
work remains to be done in this field. 

The Antillean culture is sufficiently self-centered and distinctive 
to be called unique, although the germ originally came from South 


Although the present memoir is concerned chiefly with material 
antedating written history, and the deductions drawn from it are 
objective rather than subjective, due attention should be given to the 
ethnology of the Arawak and Carib inhabiting in historic times the 
islands where these specimens were found. There is a large body of 
documentary evidence bearing on the use of some of these objects, 
especiallj^ survivals seen by the early discoverers. It is not designed 
to treat this material from the historical point of view, but a few 
general statements at the outset may clearly define the relation of 
the historic to the prehistoric. 

This memoir relates to prehistoric times, while the documentary 
evidence deals with the historic epoch: the two methods of study 
shoidd go hand in hand. All the early historians point out that 

" The reader will flnil in Mr. T. .\. Joyce's " Central American and West Indian 
Arclifeology " a valuable popular introduction to the subject here considered. 


there was a marked difference between the historic inhabitants of 
Haiti-Porto Eico, or the Greater Antilles, and the Lesser Antilles. 
The former were called Arawak, the hitter Carib. The two races 
were hostile to each other and their culture was similar but Hot 
identical. To the historians this was a fact of geographical dis- 
tribution. They paid no attention to what had been the condition 
in prehistoric times, or whether the life of the earlier inhabitants 
from Trinidad to Cuba was ever more uniform than they found it. 
They recognized, however, that the Carib were a more or less 
nomadic, while the Arawak or Tainan were a stationary people. A 
study of prehistoric material here presented supports the belief that 
the earlier inhabitants of the Lesser Antilles were even more closely 
allied culturally to those of the Greater Antilles than were the later 
Carib to the Arawak. The Carib inhabitants of the Lesser Antilles, 
as we know from both archeology and legend, had submei-ged the 
former population as far north as Vieques Island and the east coast 
of Porto Eico. They were likewise known to all the Greater Antilles, 
even to the Bahamas, but had not yet overcome and replaced a 
preexisting Tainan or Arawak population. , 

In his memoir on " The Aborigines of Porto Eico " ^ the author 
has shown, as far as possible with limited material, the characteristics 
of the culture of that prehistoric Antillean life in Porto Eico. In the 
present article he will try to indicate, mainly from archeological 
material, the culture of the Lesser Antilles before the advent of the 
Carib. While it is i^robably true that many of the older customs and 
objects belonging to the prehistoric people of the Lesser Antilles sur- 
vived among the Carib and were in use when these islands were first 
visited by Europeans, many were not. These objects of a past culture 
were obsolete and the most exhaustive examination of the literature 
fails to reveal their probable use. Notwithstanding this uncertainty, 
however, these objects of stone, clay, wood, or shell are often desig- 
nated " Carib artifacts," as if made by this vigorous nomadic stock. 
Many of them are, however, mentioned as in use at this early time 
by Carib, and as there is a larger literature on Carib than on Arawak 
ethnolog}-, dating to the fifteenth, sixteenth, and seventeenth cen- 
turies, much information may be gathered by historical methods or 
examination of documentary accounts of this race. In one or two 
instances this method is used in the following pages, but the arche- 
ological or objective method is the one generally employed. 

» Twenty-fifth Ann. Kept. Bur. Amer. Ethn., 1907. 



When the West Indies were discovered by Europeans the inhabi- / 
tants of these islands were ignorant of the metals, iron and bronze, 
■which have played such an important part in elevating the condition 
of prehistoric man in the Old World. Stone, clay, wood, bone, and 
shell were employed by the natives for utensils and implements; 
gold and copper for ceremonial purposes or for personal decoration. 
The pre-Columbian aborigines of the West Indies, like those of the 
rest of America, were practically in what Prof. Hoernes has aptly 
called the infancy of our race culture, to which the name Stone Age 
is commonly applied. 

This period of race history seems to have been universal; it was 
nowhere of brief duration. Successive steps in cultural advance- 
ment were slow and in certain localities were retarded by unfavorable 
environmental conditions. ' 

It has been estimated that the Stone Age in tlie Old World 
lasted from the year 100000 to 5000 B. C.= The American Indian 
was practically in the Stone Age when he was discovered at the close 
of the fifteenth century, and the inhabitants of a few of the Poly- 
nesian Islands were still living in this epoch a little over a centurj' 
ago. There is every reason to suppose that the parentage of the 
American Indian dates as far back as that of the Europe-Asian man, 
provided both sprang from the same original source. It is known 
from evidences drawn from diflFerences in implements that during 
the protracted Stone Age epoch man in Europe passed through 
distinct phases, which have been designated the earliest, the old, and 
the new stone epochs, before he entered that of metals. The Ameri- 
can Indian had developed into the new or polished Stone Age when 
he came to America, and had not progressed beyond it when America 
was discovered by Columbus. 

Although the Stone Age still survived in America when it was 
discovered, this epoch in the Old World had long before been super- 
seded by one of metals, showing that the Age of Stone in the Old 
and New Worlds does not correspond in time. When the New World 
was discovered Europe had been in possession of metal implements 
for several thousand years. The highest development of stone tech- 

* Journal of the Washington Academy of Sciences, Vol, V, No. 12, June 19, 191.5. 
^Practically another way of saying that the length of the Stone Age far exceeded the 
age of metals. 




nique, other things being equal, would naturally be looked for where it 
had been practiced the longest time, and it is to be expected that the 
prehistoric stone objects found in America would be superior to the 
European, known to have been made before the discovery of bronze 
and iron. 

Individual specimens of stone implements from the Old and New 
Worlds are so similar in form and technique that it is very difficult 
to determine which continent can show the better examples, but 
comparing the majority of implements from the Stone Age in Amer- 
ica with those made before the discovery of bronze and iron, now ex- 
hibited in Europe, it has been found that the former are, as a rule, 
superior to the latter. In Stone Age architecture we find a like 
superiority. The buildings constructed in the American Stone Age 
excel those of the same epoch in Europe, as will appear when we 
Tompare the stately temples of Peru, Yucatan, or Central America 
with the megalithic monuments and other buildings ascribed to the 
latest Stone Age of Europe.^ 

Character and decoration of pottery is also a fair indication of 
cultural conditions reached in the Stone Age in different regions 
of the globe. The ceramics of this epoch in America reached a 
higher development than those of the polished Stone Age of the 
Old World, as may be readily seen by comparisons of the beautiful 
prehistoric American Stone Age potteiy with that of man before 
the use of metals in the Old World.' 

It thus appears that, if we base cultural advancement on pottery 
or house building, America had reached a higher stage of develop- 
ment than Europe, even though man in the former was ignorant 
of the metals, bronze and iron. The implication is that the human 
race, found in America in 1500 A. D., had lived in a Stone Age longer 
than man in Europe, where metals had been introduced fully 6,000 
years before Columbus. 

The implements found in the West Indies are among the highest 
developed examples of this Stone Age. Many of them are the most 
perfect of their kind and rank with the polished stones of Polynesia, 
Africa, and Asia. In architecture the branch of the American 
race inhabiting the West Indies in prehistoric times had not made 
great progress, although the cognate ceramic art was well developed. 

Wliile there is little in prehistoric America to show a serial succes- 
sion of stone implements based on method of manufacture, as indi- 

' This Judgment is based on the probable form and character of the ancient houses 
of the Stone Age in Europe, from " house urns " or burial urns shaped lilie houses, or 
from the reconstructions made of walls as indicated b.v post holes and floors. These 
buildings of the European Stone Age were certaini.v inferior to those of the same epoch in 

'These examples show the weakness of relying .solely on stone, bronze, and iron In 
classification, and the futility of basing the degree of human culture on any one form of 


cated by chipping, polishing, or other superficial characters, the 
variations in their forms are great. They indicate geographical 
rather than historical cultural distribution. Certain characteristic 
forms of stone artifacts are confined to certain areas, but these char- 
acteristics are not of such a kind as to make it difficult for us to 
readily arrange them in sequence. The first step to take in explana- 
tion of different types of stone implements is naturally to define the 
areas that are typical.* 

While the different known types of stone objects found in the 
West Indies may be considered geographically rather than his- 
torically, this manner of assembling specimens in large collections 
brings out many facts which will make it possible later to determine 
a definite chronology/, and to associate types of implements with local 
conditions, thus affording an instructive study of the interrelations 
of environment and human culture. 

We can believe that certain of the stone implements found on 
these islands ai-e old, but it can not be proved that the oldest of them 
extend back to the earliest polished stone epoch. Stone implements 
made by chipping, or those having unpolished surfaces, are rare in 
the West Indies ; they have not been reported in sufficient numbers to 
enable us to say that they indicate the former existence in these 
islands of an epoch when chipped implements were the only ones 
employed. A few chipped axes have been reported from Santo Do- 
mingo and other islands, but neither there nor in other islands are 
the flint chips numerous enough to afford conclusive proof of an 
epoch, notwithstanding these implements and their chips closely re- 
semble similar objects picked up on the sites of workshops in the 
Old World. 

The discoverers of the West Indies early recognized that the abo- 
risrines of different islands differed in their mode of life, their culture, 
and their language. In early accounts we find two groups designated 
as Arawak and Carib, accordingly as their life was agricultural or 
nomadic. It was stated by the early travelers that these groups in- 
habited diffei-ent islands, the former being assigned to the Greater 
Antilles, the latter to the Lesser. 

The large collection of artifacts characteristic of the aborigines of 
the West Indies now available shows that the stone tools, pottery, and 
other objects found on the islands inhabited by the Carib are radi- 
cally different from those from islands on which the so-called Ara- 
wak lived. Students of prehistory did not at first connect this dif- 
ference with any racial dissimilarity, but ascribed all these imple- 

' The cuUure historian is concerned with the distribution of areheological objects in 
time and space or in history and geography. It is ,tor the geographer to interpret 
geography in relation to history and for the historian to translate history by the interpre- 
tation of the geographer. 


ments to the Carib. This conchision does not necessarily follow, for it 
fails to take into account the significant fact that the stone objects 
found on the so-called Carib Islands may have been made by a people 
inhabiting them before the Carib came. Moreover, this interpreta- 
tion does not give sufficient weight to the evidence furnished by the 
implements themselves, for they imply a culture quite different from 
that of the Carib as made known by historical accounts, as flourish- 
ing at an earlier date on the Carib Islands. In other words, there is 
good evidence of a prehistoric race other than Carib but related to it 
inhabiting the Lesser Antilles before the arrival of the Europeans. 
This culture is here called the lerian as that of Porto Eico is known 
as the Tainan. 

One characteristic of the prehistoric objects found on the islands 
inhabited by Carib when discovered may be mentioned in this con- 
nection. It is well known that the Arawak, like all agricultural 
peoples, are great potters, and that the ancient Carib, like nomads, 
from necessity were not. The two races probably preserved these 
characteristics in the West Indies ; and the fact that we find pottery 
objects of high excellence on all the islands inliabited by the Carib 
leads to the natural inference tliat they were made by a people allied 
to the Arawak who anciently lived on these same islands or lerian 
women and their descendants married to Caribs. 

Archeological remains left by the aborigines of the West Indies 
reveal three cultural epochs, grading into each other, which may 
indicate a sequence in time or distinct cultural stages. These epochs 
were those of the cave dwellers, the agriculturists,*" and the Carib. 
The most primitive culture is represented by objects found in the 
floors of caves or in the numerous shell heaps scattered from Cuba 
to Trinidad. A second stage is more advanced and is agricultural in 
nature, represented on all the islands, but surviving at the time of 
discovery on the larger — Cuba, Haiti, and Porto Eico; while the 
third, or Carib, stage had replaced the agricultural in certain of the 
Lesser Antilles, especially on the chain of volcanic islands extending 
from Guadeloupe to Grenada. 

Although the three stages above mentioned are supposed to follow 
each other chronologically, not one of them had completely died out 
when Columbus discovered America. The cave dwellers still sur- 
vived in western Cuba and in Haiti, and according to some authori- 
ties they spoke a characteristic language. The Arawak inhabited 
Porto Eico, Haiti, Cuba, Jamaica, and the Bahamas. 

The customs of the aborigines who left the great shell heaps found 
throughout the AVest Indies were apparently different from those 
of the natives of prehistoric Florida but like those of northern South 

*" Rome of the finest specimens of pottery evidently belonging to the agricultural epoch 
occur in sheJI heaps and caves. 


America. These people, essentially fishermen, lived on fishes, mol- 
lusks, or crabs, eking out their dietary with turtles, birds, and other 
game captured along the shores ; fruits and roots were also probably 
collected and eaten, but their main food came from cultivated crops 
of yuca planted in the neighborhood of their settlements. The 
nature of their food supply confined them to the seashore or to banks 
of rivers, where village sites occur in numbers. It is i^robable 
that the shell-heap people of the West Indies were likewise cave 
dwellers and resorted at times to rock shelters for shelter or protec- 
tion. We know from excavations in caverns that they buried their 
dead in these caves, which later came to have a religious or cere- 
monial significance. 

We may suppose that a life devoted to fishing would make men 
good sailors, and it is probable that the prehistoric Antilleans manu- 
factured seaworthy canoes, hollowing out logs of wood with the live 
ember and the stone ax. It is also evident from objects found in 
the floors of caves that the women of this epoch manufactured pot- 
tery, and as reptilean figures in relief or effigy vases representing 
this animal occur constantly, we may suppose that some reptile, as 
the iguana or turtle, was highly prized for food. Some of the bone 
needles, whistles, and ornaments of shell or wood found in shell 
heaps show that those who camped in the neighborhood were ad- 
vanced in culture, while other objects found in the West Indian shell 
heaps are, so far as techni(nie goes, equal to those of the highest of the 
Stone Age culture. It is probable that this form of culture reaches 
back to a very early date in culture development. 

One important consideration presents itself in relation to the 
shell-heap life in the West Indies as compared with that of the^ 
shell heaps in Florida and Guiana in South America. The very 
existence of the shell-heap culture on the continents and connecting 
islands would seem to slied light on the earliest migrations of 
West Indian aborigines. Unfortunately, however, the objects manu- 
factured by all primitive people in this stage are so crude that they 
are not distinctive; there is often a parallelism in their work. 
For example, pottery from widely separated regions often bears 
identical symlx)ls. even where tlie people who manufactured it have 
had no cultural connection. Consequently, although we find cer- 
tain conmion features in decorated coastal pottery of Florida and 
that of Porto Kico, this similarity implies rather than proves cultural 

The highest prehistoric culture attained in the West Indies was 
an agricultural one. It was based on the cultivation of the yuca 
{Manihot manihot). a poisonous root out of which was prepared 
a meal, from which the so-called cassava bread was made. At the 
time of the discover^' the cultivation of this plant had attained 


such complete development that Porto Eico and Haiti are said 
to have been practically covered with farms of this plant. In 
fact, -when sorely pressed by the Spaniards to furnish them 
gold for tribute, one of the caciques offered to cultivate for 
I the conquerors a yuca farm extending aci'oss the island of Haiti. 
Both Porto Eico and Haiti appear to have been densely populated, 
and the failure of the population to advance into a higher stage of 
development was due to the perishable character of the root or 
food plant cultivated. Corn and other cereals" were not extensively 
used and there was no domesticated animal. It is evident that this 
culture was built on a root food supply which was clearly a product 
of environment, and on account of this dependence merits careful 
study by the culture historian and anthropo-geograijher. 

The development of this culture varies on different islands or 
groups of islands, forming cultural centers of which the following 
can be recognized by the character of the pottery: (1) Porto Eico, 
(2) Jamaica, (3) eastern Cuba and Bahamas, (4) St. Kitts, (5) 
St. Vincent, (6) Barbados, (7) Trinidad. The differences in arti- 
facts characteristic of these culture centers of the Antilles are some- 
times small; thus, the Porto Eico area, which includes also Haiti. 
Santo Domingo, Mona, and some smaller islands, is clearly allied to 
the eastern Cuba and Bahama area. In the former we have the three 
types of stone implements — stone collars, elbow stones, and three- 
pointed idols — none of which has _yet been described from Cuba, the 
Bahamas, or Jamaica. Pottery from these islands bears rectilinear 
or curved lines ending in enlargements,^" a decorative feature which 
is absent in Jamaica. This feature does not occur in the Lesser 
Antilles from St. Thomas to Trinidad, where four different regions 
of decorated pottery can be differentiated. 

A search for a stone technique equal to that of the Greater An- 
tilles on the North or South American contiguous areas is not re- 
warded with much success. The stone collars, elbow stones and 
triangular stones of these islands are of superior workmanship 
and find their parallel on the gulf coast of Central America and 
Mexico, especially among the Totonac and Huaxtec. Here, also, 
we find enigmatic stone objects, like stone yokes and stone rings, as 
finely made as the Antillean collars and elbow stones. Their rela- 
tionship lias been suggested by several students, but their connection 
has not been made out with any satisfaction nor has it been demon- 

^ Corn iZra mayn) was introduced into thp West Indies as a food plant shortly before 
tlie advent of the Spaniards. If sufficient time had elapsed its cuitivation would have 
changed the form of cultural development based on root agriculture, unless as in the Lesser 
Antilles it liad been destroyed by Carib who were pressing in upon it with such force 
that it could not survive. 

^•^ This characteristic feature of Porto Rican pottery decoration appears on pottery 
found by Mr. Clarence Moore in mounds of northern Florida. 


strated which objects are the most ancient; wliether the West Indian 
■was derived from the continental, or vice versa, or whether both 
independently originated is one of the unsolved problems of Ameri- 
can archeolo<ry. 

The West Indies are geologically divided into two great divisions, 
known as the Greater and Lesser Antilles. The separation of the 
two is a channel, or possibly the Anegada Passage, between Porto 
Rico and the Virgin Islands.^^ The former division includes Cuba. 
Jamaica. Santo Domingo, Haiti, and Porto Rico; the latter, a chain 
of smaller islands extending from the Virgin Islands to the northern 
coast of South America. 

The antiquities of these divisions differ in many characters; for 
instance, the majority of the edged-stone celts from the (ireater 
Antilles have a petaloid or almond-shaped form, being sharp at one 
end, pointed at the other, finely polished and destitute of a groove for 
the attacliment of a handle. Ninety per cent of all celts found in 
Jamaica, according to Prof. Duerden. and, it may be added, a still 
larger percentage of those from Porto Rico, Cuba, and Haiti, have 
this petaloid form. Across the Anegada Passage, in the Lesser An- 
tilles, this proportion no longer holds true; the relative number of 
petaloid forms diminishes at a leap, and true axes take their numeri- 
cal predominance. In the volcanic islands very few petaloids occur. 
Here the points of the almond-shaped celt are replaced by wings 
or extensions — a form rarely found in the Greater Antilles, but con- 
stituting about 90 per cent of all the stone implements in these 
islands. This radical change stamps the petaloid, although it is 
represented in all the Antilles, as a northern type characteristic of 
the Greater Antilles, while the eared ax maj- be regarded as more 
strictly southern in its distribution. Shell celts are universal, but 
their relative proportion is small in all of the islands except Bar- 
bados, where they constitute 99 per cent of the total number of 
celts. A comparison of pottery and other archeological objects shows 
a similar separation of the islands into the two divisions correspond- 
ing with those above mentioned. 

The AVest Indian geographical areas are considered in the fol- 
lowing order: 

1. Trinidad. 

2. Barbados. 

3. St. Vincent-Grenada. 

4. Dominica. 

5. Martinique. 

6. Guadeloupe. 

"The Bahamas constitute a special group, the culture of the aborigines resembling 
that of Porto Rico in many particulars. 


7. St. Kitts. 

8. St. Croix. 

9. Haiti-Porto Rico. 

10. Cuba. 

11. Jamaica. 

12. Bahamas. 

The differences in prehistoric culture in these areas are mainly 
shown in their ceramics, but these variations do not always occur. 
They are mainly due to local causes, as geographical situation 
and possibly acculturation of foreign elements. The pottery of 
Trinidad should be ranked very high, both in technique and decora- 
tion, being closely related to that of the shell heaps of adjacent South 
America. It may, however, not be far from truth to say that as a 
rule there is a general similarity in pottery of prehistoric date 
from Trinidad to Cuba. Some regions of individual islands, as west- 
ern Cuba, appear to be wholly destitute of ceramic remains, and 
possibly this is due to the persistence of tribes ignorant of this art 
in these localities. 

The boundaries of the areas above mentioned overlap and con- 
verge into each other to such an extent that there is some difficidty 
in determining the limits of any one area, and it is impossible some- 
times to discover to what area some of the smaller islands should be 
referred. A determination of culture characters of some of the 
islands is impossible without larger collections and renewed investi- 

The urgencj' of a call for archeological field work in the Antilles 
was long ago expressed by M. Guesde in a " personal history " quoted 
by Prof. Mason, as follows : " In the presence of this collection 
[Guesde] one is led to ask if these wrought stones are the work of the 
Yguiris or of the Caribs, or if they would not belong to these two 
races. We are in almost complete darkness on this point." ^^ 

In the many archeological collections from the Lesser Antilles, 
embracing thousands of specimens examined, the author has not 
found a single example of the characteristic three-pointed stones,'* 
not a single stone collar, elbow stone, or stone seat, which can be 
referred without question to these islands. The fragment of a stone 
collar seen in the Norby collection at Santa Cruz, Danish West 
Indies, belongs to the Porto Rican area. Two stone collars, one of 
which is in the British Museum and the other in the Guesde coUec- 

'=.41 least two distinct cultures, probably more, existed in Santo Domingo-Haiti when 
discovered. Tiie western end of tills island, lilie western Cuba, was inhabited by cave 
dwellers ; the eastern by agriculturists. 

'^ Mason, tJuesde Collection of Antiquities, p. 734. 

'* Specimens of a fourth type of these pointed stones in the Heye collection were ob- 
tained from the Gri-nadines, hut these are somewhat different from the type of three- 
pointed stones and may not belong to this group. 


tion, i^robably also came from the Porto Kican area, although 
ascribed to Guadeloupe. 

The general character of Jamaican antiquities seems to indicate 
that the culture in that island was different from that of Haiti and 
Porto Eico, and the stone implements thus far known from there 
are certainly more closely allied to those from eastern Cuba. Stone 
collars, elbow stones, and three-pointed stones do not appear to have 
been indigenous in Jamaica or Cuba. Their absence is sufficient to 
separate Jamaica and western Cuba, culturally, from Porto Eico.^^ 

It has been difficult to clearly differentiate minor archeological 
culture areas of the Greater and Lesser Antilles, since sporadic 
specimens are found in one that do not occur in others, and the diffi- 
culties are increased by the fact that in many collections the pro- 
venience of specimens is often wrongly labeled. It is also to be 
pointed out that there is no material from several of the islands, 
making our classification of prehistoric objects and references to 
areas provisional. These areas can, therefore, only be accejited 
in a general way. 

Since three-pointed stones and collar stones are limited in their 
distribution to Porto Eico and Santo Domingo it may be taken for 
granted that this type originated there or that they are autoch- 
thonous on these islands. By the same course of reasoning the 
fishtail and winged implements, limited especially to the volcanic 
areas, as St. Vincent, Grenada, and Guadeloupe, probably originated 
where they are found buried in great numbers.^" 

The study of Antillean linguistics ought to greatly aid the arche- 
ologist in the study of West Indian culture areas. Words and 
phrases, like objects, are archeological evidences handed down from 
a remote past. 

Some light on the existence of the prehistoric culture areas above 
suggested may be shed by a study of words for animals or plants 
still current on different West Indian islands. It is instructive to 

"The more general use of caves for burials and for habitations, and the great number 
of middens, would indicate an earlier phase of Antillean culture surviving longer Id 
Jamaica than in the other Greater Antilles except Cuba. 

'° Father Labat (Nouveau Voyage aux Isles de I'Am^rique, vol. 1, pp. 142-143) de- 
scribes a custom among the Dominican Caribs of burj'ing in a cache such valualilcs as 
they wished to conceal. These have been found in caches in St. Vincent, in cutting roads 
through the country, and may be explained in this way : '" When the inhabitants fear 
pillage this is how they hide what they want to save. For such as will resist humidity, 
such as objects of iron, plates and dishes, kitchi-n utensils, barrels of wine and brandy, 
they malie a hole on the seashore 8 or 10 feet deep so that the soldiers sounding witli 
their swords can touch nothing harder than sand. After the cache is filled up and 
covered with the same sand the balance is thrown overboard so that no elevation of the 
sand may be noticed. Water is also thrown on it to solidify it, and care is fallen to align 
it with two or three neighboring ti'ees or big stones, in order to enable the cache to be 
subsequently located more easily by lining up the same marks. 

" When objects can not be carried to the seashore, holes are made in dry ground or 
among the canes; if it is in a savanna the (top) ground must be carefully lifted as when 
one lifts sod, after which cloths are put around the place where the hole is to be dug 


note that Carib place names are rare in St. Kitts and Santa Cruz, 
which is in marked contrast to other islands, as Porto Rico, St. Vin- 
cent, and Dominica, which still bear Indian names. Islands colonized 
by white settlers of English extraction rarely preserve Indian names, 
while in those settled by French and Spanish many survive. Thus 
in Barbados, settled by English, there are few Indian place names, 
while in Jamaica, which was obtained by conquest and was Spanish 
for 162 years before the English subdued it, several Indian names sur- 
vive. As, however, the present paper does not venture into the great 
field of Antillean linguistics this subject must be passed over with a 
brief mention. 

In the following pages the author considers the different archeo- 
logical culture areas in sequence, from Trinidad northward, the 
characteristic antiquities of each island being considered geographi- 


The island of Trinidad may be regarded as the gateway to the 
migration of Arawak and Carib races from South America to the 
chain of islands connecting the continent with Porto Rico and the 
other Greater Antilles. This island was the home of several tribes of 
Indians when discovered by Columbus and constant references to 
them are found in all the early writings. 

The following account of excavations at Erin, Trinidad, is quoted 
at lengtii from " Prehistoric Objects from a Shell-heap at Erin Bay, 
Trinidad " : " 

The shell heap at Point Mayaro to which the author has here called 
attention, and which he wished later to study, has been excavated 
since he left the island and has yielded many specimens, some of 

that the soil may not show on the neighboring plants. The top of the hole must be as 
small as possible and enlarged as it deepens. When the cached objects have been put in 
it is filled with earth and tightly packed down ; water is thrown on it ; the sod is also 
wet which has been lifted, carefully replaced, and the rest of the soil is carried away. 
The ground around is dampened in order to freshen the ground which has been parched. 
When clothes, laces, silks, papers, and other things which may be impaired by dampness 
are to be cache<l they are put in big coyemboucs, which are great calabashes from trees 
cut off the fourth or fifth part of their length : this opening is covered by means of 
another gourd (calebasse). and these two pieces are held together by a thread of mahot 
or agave, somewhat as the bottom of a senser is attached to its top. These two pieces so 
attached are called "coyembouc." This word, as the invention, is of savage origin. 
When the coyembouc is filled with what is desired should be put in it, the cover is 
attached with a cord and it is tied among the branches of chestnut trees or trees with 
larger leaves, which are commonly surrounded with vines, some of which are put in the 
coyembouc, which hide it so well that it is impossible to see it, and the leaves which 
cover it prevent the rain falling in It or to cause the least humidity. Thus the 
inhabitants cache their most valuable articles ; but their booty. Jewelry, and money 
they must hide themselves without witnesses, for if their negroes know they will not 
hesitate to force him to tell where it is, or the slave may rob the cache while the master is 

" Amer. Anthrop., n. s. vol. xvi, no. 2, pp. 200-220. 


■which have been described and figured by Mr. De Booy in his article, 
" Certain Archeological Investigations in Trinidad, British West 
Indies." ^^ The collection made by De Booy at Mayaro and elsewhere 
contains many more specimens than that from Erin, but they do not 
greatly differ from those here illustrated. They indicate a people 
in about the same cultural condition, allied to Tainan rather than 
Carib stocks. 

Trinidad is well adapted for the home of an aboriginal people. 
It has constant fresh water, an abundant supply of food, its moun- 
tains and j^lains being well stocked with animals, the sea affording 
an abundance of fish, moUusks, and crabs, and its soil yielding a 
large variety of edible roots and fruits. The island lies in full view 
of the coast of South America and was visible to the natives in- 
habiting the Orinoco delta. On its lee side the water is shallow, 
but landing can be made at many places in small craft. There are 
high hills in the interior, level savannas along the coasts as well as 
inland, and streams of fresh water that open into brackish lagoons. 

Early historical references to the Indians inhabiting Trinidad 
date from the discovery of the island by the great (Jenoese. As 
Columbus on his third voyage, in 1498, sailed with his companions 
along the shore of the newly discovered island which he had named 
after the Holy Ti-inity. writes Peter Martyr.'*" " from their ships the 
Spaniards could see that the country was inhabited and well culti- 
vated ; for they saw well-ordered gardens and shady orchards, while 
the sweet odours, exhaled by plants and trees bathed in the morning 
dew, reached their nostrils." Following the shore somewhat farther, 
Columbus " found a port sufficiently large to shelter his ships, though 
no river flowed into it. * * * There was no sign of any habitation 
in the neighbourhood of this harbour, but there were many tracks of 
animals similar to goats, and in fact the body of one of those animals 
* * * was found. On the morrow, a canoe was seen in the distance 
carrying eighty men, all of whom were young, good looking, and of 
lofty stature. Besides their bows and arrows, they were armed with 
shields, which is not the custom among the other islanders.'" They 
wore their hair long, parted in the middle and plastered down quite 
in the Spanish fashion. Save for their loin-cloths of various coloured 
cottons, they were entirely naked." Columbus naively declared that 
he followed in this voyage the parallel of Ethiopia, but recognized 
that the people he found in Trinidad were not Ethiopians, for the 
" Ethiopians are black and have curly, woolly hair, while these na- 

" Amer. Anthrop., n. s., toI. xlx, no. 4, pp. 471-486. Republished in Cont. Mus. Amer. 
Ind., vol. iv, no. 2. 

IS" De Orbe Novo, vol. i, pp. 132-133. 

" Tbe Orinoco Indians had elaborate shields. 


tives are on the contrary white [lighter in color?] and have long, 
straight, blond hair.'' ^'' 

According to Las Casas, who is said to have possessed accounts 
of the third voyage of the great admiral which are now lost, the 
sailors of Columbus saw human footprints on the shore of Trinidad 
and discovered implements shoving that the aborigines were lisher- 
men. As Columbus skirted this coast he observed houses and culti- 
vated fields " bien probada a labrada," indicating that agriculture 
as well as fishing was practiced by the natives. In the meager refer- 
ence to the people given by Las Casas he says incidentally that " they 
were lighter and better proportioned than those of the other Antilles, 
and wore their hair long, like the women of Castile. They wore 
variegated cloth headbands, and girdles on the loins. The men were 
armed with bows and arrows, and, unlike the inhabitants of the 
other Antilles, had [war] shields."'^ The identity of these people is 
not clear from this early account, but somewhat later they were re- 
ferred to as Arawak. 

Sir Kobert Duddeley in 1595 made a journey through Trinidad 
and lodged in " Indian towns," finding the natives a fine-shaped and 
gentle^- (sic) people, naked and painted red. 

Later. Sir "Walter Raleigh enumerated the following " nations " 
or races in Trinidad : Yaios, Amecos (Arawak) , Salvagay (Salivas) , 
Nepoios, and Carinepagotos. At the end of the seventeenth century 
there were said to have been 15 Indian towns in Trinidad, but the 
2,032 aborigines recorded as inhabiting the island in 1783 had 
dwindled to 1.082 ten years later.^^ 

In some of the early historical references to Trinidad all the 
natives are classed as Arawak.-* Thus Davies -'^ writes: "It was 
when the captain was engaged for the war against the Arawages 
who inhabit Trinity [Trinidad] Island, and to that purpose he made 
extraordinary preparations." In other references to the Trinidad 
aborigines which might be quoted the name of Carib does not occur. 

" It is not improbable that in ancient times there was frequent communication 
between the Inhabitants of the mainland of South America and Trinidad — a communi- 
cation that was kept up until quite recently, for it was only a tew years ago that canoa 
loads of Indians were accustomed to land at Erin Bay, at rare intervals, and make their 
way by an old Indian trail to the present city of San Fernajido, via Siparia, through the 
original forests. These visits are now made primarily for trade and are probably a 
survival of a custom quite common In prehi.storic times. Well-marked " Indian trails " 
can still lie followed through the forest depths. 

^ The Warrau, who lived on the mainland, have a large square shield called ha-ha, 
used in athletic sports. <See E. F. Im Thurn, Among the Indians of Guiana, London, 
1883, p. 327.) 

2= This is not characteristic of the Carib, according to Ideas current then or in later 
times. It may be noticed, en passant, that there is no mention of Carib in the early 
accounts of the Indians in Trinidad seen by Columbus, 

^ On Bryan Edwards's map of the West Indies an " Indian town " appears on the east 
coast of Trinidad. 

" The historical evidences all agree that the people of this Island were an agricultural 
race allied in culture to Arawak. 

=» History of the Carribby-Islands, 1666. 


and indeed there is no good evidence that there were Caribs on the 
island, notwithstanding several of the above-mentioned tribes are 
supposed by some authors to be divisions of " Carib." 

The nearest approach to pure-blood aborigines of Trinidad live 
at Arima, in the middle of the island; but aboriginal features can 
still be found elsewhere among the inhabitants, although the author 
was unable to learn of a person who could speak any aboriginal 
language once spoken on the island, or that there were any Indians 
of pure bhxid remaining. There survive in Trinidad numerous 
Indian place names, as Arima and Xaparuna ; but while some of 
these suggest names existing in Porto Eico and St. Vincent, they 
are as a rule dissimilar, indicating different languages. The pre- 
historic inhabitants of Trinidad were probably linguistically distinct 
from those of the other islands. 

Additional knowledge of the culture of the aborigines of Trinidad 
can be acquired either by archeological research or through survivals 
in folklore, which are very common. 

Erin Bat 

The small settlement at Erin Bay consists of a few shops, two 
churches, and a number of dwellings along a well-built road that 
passes through the town to a warehouse on the shore. Small steam- 
ers anchor at intervals a few miles from the coast, but the best 
way t-o reach the settlement is by steamer from San Fernando to 
Cap de Ville and by carriage from the landing. It can also be visited 
from San Fernando by road, via Siparia. The only accommodations 
for remaining overnight at Erin are at the Government House. 

The present population consists almost wholly of blacks and East 
Indian coolies indentured to English planters or overseers, who own 
or manage the larger estates. The vernacular is a French patois of 
peculiar construction and incomprehensible to any but the inhabi- 
tants. The plantations are large and considerably scattered; they 
produce profitable crops, mainly cocoa and tropical fruits that are 
shipped to Port of Spain for export. 

Not far from Erin there are remnants of the primeval forests in 
which game, monkeys, and tropical vegetation abound. The land 
is rich and productive, and the estates are prosperous. There are 
a few small Iritchen middens on the coast, not far from Erin, some 
of which will well repay excavation; but their isolation is a pTa.c- 
tical difficulty unless complete and systematic work be done.^" 

28 Trinidad has never been regarded as a remunerative field for archeological investi- 
gation. The first results of the author's efforts in the island were not very promising, 
but after some discouragement, excavations of a shell heap at Erin Bay, in the Cedros 
district, yielded important data bearing on the former culture of the aborigines In this 
part of the island. 

160658-— 34 ETH— 22 5 


There are several shell moiinds on the eastern coast of Trinidad 
which show fragments of pottery and other rejecta, and several 
heaps on the southern shore that are superficially composed of shells. 
In the so-called shell heaps at San Jose the shells are few and incon- 
spicuous, but in a midden at Point Mayaro. wliich covers a fairly 
hir<re area, many characteristic potsherds may still be found on the 
surface. As a rule these shell heaps are not far from the shore, but 
in several instances they lie inland.^^ 

Fragments of pottery from this region sent to the author by Mr. 
Dearie, of Port of Spain, differ from those of Erin Bay, but appar- 
ently were made by people in the same stage of culture. There is a 
small collection from this region in the Heye Museum, obtained after 
this report was completed, which contains a number of highly in- 
structive heads and other fragments. This pottery is colored white 
and purple-red, whereas that from the shell heap at Erin Bay has a 
bright red superficial slip, although the color is often worn, showing 
gray beneath. 

Ciiip-Chip Shell Heap 

The largest shell heap in Trinidad, locally known as Chip-chii? 
hill, situated at Erin, a short distance from the shore, covers several 
acres and forms a considerable elevation. Upon this mound are 
constructed the government buildings, the police station, and the 
warden's office. The author obtained from the assistant warden, 
Mr. John Menzies, ^' permission to make excavations in that part of 
the shell heap situated on Crown land, but was obliged to suspend 
work on the private land adjoining, as it could not be thoroughly 
explored without injury to the property. The specimens, although 
limited in quantity, are the most numerous known, and give a fair 
idea of the nature of the contents of a typical Trinidad shell heap. 

Chip-chip hill was first de.scribed by INIr. Collens. whose excavations 
therein were rewarded with several fine specimens, now on exhibition 
in the Victoria Institute at Port of Spain. These objects are figured 
by Collens in his Handlxjok of Trinidad,-**" and are also illustrated by 
the present author in his Aborigines of Porto Rico.^* 

Some limited excavations were also made at Chip-chip hill by Rev. 
Thomas Huckerby, of San Fernando, several years after Collens 
finished his work, but only a few fragments of pottery, now in the 
Heye Museum, were obtained. 

^ Efiforts to find evidences that man inhabited the numerous caves in Trinidad, or used 
them for burial purposes, have not been rewarded with success, although many caves, 
especially those near Pedro Martin's basin, were examined. 

''The author is very grateful to Mr. Menzies for his aid. and takes this opportunity to 
thank him .for his many kindnesses while at Erin Bay. He Is likewise Indebted to 
Mr. Dearie, of Port of Spain, for voluntary aid in the e.\cavations. 

2*« Collens, J. H. Guide to Trinidad. London, 1888. 

=» Twenty-flf th Ann. Eept. Bur. Amer. Ethn., pi. Ixxxv. 


The extent of the Chip-chip mound could not be determined, as it 
extends far into the cocoa plantation under a dense tropical growth. 
Its surface, except where cleared by the Government for the erection 
of buildinors. was covered with vegetation. Some distance from the 
hill, where a ceiba tree had fallen, the roots showed a considerable 
deposit of shells, indicating that the extent of the heap was great 
and furnishing a clew for continued excavations. 

The shells in the mound at Erin are in layers alternating with 
vegetable mold, ashes, and soil, forming a sticky mass^° that clings 
tenaciously to the specimens and almost conceals their identity. The 
terra-cotta heads, when dug out of the earth, were completely coated 
with mud, which had to be removed by washing, and by so doing 
some of the red pigment which covered them disappeared. As the 
ceramic objects had been painted after they were fired, the color is 
not permanent, and the length of time they had been in the ground 
caused it to come off even more readily. 

As mentioned, a vei'tical section of the mound exposed alternating 
layers of shells and ashes, mingled in some cases with humus and 
with frequent fragments of charred wood. Sometimes the strata 
were composed entirely of shells, but their thickness was not uniform, 
especially at~the periphery of the mound. Over the entire surface 
of the mound there was a dense growth of tropical vegetation, with 
clearings at intervals for cocoa and plantains. The fallen trunks 
of palms, live shrubs, and trees formed an almost impenetrable 
jungle, extending into the neighboring forests where the ground had 
not been cleared. On the sea side the moimd is only a short distance 
from the shore and is separated from the bay by a lagoon inclosed 
by a narrow strip of land. Near by is a spring, from which the 
shipmates of Columbus obtained drinking water in 1498. 

In their general character the objects found in the Chip-chip 
mound are not unlike those occurring in other West Indian middens, 
although they differ in special features. As is usually the case, the 
majority of the specimens are fragments of pottery, which are among 
the most instructive objects by which culture areas can be defined. 
These will be considered first. 


Comparatively little has been published on the pottery of the 
Lesser Antilles, although specimens of whole jars and innumerable 
fragments are found in various museums and private collections. 
The Heye Musuem is the richest in the world in these objects. The 
potter's art was practiced by aboriginal people from Trinidad to 

*> During the author's work in Trinidad it rained almost every day. 


Cuba and the Bahamas, but wliile there is general similarity in the 
product, there are very marked specific differences. 

The several beautiful specimens of pottery in tlie Victoria Institute 
at Port of Spain. Trinidad, two of which, through the kindness of 
the officers of that institution, were photographed, have been repro- 
duced by the author,^^ who has cpioted the description in the ap- 
pendix in Collens's Guide to Trinidad, here reprinted, as it contains 
practically all that has been published on the archeologj^ of Trinidad : 

" The discovery of some interesting Indian relics at Ei-in during 
the past month (May, 1888) is, although I had brought my work to 
an end, of sufficient importance to demand a brief notice. On the 
occasion of a recent visit of his Excellency Sir W. Robinson and 
suite to the southern quarter of the island, the Hon. H. Fowler, who 
was one of the party, observed a mound of shells. Dismounting, a 
closer inspection revealed some pieces of rude pottery, and subsequent 
excavations by Mr. A. Newsam, the Warden, led to the unearthing of 
some cajiital specimens, indicating beyond a doubt this had been the 
centre, at some period more or less remote, of an Indian settlement. 
The pottery is of two kinds, glazed '- and unglazed, the latter dating 
back to a time anterior to the discovery of the New World, for the art 
of glazing was unknown to the early Indians, nor is it likely that they 
became acquainted with it till after the Spanish occupation." 

The following specimens are figured by Collens: 

" Figure 1. A hollow stone, smooth in the concave part, forming a 
rude mortar. The Indians used a hard, smooth pebble for pounding 
their seeds and grains. 

■■ Figures 2, 3. 4. Heads of animals in burnt clay, more or less gro- 
tesquely shaped. The eyes and mouth are exaggerated, a few, broad, 
bold lines serving to bring out the most striking feature.s. In figure 
4 the head of the monkey is fantastically ci'owned. All these were 
probably deities or ornamental attachments to earthen vessels. 

" Figure 5. A well-shaped .squirrel. Perhaps a toy whistle. 

" Figure 6. An earthen bowl in fine preservation, about the size 
of an ordinary vegetable dish. With the lid, which is unfortunately 
missing, there would doubtless be a good representation of a turtle: 
as it is, the head and tail are clearly, and the limbs somewhat clumsily, 

The best entire vessel found by the author in his excavations at 
the Erin Bay midden is the shapely brown vase shown in plate 2, A. 
This receptacle was buried 2J feet beneath the surface, in a thick 
layer composed wholly of shells. Its association and situation show 
no indication that it was deposited with care, and it could not have 

»> Aboriginps of Porto Rico. Twenty-fifth Ann. Rept. Bur. Amer. Ethn.. pi. Ixxxv. 
^ The author regrets that he can not support Mr. Collens's statement that glazed 
pottery occurs in the Chip-chip mound. 


been ;i mortuary vessel, as no bones were found near by; it appeared 
rather to have been abandoned or dropped by its owner where it was 
found. The shape of this vase is an uncommon one in prehistoric 
West Indian pottery. In form it is enlarged equatorially, and tapers 
above to a rim, which, as is rarely the case in West Indian earthen- 
ware, is without handles or luijs, and below, in which region the 
exterior is slightly convex, to the base. Decoration in the form 
of incised lines appears on the surface of the upper area, but the 
under portion is smooth and witliout ornamentation. This decora- 
tion consists mainly of parallel grooves alternating with crescents, 
and circles with central dots. The walls of the vessel are thinner 
than is usual in West Indian pottery, and the surface is little worn. 
A noticeable feature of this receptacle is the base, which consists 
of a circular stand, thus rendering stability to the vessel. Similar 
bases of other specimens, being much more substantial than the 
bodies, are frequently preserved entire while the remainder has dis- 
appeared. This form of base is of common occurrence also in frag- 
ments from St. Vincent and Grenada, but is rare in Porto Rico. 

Several bowls had been so long in the moist soil of which the 
Chip-chip mound is composed that they crumbled into fragments 
when an effort was made to lift them from their matrix. Although 
the forms of these bowls vary somewhat, several resemble that shown 
in plate 2, B, which may have been used for condiments or for pig- 
ment. ^^ The walls of this vessel are thick, with smooth undecorated 
surface ; its bottom is flat. The rim shows two opposite imperfections 
that may indicate the position of heads which served as handles.'* 

A remarkably well modeled reptilian head is shown in plate 2, C. 
Its great elongation distinguishes it from the head shown in plate 
2, Z>, which is almost spherical and has the organs represented by 
inci.sed lines rather than in relief. The same general tendency to 
rounded forms is exhibited in plate 2, E, F, G, but in these the nose 
is notably exaggerated. 

The head, and especially the position and form of the nose, of the 
handle shown in plate 2, /. remind one of pottery from the Grenada 
region, a specimen of which is figured in the author's report on the 
Aborigines of Porto Eico.^'' In this instance the nose and mouth 
are indicated by hemispherical protuberances; the nostrils are rep- 
resented by parallel slits, the eyes by pits in the middle of a circular 
disk, and the lips by a transverse furrow in a circular boss. A some- 
what similar method of indicating the eyes is shown in plate 2, H. 

" Many fragments of red and green pigment were found in the mound. The majority 
of the vessels hero described are of gray or bright red ware. 

^ After pottery objects were talten from the mound they hardened considerably, but the 
handles of this vessel may have been broken from the rim previous to its recovery. 

^ Twenty-fifth Ann. Kept. Bur. Amer. Etlm., pi. Ixxxiv. 


Plate 3, .4, represents a small rude pottery rest, of spool shape, 
with flat base, very thick walls, smooth imdecorated surface, and 
somewhat flaring rim. Its size suggests that it was once used as a 
toy or as a ceremonial vessel, but it was more likely designed as a 
support for a bowl. Some beautiful pottery rests from St. Vincent 
are in the Heye collection, several of which, in a fragmentary condi- 
tion, were obtained by the author at Balliceaux. The most elaborate 
of these measures about 6 inches in height, is perforated on the sides, 
and has a face in high relief. 

The rectangular clay box shown in plate 3, 5, has thick walls, a 
flat bottom, and squatty legs continuous with the sides. Its longer 
sides bear incised S figures surrounded on three sides by a straight 
furrow. The narrow sides of the vessel are ornamented with incised 
crescents, also partly framed with straight lines. From the broken 
places at the two opposite shorter sides of the rim it would seem that 
the vessel had been provided with handles, probably in the form of 
heads, but it is also possible that a head may have been attached to 
one side and a tail opposite, thus producing an effigy vessel. Rectan- 
gular receptacles of this kind are rare in collections of West Indian 
] lottery — a fact which imparts special interest to this example. 

The object shown in plate 3, r, is a fragment of a bowl, shaped 
like a turtle, with head and tail, and the left legs drawn up to the 
side of the body. This interesting specimen is almost identi- 
cal with the unbroken turtle effigy vase figured by Collens, to 
which reference has already been made. Although nearly half of 
this specimen is absent, enough remains to enable a determination of 
its form and of the general character of the relief decoration, which 
was no' doubt identical on the two sides. ^^ ■ The head, which is not 
attached directly to the rim of the vessel but to the )ipper side, is 
rather long, with blimt snout, and mouth extending backward; the 
nostrils are indicated by pits, the eyes by slits. The tail consists of 
two buttons separated l)y grooves, and the fore and hind legs, with 
no indication of flippers, are modeled close to the body. Like many 
Antillean earthenware vessels, the walls are thick and the rim not 

The vessel shown in plate 3, D, is also supposed to be a turtle effigy, 
an almost featureless head being attached to the rim. Opposite 
the head the rim is broken, indicating where there may have been 
formerly an appendage representing the tail. This object is one of 
the few whole specimens in the collection. 

» Unlike thp clay turtle figured by Collens. this specimen has no raised rim ahout the 
base. We know from historical sources that the turtle played an important part In 
Antillean mythology, which accounts for its frequent appearance on ceramic and other 





















A , 3.5 inches; B 4.38 inches. 


In sharp contrast with the thick-walled, coarse bowl last men- 
tioned is a fragment of a vessel (pi. 3, E) which maj' be regarded as 
one of the finest and most elaborately decorated specimens found at 
Erin Bay. This beautiful example represents the highest type of 
incised decoration of which the Antillean potter was capable. It 
shows the base and practically a quadrant of the lateral decoration 
of the bowl, which was probably repeated on the missing sides. 

In plate 3, F, the form of the head reminds one of a peccary or 
wild hog. The mode of attachment to the rim of the vessel is quite 
apparent in this instance. 

In addition to the specimens of entire pottery above described, 
many fragments, some of which represent characteristic forms, were 
excavated from the Erin shell-heap. The best of these are sections 
of rims and handles, which, being less fragile, are more readily 
preserved. Their chief features will now be considered. 


Considerable variation occurs in the form of the handles of 
earthenware vessels, several of which are still associated with 
portions of the side or rim, while others show how the handle was 
attached at both extremities. Some of the handles are mere knobs 
or bosses; other examples are in the form of elaborate heads (pi. 
3, F) , the various modifications of which recall the pottery heads of 
Porto Rico and Santo Domingo. 

The handles of bowls shown in the accompanying illustrations 
(pis. 4^8) are broken from their attachments. Sometimes they 
are very simple in form, but more commonly they represent heads 
Avhich vary more or less in shape. The specimen (pi. 4, ..4) which 
has a fragment of the bowl attached is one of the simplest forms, 
loop-shaped with a conical projection near the rim. The handle 
is broad, with ample space for the fingers. In some specimens 
the handles are even simpler, as they are without the conical eleva- 
tion, while the upper end, instead of being attached to the rim. 
rises from the side of the bowl. In other examples the handle 
takes the form of a lug or knob. 

In plate 4. B. instead of a conical laiob. the handle bears a simple 
head in which the eyes, nose, and mouth are crudely indicated, as in 
other West Indian vessels. 

Plate 4, r, shows a specimen in which the head surmounting 
the handle is modeled in greater detail, and a sufficient part of the 
body of the bowl remains to show the incised ornamentation of 
the exterior surface, as well as of the handle. Incised lines unite 
at the throat and continue down the middle of the handle throughout 
its length. 


The figure of the handle illustrated in plate 4, Z>, is simihu- 
to that of plate 4, 6', but the two incised lines ornamenting it con- 
tinue along the rim of the bowl and end above an oval elevation 
evidently representing the body of the animal. The slender head 
of the animal projects upward; the eyes are small, and incised 
crook-shaped lines extend along the head and partly surround the 
ej'es. The equatorial girt of this vessel is somewhat larger than 
the circumference of the rim and is decorated with two incised 
parallel lines. 

Another variation in form of effigy handle is shown in plate 
4, £', the head represented in this case having a somewhat pointed 
snout, oval eyes surrounded by circular grooves, an open mouth, 
and projections separated by grooves on the head. This is more 
massive than the handles before described; it is not incised, and 
its breadth at the middle is somewhat less than at the point of 
attachment to the body of the vessel. 

One of the most elaborate heads ornamenting a handle parth' 
free from the body of the vessel is shown in plate 4, F. This 
handle, like the jjreceding, is thick and broad. When placed with 
the rim of the vessel uppermost the two grooves may be identified 
as lips, the crescents above them as nostrils, and the ring on the 
side as an eye. If, however, the figure is turned in such manner 
that the rim is vertical, what was identified as the forehead becomes 
the snout with nostrils and mouth. 

The handle shown in plate 5, A^ instead of being broad is small 
and rounded. It is decorated with incised lines, and the effigy por- 
tion is larger than the handle proper. The head is protuberant 
and the eyes lenticular. Although the other features of the head 
are considerably distorted, it would appear that the handle in this 
specimen extends from the top of the head instead of from the neck, 
thereby turning the mouth uppermost, as in the last example. 

In the sections of the rims of vessels next to be described no 
handles are present. Plate 5, B^ represents a rim ornamented with 
two incised, horizontal, parallel furrows, alternating with vertical 
grooves. This rim is broad and flaring, with rounded margins, im- 
parting a convex surface to this portion of the l>owl, which has a 
straight body and a flat base. 

The incised ornamentation on the example shown in plate 5, <7, 
is more elaborate than the last. In this case the rim is quite broad, 
somewhat pointed, and covered with furi'ows, indicating an elaborate 
figure which unfortunately can not be wholly determined on account 
of its incompleteness. 

Plate 5, />, exhibits a well-modeled rim, probably representing a 
turtle with open mouth and rounded eyes. The pits under the lower 


jaw are uncommon, but like other features are suggestive of a turtle's 
head. The two appendages at the sides evidently represent flippers. 

The well-modeled head indicated in plate 5, E, is attached to a 
section of the rim, but placed lengthwise instead of vertical!}', as in 
other specimens. The snout is elongated, while the mouth extends 
far backward; the eyes are indicated by pits, and a round projection 
separated by grooves appears on the forehead. 

The degree of conventionalization in these specimens is some- 
times very great, as in plate 5, F, where practically all resemblance 
to a head is lost. Here we have a disk attached by one margin to 
the rim of a bowl, which is ornamented with a rude incised design. 
A handle distantly related to the last is illustrated in plate 7, A. 

It often happens that the walls of the orifice of a flask-shaped 
bottle are modified into a perforated clay head,*' as in the specimens 
shown in plate %, A^ B, C. 

Plate 6 shows varying forms of effigy heads which served as 
handles of vessels. All of them have well-developed nostrils, eyes, 
and other facial features. The presence of nostrils differentiates 
these heads from many others and affords a hint, although obscure, 
as to the identity of the animal designed to be represented. We find 
similar nostrils in certain three-pointed stone idols from Porto Rico, 
which we have other good reasons to identify as reptiles, hence the 
conclusion is fairly logical that these heads were intended to repre- 
sent similar creatures. 

Plate 7, .1, B, are unlike any other heads in the collection. 
• The heads illustrated in plate 7, f, D, E, can not, by reason of 
their highly conventionalized character, be readily assigned to any of 
the forms above considered. 

The two projections on top of the head and the form of the eyes 
and nose of the effigy shown in plate 7, F, are exceptional. The 
crescentic mouth is suggestive of the same organ in certain unde- 
termined Porto Rican stone idols of three-pointed form. 

The unpaired nostril of the effigy shown in plate 8, A, is indicated 
by a single pit in the sununit of a conical projection; the ej'es are 
prominent and contain crescentic slits. This head, as shown by a 
fragment of the rim still attached, projected farther beyond the bowl 
than is usually the case. The flat form of the head suggests an 
alligator, but it was evidently designed to represent a mythological 
conception rather than a realistic animal. 

If superficial likenesses of conventionalized figures are regarded as 
reliable for identification, plate 8, B^ might well be considered to 
represent a shark's head, for the position of the mouth in this speci- 

^ This is tie first example of a bead from a preliistorie flasljlilce vessel from Trinidad 
or the Lesser Antilles, although common in Haiti and Santo Domingo. 


men is well below the snout, which tapers above uniformiy to its end. 
There is no doubt that the protuberances above tlie mouth were in- 
tended to represent eyes, while those near the rim of the vessel may 
have been designed for fins or other organs. No representations 
of nostrils or ears are apparent in plate 8, C, but the broad flat head 
has two eyes and a well-developed mouth. The break at the point of 
attachment shows that it was a handle of a vessel. There remains 
a considerable number of other jjottery heads obtained at the Erin 
Bay midden, some of which are too greatly mutilated f<jr identi- 

Plate 8, Z>, illustrates a clay stamp, one of a class of objects not 

unconmion in the Lesser Antilles. The face of this specimen is 

circular, witli an incised design, and was probably used either for 

___^ decorating cloth or for stamping figures 

Cr^^^^^^ on the face or body in a manner similar 
'^P^^^^/k to the clay cylindei-s elsewhere de- 
-^^^^^m scribed.''* These stamjDS are often elab- 
-^^^B^a orate. Some of those lately obtained by 
'-&^^^ JSlr. De Booy from Santo Domingo bear 
•>^ftj^^ images on their handles and rattle when 
iy§^^^ Stone Implements 

'S^^^M Stone implements from the Erin Bay 

^^^^ midden consist of celts, axes, chisels, 
^^^^ pecking stones, mortars, pestles, and 

other forms. A number of almond- 
shaped celts, like Porto Eican petal- 
oids, were collected in Trinidad. The most interesting ax is flat, with 
notches cut at opposite edges, as shown in figure 1. 

There is general similarity in the forms of the mortars found in 
the West Indies, but the pestles vary in different islands. In the 
Santo Domingo-Porto Kico area pestles commonly have handles 
decorated with animal heads or even with entire animals, but in the 
St. Kitts region they are simple unornamented cones, pointed at one 
end, circular or oval at the opposite end. but with no differentiation 
of base, handle, or head. The Guadeloupe and St. Vincent pestles 
are of the same general character as those from St. Kitts, which are 
identical with those found in Trinidad. 

There are several stones in the collection from the Erin shell heap 
that were evidently used for pecking other stones or for pounding 

s^Aboi-igines of Porto Rico, Twenty-fifth Ann. Kept. Bur. Amer. Ethn., pi. lixxvi, o. 

Fig. 1. — Notched ax. Trinidad. 




pigments or bruising roots. They are elongate, sometimes angular, 
with shallow pits on two or all four faces which served to facilitate 
handling by providing convenient places for the thumb and fore- 
finger. Circular stone disks, probably used as grinders, were like- 
wise found. 

A small, finely polished pendant (fig. 2), made of jadeite, per- 
forated at one end, was found buried deeply among the shells in 
the Erin Bay midden. In finish this beautiful specimen recalls 
certain finely polished green petaloids collected in Porto Eico and 
other islands. The stone of which these objects 
are made does not occur in the West Indies — a fact 
indicating that the pendant, as well as the celts, was 
brought from the mainland, probably from South 

Bone Objects 

Considering their occurrence in soil saturated 
with moisture, it is remarkable that bone objects 
were preserved in the Erin Baj- mound, but many 
unworked animal bones and a few bone implements 
were exposed in the course of the excavations. One 
of the latter is from an unidentified animal, and its 
flattened form resembles a spatula used in pottery 
making. Among other bone implements may be 
mentioned a tube of uniform diameter, supposed 
to be an ornament, cut off at both ends and hav- 
slit extending along two-thirds of its 


mg a 

Objects of Wood 


t. Trin- 

A fine black finger ring, similar to the rings made and worn by 
the natives in several islands of the West Indies, was found deej) in 
the shell heap. It is made from a seed of the gougou palm. An 
angular fragment of lignite of irregular form, with an artificial 
groove encircling it. was found in one of tlie deepest excavations. 


FRo.^i Other Islands 

As is generally the case in archeological studies, pottery, from 
its greater durability and variety in form, is one of the most reliable 
types of artifacts for the study of prehistoric culture areas in the 
West Indies. The Erin Bay shell heap shares with the middens of 


other islands a predominance of earthenware with effigy forms and 
relief decoration, and the incised ornamentation of pottery vessels 
from this mound is strictly Antillean. When we compare these 
specmiens with those from Porto Rico we notice certain specialized 
features which are distinctive. In geometric designs the incised 
lines do not end in an enlargement, nor are their extremities accom- 
panied by pits, as is almost always true of pottei'y from Santo 
Domingo and Porto Eico. Comparatively few elongated heads of 
reptiles are found on pottery from Porto Rico, but such forms are 
common from the shell heap at Erin Bay. The heads from Porto 
Eico are mainly grotesquely human in form. As a rule, the rims of 
the earthenware vessels from Porto Eico have approximately the 
same thickness as the vessels themselves, whereas in Trinidad they 
are often enlarged, or turned back, and are commonly ornamented 
with figures as in the pottery from Grenada and St. Vincent. 

While it has been necessary to make comparison mainly from 
fragments, it is believed that the number of characteristic forms of 
jiottery figures from this and from more northerly islands are 
sufficient to separate the two and to lead to the belief that the 
pottery from Trinidad is most closely allied to that of the Grenada 
area, as would be naturally suspected, and that it is only distantly 
related to that of the Greater Antilles.'" 

While the evidence is not decisive, it appears from the material 
available that the Trinidad pottery is nearer to that of South 
America than to any of the northern islands of the West Indies. 
This fact may be explained by the situation of Trinidad, which lies 
within sight of South America — a fact that led to an interchange of 
cultures and peoples of the two localities. 

The nearest point in South America where excavations of shell 
heaps have been made is the Pomeroon district. British Guiana, 
whence we have a few specimens of pottery. Xone of these are 
so well made as those from the Erin Bay shell mound, and there are 
otiier indications that the ceramic art had reached a higher develop- 
ment in the islands than on the adjacent mainland. 

Regarding the Pomeroon shell heaps, Im Thum reached the fol- 
lowing conclusions: "(1) That they were made not by the resident 
inhabitants of the covmtry, but by strangers; (2) that these strangers 
came from the sea, and not from further inland; and (3) that 
these strangers were certain Island Caribs, who afterwards took 

»The author has many drawings of St. Kltts pottery which shows still greater differ- 
ences In form and ornamentation. For lilseness of pottery heads from Grenada and 
Trinidad compare plate Lxxxiv, Aborigines of Porto Rico. Twcnty-flfth .\nu. Rept. Bur. 
Amer. Ethn.. and plate vii and fig. 62, De Booy, Certain Archaeological Investigations in 
Trinidad, British West Indies. 


tribal form in Guiana as the so-called Caribisi, or, as I have called 
them, true Caribs/' *" 

Attention has been called at the beginning of this paper to the 
fact that the Trinidad aborigines are not spoken of as Carib, and 
the archeological objects show no likeness to the work of this people, 
l)ut rather to that of the Arawak, who were the great potters of 
the Orinoco. 

The well-made pottery of Erin Bay suggests an agricultural popu- 
lation rather than the nomadic Carib people, and the form of cer- 
tain flat clay platters, or griddles, is not unlike those used by the 
Arawak in the preparation of meal for cassava cakes. The aborigi- 
nes who made these objects were in a stage of culture similar to 
that of a people of the West Indies before the coming of the Carib 
in prehistoric times. Pottery making is more strictly a character- 
istic of meal eaters, and as the South American Arawak were well- 
known potters, we can not go far afield if we ascribe the pottery 
from Trinidad to a kindred peojjle. The nearest South American 
l^eople to whom we would look for their kindred are the Guaranos, 
or Warrau, some of whom still inliabit the delta of the Orinoco, 
only a few miles across the Gulf of Paria, an inland sheet of water 
which separates Trinidad from the continent. 

Although Im Thurn identifies the builders of the Pomeroon shell- 
mounds as insular Carib, he gives some weight to the theory that 
they were Warrau, which theory, however, he does not discuss and 
apparently does not accept. It seems to the author that the pottery 
found in the Chip-chip mounds indicates a culture higher than that 
of the Carib, and more advanced as art products than anj' thus 
far collected from the Warrau. He regards it as a localized or 
autochthonous develojiment originally of South American origin, 
but belonging to the same great prehistoric insular culture found 
in the Antilles from South America to the Bahamas and Cuba. This 
culture liad been submerged by the Carib in some of the smaller 
islands, but persisted into the historic ei^och in the larger islands 
wliich Carib could not conquer. 

The conclusion reached from a comparison of the objects from 
the Erin Bay midden is that while there is a general likeness in 
jjottery from all the islands of the West Indies, there are special 
ceramic culture areas in different islands. It is also believed that 
the Carib had no extensive settlement in Trinidad, and that they 
came to the other islands long after agricultural people had de- 
veloped on them, or were renegades from some of the islands where 
the uncertainty of crops drove them to become marauders on others. 
They are not believed to have made permanent settlements or, as in 

" Im Thurn, Among the Indians of Guiana, p. 416. See aiso Eev. W. H. Brett, The 
Indian Tribes of Guiana, Their Customs and Habits, London, 1868. 


St. Vincent and (luadeloupe. submerged the Tainan culture and sub- 
stituted for it a mixed one. 


The artifacts ascribed to the island of Tobago, as seen by the 
author, approach so closely those of the northern part of Trinidad 
that this island is included in the Trinidad area. In these collec- 
tions occur several axes with wings on their heads and notches on their 
bodies, and a few celts of petaloid form, which were purchased by 
the author when in Trinidad in 1912-13. The majority were said to 
have been found in a sugar-cane field near Scarborough, Tobago. No 
middens are rei^orted in the various archeological references to the 
islands, and none were seen by the author in his limited visits. 


Very little has been published on the archeology of Barbados, and 
practically no attempt has been made to determine from archeological 
data the aboriginal culture of the island. References to the aborigi- 
nes occur in works devoted to the history of the island, among which 
are those of Hughes, Poyer, Schomburgk, Ligon, and others, but 
these histories deal more particularly with the colonization eijoch 
and early European history, many having been written before it was 
recognized that man lived on Barbados before the advent of the 

The opinion is generally expressed, even in the most reliable and 
complete historical accounts, that Barbados was uninhabited when 
discovered bj^ the Portuguese in 1505, and that the aborigines had 
wholly disappeared in 1626, when the English took possession of 
the island and settled it. Although not definitely stated, it is im- 
plied by several authors that Barbados never had a prehistoric 
aboriginal population, but that it was temporarily visited from time 
to time by Carib or other Indians from neighboring islands for the 
purpose of fishing or hunting. Archeological evidences show, on the 
contrary, that the island had a considerable population in prehis- 
toric times, and that the culture of this aboriginal population was 
somewhat different from that of the neighboring islands. 

The large number of implements of shell found both in the in- 
terior and on the coast of Barbados, and the extent of the several 
middens, show without question that the island had a prehistoric 
population of considerable size. Descendants of the original popu- 
lation lived in Barbados as late as the English colonization, and the 
name of the chief city of Barbados, Bridgetown, is now thought to 
be due to its vicinity to the "Indian bridge," made of logs, now re- 
placed by the well-known crossing. There is no doubt that there was 
an Indian village near Bridgetown at Indian Kiver, one of the best 


places for landing on the whole lee shore. The names Indian Eiver 
and Six Mens Bay can also be instanced as evidence that there were 
Indian residents in Barbados when these names originated.*^ 

The opinion advanced by some writers that Barbados was visited 
from time to time by Carib from St. Vincent"^ in order to raid 
the island may have some foundation. There must have been a 
motive for these visits, which were probabh' for attacks on pre- 
existing people, the agricultural race, signs of which occur in all the 
Lesser Antilles. Whetlier populated or not at the time the whites 
came, it is evident that many islanders must have lived in Barbados 
permanently before these visits, for it can hardly be supposed that 
transient visitors would have brought with them the multitude of 
implements, pottery, and like objects now found in Barbados. The 
fact that the natives had few stone implements does not mean that 
there were few people, but that there was no stone suitable for the 
manufacture of celts, axes, and the like. The implements were 
made by permanent residents from the shell which was abundant. 

On Ligon's map *^ of the island, piiblished in 1657, 31 years after 
the settlement of Barbados by \\'arner, there is figured not far from 
the place now called "Three Houses" an Indian named ''Smyago" 
carrying a bow and accompanied by a canoe " 35 feet long." The 
position on the map where the Indian is placed and the legend 
" Three Houses," which takes its name from Indian dwellings found 
there in early times, prove that men were living on the island in 
1657.** It must be confessed that this argument loses some force, 
as camels and hogs ai'e also figured, and tliese were undoubtedly 
brought to the island by white men. 

There is indicated on this same map of Ligon the name of the 
early proprietors of the island, and the legend " 5 houses " on the 
coast not far from the present estate " 3 houses," which latter, how- 
ever, does not appear on Bryan Edwards's map, where likewise is 
the legend " 16 men " not far from the bridge which appears on 
Ligon's map and apparently gave the name of Bridgetown to the 
main city of the island. On none of these early maps is there any 
indication of the Indian castle, which is not strange, as all the 
localities are not indicated. 

*' The origin o£ the name Barbados is doubtful. Some authors have supposed it to 
have been given by the Portuguese on account of (lie epiphytic plant, hangina like beards 
from the trees, but other writers have suggested that the natives were bearded. 

*2 St. Vincent has been seen from Mount Gilboa, but no one has stated that Barbados is 
visible from St. Vincent, which is quite natural and explained by the low altitude of 

" Richard Ligon, A True and Exact History of the Island of Barbados, 1657. 

"The evidence that there was formerly an Indian settlement near "Three Houses" is 
supported by the many shell chisels formerly found in this neighborhood. One informant 
told the author that he had seen bushels of these implements from that place, and that 
they were formerly ground up and thrown on the roads to improve them. 


The earliest account of the shell implements, caves, and prehistoric 
idols foimd in Barbados that has come to the authors notice was 
written by Rev. Griflith Hughes,"^ a former rector of St. Lucy's 
Parish. The Rev. Mr. Cooksey has supplemented this with a short 
article on the earliest inhabitants of Barbados, one of the earliest in 
which shell implements are mentioned, bir Robert Schomburgk's 
account of the history of Barbados, like these of John Paget (1808) 
and Dr. Hillary (1752), add little to the archeology of the island. 
According to Joseph Forte shell chisels*'' have been found in Bar- 
badian caves, over 100 being taken from a cavern 350 feet above the 
level of the sea. 

The more extended account of prehistoric material from Barbados 
in the Blackmore collection is as follows:*' "The specimens exhibited 
from Barbados have been presented to the Collection by the Rev. 
Gi-eville J. Chester, who has kindly furnished the following informa- 
tion respecting them: 'In Barbados there is no hard stone, nothing 
harder than coralline limestone ; the aborigines therefore were 
obliged to import ha7-d stone implements and weapons from the 
other islands, or from the main continent of South America. For 
ordinary purposes, however, they used implements made of various 
kinds of marine shells,** and of the fossil shells from the limestone. 
These shell implements vary in length f i"om 1^ to 6i inches : some in 
my possession are beautifully formed. In the commonest type the 
natural curve of the shell formed the handle. Disks and beads made 
of shell, and large quantities of pottery, in a fragmentary state, have 
been found associated with the shell implements. The use of an 
implement somewhat resembling a hone has not been satisfactorily 
ascertained, only one specimen out of the considerable number which 
have passed through my hands being worn down by use. The large 
number of implements discovered under rock shelters and in gullies 
proves the existence of a large native population in Barbados, and as 
shell hatchets are not found in the other West Indian islands, it is 
clear that they are of purely local origin.' " 

It is pretty generally agreed among historians that when the Eng- 
lish landed at Holetown,*" in 1625-1627, the number of Indians on 
Barbados was small, but as the islands were discovered over a cen- 
tury earlier by the Portuguese, we can not be sure that they were not 
peopled more abundantly at that time. 

•= The History of Barbados. London, IT.iO. This article contains a plate with 
illustrations of shell implements and an idol. 

" Note on Carib Chisels, Journ. Anthrop. Inst. Ot. Britain, vol. xl, pp. 2-3. 

" Stevens, Flint Chips, pp. 235-230. 

<« Found also in many islands, but most abundantly In Barbados. 

"The site of their landing is now indicated by a monument liearing an appropriate 


The lee coast of Barbados is a flat plain extending from highlands 
in which arise small streams of water which flow westward to the 
sea, the mouths being generally closed by extensive sand barriers and 
beaches. Ordinarily the water of these streams is held back in shal- 
low pools by these bars, but when abundant water fills the river it 
flows over these barriers. In places, as at Freshwater Bay, the fresh 
water having percolated through the jDorous soil, finds its way below 
these bars and bubbles up in the sea along the shore, making the 
Avater fresh. 

The plains on the west side of the hills, especiallj^ near the shore, 
are ideal places for Indian camps. Many pottery fragments and 
other evidences of Indian occupation are seen, but well-defined shell 
heaps of great height can rarely be traced at the present day. 

Near St. Lucy's Parish, in the central part of the island, there are 
steep, well-marked cliifs in which are instructive caves or cave shel- 
ters, common elsewhere on the island, and remarkable fissures called 
clefts show overhanging cliffs. The aboriginal implements found 
here indicate that they may have sheltered early man. 


Middens, or sites of aboriginal settlements, are found at various 
locations on Barbados, occun'ing inland as well as on the coast. We 
have records of archeological material from every parish in Bar- 
bados, but the following localities are the best known: 

1. Small gully near St. Luke's Chapel. 

2. Indian River. 

3. Freshwater Bay on the border of St. Michaels and St. 

James Parishes. 

4. Codrington Estate Springs. 

5. Three Houses. 

6. Marl Hill. 

7. Speightstown. 

8. Holetown. 

9. Maxwells. 

10. South Point Lighthouse. 
The most productive midden for collectors of " Carib antiquities " 
in Barbados is situated on Indian Eiver, a few miles north of 
Bridgetown. This midden is rather a series of village sites than a 
single mound. It can be readily visited from Bridgetown by using 
the tramway to Fontabel, the terminus of which is a short distance 
from the locality where the majority of objects were found. The 
mound at Indian Eiver has yielded many aboriginal specimens, the 
most complete collection of which is that gathered by Mr. Taylor, 
of Port of Spain, Trinidad. 
160658°— 34 ETH— 22 6 


The site of the settlement at Indian Kiver is characteristic of those 
along the west shore of the island. Indian Eiver is nothing more than 
a small brook hardly able to wash its drainage from its o^Yn mouth. 
It does not empty directly into the sea, but spreads out at its mouth 
into a lagoon, shut off from the coast by a narrow strip of sand 
forming the coast line. Aboriginal objects, mainly fragments of 
Indian pottery, occur in this neighborhood. They are found in most 
abundance sticking out of the bank at a point near " Old Fort," and 
many specimens are picked ujd on the surface of the groimd in the 
neighboring field. 

Following along the river toward its source we find a low, flat 
plain of rich soil capable of cultivation, in which occur manj' frag- 
ments of ancient potterj'. Although no great deposits of shells large 
enough to be designated shell heaps were discovered, the whole plain 
shows evidence of habitation and contains several home sites, but the 
field has been so long cultivated by white farmers that the midden 
form and the sites of the houses have been almost wholly obliterated. 

There was an aboriginal settlement at Freshwater Bay. near the 
road, only a few miles north of the city. The place takes its name 
from springs of fresh water that bubble up along the coast, forcing 
itself through the salt water along the shore, and is an ideal one for 
an aboriginal settlement. 

The author visited with Dr. John Hutson, of Bridgetown, an in- 
teresting undescribed midden in the marly hills, not far from the 
cove on the northern end of Barbados. This mound was situated a ■ 
short distance from the seashore on the side of a depression sloping 
downward to an inlet that may have served as a landing place. It 
was a barren place, with ver}' little soil, but many fragments of pots, 
legs of flat bowls, and two or three pottery heads were found. The 
soil was scanty, probably worn away, so that these fragments and a 
few broken shells were all that remained of human occupation. 


Several of the "West Indies are known to have caves used by pre- 
historic man. These natural caves were well adapted for shelter or 
protection from the sun or rain. Thus far no considerable number 
of artificial caves have been recorded. On the author's visit to Bar- 
bados he inspected a number of caves that bear every evidence of 
having been excavated by the hands of man. 

These artificial caves, which remind one of those in the Canary 
Islands, are described bj'^ early writers, but are not commonly known 
to modern students of Antillean antiquities. 

The few prehistoric objects found in natural caves or cave shelters 
in Barbados are ample proof of their former occupation by abo- 


ligines, but the larger number occur either in the talus of earth in 
front of these caves or the hills above, being rarely found in the 
floors. The best made of these caves are situated on the northern 
end of the island, in vSt. Lucy's Parish, at Mount Gilboa, but there 
are many natural cave shelters in the gulches so. characteristic of 
Bermudian geology. 


Mount Gilboa is a conspicuous hill when seen from St. Lucy s 
church or rectory, and resembles the precipitous promontories so 
common in countries where there are evidences of great erosion. A 
double line of caves, situated one above the other, can readily be 
approached from the neighlioring road. From a distance they re- 
minded the author of the cavate houses of the Eio Verde in Arizona. 
Although the walls of these entrances are more or less broken, there 
was in one instance a rude step cut in the stone floor. A large field of 
sugar cane, in the soil of which a few fragments of Indian pottery 
were found, covered the top of the cliff. 

The traces of artificial steps cut at the entrance of the Indian 
caves at Mount Gilboa indicate a former occupancy, and the tradi- 
tion current in the neighborhood assigns them to the Indians. Re- 
garding specimens of aboriginal handiwork found in the Gilboa 
caves, Eev. Griffith Hughes says: "Till they came to a large con- 
venient Cave under an Hill, called Mount Gilboa, in the estate of 
Colonel John Pickering; where I found several of their broken 
Images, Pipes, Hatchets, and Chissels." ■''" A negro woman, who 
lives in the plain near the caves, told the author that shell chisels had 
been found within her memory on the talus below the caves and the 
author picked up a fragment of a bowl of aboriginal make near by. 

Artificial Excav.4Tions 

The artificial excavations in Barbados ascribed to the aborigines 
are more or less problematical. Thej' differ in form and character 
from natural caves and their true nature is not known. They are 
not accepted as aboriginal work by all historians. 

The three supposed aboriginal excavations visited by the author 
are: (1) Indian Castle: (2) Indian excavations at Freshwater Bay; 
(3) Indian excavations at Indian River. 


The so-called Indian Castle is situated northeast of Speightstown, 
on the Pleasant Hill property, about 3 miles due east of Six Men's 

" Hughes, op. cit., p. 7. 


Bay. As one leaves 'Speightstown the road rises gradually to a hill 
and passes the castle, the entrance to which appears on the right-hand 
side about lU feet above the road. From its elevation the road has 
been cut down to its present level, which necessitates leaving the road 
in order to enter the cave by a slight climb on one side. The entrance 
to the cave is thi'ough an archway with a kej'stone, on which a figure 
is carved in relief. Both entrance and arch have their walls so 
smooth that they appear to have been made by metallic implements; 
their angles are well made and the walls are perpendicular. The 
general form of the chamber reminds one of a beehive tomb. There 
are recesses on each side wall and small niches in the rear wall facing 
the observer as he enters the chamber. The floor is level, slightly 
elevated above the entrance passage, and there is an opening in the 
right-hand wall which communicated with a well with slanting sides 
and floor lower than that of the main chamber. This well is open to 
the sky above and externally at its base b}' a passageway entered from 
a side hill, recalling a limekiln. The whole character of this exca- 
vation, especially its conical apex, led the author at first to ascribe it 
to Euroiaeans. He accepted the opinion that it had been constructed 
for a limekiln. It is to be said, however, that the walls are compara- 
tively smooth and the angles and arch so well cut that it seemed to 
have been constructed with more care than is usual with these struc- 
tures. The theory that it was a place of refuge, for storage, or pos- 
sibly a chapel, seems to have something in its favor. This beehive 
subterranean chamber has borne for several generations the names 
" Indian Cave," '" Indian Temple." " Indian Castle " ; and the ad- 
jective " Indian '" must be considered and explained away unless it 
was made by aborigines. Several old residents affirm that this room 
has always been called the "Indian Cave" or "Indian Castle." 
This name was current in 1750. as shown in the following quotation 
from Rev. Griffith Hughes : 

"As there is a very commodious one [cave] in the Side of a neigh- 
bouring Hill, called to this Day the Imlian Castle, and almost in a 
direct Line from Sir Mens Boy, and not above a Mile and an half 
off, in a pleasant Part of the Countrj', it is more than probable (espe- 
cially as there was no other so near, and so convenient), that they 
should i^itch upon this, being upon several Accounts very commo- 
dious; for, as the Mouth of it faced the West, and. being under the 
Shelter of an Hill, was secured from the AA^'ind and Rain, and even 
from Danger by Hurricanes, and as the Entrance to it is so steep 
and narrow, that, upon Occasion, one Man may defend himself 
against an hundred, it may be justly called their Castle. But what 
made this place more complete ... is an adjoining clayey Bottom, 
where they dug a Pond . . . which Place is, and hath been, since 
the Memory of the oldest Neighbours alive, call'd the Indian Pond." 


''Among several broken Frainnents of Idols, said to be dnp tip in 
this Place." continues the Rev. Mr. Hughes, '• I saw the Head of one, 
which alone weighed above sixty Pounds Weight. This, before it 
was broken off. stood u^wn an oval Pedestal above three Feet in 
Height . . . The Heads of all others that came within my Observa- 
tion, were very small : One of these . . . exceeds not in weight fif- 
teen Ounces; and all. that I have hitherto seen, are of Clay burnt." '^^ 

One or two more heads, former handles of pottery objects, have 
been found at the settlement near the cave. 


If Indian Castle were the only artificial excavation in Barbados it 
must be confessed its very exceptional character would have great 
weight, but artificial rooms dug in the rocks also occur at Freshwater 
Bay and at Indian Eiver, both of these being known as Indian ex- 
cavations, although they have a distinctive character. It may be 
noticed that remains of Indian village sites likewise occur near them 
and aboriginal objects have been found in the immediate vicinity .^- 

As there are remnants of an old fort not far from the Indian ex- 
ca-vations at Freshwater Bay. the theory that these excavations are 
" magazines " has been favored by several writers, but this explana- 
tion would hardly hold for the similar structures on Mr. Belgrade's 
property at Indian Eiver. where no indications of fortifications 

The general form of these excavations is rectangular and they 
measure several feet deep. They consist of several rooms hewn out 
of the rock and arranged side by side, communicating witli eacli 
other, sometimes having alcoves or niches in their walls. On the 
hypothesis that they are subterranean habitations we may suppose 
them to have been formerly roofed and that the entrance to them, 
which is not otherwise apparent, was a hatchway in the roof. In 
similar excavations at Indian River there was a side entrance 
through the perpendicular bank of the neighboring stream.^^ 

^Miile the nature of these excavations is decidedly problematical 
there seems no good reason to doubt their aboriginal character. 
They have from the earliest times been known as Indian excavations, 
and it would be strange if, after having been so called for so many 
years, they are not of Indian manufacture or associated with the 

" Hughes, op. cit.. pp. 6-7. 

^ Magistrate Sinkler, of Port of Spain. Trinidad, in a figure in liis Handboolj of 
Bariiados designates these excavations as " Carib graves.'* They have also been called 
magazines of the neighboring fort, but in this memoir the author regards them as Indian 
pit houses. 

" The rocli is hern so soft that there was little difficulty in excavating holes of this 
nature with shell implements. 


aborigines.^* It vrould certainly be possible for Stone Age man to 
have excavated them as easily as for Indians of Arizona and Mexico 
to dig out the well-known cavate dwellings of the Verde or Eio 
Grande Valleys. 

Certain depressions, which have a marked artificial appearance, 
occur at various localities in Barbados and are called Indian ponds. 
There is an estate known as Indian Pond which would cei'tainly refer 
them to Indians. One of these Indian ponds, situated near Mount 
Gilboa, is mentioned by Rev. Griffith Hughes, and there are other 
similar excavations in different parts of the island. 


The collection of Barbadian prehistoric objects in the British 
Museum is one of the most important known from this island. Dr. 
John Hutson, of Bridgetown, has a considerable collection and there 
is a cabinet of antiquities at Codrington College. The greatest as- 
semblage of prehistoric objects from Barbados was made by the late 
Mr. Taylor at Indian River and contains several whole pieces of 
pottery and others slightly broken, besides a number of pottery 
heads and fragments. 

Among the whole pieces of pottery there is a globular bowl like a 
teapot, with snout on one side, reminding one of the form called the 
"monkey," still used by the blacks in the West Indies, and one or 
two i^latters of somewhat exceptional form. The pottery heads have 
characteristic forms, but perhaps that shaped like the head of a shark 
is the most unusual. 

Mr. Taylor's collection has several shell objects, among which 
may be mentioned perforated disks and cone-shaped perforated ob- 
jects recalling spindle whorls. 

Among the problematic objects are two hourglass-shaped objects, 
concave at each end and narrowed at the middle, which were prob- 
ably used as rests for pottery. A stamp of disk shape, having a 
handle in the middle and a design on one face, resembles pottery 
stamps in the Heye collection, many of which came from St. Vincent. 
These are flat angular shell plates decorated on their faces with 
incised lines. Some are perforated near the border, while others are 
without perforation. 

Among stone implements may be mentioned a ball girt with 
grooves crossing each other at right angles. The few stone celts 
resemble the Scandinavian type, but petaloid celts also occur. The 
finest specimen is a well-made shell fetish having a head finely 
carved at one end and a knobbed extension at the other. 

^' The fact that Indian implements have been found 'in some of these caves shows that 
the aborigines utilized the cave shelters and natural caves of Barbados. 


One of the most instructive specimens (see pi. 87 C, D) from Bar- 
bados is owned by Mr. Conneil at St. Kitts. It is made of clay, one 
end enlai-ged in the form of a head, with a neck prolongated into a 
handle tapering uniformly to a point. The enlarged end bears eyes, 
nostrils, and mouth, and its identification as a rude head is uncjues- 
tionable. This specimen is so highly conventionalized that deter- 
mination of its use is not possible, but it resembles a stone baton 
from Porto Rico described in the author's article on Elbow StoneS.-*" 
Both were possibly carried in the hand as a badge or for cei-emonial 

A large collection of fragments of the aboriginal Barbadian pot- 
tery was obtained at Marl Hill on the northern end of Barbados. It 
is a coarse red ware, showing no signs of painting, the surfaces 
appearing to be much eroded. Judging from the number of clay 
cylinders with attached fragments of bowls from Marl Hill, the 
general form of the dishes seems to have been flat or disklike with 
raised rims. The rims of West Indian aboriginal pottery are often 
decorated with finger prints and their walls with indentations — a 
mode of ornamentation still practiced by negro potters of Xevis and 
elsewhere. A common fomi of coarse potteiy was a flat dish, gen- 
erally circular, with the edge turned up into a low ridge, imparting 
a T shape to a section. These are supposed to have been used in 
frying cassava cakes. 

There is nothing in Barbadian archeology thus far brought to light 
to indicate that the prehistoric people of that island were less highly 
developeil than those of Porto Eico or Trinidad. Xo cause has yet 
been discovered for its depopulation so early in history. The island 
is not volcanic and we have no intimation that a convulsion of nature 
drove awa}' its prehistoric people or forced them to abandon agricul- 
tural pursuits. The island was too isolated to have been frequently 
raided. It has had several severe hurricanes in historic times, three 
of which have done much damage, but none of these could have 
driven away the inhabitants. 

One of the important questions in Barbadian archeology is the pos- 
sibility that there were once cave dwellers or aborigines who exca- 
vated rooms in the soft calcareous formations which compose a great 
part of the island. While there is no doubt that natural-cave dwell- 
ers existed in the Antilles at the time of their discovery, it is not 
so evident that the aborigines excavated their houses out of the rock. 
As shown in Barbados, howe^'er, we have artificial excavations, which 
have received the names " Indian excavations " and " Indian caves." 

"" Porto Rican elbow-stones in the Heye Museum. Amer. Anthrop., n. s. vol. xv. Xo. 3, 
pp. 435-459, 1913. 

^Mr. Chester states (Stevens, Flint Chips, p. 236) : " I have also a small and beauti- 
fully formed implement in the shape of a knife, made of yellowish alabaster, and a kind 
of a stamp of the same material." 


The author is unable, after an examination of these reputed works of 
aborigines, to decide whether they were made by aborigines or by 
later inhabitants, but he has no doubt that they are artificial. 


The islands included in this area are those called the Carib Islands, 
par excellence ; that is, these are the islands on which the Carib cul- 
ture had submerged a previous^ existing agricultural or " Tainan "' 
epoch, and replaced it with their own. As, however, the natives of 
these islands, at the time of the coming of Europeans, were de- 
scended from Tainan women, the men being Carib, the resultant 
culture was essentially agricultural, especially as descent was matri- 
lineal, and most of the ai'ts and industries were due to women's 

Those artifacts that are treated under this area were not all col- 
lected in St. Vincent, many having been obtained from the Grena- 
■ dines, Grenada, and elsewhere, but as they all have common features 
in their local differences the}' belong together.^" 


The antiquities from the island of Grenada in the Heye collection 
show that the culture of this island is closely connected with that of 
St. Vincent, although it has also relationship to Trinidad and South 
America. Like the other Lesser Antilles. Grenada shows evidences 
of a shell-heap culture, an agricultural culture, and also a true Carib 
culture. Prehistoric objects from Grenada closely resemble those 
from St. Vincent, although they have some affinity with Trinidad. 
We find petaloid celts, typical St. Vincent axes, and potterj' not un- 
like that from Carriacou. 

One of the accompanying figures (pi. 9, ^4) shows a perforated 
object, another (pi. 9, B), an ax with curved cutting edge and 
elongated shank as if intended to be inserted in a handle of wood. 
Both of these specimens are in the Berlin Museum, the former labeled 
" Carriacou, Grenada." 

Between Grenada and St. Vincent there are many small islands, 
some of which have yielded interesting archeological specimens, but 
the majority have not yet been explored. The author visited Beijuia 
and Balliceaux, but was unable to cross the channel to Battowia, 
which is one of the most instructive of this group, both from its 
geographical position and the archeological remains found in one of 
its caves. 

" The majority of these objects, which number thousands, were obtained by Mr. Heye 
from Rev. Thomas Iluclierby, whose collection was the largest ever made in the Lesser 
Antilles. There are still many more, mostly duplicates, in public and private collections 
on these islands. 

fewkes] culture akeas in the west indies 89 


Bequia. an island near St. Vincent, has several kitchen middens 
from which various forms of stone implements, frafrments of pot- 
tery, and other objects have been added to the Heye collection. 
These were mostly purchased from natives and are like those of St. 
Vincent. So far as the author can learn no systematic archeological 
excavations have ever been attempted on the island. 


The island of Battowia is celebrated for its Indian caves, which 
have furnished several instructive specimens of aboriginal life. It 
lies east of Balliceaux, from which it is separated by a narrow chan- 
nel, which at the stormy time the author was at Balliceaux was im- 
possible to cross without some danger. There are several cabins on 
the lee side of Battowia inhabited by negroes, who venture across 
the dividing water at almost all seasons of the year. These primitive 
people, who are generally employed in raising cotton, were the la- 
borers upon whom the author relied in his excavations at Banana 
Bay. The best known of the objects obtained from a cave in Battowia 
are the wooden turtle ^'' found by Ober and a duho, which has not, to 
the author's knowledge, been described or figured. 


After the Carib war m St. Vincent,"^ the most hostile of these 
Carib Indians, called the Black Carib,'^" were removed from St. 
Vincent to a small island. Balliceaux, from which they were later 
transported to Ruatan Island, off the coast of Honduras. Their 
Balliceaux settlement, now abandoned, was situated on the lee side 
at a place called Banana Bay, and is marked by walls of a well near 
the mouth of an arroyo- These walls are European in origin and 
resemble those found elsewhere in the West Indies. The cemetery 
of the Carib settlement was easily found, and from it several Carib 
skulls and some fragments of pottery were obtained. It extends 
along the beach a few feet above high-water mark, and is small, the 
burials being shallow. 

A general study of the mound at Banana Bay in Balliceaux indi- 
cates that the midden was not inhabited for a great length of time, 
and there is every evidence that it is comparatively modern. The 
layer of soil which contains artificial objects is not more than a 
foot thick; the sea has washed into the bank under the midden along 

°' Aborigines of Porto Rico, Twenty-fifth Ann. Rept. Bur. Amer. Ethn., pi. xc, flgs. a, a'. 
^ See "An Historical Account of the Island of St. Vincent," by Charles Shcppard, 1831. 
5" Said to be descendants of Negroes and Carib, the former saved from a slave ship 
wrecked on Bequia. 


the shore, exposing one or more skulls and a few skeletons, some of 
which were removed by the author."" These skeletons were interred 
in the" contracted or "embryonic" position and were accompanied 
by broken pottery, shells, and fragments of charcoal and ashes, but 
no whole jars were found. The place is now uninhabited and over- 
grown with manzanillo and other bushes, but none of the trees show 
marks of great age. The author's excavations verify the historical 
and legendary account that Balliceaux was inhabited by aborigines 
and that the Black Carib probably lived at Banana Bay after the 
Garib war in St. Vincent. 

St. Vincent 

The three islands, St. Vincent, Bequia, and Balliceaux, seem to be 
related in all archeological considerations, the objects from them 
being practically identical. Stone axes from St. Vincent are found 
in almost every museum which makes any pretension to a West 
Indian collection. 

The Heye collection has over 3,000 specimens from St. Vincent, 
mainly collected by Rev. Thomas Huckerby, who was Methodist min- 
ister for several years at Chateau Belair. He likewise, through 
agents, collected in Balliceaux and Bequia, large islands near by, 
many specimens which he also sold to Mr. Heye for his collection. 

°° Evidently the bodies of tlie dead at Banana. Bay were Iniried in ttip same way as 
those described by Du Tertre (Histoire des Isles des Christophe, etc., p. 455) : "As soon 
as one dies the women take the body, wash and clean it with great care. They point it 
with roucou from the feet to the head, greasing the head with palm oil, comb them, dress 
their hair, and arrange as decently as if lh<'y were going to a solemn assembly ; then 
they wrap them in a new bed of cotton which no one has ever slept in. They make the 
grave where they are to be buried in the same house where they have died, or they build 
one for the express purpose, never burying the dead without covering them with earth, 
nor omitting any ceremony they are accustomed to have wherever they happen to be." 

In an account of the burial of a child we read (op. cit., p. 456) : " They asked us tor a 
little abandoned ' casa ' house in our garden, which we gave them, and they immediately 
all set to work on the house and put it in as good condition as though entirely new. 
They made sepulture of their child in the following manner and with these ceremonies : 
They made a grave in the middle of the house, round, and 3 or 4 ifect deep. They placed 
in it the child prepared and arranged as I have said and wrapped in its cotton bed. 
They placed it seated on its heels, the two elbows on the two knees, the head resting on 
the palm of the two hands. Then all the women sat around the grave and commenced to 
sigh strangely ; then they intoned a sad and painful song. This song was divided into 
sighs and often cries in a loud voice with the e.ves turned to heaven. They shed so many 
tears that it would have saddened the hearts of the most hardened. The husbands were 
seated heliiud (he wives, bathed in tears in imitation of them. They emiiraced them with 
one hand as though to console them and caressed them with the other. During this time 
a man filled up the grave with the end of a board, from time to time the women threw iu 
earth. After these ceremonies (which lasted a good hour) the women buried all the 
valuables of the dead person which consisted of certain little baskets, cotton thread and 
other little bagatelles on the grave." Referring to this method of burial Labat adds (vol. 
VI. p. 10.3) : " I learned during my sojourn in Dominica that when the master of a house 
came to die that he was not buried in the corner of the house, but in the middle, after 
which the bouse was abaniloned and another was built in a different locality without the 
thought ever occurring to any one lo return and lodge in that place. I have sought with 
care the reason of this ceremony so extraordinary without having been able to discover 
anything else than that It was an immemorial custom with them." 


There are also specimens from St. Vincent in the Berlin Museum 
collected by Mr. Huckerby. During the author's stay of about six 
weeks at St. Vincent he visited several refuse heaps, prehistoric 
mounds or kitchen middens, on both the windward and leeward 
coasts. An enumeration of a few of the most important of these is 
given below, but there are many others of smaller size that are not 
considered. Through the kindness of Mr. Huckerby he saw several of 
the St. Vincent pictogiaphs and visited the middens at Fancy and 
elsewhere, where a number of strange artifacts are said to have been 
found. The pictographs of St. Vincent have been well described 
by Mr. Huckerby." 


Refuse piles and other evidences of former occupation by the 
aborigines are found along the leeward coast of St. Vincent from 
Kingstown to the extreme northern end of the island, especially 
wherever there were convenient landing places or where valleys 
opening to the sea presented available land for cultivation. They 
are abundant at Barrouallie, Petit Bordel, and Chateau Belair, 
in which neighborhood we often found bowlders with pictographs 
and other evidences of past occupation. There are several middens 
on the windward side, as at Argyle, Stubbs, Overland, and Ouria. 
The volcano Soufriere has, however, covered with successive erup- 
tions of ashes most of these evidences of village sites in the northern 
end of the island, which has been designated on maps since 1733 
by the name of the Carib country. 

The midden at Fancy, designated on Bryan Edwards's map as 
a Carib settlement, lies in the Carib country at the extreme northern 
end of the island. It is extensive, but has been somewhat modified 
in form by the last eruption of Soufriere. A small stream flowing 
past the Estate House at Fancy has cut its way down through the 
soft formation, exposing a bank in which were gathered many frag- 
ments of pottery and worked stones. The top of a low bluff, near 
where this stream empties into the sea, is covered by a Carib ceme- 
tery. Here the stream has encroached on the bank, exposing skele- 
tons of the former natives and washing out human bones that are 
strewn along the l)ase of the bank. 

The midden at Stubbs, situated on the windward side of St. 
Vincent, is one of the largest in the island ; but as its surface is now 
almost wholly under cultivation, digging in it M-as not feasible, 
as it would disturb not only cultivated fields, but also the founda- 
tions of inhabited houses. Fragments of pottery are common along 
the shore where the bank is eroded by the sea, and stones showing 

s'Amer. Anthrop., n. s. vol xyi, pp. 238-244, 1914. 


evidences of having been worked by human hands can be picked up 
in the bed of a neighboring stream. All along the high bluffs on the 
sea side and in the bank of an adjacent inlet layers of pottery occur 
and several fragments of bowls were picked out of the cliff. Super- 
ficial indications show great possibilities when systematic excava- 
tions are made at this place. One or two mortars with surfaces 
hollowed out, but too large to be moved any considerable distance, 
were seen lying at the base of the bluff or on the top. Not only 
pottery fragments, but likewise shells, fragments of tests, and claws 
of crabs occur with human bones mingled with the fragments of 

The Oberland midden lies near Oberland village and is approached 
by a good road from Georgetown. This portion of St. Vincent suf- 
fered greatly from the eruption of Soufriere volcano in 1902, and 
many estates were destroyed, the inhabitants being asphyxiated by 
poisonous gases. A road cut through the bank in descending to a 
small stream exposed a section of the midden and revealed terra 
cotta or pottery heads, some of which have been washed into the 
ravine below by rains. This site sliows evidences of a settlement 
of considerable size, and would well repay systematic excavation. 

The surface of the Argyle midden has been cultivated many years 
and is now covered by fields of arrowroot. It lies to the left of the 
road shortly before crossing the bridge over the Yambou River going 
north, and can be followed for some distance on the way to the 
Yambou pictographs. Xothing of great importance has been ob- 
tained from this midden, although fragments of pottery are not rare 
on the surface. 


Following up the river to the narrow defile. Yambou Pass, a trail 
leads to bowlders on which were cut some of the finest pictographs 
in St. Vincent."^ They overlook the beautiful stream which here 
flows between two high cliffs amidst fascinating scenery, with tall 
l^alms and other tropical vegetation. The valley at this place is 
sparsely cultivated, and to reach the Yambou pictographs one has 
to cross a ditch several times which feeds a sugar-cane mill lower 
down the valley. It is, however, possible to drive directly to the 
pictographs or to within a short walk, although the road is obscure 
and ends rather abruptly. 

Pictographs recorded by Mr. Huckerby occur in St. Vincent at the 
following localities: (1) Pass leading into Mesopotamia Valley at 
Yambou Pass; (2) Layou; (3) Villa; (4) Buccament Valley; (5) 
Barrouallie; (6) Petit Bordel. 

"= These are figured by Rev. Thoma.s Huckerby in Amer. Anthrop., n. .si. vol. xvi, pp. 



Geological differences between the islands of Barbados and St. 
\'incent are great, and the culture of j^rehistoric man in the two 
may have been equally divergent. This diversity' is reflected not only 
in the form and character of the implements made in the two islands, 
but also in the material of which they were made. The former does 
not furnish hard rocks for implements, the prevailing rock being 
coral limestone; in the hitter the rocks are volcanic, very hard, and 
suitable for fine implements. 

Stone axes with extensions on the heads are characteristic of 
the zone, including Grenada, St. Vincent, Santa Lucia, Martinique, 
Dominica, and Guadeloupe, but are not found in the Greater Antilles, 
Barbados, or Trinidad. These implements are, in most instances, 
not very sharp on their edges and are only rarely pointed at the 
head, true almond-shaped or petaloid implements being rarely 
found. This culture area is one of the best known for a peculiar 
type of " Carib stones " well represented in different museums in 
Europe and America. They are often found in caches, suggesting 
either unfinished or ceremonial implements." 

The island of Santa Lucia, which has also yielded many artifacts 
in the Heye collection, was not visited by the author, but from a col- 
lector he has learned that the middens resemble those of St. Vincent, 
the islands being in sight of each other. The artifacts from that 
island are mainly stone axes and fragments of pottery. 

The main feature of the stone axes from St. Vincent is an exten- 
sion or ear on each side of the head, which imparts to it a variety 
of forms, as notched, indented, and serrated or forked. This type, 
preeminent in the St. Vincent zone, is well represented in collections 
from Guadeloupe and Dominica, and to an extent from (irenada, but 
it is sporadic, not occurring in St. Kitts. Barbados, Trinidad, or 
other contiguous regions. 


The prehistoric stone implements from the St. Vincent area may 
be classified into divisions as defined in the following pages : 
Celts and axes. 


Axes and chisels. 

Axes with caps. 

Grooved hammers and axes. 

Asymmetrical axes. 

"The collection of C.irib stone implements in the public library of St. Vincent has a 
few forms of the curved flat objects in which we find a continuation of the notch forming 
a projection that is unfortunately broken, but there is a much larger collection in the 
Heye Museum. 



Implements of crescentic form. 
Eared axes. 
Engraved axes. 
Problematic stone objects. 
Grinding implements and pestles. 
Stone fetishes, amulets, and idols. 
Enigmatical objects. 

Celts and Axes 


There are a very fevr examples of the fii'st group of stone imple- 
ments or true petaloid celts in collections from the St. Vincent region, 
and a much smaller number of these have heads or figures engraved 
upon their surfaces. Their general form is almond-shaped, identi- 
cal with those from Porto Rico. They may be characterized by a 
sharp edge at one end and a point at the opposite end. These celts 
are supposed to have been once set in a wooden handle or to have 
been carried in the hand without any such attachment. Xo speci- 
men of the so-called monolithic type of petaloid. or those with 
handle as well as blade made of one stone, has yet been recorded from 
the Lesser Antilles, although several are known from the larger 


The second group of stone implements, or axes and chisels, differ 
from petaloids in the absence of a pointed tip. which is. as a rule, 
rounded into a head. While in the first group no head is differenti- 
ated from the body or shaft of the blade, and there is no groove 
surrounding the implement for the attaclmient of a handle, in this 
group there are notches in the margin that may haA^e served for that 
purpose or grooves to which a handle was attached. Plate 9, A, 
represents an unidentified perforated stone object. The edges of 
some of these axes (pi. 9, B, C) are often so blunt that they could 
hardly be classified as cutting implements, although thej' may 
have been used for hollowing out logs for canoes after fire had 
reduced the interior of the log to charcoal. D and F are typical 

The stone chisels {pi. 0, E), of which there are a few, are longer 
and narrower than the axes, being beveled at one or both ends into 
a cutting edge, but these implements are often jjointed at one or 
both ends. The pointed specimens are sometimes flat on one side 
and curved on the opposite side, although many are curved on both 
faces. "When the edges of these chisels are squared they often bear 


projections on one or both borders, but they never have raised ridges 
or encircling grooves, separating the head from the body or shaft. 

The opposite margins of a number of chisels belonging to a well- 
defined group are indented, but these indentations are only rarely 
connected by encircling grooves, indicating a point of attachment to 
a handle. This type is v.ery numerous and assumes a variety of 
forms, but the many modifications included in it differ mainly in the 
shape of the head and shaft, as seen in profile. Thus the head may 
be extended laterally into two ears like horns, as viewed from one of 
the flat surfaces, or may be decorated with carvings on their edges. 
The blade is sometimes perforated, and the head often assumes a 
fishtail shape or bears a crest on its terminal margin. 

In one of the modifications of stone implements enumerated a 
head (pi. 9, G) is still further differentiated from a shaft by a 
shallow encircling groove that connects the two marginal indenta- 
tions above mentioned. While this groove is ordinarily more pro- 
nounced on the margin, it is often so shallow on the sides that it 
is almost invisible and difficult to trace throughout its whole length. 
Its breadth may vary, but tl\e head is always clearly indicated. 
Paired and unpaired projections sometimes occur on the margins 
of the shaft, as may be seen when these implements are laid on their 
flat side. In one of the numerous groups an asymmetrical outline 
is brought about by lateral extensions. 

A normal ax with the head perfectly symmetrical on both margins 
is shown in plate 9, G, but the form of the head is almost triangular. 
In this implement the marginal indentations are so shallow that the 
general shape approaches that of a petaloid or almond-shaped celt. 
An examination of the figure and a study of the character of the 
marginal indentations sometimes shows that they are in all prob- 
ability secondary in manufacture. 

In the specimen seen from obverse and reverse surfaces, from 
which plate 9, //, was made, we have an approach to a ceremonial 
celt, or one with a figure engraved on its surface, but of a form 
quite unlike any yet figured. This specimen has two projections, one 
on each side of the blade, while on the head there is cut an oval 
incised figure, in which the eyes, nose, and mouth of a human face 
can be readily seen. The incised lines of the face of this specimen 
have been more or less deepened since it was found, but the fresh 
markings follow the original engraving and are readily detected. 
The aboriginal character of the head of the celt is so evident that 
this specimen, although unique, is regarded as of veritable Indian 

The Heye collection possesses a large number of flat stone im- 
plements of triangular shape, often sharpened on one side by bevel- 
ing. The other margins and front are rounded, and the specimen 


was evidently used for scraping skins or cutting fibers, sticks, or 
other material. Two stone implements of this type with straight 
edges are shown in plates 9, /, and 10, >4, while in the third, illus- 
trated in jilate 10, 5, the cutting edge is almost semicircular. These 
are supposed to have been used as cutting implements, and to have 
been held in the hand in such a manner that the straight edge was 
opposed to the palm and the circular edge free. Their form sug- 
gests the semilunar slate knives of certain of the aborigines of New 
England, who, like all primitive peoples, endeavored to have a good 

The notched edge of the implement in plate 10, C, suggests a saw or 
a scraping implement, like the semilunar knife in figure B. 

The object shown in plate 10, Z>, has a semicircular form, with its 
convex side so rounded that a section takes the form of a crescent. 
The concave surface, on the contrary, is flat, and the end of this 
latter region is prolonged into a crescent horn, which is pointed, the 
other extremity, or that shown on the right, being almost globular 
in form. 

One of the most characteristic implements of semicircular shape 
from St. Vincent is represented in plate 10, E. This implement may 
have been a spear, its length from one point to the opposite being 
greater than its breadth. In this imj^lement the curved or cutting edge 
shows evidence of having been chipped after the polishing shown 
on the two sides and the cutting of the straight edges. Plate 10, F, 
like the last mentioned, exhibits marked evidence of chipping, which 
is here confined to the poll or head and sides, the curved edge being 
comparatively smooth and sharp and destitute of any signs of second- 
ary chipping. 

The breadth of the ax, plate 10, G, from one end of the cutting 
edge to the other, is slightly greater than elsewhere on the blade. 
Its margin is notched at those points, which imparts an unusual 
appearance to the whole implement. As prehistoric perforated axes 
are very rare in America, this fact gives more than usual interest 
to the specimen shown in plate 10, 77. The perforation in this 
specimen is at right angles to the surface, or from one flat surface to 
another, not from one edge to the opposite as occurs in those from 
the Stone Age of Europe. 

The unusual stone object represented in plate 10, /, is unique 
among stone implements from the Antilles. Unfortunately, it is 
broken, and the specimen seems, when entire, to have had a projec- 
tion at that point. This implement is, however, wholly different in 
form from the double-l)laded ax figured and described by Prof. 
Mason in the following lines: "A double-edged, grooved blade, of 
light brown color. The form is common enough elsewhere, but cer- 

-nif^. *^^^4ViTSft4rA)Qijrfni^> ■-■-■-"'-i-ti-ii trtiWjJtlj'' ■ *-1 ifr'tf D»'^i ^^**-*^ - ' ^ " ^^* 


tainly it seems to be the first appearance in this area of an ax with 
both ends alike." "* It is needless to say that the use of this imple- 
ment must remain in doubt until other examples are brought to 

There are represented in jilate 11 a number of artificially formed 
flattened stones which generally have their surfaces more or less con- 
vex and tlieir margins rounded. The characteristic features of one 
type are indentations, one in each margin, but there is no encircling 
groove comiecting them by which the ax was hafted to a handle. 
There are two divisions of this type, the first group including those 
in which both sides of the notched area are of about equal size, im- 
parting a dumb-bell shape as seen in profile, and a second (Z>) where 
they are unequal. The former is well shown in a specimen (C) in the 
Heye collection, as are all those considered under this heading. Plate 
11, A, represents a chisel and B a celt-like form with point cut off 

No difference in size between the two halves of the object appears 
in plate 11, 6', where both parts are crescentic when seen in profile. 
The surface of the notch is convex and has angular edges. 

The general form of the specimen, plate 11, C, is like a dumb- 
bell, the two halves being about spherical, and so flattened on the 
sides that the length is less than one-fourth of the wi<lth of the 
longer diameter. The general form of plate 11, L>, is circular, but 
its upper half is much reduced, as compared with the lower. The 
indentations separating the two are deep and the specimen is a com- 
paratively thin object with blunt edges. Its nearest ally, belonging to 
the same type, in which two sides are slightly incurved and two re- 
main flat and oblong, is rectangular in form. 

In plate 11, Z>, the margin notches are very deep, imparting a spool 
shape to the implement when seen in profile, although the curves of 
the upper and lower halves differ somewhat in size. 

The objects represented in plate 11, F, G, are stone implements 
having more or less rectangular profiles, angular sides, and flat, 
luidecorated surfaces. It is not necessary to assiune that they were 
ever furnished with special handles; they were more likely held in 
the hand for the use they served. Although rudely made, there is 
no doubt that E and H were artificially fashioned. Their forms 
are not synmaetrical. In the implement figured in plate " 11, /, 
there is a marked triangular form, which, but for other features, 
would be considered among the group designated as triangular 

Plate 11, •/, represents a stone knife and A', a petaloid celt. 

"Mason, The Guesde Collection, p. 789, Gg. 109. 
160658°— S4 ETH— 22 7 


The essential characters of plate 12, A, B, C\ D, are such as to ally 
these implements with the same type, their differences bein^ mainly 
in details. The blades (A, B) liave practically the same general out- 
line, one (pi. 12, B) being sharper than the other. Plate 12. F, has 
the ears turned upward. The margin of the blade is somewhat flat- 
tened. The use to which these imi^lements were put is doubtful but 
they may have been used for grindere. 

Plate 12, J, represents a perforated ax, whde A' shows a groove 
for hafting. 

In plate 12, i, a head m the fomi of a knob is well developed. 

Plate 12, M, N, represent implements in which the head is ex- 
panded slightly on each side, the ends of the projections becoming 
pointed. The specimen K has a well-developed shank, the edge of 
which is curved, in places relatively sharjj. It is one of the best- 
formed and most carefully made of all the implements of this type. 

Modifications of different parts of these specimens are common; 
thus a still further development of the head and an extension of the 
two exti'emities is found in the implement, plate 12, 0. Its cutting 
edge is curved to such an extent as to be the same as the general 
curve of the margin of the blade. 

In plate 13, A, there is a well-marked separation of the cutting edge 
and the sides of the blade. Plate 13, 5, has a form similar to the 
last, but differing from it in details whidi are apparent. 

The marked feature of plate 13, C, is the comparatively great 
development of the head as compared to the rest of the implement, 
and a marked secondary beveling of its edge, the original form being 
practically identical with other members of this type, except in the 
relatively deep lateral notch. 

The implement represented in plate 13, D, differs from all others 
belonging to the group of notched axes in having its blade developed 
to a relatively much greater size at the expense of the head, which 
is comparatively small. The form of the- implement, when seen in 
l)rofile, is such that its edge is continued by a gentle curve into the 
sides of the blade — a feature that might well be compared to an 
almond-shaped or petaloid celt. The convex head is roughly indi- 
cated by the two shallow notches. 

In plate 13, E, is represented a fine ax, the blade of which, when 
seen in profile, is almost circular, while the head is lenticular, con- 
tinued on each side into a sharp point. The distinguisiiing charac- 
teristic of plate 13, F, is the pi'esence of a circular pit situated in the 
middle of one side of its blade. This pit is deeply and symmetrically 
sunken and has a smooth surface. The edge of this ax is much 
broken and the head is slightly notched. 




B, G inches. 




Of 5.;i8 inches: 7, 6 inches. 




Ff 2.SS inches; 11, 2.2j iuclies; /, 4.7.3 inches. 




Ff 7.0 inches. 




B, S.25 inches. 



7 f" 



There is the same relative predominancci in size of the shaft over 
the head in this as in previous specimens, while tlie fonn and the 
relatively greater width of the shaft is shown in the specimen figured 
in plate 13, G. In plate 13. //, the head is more sharply pointed 
tlian the others, but their general form is identical. 

The head of the circular implement, plate 14, A^ is narrower 
than the blade and separated from it by sj'mmetrical marginal in- 
dentations, presenting one of the best-known examples of a type 
which has comparatively few members. Plate 14. 5, 6", Z>, £', are 
provisionally placed in this group, although in plate 14, £", there is 
a pronounced want of symmetry in the two sides. 

The features separating plate 14, £", F^ from the different mem- 
bers of the eared type are small, mainly specific rather than generic. 
Plate 14. (r, representing a typical notched-edge ax, unlike most of 
the siDeciniens thus far referred to, has a cutting edge. The essential 
feature of an ax the head of which is notched at the edges is shown 
in jilate 14, Z>, and its relatively greater breadth of head compared to 
the shaft is evident from the illustration. 

The implement i-epresented in plate 14, F, has a deep groove in 
the head and a shallow indentation on each margin. The former 
is deep and broad, whereas in plate 14, (?, where it again occurs, 
it is small and semicircular in form. 

The single notch on the head of plate 14, G^ is replaced by three 
notches in plate 14, H^ which is unique in this particular feature. 
These grooves become quite deep in plate 15, A^ while in plate 15, i>, 
the median groove of the head is more pronounced than the twa 
lateral ones. In plate 15, C\ Z>, £", there is a return to a tendency 
to a groove sepai'ating the head of the ax from the blade or a hafting 
for a handle. 

The tendency to introduce a groove between the head and blade of 
the ax appears also in plate 15, £', F, whose margin, as seen in profile, 
Itecomes a waved lijie with alternate projections and furrows, the 
latter most strongly marked on the head. The margin of the ax 
becomes rectangular in one si^ecimen, shown in plate 15, G ; there are 
furrows cut on the head of plate 15, H^ I. Figures H and / must be 
classified as aberrant forms of a tj'pe differing from the stone imple- 
ments with indented edges, but having common features which justify 
their being i^rovisionally placed in this group. 

In considering the shapes of plate 16, ^4, B, we are reminded of 
the forms of bronze axes so common in the Old World. They have 
the same symmetrical form and the sharp edge, showing that they 
are im|)lements used in cutting. 

The implement shown in plate 16, Z>, is dumb-bell shaped, when 
seen in profile, and has a median groove, the two halves being about 
uniform in size. 


In the object shown in jjlate IC, E, one half is larger than the 
other, suggesting a head and blade. There is, however, no encir- 
cling groove. 

In jjlate 16, C, F, there is an approach to an encircling groove and 
a head distinct from the blade. The top of the head of E is indented 
and the right side flattened, perhaps broken. 

Plate 16, G, represents a similar implement with two bevels on 
the head, which impart a triangular shape to this region of the 
implement when seen in profile. 

The head in plate 16, H, is rectangular and extended; in plate 
16, /, it is concave on the top. 

In plate 17. A. the head has a point on the apex, the outline of 
the blade being circular. The implement, plate 17, B, has a ferrule 
near the head extremity, the apex being flattened, and in plate 17, C, 
the apex is concave. Plate 17, D, has a broad, flat apex, which in 
plate 17, E, is incurved, forming earlike extensions. 


There are in the TTeye collection several specimens which, when 
seen in profile, resemble a petaloid celt with a cap perched on the 
pointed extremity. The rim of this cap seen on the margin over- 
hangs the surface of the blade, forming a low ridge, which is the 
upper rim of the groove, by which probably the handle was pre- 
A-ented from slipping over the pointed head of the ax. The simplest 
form of this tj^pe. shown in plate 17, F, has, in addition to the cap, 
a secondary groove situated just below it. Another form of head is 
shown in plate 17, O. In plate 17, //, there are indications of two 
similar supplemental encircling grooves, pronounced on the margins 
as notches. In this specimen the top of the head has become more or 
less flattened and slightly rounded, its end being cut off so that it is 
IDarallel with the groove, instead of being continued into a point. 
A like feature appears on the side of the ax, plate 17, /, where there 
are indications of three supplementary notches, parallel to each 
other on the flat caplike structure on the end of the. implement. 
There was probably still another notch in that portion of the head 
which is now broken off. 

In plate 18. .4, we have a typical form of ax without cap, but with 
head broad ; one edge straight, the other curved. 

The apical cap of plate IS, 5, is more globular in form than the 
last, and is separated into two regions by an apical furrow, absent 
in plate 18, .4. While the blade of this ax is destitute of the sym- 
metrj' ordinarily found in this type, the groove of the handle 
forms a true caplike head, a distinctive feature of the group. The 










t', o inches E, fsjnclics; F, 2,5 inches; G, 9.13 inches; H, 4 inches. 







iJ. 7.75 inches; C, S.44 inches: E, 6 inches; J, 5.25 inches; L, 5.5 inches. 




B, 4.38 inches; C, 6.63 inches; ii", 6.5 inches; //, 7 inches. 




.1,9 inches; B, 4.3 inches: O, S inches; E, 7.6 inches; F, 7 inches; 0, 4.3S inches; H, 7.25 inches. 




A , 9.25 inches; C, 0.7.5 inches; X*, 7. J incties; E, inches; F, .j..j inches. 


cap of plate 18, C, has its apex not symmetrical but turned to one 
side. A similar absence of symmetry likewise appears in the margins 
of the blade, which are convex on one side and straight or slightly 
concave on the other. The cap of plate 18, D, has two apical furrows 
dividing the head into unequal lobes. 

Two specimens in which the apical furrow has become so deep 
that two extended lobes have elongated into horns or ears are figured 
in plate 18, E, F. In figure F^ these ears assume a fishtail form, and 
in E the lobes are curved, distinctly resembling the ears of a rabbit. 

The head of an ax of the cap type, shown in plate 18, (?, has a 
notched or serrated rim, the sawlike margin appearing especially 
pronounced on tlie right aide, imparting to the whole implement a 
conical or triangular form when seen in profile. 

Plate 18, H, represents another ax of the same type, the head being 
broken and the blade missing. In essential features the cap of this 
implement recalls that of plate 18, 6^, from which, however, it dif- 
fers in dimensions and number of notches, the workmanship on it 
being so coarse that it appears to be an unfinished specimen. We 
repeatedly find similar implements of this form in caches brought to 
light by construction of roads across the island or by inroads of the 
sea on exposed coasts. It appears that the natives blocked out these 
implements and stored them for future use or for barter with those 
who lived on islands where there was no stone adapted to the manu- 
facture of implements. 

Plate 18, /, is an imidentified stone implement of rectangular 

Plate 19, 4, is a rectangular ax with sharpened edge; plate 19, G^ 
represents an ax of simple form ; but the blade, plate 19, B, is not 
unlike a modem ax in shape. 

It is not possible to identify the use of the implements shown in 
plate 19, Z>, E^ but it is supposed that they once had handles. 


Four stone hammers from the Heye collection, shown in plate 19, 
F-/, represent typical forms of these impalements from the Lesser 
Antilles. In their general outlines they correspond closely with 
those of Porto Eico, all being deeply grooved for hafting, rounded 
above and below. They were evidently battering or pounding imple- 
ments. From the depth of the groove it is evident that a wooden 
handle was formerly firmly lashed to them, either tied by cords or bent 
around their body, filling the encircling groove, which insured its 


The shape of the hammer sliown in plate 19, G^ approaches the 
form of an ax, its deep groove and slightly projecting cutting edge 
flaring. The groove in plate 19, //-A', is pronounced, evidently for 
the attachment of a handle. A marked feature of ax J is the flat 
head, which, as shown in the accompanying figure, also has angu- 
lar projections. 

Plate 20, J., has two lateral birdlike projections on the top of the 
head, but the groove for hafting is less pronounced than in the 
preceding. Plate 20, B^ C, represent grooved axes with prolonged 
extensions on the tops of their heads, where there are also notches. 
The extent of the prolongation of the head is here so great that we 
can hardl}^ suppose that the notch played any great part in lashing 
the blade to its handle. 

The groove of an ax in the Heye collection, represented in plate 20, 
Z>, is well marked and the beginning of the differentiation of a blade 
or body from the shaft is apparent. 

The implement figured in plate 20, E, has the sides of the head 
prolonged into two projections, like horns or ears. Its head is sepa- 
rated from the rest of the implement by a well-marked groove, the 
body being divided into a clearly defined shaft and blade. In plate 
20, F, the head has been reduced to a low narrow ridge above the 
encircling groove, there being in this specimen no line of demarcation 
between shaft and blade, while in that shown in plate 20, G, there 
are rectangular elevations on the head, which is perforated. This 
object shows no want of symmetry on its two margins. The differ- 
entiation of shaft and blade is quite evident in plate 20, H. The 
head furrow is here less pronounced than in G, where the groove is 
deep and has about the same breadth all around the implement. 

The grooved ax, plate 20, /, is well made, showing head in profile 
with ridges curving on the groove. The body, or the blade, is 
crescentic in form, and its upper diameter is about equal to that of 
the head, l)ut less than that of the shaft, which broadens some- 
what below the groove. The margins of the shaft are angular. In 
the middle of the side of this specimen a pit has been sunk in the 
surface, and there is still another such depression situated on the top 
of the head, both recalling the notches in previous specimens. 


Axes with or without grooves for hafting, but with an extension 
on one edge imparting to them an asymmetrical form, are quite com- 
mon in the St. Vincent-Grenada area. In its simplest form (pi. 20, 
A'. L) this want of symmetry is not very pronounced, consisting of a 
slight projection on the side, almost imperceptible without close ex- 
amination. From this we pass through intermediate forms to those 


at the other end of a series in which the lateral projection has de- 
veloped to such an extent that it has completely modified the form 
of the ax. 

One of the simplest of these asymmetrical axes in the Heye collec- 
tion is represented in plate 21, A. The want of symmetry in this 
specimen appears in the left border, the right margin of the ax, 
when placed in a nomial position, being simply curved, the left 
slightly pointed. 

In the next illustration, plate 21, B, representing a somewhat more 
develoi^ed ax, the asjTnmetry of the left margin is more pronounced 
and indicated by a marked projection. Plate 21. C, has this 
feature still more developed, for in this specimen the right margin is 
almost straight, while the left is curved inward, terminating in a 
projection. Unlike the preceding, the specimen ^4 is girt by a well- 
marked encircling groove for hafting, which, combined with the 
other features, imparts to it the conventional tomahawk form. Ex- 
cept that it is relatively longer and narrower, the specimen rep- 
resented in plate 21, C, is not unlike that shown in plate 21, A. 
The asymmetrical feature is well marked in this, and also in those 
represented in plate 21, />, E, although in the last mentioned all 
sign of a groove for hafting is absent. 

While the ax. plate 21, F, belongs to the same type as those de- 
scribed above, tlie shape of its head is somewhat different. Here we 
have a groove on the top of the head, evidently designed for lashing 
the ax to a handle. Tlie same general outline exists likewise in 
plate 21, //, but in this specimen the single groove on the head is 
replaced by three, and the groove encircling the head of the ax and 
separating it from the body is more pronounced. Plate 21, G, has a 
groove on the head for attachment of a handle. 

The specimens thus far mentioned have an enlargement at the end 
of the shaft forming a head. Not so, however, the next specimen- 
represented in plate 22, -1, where the shaft simply tapers to the end 
and is bent backward, forming a distinct curve, bringing the pro- 
jection on one side of the ax at the extreme left end of the cutting 
edge of the- blade. In the next specimen, plate 22, B. this extension 
has become still more prominent, for although the pointed end of 
the blade has become somewhat enlarged, the projection imparting 
the asymmetrical form almost equals that of the ax in length. 

This lateral extension of the left margin of the ax has thus far 
been confined to the lower end or the middle of the implement, but 
in plate 22. (', Z>, it has shifted its position and is here found near the 
middle. The result is that the cutting edge of the ax has been ex- 
tended on one side, whereas in the preceding specimens the length of 
the cutting edge is about equal on each side of a median line. 


The lateral extension which imparts the asymmetry to these axes 
has been abnormally developed in plate 22. E, and would have been 
still further extended in the specimen, plate 22, F, were this speci- 
men entire. 

Closely connected with the asymmetrical blades already men- 
tioned are a few of somewhat aberrant form which have the same 
peculiarities, but are modified in a somewhat different manner. 
Among these may be mentioned plate 22, G. 

Plates 22, H, and 23, A, B, represent other aberrant forms of 
implements, each of which have pecularities, but all evidently belong 
to the same type as the above. Of these specimens, that represented 
in plate 23, B, departs so much fi'om the normal form that we may 
well doubt whether or not it belongs to the group. Its outline is, 
however, asymmetrical. 

The ax. plate 23, C\ is regarded as one of the best made of the 
asymmetrical type, for not only is its cutting edge continued into a 
projection on the left side but its termination is turned upward, 
imparting a characteristic form to the head and body of the blade. 
Instead of being rounded, as on the right edge, three planes are cut 
on the left margin, one in the middle and one on each side, forming 
a kind of chamfering which differentiates it from the axes previously 

In plate 23, D, both the right and left margins, instead of being 
rounded, are bounded by flat planes, imparting to the two sides of 
the blade, as seen in profile, very different outlines. The extension 
on the left margin is exceptional. 

A somewhat similar difference can be seen in an implement found 
in the Berlin Museum, in which this difference in the two margins 
is even more marked than in any other asymmetrical ax. It repre- 
sents an implement allied on one side to the asymmetrical forms and 
on the other to those with projections on the head, features so marked 
that we may consider it a connecting link between those character- 
ized by these two i^eculiarities. 


The variety of shapes assumed by stone tools in the St. Vincent- 
Grenada area appears in the following figures, one of the most per- 
fect forms of which is that represented in plate 23, (t. Here the borders 
of the shaft are angular and the cutting edge is curved and sharp. 
The size of this specimen, or more especially the angular character 
of the border of the shaft, points to the probability that it formerly 
had a handle which may have been attached to it longitudinally, or 
it may have been inserted in the end of the handle like a chisel. The 
implement shown in plate 23, F, is one of the most effective cutting 
tools yet described from the Antilles. The margins are angular 


and both the upper end and terminal edges are sharpened. It is the 
author's belief that this stone implement was inserted in a slit 
formed in the extremity of a handle and firmly bound in place by 
cords. Forms similar to the last mentioned are shown in G and £', 
which are broader than the last and sharpened at the end. There is 
a want of symmetry in the two margins of H, the left being shorter 
than the right. The end opposite the cutting edge is here pointed like 
a petaloid stone. 

The common form of tool is shown in plate 23, F, G, TI, in which 
we recognize the contracted shaft, which was probably fitted to a 
handle, and the more or less curved edge. 

We have in plate 24, A.B, tool-formed implements in which the two 
sides of the shaft ai'e not convex, as is usually' the case, but are 
slightly concave when seen in profile. The profile of plate 24, B, 
would probably have been a complete triangle but for the fact that 
one point has been broken. The curved side of this triangle is 
sharpened and probably served as the cutting edge of an implement. 
The implement C is suth tliat it could readily serve as a spear point, 
but it may have been used as a gouge for cutting wood previously 
charred or otherwise softened by fire — a custom ascribed to Carib 
when they cut down trees or dug out cavities in logs for canoes. It 
is said that in making canoes they first burned a hollow in a log 
with live coals and then scraped it out with stone chisels, and some 
of the stone implements we are now considering may have been used 
in the way indicated. 

The form of plate 24, Z>, is almost rectangular when seen in profile, 
although there is a slight difference in width of the butt and cutting 
edge. In the implement shown in plate 24, E, a handle has been 
formed by a contraction of the diameter above the end. The same 
reduction in size occurs in plate 24, F, but in it the handle is some- 
what shorter and the cutting edge has a circular form. 

The implement represented in plate 24, (?, is a good tool with 
scjuare margins, tapering uniformly and slightly curved, making it 
a most effective cutting tool. 

These chisels are sometimes elongated in form, as shown in the 
specimen in plate 24, /. This otherwise perfect implement is un- 
fortunately slightly chipped on the cutting edge, but it is sharp and 
not too thick. The handle is round, terminating in a blunt point. 
Like the preceding, it shows evidences of having been formerly tied 
by its short handle to a stick or stave to increase its effectiveness. 
The object shown in plate 24, H, is problematical. 

The tendency in all these tools is to become pointed at one ex- 
tremity, as in plate 25, A, B, (\ and to become broader at the oppo- 
site end. imparting a well-marked spatulate form. 


The broad cutting edge of plate 26, A, is slightly curved but sharp, 
although the shaft tends to be angular on each margin. The finest 
examples of a Ijeveled cutting edge occur in plate '26, ^4., D-L. These 
tools are angular and elongated in shape, of about uniform size 
throughout their length, their thicloiess being about one-fifth of their 
breadth. The butt, when seen from one extremity, has a rectangular 
form and the cross section throughout its length is uniform. The 
use of stone objects represented in plate 26, 5, C, is problematical. 

In plate 26, G, we have represented an angular implement, sharp- 
ened but not beveled at its cutting extremity, and squarely formed 
at the opposite end. 

In plate 26, H, we find the tendency to angular edges very pro- 
nounced, and in plate 26, /, J, the same figure is evident, although 
the implement is much elongated and tapers slightly to each end. 

In plate 26, /i', the cutting edge is almost straight, but the ax 
shown in plate 26, Z, is slightly curved, its two margins rounded 
rather than angular, and the tip blunt. 

Plate 27, ^4, B, may be regarded as tools of typical forms, and in 
jjlate 2", C, we have an ai^proach to the chisel form which often 
occurs in all collections from different West Indian islands. Plate 
27, D, represents a tool, the cross section of which is square, the 
edge sharpened, and the tip flat. A similarly formed chisel is 
shown in plate 27, E, the same type, almost square in profile, appear- 
ing in plate 27, F. The chisel, jjlate 27, (?, is beveled on two oppo- 
site sides, diminishing to a point at the opposite extremity. The 
specimens represented in plate 27, U, I, are tools of the types men- 
tioned, whose forms are somewhat modified in detail, but still pre- 
serve the same general features. 

The imjilements shown in plate 28, A, 5, have been designated 
chisels, and are probably, more strictly speaking, fomis of cutting 
implements, rather than celts or wea^jons. 

In plate 28, 0, Z?, E, we have different fonns of stone cutting 
implements, cleaver shape, but with slight projections on the head, 
on each side of a median notch. Although their outlines vary con- 
siderablj' they jDreserve the same general form, usually having a 
cutting edge. 

Plate 28, F, from the Berlin Museum, is noteworthy on account 
of the relative sizes of the head and blade. 

A modification in the cutting edge is introduced in the two tools 
represented in plate 28, G, II ; the angle to the axis of the implement 
is slight in G, but more acute in H. Plate 28, /, shows a passage 
from this t_ype into the grooved implement shown in this specimen, 
while plate 28, t/. A', L. are tools having the same form, but made of 
shell. These implements are not as connnon as those of stone and 



E H 


.4 , 3.88 inches; H, 2 inches. 



A, 5.25 inches; F, 2.5 incb(_'s. 







F I 





F, 5 inches. 




.4 , 5.5 inches; B.SiiK-hcs; f, 9.5 indies; i>, 'j.sinclios; E, s inches; F,'J inclies; G, 6.3.Sinclics; //, 7.25 inches; /, 7.5 inches; 

J , u inclies; A', 1) inclies. 


are rarely collected on the island of St. Vincent. Shell was the only- 
hard substance available in Barbados for the manufacture of cutting 
implements, and axes made of this material occur in great abundance. 

Implements of Cbescentic Fobm 

Certain crescentic stones, called sacrificial knives,''^ generally con- 
vex on one edge and concave on the other, commonly sharpened on one 
point, are among the best found in St. Vincent. These assume a 
variety of modifications, well illustrated by specimens in the Heye 
collection. Although the majority of these forms were found in 
St. Vincent and Grenada, the type is not limited to these islands. 
The simplest forms of sacrificial knives are shown in plate 29, 
A, B, C. The crescentic shape is somewhat modified in ^4, and a 
projection which may have served as a handle arises from the 
concave edge. It is popularly believed that these curved implements 
are knives used in cutting out the hearts of victims in human sacri- 
fices, and following out this erroneous idea certain large bowlders 
bearing pictographs are called altar stones. 

These crescentic implements are commonly sharp at one point and 
blunt at the opposite. They may have been used in cutting fish, 
meat, or even human bodies, suggesting sacrificial knives. They are 
commonly flat at two opposite sides, rounded, often blunt on the 
edges, but they almost invariably terminate in a cutting edge or a 
sharp point. Their shape varies from a slightly crescent form to the 
spiral; sometimes their handles are straight prolongations, terminat- 
ing in a curved extremity. No historic authority can be quoted from 
accounts of the aborigines of the West Indies that they sacrificed 
human beings, but there is abundant proof that they removed flesh 
from the skeletons of the dead, even their own relatives, in their 
mortuary ceremonies. 

Various other forms of sacrificial or ceremonial knive-s are figured 
in the series represented in plate 29. One of these, plate 29, C, 
has a semicircular cutting edge like the skin scrapers manu- 
factured by some of the North American Indians. Figure G 
represents a most instructive type, in which the implement is en- 

"' The crescentic form of stone implement, lo<'all.v called sacrificial knives, is not ver.r 
common in St. Vincent, but a few fine specimens are known to exist. (See author's picture 
of pictographs on a large bowlder in 2.5th Ann. Rept. Bur. .\mer. Ethn., pi. Ixii, upper 
fig.) The best known of these is owned by Mr. Patrick Huggins, an old resident of this 
Island, whose ancestors he affirms received it from a " Carib chief." who said it had been 
used for sacrificial purposes. Local collectors in St. Vincent are accustomed to call 
stones bearing pictographs " jumbi " stones or altars, and the.v say that sacrifices of 
human beings were made on these altars with stone implements of crescentic form, but 
none of the early contemporary accounts support this statement. The crescentic type 
may be the curved knives mentioned by Labat as the instruments with which the flesh of 
men devoured in cannibal feasts was cut into pieces or scraped from the victim's bones. 



[ETH. ANN. 34 

larged into a scroll at one end, while the opposite extremity has a 
flat tool-shaped edge. It is evident that the enlarged scroll was 
so shaped in order to fit into the pahn of the hand, thus enlarging 
the handle and giving an opportunity to grasp the implement firmly 
while it was being used. 

The following figures (£', F, H, I) represent other forms of sacri- 
ficial knives whose curved ends have been enlarged into handles or 
disks, evidently better adapted for grasping in the hand. A common 
feature of these Jinives is a notch in the periphery, which in two in- 
stances (./, A') becomes quite 
prominent. The firet of these 
scroll-shaped knives to be men- 
tioned was a fragment illus- 
trated in the author's report on 
the Aborigines of Porto Rico."" 
At the time this rejjort was 
written the complete form was 
unknown. The first unbroken 
specimen of the" type of sac- 
rificial knife was described by 
Mr. T. A. Joj'ce in his account 
of prehistoric implements from 
the "West Indies in the British 

The sjjecimens figured in 
plate 29, /, /, /i, resemble sac- 
rificial knives in some particu- 
lars, but differ from them as 
follows : The inner edge of these 
specimens is almost straight, 
the other cuiwed, the two being 
separated by a shallow notch, imparting to the implement a form 
resembling an ax with sharpened edge on one side. 

Eared .\ses 

All the members of this type of stone implements possess two ex- 
tensions, one on each side of the head (fig. 3). These projections 
sometimes resemble forks, and at times impart to the head of the 
implement the form of a fishtail. In other specimens they take the 
form of simple rounded knobs, recalling incipient hoj-ns. The body 
of tlie specimen shown in plate 30, />, is perforated. As a rule, as in 
plate 30, /, the groove for hafting is absent in implements of this type, 


Fig. 3. — Ell red ax frum Guadeloupe, 


'« Twent.v-fi(th Ann. Kept. Bur. Amfr. Ethn., pi. xxiil, *. 
" Journ. Boy. Anthrop. Inst., vol. xxxvii, p. 418. 


but in figure E this furrow, although shallow, is well defined. The 
specimen, plate 30, E, is exceptional in possessing two angular ridges, 
extending one on each margin of the sides, almost half the length of 
the shaft. Plate 30, F, G, show prominent typical earlike appendages 
to the head, characteristic of the type, but in plate 30, H^ /, these lap- 
pets are small, separated by a slight shallow notch. These ears, 
which in the preceding illustration, plate 30, G, are prominent, suf- 
fer a reduction in size in a specimen not shown, where they appear 
as lateral projections and low prominences, one on each side of the 
middle line of the ax. The two margins of this ax are almost paral- 
lel, the cutting edge being slightly cur\-ed. In none of the specimens 
is there an indication of a groove for hafting. 

The form of implement shown in plate 30, /, departs slightly 
from that of the typical eared ax, the prominent characteristics of 
which are a broad, square incision on the poll at the middle line, and 
the recurved ears. 

Plate 31, 4, shows a fine ax with prominent lateral projections 
from its head, and two rectangular elevations replacing the ears on 
the heads of specimens already described. Appendages of the same 
fonn appear in a specimen (pi. 31, F) where the upper notches are 
represented, when seen in profile, by a waved line. The margins of 
the body of this ax are planes. 

In the ax shown in plate 31, B^ the two projections of ears are 
confluent, having a perforation or opening Ijetween them. The edges 
of this opening are not beveled but rounded, having been smoothed 
on their surface, betraying the original lines of formation here 
suggested. There is visible in this specimen a rectangular ridge 
situated on the upper rim of each ear or lateral extension similar 
to those found in other axes of the same type. The margins of the 
blade in this ax are parallel and somewhat angular, but the sharp- 
ened edge or the cutting portion is slightly curved. 

The specimens, plate 31, C , D, E. may be called eared axes. The 
projections from the head of D are here bounded by flat planes and 
are not curved, and the median depression on the head is rectangidar. 

Engbaved Axes 

The head of the specimen in plate 32, ^4, is somewhat broken on one 
side, and the surface of the liody of the blade, lielow the groove, 
decorated with incised lines arranged in triangles. Plate 32, B, bears 
a number of parallel indentations on the edges. 

If the ax shown in plate 32. C, be viewed in profile it will be 
seen to bear, instead of two ears or horns, two grooves for hafting a 
handle to the implement. 


As the upper portion of the head of the specimen, plate 32, />, 
is much broken, it is very difficult to determine its original form. 
The general appearance of the unbroken portion would indicate a 
perforation. The instructive features in this specimen are the figure 
incised on the surface of a depressed area, situated in the middle of 
the side, and a pointed arch, which lies within the same area. These 
incised figures in St. Vincent axes remind one of the ceremonial celts 
of Porto Rico. 

Plate 32. A', shows a broken ax somewhat similar to the last except 
that the dumb-bell design which occupies the center of the latter is 
here replaced by a perforation. The right-hand side of the blade is 
bi'oken, and when whole we probably have the anomalous feature of 
an ax with its blade divided in its length in such manner as to present 
two cutting edges. This remarkable feature in the blade is no less 
strange than that of the head of the same implement, which is unlike 
that of any known ax. 

In plate 32, F\ where we have type features of the group of imple- 
ments with projections on the head, there is a well-marked shaft, 
narrow blade, and well-marked ears on the head, in the form of 
angular extensions, one on each side of a median furrow. 

Problematic Stone Ob.iects 

Plate 33. A, is melon-like, and, like the others, enigmatical so far as 
use is concerned. Although these specimens ai'e made of a hard 
stone their forms recall certain objects made of pumice stone found 
at the hamlet called Fancy, on the north side of the great St. Vincent 
volcano, Soufriere. 

The specimen illustrated in plate 33, Z>, with several features 
ascribed to stone implements, is exceptional in possessing a peculiar 
beveled edge which is shown on the right lower side. Its opposite 
side, not shown in the figure, has an identical form to that shown in 
the illustration. 

Plate 33, C\ so far as fonn goes, to all intents and purposes repre- 
sents a pestle, but unlike all grinding implements thus far described, 
it is rectangular instead of oval or circular when seen in cross sec- 
tion. The general appearance of this implement recalls a stamp or 
rubbing stone, but, although this resemblance is heightened by the 
form of a handle, the majority of other features place this specimen 
in the group we are considering. 

The two globular stones of irregular form shown in plate 33, D, E, 
have artificially worked surfaces and resemble each other in the 
common feature, a slight indentation on the surface, by which a 
pointed projection recalling a tooth is brought into relief. 




B, 6.3 iiuhes; D, 3.31 inches; E, 9.5 inches. 



B, 3.25 inches; D, 8.3 inches. 



A, 5.0 inches; B, 7.4 inches: E, 4 inches. 




D, 4.:is inches; E, inches. 




D, 5 inches; F, 9.81 inches; /, 4.25 inches. 




A , 3.2.3 inches: 11, ,0 inches 


The problematical implement of irregular shape, plate 33, F, is 
artificially worked, evidently for some unknown purpose, while that 
illustrated in plate 33, G, has six regular faces of unequal size. 

Another conical stone, plate 33, 77, also has six faces which are 
slightly convex, no two of which have the same length or width. 
Plate 33, /, was drawn from a stone specimen of hemispherical form, 
flat on one side and convex on the other. Like that shown in plate 
33, J, it recalls an unfinished ax, being irregular, hemispherical or 
ovate in form. 

The last-mentioned fonn is preserved in plate 33. A', the convex 
siu'face of which is irregular throughout. This object might perhaps 
be better described as an oblate spheroid with flat surfaces. 

The general features of these stones are the slightly convex sur- 
faces which reappear in the dumb-bell shaped stone, plate 33, £, 
which has a flat surface on one side and a sharp cutting edge on the 
opposite. Although in form this specimen reminds one of an ax, it 
is unlike those thus far considered. 

Plate 33, J/, represents an implement of cubical form with two 
opposite faces of circular contour and a slight median constric- 
tion in the narrow edge. This object has all the outward appearance 
of a grinding implement, and may have been used to bruise roots, 
vegetables, or pigments. 

The form .1 (pi. 34) is somewhat more elongated than that last 
mentioned, having surfaces smooth and slightly convex, bounded b.v 
flat facets. 

Among other problematic stone objects of the same general type 
there are certain forms, one of which is shown in plate 34, B, 
which resembles a small whetstone, flat on one side, convex on the 
other, and beveled into a cutting edge at one end. Unfortunately the 
opposite extremity is broken, but its general form suggests that it 
terminated in a point. The form of this implement is like that of a 
chisel, but it difi'ers from chisels found in Porto Eico, Santo Do- 
mingo, and the other Greater Antilles. 

The boat-shaped stone object, plate 34, (\ has its two opposite flat 
faces connected by a flat plane, the width of which is uniform and 
equals the distance between the faces. This implement terminates 
at each end in beveled edges, which, however, are not adapted for 
cutting, although it is not impossible that they may have been used 
in fashioning clay or pottery objects. 

The continuation of the plane around the border of the whole im- 
plement found in plate 34, C, is also a marked feature of plate 34, D. 
A flat stone with crenelated border, plate 34, £", may have been used 
by potters in rubbing down pottery to the desired form. 



[ETH. ANN. 34 

-restle seen In 
(3.5 inches.) 

Gbindinc Implements A^•D Pestles 

There is considerable varietj' in the form 
of pestlesi and mortars from the St. Vincent 
region. Some of the former are oval or 
sphei'ical stones, slightly concave, sometimes 
with equatorial grooves. They often have 
the head, handle, and base differentiated, but 
the head is not, as in the Santo Domingo 
pestle, carved in high relief, but incised on 
the point of the handle. The forms, which 
are conical, are distinctly characteristic of St. 
Kitts, although they occur on all the islands 
from Porto Rico to Trinidad. 

The object plate 

34, F, is of un- 
known use and is remarkable on account of 
the face cut on one end. Plate 34, G. is 
placed among pestles on account of the simi- 
larity in its form 
to these imple- 
ments. Plate 34, 
//, has a pestle- 
like form and / 
and / are objects 
of unknown use. 

Fig. .5. — Stone postk' with 
face. (3.31 inches.) 


6. — Stone pestle with 
eyes and mouth. 

Plate 34, A', is a 
pestle with globu- 
lar end and handle 
slightly enlarged 
at its termination. 

Different objects from the Lesser Antilles 
are represented in figures 4^9. Figure 5 
represents a pestle, the end of which is pro- 
longed into two knobs or ears and the oppo- 
site grinding surface slightly enlarged. It 
has a face carved on the handle, with eyes, 
nose, and mouth well represented. In many 
respects this pestle recalls those from Porto 
Rico. The object shown in figure 9 has the 
form of a pestle, but the opening or depres- 
sion at one end would indicate that it was 
used for some other unknown purpose. 
Figure 6 is a pestle with eyes and mouth 
represented at one end, while in figure 7 




these organs are simply pits or depressions. In 
figure 8 the face is in relief. Both figures 7 and 
8 have a transverse perforation, which would 
seem to indicate that they were not pestles, but 
were suspended, possibly as ornaments. 

On plate 35 there are figured a number of 
pestles which varj^ in shape, all but one {B) 
having the conventional form. That repre- 
sented in ^4 has a circular base and the point of 
the handle turned to one side. In C the form is 
angular, a rare condition among pestles, but D 
has the i-egular conical form. The point of the 
handle is cut off by a flat plane in E and in F 
the whole imjolement is pyriform. The jDestle 
shown in O is bicornis, and 
11 also originally had two 
horns, one of which is 
broken. / has a constric- 
tion near the base and 

^ 'SsL ^^^ ^^^® *'°1^ °^ ^'^^ handle is 

1 \^'I^SK rounded. 

Fig. 7. — Stone object in 
shape of pestle. (3.88 

Stone Fetishes, Amulets, and Idols 

The St. Vincent area has furnished a few 
fetishes that illustrate the idolatiy of the Lesser 
Antilles. Rep- 

•ilead and han- 
dle of broken pestle. 

resentations of 
idols that exist 
in public or 
private col- 
lections from 
these islands 
are not as well 
made as those from the Santo 
Domingo-Porto Rican area and 
the characteristic forms of those 
known from the two areas are 
different. The only examples 
of three-pointed stones are two 
or three specimens belonging to 
the fourth type which were col- 
lected in Grenada. There are 
only a few of these, and I doubt 
whether those known came 
from St. Vincent or from the 
Greater Antilles. 

160658°— 34 eth— 22 8 


Pestlt sUupcd problematical object. 



[ETH. ANN. 34 

The various stones illustrated in figure 10 include forms of pend- 
ants, triangular stones of unknown significance, and ornaments. 

One form of amulet (pi. 36, A) in the Heye collection resembles 
in several particulars a specimen in the Vienna IMuseum figured by 
Dr. Heger (see fig. 61). This specimen resembles in fonii a wonn or 
centipede and has appendages to the head, the body being divided into 
segments by grooves. 

The amulet in the Heye collection shown in plate 36, B, is rec- 
tangular in shape, much broader than wide, with a wing on each 
side of the body. The head of this specimen has two perfo- 
rations, and is separated 
from the body by a slightly 
curved groove. There are 
similar jjerforations near the 
upper edges of the wings. 
In the center of the body 
(if the amulet there is a 
pit surrounded by a circle, 
from which extend curved 
lines, indicating appendages 
with some likeness to legs. 
The form and markings on 
this specimen suggest a con- 
\entionalized animal, as a 
bird, while the position of 
the marginal pei'f orations 
indicates that it may have 
been used as a pendant. The 
surface is smooth and highly 
polished, as if worn or much 
handled by its former owner. 
The form of the amulet (pi. 
36, C) from the Heye collec- 
tion approaches that last mentioned, but the different parts are more 
conventionalized. This amulet, probably a fetish, is quadrangular in 
form, slightly curved on one surface and almost flat on the oppo- 
site. At about one-third its length the specimen is crossed by a 
deep groove extending from the margin to the middle of one side; 
other deep grooves mark off a triangular figure that may have been 
intended for a head. It is probable that this stone was used as 
a pendant attached to a necklace, serving as a fetish for personal 
l>rotection or as an ornament. 

The small, conical, well-worn, and perforated stone (pi. 36, D) 
recalls the stones shaped like cones above described, but differs from 

10._Stone pendants. (3, 1.38, 2.44, 1.88, 
3.31 inches.) 


them in having a sulcus or groove cut in one side, but not extend- 
ing around the rim of the- base. This specimen, like the last men- 
tioned, was probably a fetish or worn as an amulet on-i necklace. 

To the same type as the stone objects above mentioned belongs 
a thin semicircular stone object which has two perforations, one at 
eacli end of a scroll shown on each face. The margin between these 
scrolls becxjmes straight instead of curv'ed. The fomi of jjendant next 
to be mentioned consists of stones with perforations on their margins. 
The simplest forai of perforated stones used as pendants is seen in 
plate 36, Z>, where we have an oval nodule with a perforation extend- 
ing through the object. This nodule is made of the hardest kind of 
rock and its perforation shows a degree of skill in the use of boring 
implements which is not surpassed in work of it."? kind. 

The specimen illustrated in the two following figures (pi. 36, 
£", G) from the Heye collection, is made of a hard stone, cut in a 
triangular form, perforated with inciseel decorations on both sur- 
faces. A perforation for suspension of this specimen is large and 
regularly beveled. 

Enigmatical Objects 

The author approaches a consideration of these singular objects 
with some trepidation, for while they are the most exceptional forms 
reported from the Antilles they are not unlike certain stone objects of 
undoubted Indian manufacture found elsewhere in the West Indies. 

His first introduction to them incited a keen desire to see the 
locality in St. Vincent where they were said to have been found 
and discover others in situ. Through the kindn^ess of Mr. Hej'e, the 
author, accompanied bj" Rev. Thomas Huckerby, made a \-isit to the 
locality, but, either because his time was too linrited or from other 
reasons, no additional specimens were obtained. The objects figured 
in the accompanj'ing plates were the only specimens of the type 
examined. The majority are said to have been found at the settle- 
ment called Fanc}', on Fancy River, 2.50 yards from the sea, by Mr. 
Morgan, from whom. Mr. Huckerby obtained them and afterwards 
sold them to Mr. Heye. They were exposed in digging a roadbed 
from Shipping Bay to Fancy. The objects have a red or dark gray 
color, sometimes with patches of black, and are made of a soft vol- 
canic tufa that readily crumbles, especially under moisture. They 
show a great variety of form and a number of plates (pis. 37-61) of 
the more striking ones are here given for comparison. 

These objects are unique and unlike any Antillean objects known 
to the author. It is impossible for the author to interjjret their use, 
as his Imowledge of the circumstances under which they were found 
is limited. It has therefore seemed justifiable to give what might be 
considered a superabundance of illustrations to guide future arche- 


ologists in studies of these fonns to which the author is able to add so 
little definite information. 

The following mention of these objects by Mr. George H. Pepper 
is published in an account of " The Museum of the American Indian, 
Heye Foundation.""* 

" Of still greater interest," he writes, " to the student is a series of 
both well-known and fantastic objects made from a metamorphosed 
volcanic scoria. There are several hundred specimens, all of which 
were found in a restricted area near Fancy at the base of the volcano 
of La Soufriere. Nothing like them has been found in the adjacent 
islands, and it is quite probable that they were made and deposited 
at this place as votive offerings in way of propitiation to the god of 
the volcano." 

Eegarding their age the Eev. Thomas Huckerby, from whom Mr. 
Heye purchased these objects, writes as follows: "I think that the 
specimens indicate, a verj' old civilization. Probably they take us 
back beyond the Carib occupation. As suggested in a previous letter, 
it is probable that these specimens and the people who used them 
were covered up by the ejecta of a prehistoric eruption of the 
Soufriere. * * * " 

The autlior is unable from the scanty evidence available to deter- 
mine either the age or genuineness of these objects; but he woidd 
judge from a personal examination of the site where they are said to 
have been found that they date back to pre-Carib times. No object 
was found on the site that resembles them, so we arc obliged to rely on 
the testimony of the collectors for their authenticity. 

These objects show artificial working and appear to belong to a 
type. Their variety of form may be best illustrated by considering 
in turn several of the most common representatives. Their three- 
sided form is prominent in plate 37, A, somewhat resembling in out- 
line the three-pointed stone idols of the fourth type partially finished, 
their points being rounded and without superficial decorations or 
carved heads. 

In plate 37, /?. we have a similar stone, possessing three rounded 
points, two of which are extended in such a way as to resemble 
wings, arising from a spherical middle region that may be desig- 
nated the body. 

Til is modification of two points into forms of wings has gone 
still further in plate 37, C, where the body has taken on an angular 
or rectangular form. • 

The specimen shown in plate 37, D, resembles an implement with a 
groove for hafting and a pointed extremity; the other pole being 
lens-shaped. It resembles in profile the other specimens, but its use 
is unknown. 

«» Geographical Review, vol. li. no. 6. p. 411, 1916. 





A , 3.U iiiL-hes; B, 2.69 inches; C, l.l:i inches; D, I.IH inches; F, 2.91 inches. 



B, 3.S1 inches; C, 4.31 inches; /, 7.44 inches; A', s inches; L, .J.J inches; M, 2.75 inches. 



A , 4.5 inches; C, 2.13 inches; D, 2.5 inches; B, 2.88 inches; F, 2.63 inches. 



.4, 3.31 inch"s; B, 3.53 inches. 







A , 3.5 inches; B, 4.13 inches; C, :!.19 inches. 











A , 4..j(i inches; C, ?,Ai inolies; D, 3.5 inches; £, 2.13 Indies; F, 2.5 inches. 




A , 4.1'J inehcs: B, 4.19 inches. 




C, 3.44 inclies; E, 5.44 inches; F, 4.63 inches. 




.4 , 3.5 inches; F, 3.N8 inches. 




,•1 , 3.31 iiuhes; B, 2.56 inches; C, 2.63 inches; F, o inche."; H, 2.44 inches; /, 3.5 inches. 




B, 5.31 inches; C, 3.jG inches; D, 2.75 inelies; E, 2.75 inches. 






A , 4. la iin-hes; It, .i..5 iiRhcs; C, :. inches. 



A . a.ti'J iuflies. 



A , i.oti inches; £,3.5 iuchcs; C, 3.19 iuchcs. 




A , 1 inches; C, J.T.J inches; D, 'i.'o inches. 



A , 5.13 inches; B, 3.SS inches; C, 2.81 inches. 




C, 4.25 iucbos. 



.1 ,3.19 inches; B, 2.7.J inches; C, l.:is inches. 




A , G.7o inches; B, 6.63 inches. 



A , 4.25 inches; £,4.00 inches; C, 2.5ij inches; 2), 3.19 inches. 






,4 , 'AM inches; B, :!..')i) inches; C, .S.o inches. 



A , 4.7.') iuches; B , 2 iiiclie.s. 




B, 2.0 inches; t, 3.75 inches; D, 3.00 inches; //, 3.19 inches. 


Plate 37, E, represents « globular stone having a projection which 
is bifurcated at one end in such a way that it might easily be mis- 
taken for an unfinished pestle. 

Specimens shown in plate 37, /•', G, 11. are modifications of the 
preceding, possessing conical projections or extensions, one on each 
side. The lower edge of plate 37, //, like all members of this group, 
shows asymmetry- in form. The object illustrated in jjlate 37, /, is 
elongated and bears two deep equatorial grooves separated by a 
ferrule ; one-half of the specimen is oval, the other more pointed. 

Plate 37, t/, recalls certain prehistoric objects called " banner 
stones," often seen in collections of prehistoric objects found in the 
United States. The body of this implement is elongated, pointed at 
each end, and with an extension like a wing on each side. It reminds 
one of a bird or some animal form. 

The shape of the object shown in plate 37, A', is that of a paddle, 
with two parts, the bhide, which when seen from one side is rec- 
tangular with rounded angles, and the handle prolonged to a point. 
The breadth of the flat blade is about equal to its thickness; but 
the handle has rounded angles and the point of its union with the 
blade is indicated by a deep groove, forming a triangular figure, one 
angle of which is situated a little to one side of the middle of the 
blade. The use of tliis implement is not known, as there is not 
sufficient evidence to prove that it was employed as a cutting, bruis- 
ing, or grinding implement. 

The specimen shown in plate 37, Z, has the form of a disk, girt 
with two grooves, one of which incloses a rounded projection oc- 
cupjdng the central part of the upper hemisphere. The shape is 
symmetrical, the surface convex, but there is no evidence visible 
showing its use as a poimding or grinding implement. 

The form of the stone objects shown in figures .1/, .V, is unusual, 
but the objects are artificial, evidently belonging to the same group as 
those referred to above. 

Some of the specimens when found were in a rather soft and 
pliable condition and had to be carefully handled until they were 
dry. As corroboration of their aboriginal origin it shoidd be 
pointed out that some of the simpler forms of these objects resemble 
closely stone objects from the island of St. Vincent. This is more 
especially true of their likeness to certain ax-formed specimens in 
which a blade, poll, and surrounding ridge is well marked. Some 
of the bowls also have lugs that recall bowls and cups of burnt clay, 
while the heads have a distant resemblance to the heads of pottery 
objects. Some of them are symmetrically formed and covered with 
complex ornamentation of incised spiral and rectilinear forms. Cer- 
tain motives are prominent and modified in a way that shows skill. 
They are remarkable enigmas and most difficult to interpret. 


It is hardly necessaiy to consider individually each of the many 
forms of these strange objects of the Heye collection, and almost im- 
possible to make a satisfactory classification with hard and fast 
divisions separated by strict lines of demarkation. In a general way 
it may be said that the collection contains imitations of animal heads, 
vessels of various forms, circular-lobed disks, and objects like mortars 
with animal heads appended on one or both margins, recalling in that 
respect pottery bowls or vases. 

Among a great variety of animal heads of grotesque form several 
specimens distinctly resemble human heads, and among animal 
forms the bird, turtle, or some reptile is recognizable, but the re- 
semblance is so aften so imperfect that close comparison with stone 
or potterj' heads from St. Vincent and other West Indian islands 
is very difficult. They are highly conventionalized productions. 
Some of them have a close likeness to the carapaces of crabs and 
one or two are fish-like in form. Lateral extensions on the borders 
of others recall wings of flying animals. These lateral extensions 
impart a triangular form to the object when seen from above, the 
body being represented by an enlargement with a well-defined head 
with mouth and eyes at one end. A quite distinctive forin is circular 
or oval with lobes separated by depressions extending from center 
to circumference. In many these lobes are regular, but in a few 
one or more of these bodies are absent, making an asymmetrical 
form consisting of a central circular region around which is grouped 
a ring of enlarged bodies. These bodies often do not extend com- 
pletely around the periphery, imparting a more or less U shape to 
the object. The number of lobes varies ; several have seven or eight, 
others a smaller number. 

Several specimens have a form that might well be likened to the 
whorls of a seashell, consisting of a coiled body terminating in a 
spine with an open end such as occurs in the ordinary conch shell 
or other univalve moUusca. Comparisons with animal forms are, 
however, not very close, for the objects are too crude to allow accurate 


The potter}' from St. Vincent, Carriacou, (Grenada, and other 
islands of this culture center is judged largely from fragments; no 
whole specimens occur in known collections from these islands. These 
fragments, however, especially those from Carriacou, one of the 
Grenadas, are characteristic and have distinctive features that readily 
separate them from those of Trinidad, St. Kitts, Porto Rico. ancL 
Santo Domingo. The larger number of pottery fragments from 
St. Vincent belong to the so-called red ware, which there is every 


reason to suppose was painted with red pigment. In numerous in- 
stances this paint disapj^ears if the fragment is moistened. 

Among different forms of pottery indicated by the fragments 
may be mentioned bowls, bottles, cups, griddles (flat plates), vases 
with or without snouts like the so-called monkej' jars or modern 
teapots. As is the case with prehistoric pottery from the other "West 
Indies, there are many forms of effigy vessels and a considerable num- 
ber with rims adorned with animal heads serving as handles or legs. 
There are several fragments of the bases of vessels or basal clay 
rings, which, being more massive than the remaining walls of the 
bowl, remain unbroken with fragments of the bowls. One of these 
is shown in plate 6'2. -1, which, unlike the majority, is decorated with 
an incised zigzag figure. 

While the lips of the majority of these vases and bowls which are 
not decorated with heads are plain or without distinctive markings, 
certain of them are decorated with a succession of impressions evi- 
dently made with the thumb-nail or finger when the clay was soft. 
One fragment (pi. 6i, H) has a series of holes around the rim, 
evidently made with a round stick or bone. Nothing like a roulette 
was found, although there are several clay stamps. By far the 
largest number of bowls have clay heads attached to the rim, placed 
either vertically or horizontally, according to the shape of the 
vessel, the latter being mainly confined to certain griddles or flat 
plates evidently used in frying cassava bread, so placed that the 
cook may handle a hot dish without burning her hands. 

Plate 62. (\ shows a fragment of pottery with a small attached 
snout similar to those found on modern vessels from St. Kitts and 
Nevis. '^ A double bowl, one of the component parts of which is 
unfortunately broken, is illustrated in plate 62, D. This object has 
a flat base, thick walls, and is undecorated. A hole leading from 
one bowl to the other characterizes this object. Its size and general 
appearance suggest a receptacle for paints, and it may possibly have 
been for red dye used in painting the body, as recorded in the early 
accounts of the inhabitants of the Lesser Antilles. 

The relief decoration shown on a fragment of a bowl (pi. 63, A) 
is typical of the style of ceramic decoration constantly employed by 
the prehistoric Antilleans. The fragment represents a section of 
the rim. the ornamentation being on its outer wall. 

The decorated handles of Grenada bowls are more massive than 
those of Trinidad, and are commonly two in number, situated on the 
end of the longer diameter near the rims or the sides. Plate 63, B, 
illustrates the fonii of one of these handles and a fragment of the 
vessel, upon which is the head of some animal, possibly a bird. There 

""The ancient potters* technique still survives in these islands. 


are several similar fragments in the Heye collection. The handle of 
plate 63, C\ is decorated with a head bearing a toothed structure 
that may be likened to a bird's head. The representations of the 
eyes in this head are quite different from those of other birds' 
heads. In plate 64, A, Tre find one of the flat handles of a vessel 
modified into a head of peculiar shape. This figure shows the head as 
seen from above, and from a section of the rim it is evident that the 
vessel to which this head was attached was a flat circular platter of 
clay. This wasi, in fact, the handle of one of those flat dishes or 
griddles on which the natives formerly fried their cassava cakes — a 
cooking utensil now replaced by an iron plate. 

The square, angular head of this specimen is unlike those already 
described, and has a peculiar incised decoration not common in West 
Indian pottery. In plate 64. i?, we have another handle of a vessel, 
from the island of (irenada, which is modified into a head. Although 
this sisecimen is rather roughly made, tlie outlines of mouth, eyes, and 
nose are well indicated by incised lines. Plate 65, .1. i-epresents 
another handle of a cooking dish like a head, seen from above. The 
presence of eyes and mouth and a circle commonly found on the 
middle of the foreheads of animals, which is one of its most striking 
features, leads to the identification of this as a head of an animal. 
This head is surrounded by a row of holes which may have been 
either for insertion of feathers or purely decorative. 

In i^late 65. /?, is shown a comparatively large fragment of the 
rim of a vessel, in which the face or head has well-made eyes, nose, 
and mouth. The meaning of an appendage on each side of the head 
of this specimen is not easy to interpret. These appendages occur 
in pairs — one pair on the forehead, the other in the position of the 
ears. It is not unlikely that they represent the two arms or legs of an 
animal form, or are appendages of a human being. 

Plate 65, C\ represents the handle or lug of a flat cooking dish, evi- 
dently a griddle. 

Various forms of handles modified into heads are shown in plate 
66, A, B, C. They are parts of griddles or clay disks used in cook- 
ing, and while differing in details appear to represent like animals. 
The effort made by the artist to enlarge the size of the handles by 
an extension of the tops of the head or prolongation of the snout 
imparts a ludicrous expression to the faces of two of these figures. 
Both of these are large fragments of flat, shallow vessels to which 
they were formerly attached. 

A unique form of pottery rest with a face on one side, and various 
other relief designs, appears in figure 11. 

The burnt clay head, plate 66, 6', evidently broken from a bowl or 
other piece of pottery the shape of which is unknown, has a project- 



,4 , 3.4 inches; D, 4.25 inches. 










.4, 5.4-1 inches; D, '■'• inches. 






A , 7 inches; C, 5.31 inches. 



G, iaches; /, S inches; J, 0.94 inch; K, 1 inch. 



^,1.31 inches; C, 1 inch; fl, 2.25 Inches; E, I.u inches; J?, 2.19 inches; G, l.SS inches. 




/I, 4. 75 inches; £,5.5 inches: C. e. 88 inches; i),6inches; £, G.25 inches; /•', 0.5 inclies. 




ing snout and prominent eye. The head, plate 66, D, resembles that 
of a turtle and is more angular than the head of a bird. The upper 
jaw, which resembles a beak, extends beyond the lower and temiinates 
in nostrils. The exceptional feature of this head is a row of holes 
at its attachment to the body, i-epresenting the flat dish of which it 
was one of the handles. 

Plate 67, A, is likewise identified as the head of a turtle. It has 
an enlargement liack of the head near its attachment, in which are 
pits corresponding to the row of holes described in plate 66, D. 

Plate 67, 5, represents a head with a pointed snout, resembling the 
wild hog, or peccary. The nostrils have the form of simple projec- 
tions and the ears are indicated by raised ridges near the perforated 
eyes. In the mid- 
dle of the forehead 
there is a rounded 
elevation with a 
central pit — a fea- 
ture recalling that 
found in three- 
pointed stones rep- 
resenting reptiles. 

The head, plate 
67, C\ is more or 
less simian, and 
its form recalls 
those from Gre- 
nada figured by the 
author in his "Ab- 
origines of Porto 
Rico."'° A re- 
markable feature 
of this head is the pair of disk-shaped appendages just above the 
level of the ej'es. The same or similar organ is shown also in plate 
67, J?, where the monkey fac€ is even more striking than in C. Plate 
67, D, has the appendage between the eyes in tiie fonn of a low cone 
or horn, while the highly conventionalized form, plate 67, F, defies 

One of the largest and best-preserved fragments of Grenada pot- 
tery, much reduced in size, is shown in plate 68, A. The patches of 
pigment with which this object was painted can still be seen adher- 
ing to its surface. This illustration represents the top of a jar the 
sides of which are missing, but the flat, heart-shaped base from 
which the head rises is a cover to the neck of a flask or vase with 

cylinder with 

relief. (6.13 

'"Twenty-fifth Ann. Rept. Bur. Amer. Ethn. 


narrow orifice. The details of the head are evident from the ilhis- 
tration. but an unusual feature is the connectin<j band whose outlines 
inclose the pit representing the eye and the opening of the nostrils. 
The use of a vessel of the form indicated by this fragment is not 
apparent, and the animalistic head is unlike that of any other known 
to the author. 

The fragment, plate 68, B, shows a handle and part of a spoon 
collected in the Grenadines. Many fragmentary specimens of like 
objects occur in the collection. It would appear that the frag- 
ment, plate 68, f, is part of a cylindrical vessel pointed at one end 
and probably modified into a head at the opposite end ; but whether 
this vase was an effigy vessel or not is not clear. The part figured is 
exceptional in form and unlike any known bowl from the West 


The perforated stones for pendants found in St. Vincent and 
Grenada closely resemble those from the island of (Guadeloupe, but 
are as a rule somewhat better made. A number of these pendants 
are in the Heye Museum and were collected by Rev. Mr. Huckerby, 
of St. Vincent. 

In their simplest form they are stone nodules perforated near the 
border, as shown in plate 69, A^B, C. These differ mainly in form, 
the last mentioned, plate 69, f , being shaped like a human kidney. 
Plate 69, B, represents a spherical pendant and A a pendant of ir- 
regular shape, the perforations in both instances being angular and 
not beveled. 

In the specimen illustrated in plate 69, Z>, we find a pendant in 
the form of an irregular nodule, the perforation being beveled on 
each end. There is no question that the hole was for suspension by 
a cord. This nodule is well polished, except at one point where its 
surface shows marks of pecking. The surface of the heart-shaped 
pendant, plate 69. E^ is crossed by grooves spirally wound about it, 
the perforation having beveling on each end. Plate 69, F, represents 
a pendant, rectangular in form, made of greenstone, its outlines re- 
calling those of an ungrooved ax. The perforation is minute, not 
beveled. The stone figures, plate 69, G, H, /, are made of slate, 
pointed at each end, rounded on one surface, and flattened on op- 
posite margins. The perforations, wlien present {G, /), are situated 
near the edge about midway in its length. In plate 69, H, we have 
representations of crescentic stones, so allied to the former that they 
appear to connect them with more aberrant types. 

One pendant, plate 69, «/, in the St. Vincent collection was at- 
tached to a necklace or other foreign body by a groove at one pole, 
Avhich is pointed. 


• A somewhat larger stone pendant is perforated at a point about 
one-third its length. This object, shown in plate 69, Z, is flat on 
the surfaces and is rounded at one end, which recalls the cutting edge 
of an ax. The other extremity tapers, narrowing gradually to a 
blunt point. 


The three shell objects, plate 69, M. X, O, collected in St. 
Vincent, are made from the tip of the giant conch shell, and were 
probably implements, possibly used by the aboriginal potters. Each 
is prolonged into a handle at one end, while the opposite end is 
flattened, imparting to it the form of a spatula. Similar specimens 
have been collected by the author in Porto Rico, and others have been 
seen by him in collections from Barbados, where, from necessity, 
implements of shell replace those made of stone of an almost identical 


The common St. Vincent form of stamps for printing are disks 
with short handles, or a cylinder with superficial incised figures. 
Three of the four with designs upon them are here figured (pi. 70, 
A, B^ C). A unique figure representing the human face is shown 
in A; the geometric designs, B and <^\ are more common. This type 
of pottery stamp is rare in the Porto Rico area, where it is replaced 
by a cylinder, several specimens of which have been described by 


There are several small perforated stones and terra-cotta disks of 
the same form, collected at Grenada and St. Vincent, in the Heye 
collection. One of these, plate 70, Z>, is made of stone and un- 
deco'rated. The two opposite surfaces of this specimen are flat and 
its margins square. The specimen, plate 70, £", is conical in shape, 
its base being flat and the sides sloping gradually to the frustum, 
which is almost wholly occupied by the perforation. 

The perforated stone shown in plate 70. F, is double, the conical 
sides imparting to it a resemblance to spindle whorls from prehistoric 
Mexico. A similar shape appears in an unperforated stone, plate 70, 
G, which also has the lenticular form reproduced in plate 70, 77. 

These perforated disks with convex or conical sides are identified 
as spindle whorls, but the use of the lenticular stones is not certainly 


The author remained a few hours in Dominica and saw one or two 
collections from the islands, but is unable to add any information 


to what is known of shell mounds or middens on this island. There 
are opportunities to study the ethnology of the Carib who still 
survive in Dominica and still speak their aboriginal language. As 
historical accounts of the ancestors of the Carib in Dominica are not 
voluminous, archeological field work among them would be amply 

The archeology of Dominica is illustrated by the numerous char- 
acteristic objects purchased from Mr. Huckerby, and there are sev- 
eral collections on the island, the objects in which resemble those 
from the neighboring island of St. Vincent. 

The museum in the library at Roseau. Dominica, has a collection 
loaned by Dr. Nichols, which is one of the most complete examined, 
and contains typical prehistoric specimens from the island. These 
objects are mainly stone implements, which as a rule resemble 
those from neighboring islands, although there are two or three 
specimens which are quite diiferent in form from any others from the 
St. Vincent-Grenada area. Several petaloid celts were seen, but the 
majority of the axes are similar to those so widely distributed 
throughout the West Indies. The majority are peculiar to the chain 
of volcanic islands to which Dominica is geologically related.''^ The 
most characteristic implements are the rubbing and grinding stones. 

Dominica is one of the few islands still inhabited. by a consider- 
able number of the natives who speak Carib and preserve some of 
the old Indian legends and many place names taken from their 
aboriginal tongue. Many stories connected with a lake in the 
middle of the island are survivals of Indian myths, and a large body 
of folklore of like character is current. 

The sacred lake of Dominica is regarded with superstitious dread 
by the natives and has much folklore connected with it, which are 
evidently survivals of old Indian legends. 

Notwithstanding all efforts to crush it out. there still survive in 
some of the Antilles remnants of the old negro culture called the 
voodoo, brought by slaves from Africa. The so-called priests of this 
superstition make use of all prehistoric objects in their rites and are 
said to sprinkle Carib celts and axes with blood in their orgies. It is 
not as complicated as in Haiti, and is generally used by the Obia men 
in laying spells or in medical practices, which the common people 
cling to with great obstinacy. The power of these Obia men on the 
poor blacks is a very serious obstruction to the uplifting of economic 
conditions in all the West Indies. 


" A collection of prehistoric objects at Fort do France, Martinique, is represented to 
contain several exceptional forms, but the author did not have the opportunity to 
examine tbem. 


Stone Implements 


The axes from Dominica are mainly without grooves for hafting. 
but one of those in the museum at Eoseau is girt by a deep furrow, 
and another has deep notches in the margin like those from South 
America. A unique specimen is flat or slightly curved on each 
face, which is smoothly polished, recalling the hatchet found in 
the Erin shell heap at Trinidad, in the mounds at St. Kitts, and 
common in British Guiana (fig. 12). 

Fig. 12. — Ax with marginal notchps. 

Another exceptional stone implement from Dominica, likewise in 
the museum, is rectangular, with the four angles continued into fish- 
tail extensions, one of which is broken. In the middle of one face of 
this specimen there is a i^it, apparently the beginning of a perfora- 
tion. There is another unique specimen in the Dominica Museum in 
the form of a flat stone about 9 inches long. The collection in 
Dominica contains also one stone grinder resembling those from St. 
Kitts, another from Tobago, and a jierforated stone from Nevis. 

In the Nichols collection in the public library of Dominica there 
were two flat stone implements, the remarkable feature of one of 


which was the double crescentic form, united midway in the two 
convex surfaces, the point of union being perforated. 

In a form (pi. 71, ^-1) resembling celts of the northern Antilles, 
except that the body is less tapering, the blad^. when seen in profile is 
lenticular. This is supposed to be the same as that figured by Prof. 
Mason, whose description applies to this object. He says : " A bell- 
shaped blade of brown patina and elongated body. It is difficult to 
conjecture how such a blade could be fastened in a haft. There are 
found in the Antilles frequently implements for smoothing, shaped 
like this specimen inverted. This form with the edge at the small 
end is unique."' '- 

It is difficult to tell to what class of implement Mason refers this 
specimen, but it is here placed, not among grinders but among celts. 

The various objects obtained from Dominica belong to the same 
type as those used by an agricultural people, and probably belonged 
to a race antecedent to the Carib. This points the same way as 
the archeological material from the other Lesser Antilles, and sup- 
ports the theory that the original inhabitants of these islands had 
a kinship with those of Santo Domingo and Porto Eico. 

The existence of an agricultural race allied to the " Carib " in 
the island of Dominica is thus stated by Davies : '^ " The Savages 
of Dominico affirm, that it proceeds hence, that when the Car- 
ribians came to inhabit these Islands they were possess'd by 
a Nation of the Arouagues, whom the}^ absolutely destroy'd, save 
only the Women, whom they married for the re-peopling of the 
Country; so that those Women having retain'd their own Language, 
taught it their Daughters, and brought them to speak as they did; 
which being practised to the pi'esent by the Mothers towards their 
Daughters, their Language came to be different from that of the 
Men in many things. ... To confirm what we have said 
concerning the cause of this difference of Language, it is alleg'd, 
That there is some conformity between the Language of the 
Arouagues who live in the Continent, and that of tlie Caribbian 
Women: But it is to be observ'd, that the Caribbians of the Con- 
tinent, as well Men as Women, speak the same Language, as having 
not corrupted it l)y inter-marriages with strange Women." 

In the following quotations'* from the same author we have the 
legend of the introduction of cassava as the food plant. 

" They say then. That their Ancestors were poor Savages, living like 
Beasts in the midst of the Woods, without Houses or places where 
they might retreat, living on the Herbs and Fruits which the Earth 
produe'd of itself without manuring; whilst they were in this 

'2 ifason, op. cit., p. 753, fig. 24. 

"The History of the Caribby Islands, p. 261. 

" Ibid., p. 2S7. 


miserable condition, an old man among them, extremely weary of 
that brutish kind of life, wept most bitterly, and, orewhelmVl with 
despair, deplor'd his wretched condition ; whereupon a Man all in 
white appear'd to him descending from Heaven, and coming neer, he 
comforted the disconsolate old man, telling him. That he was come 
to assist him and his Countrymen, and to shew them the way to 
lead a more pleasant life for the future; That if any of them had 
sooner made his complaints to Heaven they had been sooner relieved ; 
That on the Seashore there was abundance of sharp Stones, where- 
with they might fell down Trees to make Houses for themselves; 
And, That the Palm and Plantine Trees bore Leaves fit to cover the 
Eoofs of them, and to secure them against the injuries of the Weather ; 
That to assure them of the particular care he had of them, and the 
great affection he bore their species, beyond those of other Creatures, 
he had brought them an excellent Eoot, wherewith they might make 
Bread, and that no Beast should dare to touch it when it was once 
planted ; and that he would have them thenceforward make that their 
ordinary sustenance: The Caribbians add further. That thereupon 
the charitable unknown person broke a stick he had in his hand into 
three or four pieces, and that giving to the old man, he commanded 
him to put them into the ground, assuring him that when he should 
come a while after to dig there, he should find a great Root : and that 
any part of what grew above-ground should have the virtue of pro- 
ducing the same Plant; he afterwards taught him how it was to 
be used, telling him the Root was to be scraped witli a rough and 
spotted Stone, which was to be had at the Seaside; That the juice 
issuing by means of that scraping was to be laid aside as a most tlan- 
gerous poison: and then with the help of fire a kind of savory Bread 
'might be made of it, on which they might live pleasantly enough. 
The old man did what had been enjoin'd him, and at the end of 
nine Moons (as they say) being extreamly desirous to know the suc- 
cess of the Revelation, he went to see the pieces he had j^lanted in 
the ground, and he found that each of them had produced many fair 
and great roots, which he disposed of as he had been commanded: 
Those of Diminico. who tell this story, say further, . . . But in 
regard he went not to look what became of them, till after the expira- 
tion of so long a time, the Manioc continues to this present all that 
time in the ground, before it be fit to make Cassava of." 
The following quotation is likewise from Da vies : '■' 
" This Island [Dominica] is inhabited by the Caribbians, who are 
very numerous in it. They have a long time entertain'd those who 
came to visit them with a story of a vast and monstrous Serpent, 
which had its aboad in that bottom. [Borlin's Lake?] They af- 

" Ibid., p. 15. 


firmed that there was on the head of it a very sparkling stone, like a 
Carbuncle, of inestimable price; That it commonly veil'd that rich 
jewel with a thin movin<T skin, lilce that of a mans eye-lid; but that 
when it went to tlrink, or sported himself in the midst of that deep 
bottom, he fully discover'd it, and that the rocks and all about re- 
ceiv'd a wonderful lustre from the fire issuing out of that precious 

The story of the Man in White from heaven or the sky who 
brought the blessings of a higher culture has a familiar sound in 
American aboriginal mythology, and naturally reminds one of the 
Quetzalcoatl mjiih of the neighboring Central America. He brought 
them the gift of the national food plant, the yuca, and was no doubt 
worshipped on that account. 

Not less aboriginal is the storj^ of the (h-eat Serpent who liAed in 
the bottom of the lake and wore the sparkling stone in his head which 
lighted uj:) the cliffs. This might well be a myth of the Sky God, 
who was worshipped in the form of a snake. 


Less is Imown of the archeology of Martinique than of any other 
West Indian island. The specimens from that island in different 
collections differ so little from those of Santa Lucia and Dominica 
that the island is put in the same culture area, although its relation- 
ship to Dominica and Guadeloupe is closer than to St. Vincent and 
the (irrenadines. 

There is a small public collection at Fort de France which was not 
visited by the author. The great volcanic activity of Mont Pelee 
shortly before the author's visit had about depopulated all the 
northern region of the island. 


The most aberrant island, archeologically speaking, of the St. 
Vincent culture area, is Guadeloupe, from which there are few, if 
any, specimens in the Heye collection. In order, however, to give an 
adequate idea of the objects from this area the author made new 
studies of the famous (xuesde collection now in the Berlin Museum 
fiir Volkerkunde and has embodied the results of his work in the 
following pages. It was found necessary to prepare new illustra- 
tions'" of the Guesde objects, which are also introduced into this 

" These illustrations were prepared by Mr. von den Steinen with permission of the 
director of the museum. 


GuESDE Collection 

The most important collection from the Lesser Antilles, next in 
size to that owned by Mr. Heye, is the celebrated Guesde collection 
made many years ago in Guadeloupe and now in the Berlin Museum 
fiir Vulkerkunde. This collection was brought to the attention of 
the scientific world by the late Prof. Otis T. Mason, whose account of 
it will remain an important pioneer work on the antiquity of the 
Lesser Antilles, as that on the Latimer collection is the first of im- 
portance on the antiquities of Porto Eico. There are about 600 speci- 
mens of the Guesde collection in Berlin, of which about a third were 
described by Prof. Mason. 

Mason thus refers to the nature of the material used in the prep- 
aration of his monograph on the Guesde collection: "The edi- 
tor of this monograph sincerely regrets that he has not the 
specimens before him; but it was impossible to transport with 
safety so many valuable objects to Washington, and equally impos- 
sible for the editor to make the journey to Guadeloupe. Fortunately 
M. Guesde has painted in water colors, with scrupidous care, all of 
the examples figured, preserving both the color and the size. The 
omission of the thickness would somewhat mar the description in 
many cases were we not familiar with the two typical forms of 
blades so frequently figured here."" Many of these specimens are 
described in the following pages with objects in the Heye Museum ; 
the general type of antiquities from this- island are so closely related 
to those of St. Vincent that they are included in the same area. 

The following quotation from M. Guesde, who made the Guesde 
collection, is copied from Mason ^"^ and gives a good idea of the nature 
and variety of objects found at Guadeloupe. 

" From my youth I have alwaj's been deeply impressed with what 
I have read about the Caribs. The sight of the stone objects which 
once belonged to these primitive inhabitants of the Antilles produced 
an indescribable impression on me. 

"As 3'ears went by the stronger became my desire to collect together 
all that the soil of (niadeloupe might contain relating to the Caribs. 

" I accordingly went to work in the year 1866, and after 18 years 
of constant research, never allowing myself to be discouraged by 
any difficulty, I have the satisfaction of being able to exhibit to 
ethnologists this collection, which I believe to be more complete than 
all others now existing, in Paris as well as in America. 

"My collection includes roughly worked stones indicating an in- 
dustry in its infancy ; and others, on the contrarj^ which are brought 

"Mason, op. cit., p. 732. The closing Unes account for tbe fact that several rubbing 
stones were identified as blades. 
"■ Op. cit., pp. 733-740. 

160658°— 34 ETH— 22 9 


to such a degree of perfection that it would be difficult to improve 
on them, either in design or workmanship. 

" It is necessary to state the fact which permitted John Lubbock 
to class the aboriginal inhabitants of the American islands among the 
neolithic peoples; it is because the stone is always polished. There 
is not a single relic formed solely by being chipped, for those rare 
pieces (axes or chisels) which jjresent such an appearance also have 
the surface very well polished. Besides, these volcanic stones can not 
be worked by chipjiing, like flint, quartz, or obsidian. 

" We come across axes so small that we ask ourselves if they were 
not used by pygmies, and these alongside of others so large and heavy 
that we dream of Titans, and no longer of men like ourselves. 

" In addition to all these relics, which I have gathered from the 
ground in all parts of the colony, both on the seashore and in the 
interior, and at altitudes of from 200 to 900 meters, enormous stones 
covered with strange designs are found, especially in a single quarter 
of Guadeloupe proper. The dimensions of these stones vary con- 
siderably. In some the drawings are so high up that it is difficult to 
reach them ; in others they are near the ground or buried under the 
surface. They are scattered without order about the country and in 
the beds of the rivers. At St. Vincent also, the last refuge of the 
Caribs, stones with inscriptions on them are found in the beds of 

" It is now very difficult to find wrought stones in the ground. 
Here and there the plow or the hoe turns up some occasional frag- 
ments. These stones lie in fact in the arable layer or stratum, and 
this has been so well worked that everything it contained has been 
brought to light. New clearings alone would favor the collector. 
In the deep strata would other things belonging to an earlier race 
be found? In the case of Grande Terre it would be impossible, for 
as soon as we have passed the vegetable mold we reach calcareous 
rocks, Madreporic formations containing numerous fossil shells and 
dogfish, which preclude all idea of the presence of man. It appears 
to me more probable in the case of Guadeloupe, which is of more 
ancient formation, and which must at all times have offered more 
resources to man. 

" However large may be the number and variety of the types 
which I possess, I still consider my work incomplete. It constitutes 
only the prolegomena of what I would wish to accomplish. 

" In the presence of this collection one is led to ask if these wrought 
stones are the work of the Yguirh or of the Caribs, or if they would 
not belong to these two races. We are in almost complete darkness 
on this point. It is necessary to throw some light on the subject. 
This could be done only by visiting all the Lesser Antilles, which 
were already occupied by the Caribs on the arrival of Columbas; 


the Greater Antilles from Porto Eico to Cuba : and Trinidad, which 
is but a fragment recently detached from the continent ; by gathering 
carefully in each island all the wrought stones which would certainly 
be found there ; by studying with the utmost care the inscribed stones; 
by classifying separately the inscriptions and relics according to 
locality; and finally by comparing the whole together in order to 
determine the points of relationship. 

" Having completed this first labor in the (Ireater and Lesser 
Antilles, it would be necessary to collect together the relics from the 
soil of Guiana and, taking them as types, to compare them with those 
of each Antille separately. Then onlj' could we come to some con- 
clusion. We would have laid open to us in fact the now silent his- 
tory of these aboriginal inhabitants. 

" I have been able to obtain some pieces from Porto Eico, as fol- 
lows: 1st. Celts of all sizes, in general well polished, but some with 
a fine brilliant glazing. 2d. A mortar representing a bat — a very 
curious piece which must have recjuired long months of labor. 3d. 
An idol representing a man lying on his belly and supporting a 
mountain on his back. A very remarkable peculiarity is that the 
legs are bent as if in the act of swimming. I think that this idol is 
the personification of some marine deity, protector of an island. 
4th. An enormous necklace, covered with inscriptions on one of its 
lower surfaces. This necklace was evidently slung over the shoulder 
like a hunting horn. ."ith. The lower part of another necklace, but 
without any in.scription. 6th. A small netting needle. 7th. Some 
remains of pottery (heads of men and monkeys modeled with great 
boldness, evidently forming cup handles) and the upper rim of a 
cup which must have been of great diameter. Some of these frag- 
ments of pottery still bear traces of a fine red glazing. 

"I must acknowledge that during two sojourns at Porto Eico — 
one of six and the other of two months — I never came across an ax. 
Moreover, there is not a single ax in the superb collection presented 
to the museum at Washington by Mr. G. Latimer, and which is 
entirely from Porto Eico. The abundance of axes in the Lesser 
Antilles and their complete absence in Porto Eico would seem to 
indicate a difference of race in the inhabitants of these different 

" I have been able to obtain five perfect celts and four fragments 
from Martinique ; one single celt — but very remarkable for form and 
polish — from Dominica ; two celts and three axes from St. Lucia ; 
and one celt from Santo Domingo (the Hispaniola of Columbus). 

" No typical difference can be established between the celts, whether 
they come from Porto Eico or from Martinique, Guadeloupe, Domin- 
ica, and St. Lucia. 


" Now, since the strata of the Lesser Antilles do not contain the 
material used in some of these celts, it is certain that they were 
not made where they were found. Should we not, therefore, infer 
from this that thej' all have the same origin, that they all come 
from the continent or from the Greater Antilles? 

"I have in my possession a club {haton) from the Galibis of 
Dutch Guiana. This club has a certain age. The wood, of a red 
color when freshly cut, has assumed a very deep black hue. The 
cotton thread around the handle is very dirty. The weapon has 
seen service. This club is exactly like those used by the Caribs of 
the islands, and which Father Dutertre has described; but the 
peculiar part of it, the thing that gives it an enormous interest, is 
the green celt fixed in its lower extremity. Now, this celt resembles 
all those which I have found in Guadeloupe and the other islands. 
Is it of modern manufacture? Is it not rather the work of the 
first inhabitants of the continent? Has it not been found in the 
soil and used by its discoverer? I would decide without hesitation 
in favor of the latter hypothesis, for it is covered with a patina, 
which only a long continuance in the soil could give it. 

" Here is another fact which seems to prove that the Caribs of 
Columbus and of Father Dutertre are the same as those of Guiana. 

" The exterior distinguishing color is not always that of the 
stone of which they are made. The color, which is black, red, yellow, 
brown, or bluish, partakes essentially of that of the soil from which 
they were taken. Those from Grande-Terre, whose calcareous soil 
is covered with a thin layer of black and compact vegetable earth, 
all have the colors more or less dark — brown, red, black — while 
those from Guadeloupe proper, whose soil is covered with a thick 
layer of more or less ferruginous red earth, ha^e the tints lighter. 
. Yellow specimens are numerous there. Many of them have preserved 
their normal tint. These are the ones found near rivers. Contin- 
ually washed by their waters, they have not acquired the coating 
of rust with which those buried in the ground are covered. 

" So true is the above that every fresh break shows the interior 
of the stone to be of a different color from the exterior. 

"All these rocks are volcanic, and are naturally either black, 
blue, or green. 

" This peculiarity does not generally exist in polished celts. The 
glazing has unalterably fixed the color of the stone. They have, in 
consequence, remained free from all oxidation, and appear .as if just 
from the hands of the workman. 

'•'■Axes. — Axes are more numerous than all the other pieces. That 
may be readily understood, the ax being of prime utility to man. 
Some are long and narrow, others short and wide; some are very 


flat, others very thick; some are very small, -while others are of 
enormous size and weight. I have two weighing, respectively, 4 
kilograms 750 gi-ams and 4 kilograms 775 grams. Some are of 
very simple construction, merely the natural stone of appropriate 
form, which a little working transformed into an instrument ; while 
others, on the contrary, are true masterpieces, which will bear com- 
parison with those found in Denmark only. The latter are very rare. 
They were evidently used for purposes of parade, for it can not be 
allowed that the author of such a work would have exposed it to be 
broken at the first shock, thus losing the product of the labor of 
several months — I might even say of several years. 

" The ax admits of four distinct parts — the head, the neck, the 
blade, the cutting edge. 

" The head is sometimes round, sometimes flat, sometimes very 
small, sometimes as large as the blade. Some axes have one or 
several transverse grooves, some have none at all, others a single longi- 
tudinal groove. The last are very rare. Pierced axes are very rare. 
The holes served, if they offer any assistance, to fasten the stone to 
the handle; if not, to suspend ornaments. The head played an im- 
portant part in attaching the ax to the handle, for there can be no 
doubt that all these axes had handles. The small as well as the 
large ones were fixed on a wooden handle by means of cords made 
of cotton or mohot. 

" The neck is more or less lengthened. Sometimes it is formed by 
lateral notches only, but generally by a circular depression. 

"The blade varies considerably in form, length, and thickness. 
There is no proportion between it and the other two parts. 

" The edge is more or less distinct. In some axes it is so perfect 
that one would think they had been sharpened the day before. 

" I have three double-edged axes — two of moderate size, the other 
very small. I have four axes of which the head is prolonged into 
a long tail, and which resemble, one of them especially, that which 
has been termed Montezuma's ax. I have also a certain number, 
both small and very large, with a slight but decided protuberance on 
the lower part of one side of the cutting edge, which suggests the idea 
of a tool appropriated to some special purpose. Finally, I have some 
axes with the blade curved like that of a cimeter. These are rare. 

" Celts. — Celts vary much in form, size, and color. Some are 
slender with a sharp point, others are massive with a blunt point; 
some are broad and flat, others narrow and deep; some reach enor- 
mous proportions, while others are very much reduced in size. 

" Celts are scarcer than axes in Guadeloupe. Most of them are 
made of a handsomer, harder material than that used for axes, 
such as serpentine, jade, or jadeite. The fine glazing of the stone^ 


also, is found only in celts. I have some, large and small, made of 
the volcanic stone used ordinarily for axes. These are very well 
polished, but not glazed. This handsome glazing gives an exalted 
idea of the industry of these savages, for it could not be done better 
in our days. 

"The Caribs made use of the living forces of nature to fix the 
celts on the wood. But to introduce a celt into a young tree and 
let the tree grow till the resistance was sufficient, required many 
years. I believe, therefore, that they rarelj' had recourse to this 
process. They evidently followed the same method employed by 
the Canaques and other savages of the present century ignorant of 
the use of metals, whose celts do not differ from those found in our 
islands. This method consisted in fixing the stone by the aid of 
very fine cords in a socket prepared in the wooden handle. 

"I must not forget to mention the shell celts. These are not 
made of living shell, which would not have been hard enough for 
the purpose, but of fossil shell. They are very rare. They were 
extracted from the outer edge of the Stromhus g'lgas, very common 
in the Caribbean Sea. 

" It is to be supposed that the glazed celts were rather warlike 
weapons than instruments of labor, for they offer more resistance in 
proportion to their size ; and we know besides that the savages used 
in war whatever had most value in their eyes. The very large- 
sized celts must have served as wedges in splitting trunks of trees. 

'■^Casse-tetes. — The casse-tete type is furnished by a stone, either 
round or with bilateral facets, in the center of whicli is a more or less 
deep groove for the wooden handle. One can easily conceive the 
power of such a weapon wielded by a muscular arm in hand-to-hand 

" Some are more perfect in form than others. Everyone was free 
to fashion so important a weapon as best suited him. 

" But what astonishes the observer is the small size of one of these 
relics. Evidently it could have been only an amulet, worn with 
the idea of preserving its owner from the blows of the weapon it 

" Other casse-tetes were used without handles. Only two types 
figure in my collection. This weapon had not the value of the 

" Pestles, grinders. — Pestles and grinders are of various forms 
and sizes. My collection includes a certain number of them. I 
possess a single specimen, which was used with both hands. 

'■'■Mortars. — Mortars are not very numerous. This is explained by 
the fact that any hard stone which was flat and smooth would take 
their place. The complete mortar could have been only an article 
of luxury belonging to a cacique. 


" Shall I designate as mortar that rounded concave stone with 
regular grooves descending from the central point to the rim? 
Although quite hollow on its lower surface, I do not think it could 
have been anything but the lid of a large vase, grooved or fluted 
in like manner. In fact, this mortar would have had no fixed posi- 
tion. It could not remain stationary in the position necessary to 
make use of it. Or should we not rather think that the maker of 
this piece wished to represent a miliform cactus so common in the 
Antilles? And in this case should we not rather class it among tlie 
idols ? 

'■'■Dishes. — There are but two dishes in my collection: 1st. A large 
one of rude workmanship. The concavity only is polished; the ex- 
terior rough and very irregular. 2d. A small one of very remarkable 
finish. It is in fact very well polished on all its inner and outer 

^^ Harpoon. — One single harpoon, slightly broken at the three 
extremities. The absent parts can, however, easily be restored in 
following the lines traced on the body of the piece. This instrument 
is very remarkable. 

" Hool-s. — I have two hooks very different in form. Both are a 
little broken, but easy to reconstruct by following the method in- 
dicated above. 

^^Awls. — Awls are rare. My collection includes only two of them, 
but I must state that the material employed is harder than that of 
the ordinary tools and instruments. 

" Chisels. — Chisels are numerous and of various forms and sizes. 
The basil of the cutting edge is perfect. Some of them are made 
of the same material used in the fine celts, and, like the latter, have 
the handsome glazing mentioned above. 

" Yases.- — I have only two vases. One is of guaiacum. The handle 
is perfectly isolated from the body of the vase. This piece is of 
very great interest. As the guaiacum is incorruptible, we need not 
be surprised that it has come down to us. It was found at Bertram 
Creek, the last quarter of Guadeloupe inhabited by the Caribs. Its 
edges are worn and hacked, and bear evidence of having been a 
long time in the earth. I have seen a small tortoise of the same 
wood found in a cave at St. Vincent. 

" The other in my possession is of stone. It is an astonishing piece 
from its general I'egularity and its contour. 

" Shall I class among the vases that small cup with a rather long 
spout? It rather resemliles a spoon, and I think that it might be 
designated as such, taking into consideration the break, which leads 
us to suppose that a prolongation forming a handle formerly existed. 


'"'•Netting needles. — There is one small netting needle, very well 
made and very regular, which evidently served to net cotton, and 
two other larger, more massive ones, which served to prepare cords. 

'''■Idols. — The idols are six in number : 

" 1st. One representing a man extended on his back, the legs bent 
under him, the arms applied to the chest, the head covered with a 
cap, the sexual organs very conspicuous. It is well finished and 
must have cost years of diligent labor. 

"2d. One representing a man on one face and a monkey on the 
other, is very interesting. It was found at Matouba. The work 
on this statuette is rude. The hand that made it was wanting in 
skill. But what shall we say of the genius which inspired this 
combination of man and monkey ? Should we not consider Darwin 
only a plagiarist? 

" 3d. Another found in Guadeloupe, of the same type as that froiu 
Porto Rico, but much larger and so rough that it can not be de- 
termined what it represents. The undersurface is slightly concave. 

" 4th. A small granite pyramid, with three grooves or furrows 
on its lower part. It was found on the island of Desirade. 

" 5th. A head with two faces surmounted by a Phrygian cap. This 
head was to be fixed on another stone or a piece of wood forming 
the body of the idol, for it is much too heavy to admit of the sup- 
position that it was carried in the hand. I have vainly searched 
for this complemental lower portion at the place where I fountl the 

" With this last idol we must place an ax and one other piece, 
both having lines identical with those of the idol head. I think they 
represent faces. 

^'■Amulets. — The principal amulet is of carbonate of lime in bladed 
crystallization. It represents a mahoya (evil spirit), with bended 
arms and legs, and the virile organ in a state of action. The 
shoulders are pierced posteriorlj^ to allow of the suspension of the 
amulet. The other amulets ai-e medallions of different sizes, more 
or less round, all pierced with a small hole to admit of suspension. 
I have a single small crescent of stone, an evident representation 
of the caracoli of metal. This cre.scent must have been set in wood, 
unless it was provided with a cotton string terminating at each ex- 
tremity in a small cord for suspension. 

"Z>?'.?^'s or qvoits. — I have six disks, large and small. One espcr 
cially is a very remarkable piece of work. There is no doubt about 
the determination of these relics. The Caribs played quoits. 

'•''Edicul-e. — A small monument having handles on eacli side; on 
top of the handles a platform disappearing under a vault. There is 
a hole in the middle, presumably the ])lace for an idol. This relic 
is very curious, and reminds one of the ^lexican teocalli. 


'■'■Chisels of shell. — Besides the various stone tools my collection in- 
cludes a series of very fine chisels extracted from the outer edge of 
the Sti-ornhus gigas. This part of the shell is very thick and harder 
than stone. It is certain that the Caribs did not use the living 
Strombus, but were careful to take the fossil Stromhi, which had 
in time acquired the hardness of ivory. 

^'■Stone for 7nal:ing aj'es. — I have in my possession a very interest- 
ing stone, which has inscribed on it the use for which it was in- 
tended. It has concavities on three of its surfaces. It is evidently a 
kind of grindstone, on which stones were rubbed in order to shape 

" Since writing the above I have had the good fortune to discover 
in Grande-Terre, in a piece of ground which had not been plowed 
for 60 or 80 years, two tools of flaked flint — a knife and hacking 
Iniife. This discover}' somewhat modifies the theory held to this 
day by writers on America that flaked flint does not exist in the 

" It is very evident, however, that these two flints were not dug 
from the soil of the island and then flaked h\ their possessor, for 
this stone does not exist in Grande-Terre or Guadeloupe in a state 
of nature. 

" These two flaked flints establish, in an irrefutable manner, the 
fact of a migration of men from the valleys of the Orinoco toward 
the islands." 

The account given by Prof. Mason of the Guesde collection was 
based on the excellent figures in an albxmi and not from direct obser- 
A'ation of the specimens themselves. Since this report was pub- 
lished tlie Guesde collection has jjassed into the possession of the 
Berlin Museum fiir Vtilkerkunde, and in order to familiarize him- 
self with the typical features, tlie author visited Berlin in 1913 and 
prepared new drawings,'* some of which may be of assistance to 
future students. 

There are many more specimens in the Guesde collection than 
those thus far described. The collection numbers not far from 600 
specimens, 190 of which are figured by Mason in his monograph. 
M. Guesde apparently did not prepare drawings of some of the most 
typical forms, and some of those which ^Mason figures were not in 
the collection studied by the author. The majority of the specimens, 
however, were examined and their special features noted, and the 
more typical forms, several of which do not occur in any other collec- 
tion, are here illustrated. It is hardly necessary for me to state that 
the Guesde collection is by far the best, as it is the only one of size 
from the island of Guadeloupe. 

" These drawings were made from the objects themselves by Herr Ton den Stelnen, of 



Mason divides objects in the Guesde antiquities into tlie follow- 
ing groups: (1) Unpolished implements; (2) polished blades with- 
out haft-grooves; (3) faces continuous, sides incurved or notched; 
(4) butt distinct, faces not continuous; (5) blades with hooked 
edges; (6) blades with encircling grooves; (7) hammers, grinders, 
and polishers; (8) perforated stones; (0) ornamental forms. 

The author has adopted one or two of these groups, but certain 
features unrecognized as of importance l>y others has led him to 
remodel the existing classification of these AVest Indian objects. 

In the author's classification of Guadeloupe antiquities the first of 
Mason's groups is wholly eliminated and does not appear as a dis- 
tinct type ; his second group practically includes petaloid celts, while 
his third and fourth groups are broken up into minor divisions. 
His fifth group is regarded as comjiosed of two distinct types, blades, 
and grinders, the appropriate place for the latter being his seventh 
or the author's sixth group. 

The stone objects from Guadeloupe studied by the author are here 
considered under the following headings : 

1. Axes with regular margins. 

2. Axes with asymmetrical margins. 

3. Eared axes. 

4. Engraved axes. 

5. Perforated axes. 

6. Anchor axes. 

7. Incised and perforated stones. 

8. Problematical stones. 

9. Mortars. 

10. Pestles, grinders, and hammers. 


The greater number of axes from Guadeloupe belong to this type 
or those stone implements with regular margins. Their form ap- 
proaches closely that of similar axes already described as found in 
great numbers in the islands of St. Vincent and Grenada, to which 
area Guadeloupe is allied in pottery and bone and shell objects. 
Only a very small percentage of petaloid implements occur in Guade- 
loupe, while in Porto Rico the majority belong to this group. 


The three axes, plate 71, A, 5, C, ascribed to the Guadeloupe area, 
seem more characteristic of the St. Vincent area. 

Plate 71. C, has on one side a rectangular projection and a slightly 
incurved edge on the other. What appears to be a like object is 
represented by Mason in his figure 46 and described as follows: 
"A curiously formed blade of dark color and highly polished. It is 




not altogether unlike figure 45, the chief peculiarity being the pro- 
jection upon the upper side. This characteristic does not appear on 
any other specimen in the collection." '^ 

Several specimens illustrating the asymmetrical type of stone axes 
have been figured by other authors, among whom may be mentioned 
Joyce and Mason, although no one has yet differentiated this type 
from those of more regular forms. The present specimen, plate 71, Z>, 
is one of the most striking and is thus described by Prof. Mason : 

"A grooved blade of dark brown 
color and fine polish. The butt 
wedge-shaped and rounded. The 
hafting space is a complex affair, con- 
sisting of four parts, two narrow- 
faced grooves, a groove on the lower 
side a little wider, and a long, wide 
notch on the upper. The section of 
the groove is rectangular. The same 
idea of a shoulder on one side of the 
blade may be studied in a specimen 
from Mennithorpe, Yorkshire, Eng- 
land. This latter one, however, is 
very rude and far behind the Guesde's 
example. (Evans' 'Ancient Stone 
Implements,' fig. ft2.) This blade lashed to a shouldered handle 
would be a very effective tool or weapon. From Marie-Golante. 
Length, 6 inches ; width, 2j% inches." '» 

Plate 71, E. F, are remarkable objects. The former has the head 
ornamented with an 8-shaped design; the latter, from St. Vincent, 
has an incised design on the surface. 


13. — Ear-ahaped 

blade. (3.6 


The simplest form of eared axes from Guadeloupe are not unlike 
many from the St. Vincent area. One of these is shown in figure 13. 
The edges of these could hardly be called cutting edges, and the 
implement may be unfinished. 

The following quotation" gives Prof. Mason's comments on this 
specimen : 

" A hoe-shaped blade, of the double-beaked variety and light, mar- 
ble color. The beaks are reduced to the simplest form and divided 
by an emarginate curve. The lateral notches are not separated from 
the other parts, their lines being continuous from beak to beak. The 
highly polished and finished condition of this specimen separate it 
from the agricultural class, although its shape is that of the planta- 
tion hoe. A similar but clumsier butt is seen in Im Thurn's volume 

" Mason, op. cit., p. 762. «> Mason, op. clt., p. 793. " op, clt., p. 755, fig. 31. 



[ETII. ANN. 3* 


specimen was 


(Timehri III. pi. VII, fig. 2). His blade, also, is nearly rectanini- 
lar. Width. 3^% inches; width of neck, 1^% inches." 

The larjre ax belonging to the (iiiesde collection, now in Berlin, 
is a fine implement, and. although unfortunately broken, might well 
be regarded the most jierfect known eared ax. Two unique features 
distinguish this ax — viz., the decoration on one surface and the pro- 
jections on the margins. One of these appendages has been broken, 
and its former shape is indicated by a dotted line (see fig. 22). The 
other, still intact, is perforated by a minute hole which we have every 
reason to believe was also present in the projection on the opposite 

figured and described by Prof. ]\Iason.*- 
It appears that the e 1 a b - 
?;-"°^, orately engraved figure on the 

surface of the blade was over- 
looked by him, as in fact it 
was by others, and was only 
detected by the author after 
moistening with a damp rag, 
when the engraved design, 
which is more or less frag- 
mentary, was brought to light. 
It is evident that the surface 
of the blade is much eroded 
to have so effectually ob- 
scured the engraving upon the 

There are several specimens 
in diiferent museimis which re- 
::::* semble that here figured (fig. 
>'' 14). The nearest approach to 
it, possiblj' the same speci- 
^^•* men, is thus described ^' by 
Prof. Mason: "This beautiful 
blade, up to whose form the last few specimens have been 
leading us, is of a dark-green color, and presents some interesting 
characteristics. The butt resembles two eagle heads facing outward. 
The long haft-space or neck widens gi-acefully outward to where it is 
joined to the sides by abrupt shoulders. The faces are highly pol- 
ished and continuous over the entire specimen. The lower side of the 
edge has been broken and reground. Length, 9^ inches; greatest 
width of blade, 4i'v inches: greatest width of haft-space. 2yV inches." 
The ax next considered (fig. 15) is referred to by Prof. Mason ^* 
in the following quotation: "A two-beaked blade of blackish-drab 

" Op. (-it., p. 770. flc. 76. "Op. oit., p. 758. fig. 37. " Op. cit., p. 798, fig. 128. 




'ft,-.:;- " 

Fig. 14. — Eared stone implement. 




color, and perfectly smooth. The lines 
of this specimen are everywhere bold and 
graceful. The slender beaks, high crests, 
and other characteristics are verj' taste- 
fully combined. Length, 6^ inches; 
width of edge, 2^^ inches." 

A related specimen (fig. 16) is de- 
scribed by Prof. Mason ^^ in the following 
account: "A massive two-beaked blade of 
mottled, marble-colored stone. The dis- 
tinguishing feature is the ridged, seal-like 
depression between the beaks. Length, 
11]% inches; width, 6t% inches." 

The slight modifications in the course 
of the cutting edges that distinguish 
some of these specimens are not deemed 
important enough to illustrate by sepa- 
rate figures. The same holds 







Fig. 15. — Stone implement with 
two beaks. (5.8 Inches.) 

^??'®'^:-.. y^"- 

gar ding 






16. — Bared stone Implement, 


the relative length and 
breadth. The general characters 
are brought out in Prof. Ma- 
son's description^" of an ax (fig. 
17) similar in all essentials to 
the typical eared variety: "A 
massive and graceful blade of 
dark sooty-brown patina. It is 
in perfect preservation, highly 
polished, and almost perfectly 
symmetrical. The butt has the 
double eagle head, the crests form- 
ing a gradined depression in the 
center. The haft-space or neck 
has nearly parallel sides, con- 
nected with the body by shoulders. 
The sides spread rapidly out- 
ward to meet the broad, finely 
curved edge. Length, 11t% inches; 
width of edge, 7^ inches; top 
of blade, 4 inches; width of 
shank, 3 inches ; width of butt, 5^^ 

s=Op. cit., p. 798, flg. 129. 

""Op. cit., p. 759, flg. 38. 


The ax shown in figure 17, from the Guesde collection, now in the 
Berlin Museum, may be considered one of the most highly developed 
of the eared axes. This blade has all the essential features of an 
eared ax, but its form is regidarly spatulate and its edges crescentic. 
The margins of the shaft are parallel, the head projections extending 
outward, recurving slightlj' at their extremities. Small rectangular 
elevations are to be noted on the upper edge of the projections, and 
these are separated by a median notch, the walls of which are straight 

and more or less angular. 
This fine symmetrical speci- 
men was probably used ceve- 

:i^"' monially, for it presents no 

M evidence of having been at- 

;,:^ tached to a handle, and the 

blunt edge shows no sign of 



, -•?< 

-v.^ use. ° 

r'^:S^ Only one known eared ax, 




that is here described, has a 
■% perforation on one edge of 

J^ ;• ' , V: ■ ■ - >:^ the blade and in the head 

^^i'./;' '; : , ' : % as well. This specimen (fig. 

^^i'^i^^i-.:.^:^- -.''r- ■. :' , ■-■ ^-^ 18) was first described by 


|P^^?ivVj^Vp- -V :- ! :--im Pi'of. Mason " in the follow- 

fi0i'0^-M'^W-'^:^^''^ly -■.^■■:^^ ing lines: "An elaborate 

^^'^S:^-ii>^^?-- : : 'i'S^ blade of deep brown color. 

fti^-ii^-^^t^y^'S-^-^-:-}:^:'' ■ ' ': \;^ This specimen really belongs 

^; r;/^;:}.vY);:;\ ^ ■ ":" ■ ^M to three of our classes. The 

^'^'^^/•y.'V;. ^ ,-i , \:Xi^ butt is two-beaked and per- 

^;,^;~ "•'/:;, . :/:~0^^ forated, the beaks with long, 

''^*'>iavw. . J,r0^ prominent crests. There is 

■'■>.w^-)-i;&i;:i&4;V;.vr."i-''-!'-'-' j-,q -^^jje extension of these 

Fig. 17.— Massive stone eared Implement. (11-2 beaks, however, and the long, 
inches.) , • 1 i!i 1 

tapering haft-space or neck 
is abruptly shouldered. The body is of the meat-chopper form to be 
seen further on. Its upper side has the countersimk perforation to 
be observed on several specimens in this collection. Length, Gj^jf 
inches ; width, 4^ inches." 

The peculiar serrated edges of the head in this specimen (fig. 19) 
are rightly interpreted by Prof. Mason as ornamental, but in the 
following reference to the object he fails to mention the rarity of this 

"Op. clt.. p. 759. flg. 39. 

'■ .Vttention may he called to the fact that multitudes of similar implements have been 
found in caches, as if hidden in this way intentionally, either for concealment for subse- 
quent use. or for some other purpose. It has been suggested that they are partially 
finished implements, which seems improbable, as most of them are perfectly made and 
Bymmetrlcal in all their parts. 




form of decoration, which, so far as known, is confined to two or 
three specimens. In his account '*'^ of this specimen Prof. Mason 
writes: "A beautifully polished blade of light brown color. It is 
meri-shaped. The butt is gently rounded, bounded by a ridge, curved 
transversely in a ' line of beauty,' and ornamented with nine nuimmi- 
form protuberances. The other elements form one continuous and 
graceful outline, save a slight fracture on the right extremity of the 
edge. Length, 6y% ^^. .^^ .^.^.„. 




••■;> \^^,^ 


'C--y'''-- ' •; 

inches; width, 4i\ 
inches; width of neck. 
2i\; inches." 

A careful examina- 
tion of the specimen 
(fig. 20) supports the 
conclusion of Prof. ^la- 
son as to its beauty 
and polish. His com- 
parison of the head 
to an opera hat is j^ar- 
ticularh' approi:)riate, 
as in this feature it 
stands alone among all 
the axes examined by 
the author. Prof. Ma- 
son's description of ^M:'- 
this object is as fol- #%^ 
lows : ^'' "A finely pol- 
ished blade of brown 
color. This is one of 
the most beautiful 
specimens in the col- 
lection. The butt has 
a bounding ridge very 
prominent, the curved surfaces above and below nearly alike. Two 
gradines above this are carved in the shape of- an opera hat or the 
sheath of the lace palm doubled in and dented on the top. The unlike 
sides are very Avell seen here. Length, 5]^ inches; greatest width. 
4j>5 inches." 

n'Siy-)"^' ■■ 

Fig. 18. — Perforated eared implement. (6.2 inches.) 

»Op. cit., p. 767, fig. 58. 

" Op. cit., p. 772, fig. 65. 



[ETH. ANN. 31 

The term ax is here limited to the implement with poll, shaft, or 
blade, all combined, or with one or the other part absent. The poll 
or head is jjresent in some form or another, sometimes simply blunt 

Fig. 19. — Stone implement with indented head. (6.2 inches.) 

or rounded, convex or concave, often with ear-like lappets or vari- 
ously formed lappets, indentations, or perforations. The shaft or 
shank may be indicated either by incisions on opposite edges or 
by grooves, narrow or broad, sometimes with a ridge on each side. 




The shank is never perforated like the European stone ax for the 
insertion of a handle, altliough, as we shall presently see. it is in 
one instance decorated with an incised figure. The body of the ax 
may be circular; when seen in profile it is elongated, oval, asymmet- 
rical, and sometimes perforated near the edges. The grooved ax is 

Fig. 20. — Finely polished stone Implement. (.5.1 Inches.) 

A'ery rare in collections from the Antilles, there being only a dozen 
specimens in the Hej'e collection. All of these are from the Lesser 
Antilles, the majority from those islands midway between Trinidad 
and Anegada Passage. The few known specimens recall, in form 
160658°— 34 ETH— 22 10 



[ETH. ANN. 34 

and existence of grooves, the ordinary grooved axes of North 
America, but we can not suppose that they are significant enough 
or sufficient in number to modify the theory that the Lesser Antilles 
are archeologically cemented with South rather than with North 

As the petaloid celt is the most abundant implement in the Greater 
Antilles, to which the engraved celt is confined, the ax is typical of 
the Lesser Antilles from Anegada Passage to Trinidad. Petaloid 
celts also occur in the Lesser Antilles, and the axes of the Carib- 
bean islands occur sporadically in the Greater Antilles, but the 
celt and ax are practically limited in their distribution in the way 
indicated. As in the case of the petaloid celt 
we have specimens with heads or figures en- 
graved on one face, so we find engraved axes 
or those in which figures are incised on their 
faces. These figures are. however, geometric, 
and in no instance known to the writer are faces 
or human outlines depicted. We find, however, 
instances of an animal head cut on the poll 
of an ax which seem to be a connecting link be- 
tween the petaloid with the head or face and 
the ax with geometrical design. This group is 
represented by the so-called fish idol from Cuba 
(pi. 93, J.), first figured by Poey. The general 
shape of the body recalls those of several axes 
in the He3'e collection, the only difference being 
in the fish head cut on the pointed end. It will 
be described under engraved celts. 
Figure 21 represents a rare form of ax found in the Guesde collec- 
tion in the Berlin Museum, in which the two opposite surfaces are 
decorated with incised geometric figures which are duplicated. Be- 
ginning with the notch on the margin of the implement they fonn 
graceful curves, recallino; scrolls on a ceremonial baton, which will 
be described later in this article. 

This specimen is described by Prof. Mason, who writes : '^ " A highly 
ornamented specimen, one portion of which is plain, resembling the 
edge of a cleaver; the remainder is covered with ornament. Let us 
imagine this to be a stone ax, the most beautiful in the world. The 
following characteristics claim our attention : The hafting notches 
are extended, that on the upper part liy a narrow gutter almost par- 
allel with the edge; that on the lower part sweeping outward in a 
curve which combines the lower portion and both faces in a continu- 
ous pattern. This is assuredly M. Guesde's jewel in the ax class. 
Length. 5i inches." 

Fiti. 21. — Inci-st'd ax 
from Guadeloupe. 
(5.5 inches.) 

»' Op. cit.. pp. 824-825, flg. 196. 




Some of these engraved axes did not show the incised figiii'es on the 
illustrations used by Prof. Mason in describing the specimens. 


Several of the large ax-shaped stones from Guadeloupe are perfo- 
rated at or near the upper margin of the head (fig. 22). suggesting 
that they were suspended, perhaps ornaments worn about the neck. 

Fig. 22. — Incised decorated stone implement. Gue.sde collec- 
tion. (9.5 inches.) 

The Guesde collection has several of these axes with perforated 
heads. Specimens from St. Vincent are fewer in number. Prof. 
Mason thus describes one of these specimens : ^^ 

" A finely-polished blade, of brown color. The general outline is 
that of a shouldered hoe-blade. The edge is quite regular, the taper- 
ing sides nearly alike, the neck symmetrical, and the faces continuous 
nearly to the perforation. The butt is flared out at the sides like 

«0p. cit., p. 757, fig. 36. 



[ETH. ANN. 34 

a crutch, the concave of which is occupied by a narrow, perforated 
ridpe. With this should be compared a specimen from St. Vincent 
(Timehri, I, p. 264, fig. 3). The latter is more ornamented on the 
upper border, but the body falls far below that of M. Guesde's 
specimen. A splendid example from St. Lucia is also in the collec- 
tion of Mr. Cropper. (Timehri, I, p. 263, fig. 2.) Length, 7.2 
inches; greatest width. 4 inches." 

Figure 23 is the same object as that mentioned by Prof. Mason 
without comment, save that he refers to the " concave grinding " as 
very uncommon. There are only two specimens known to have the 
blade bifurcated: it is doubtful whether we can properly speak of 
this concavity as an edge, and so peculiar is this remarkable specimen 

that we may require a new group for its 
reception. Prof. Mason mentions it as 
follows : °^ "A grooved blade of dark 
brown color. It resembles figure 103, ex- 
cepting that the butt is more distinct and 
the sides divergent. The edge is much 
worn by use, and the concave .grinding 
very uncommon. Length. 4 inches ; width 
of blade, 'dj\ inches." 


There are several stone implements 

*^^^^^ which from their shape are called anchor 

^^^^ axes. They are made of one stone with a 

F.G. 23.-AS with bifurcated ^.^j^tj.^] g^aft modified into curved exten- 
blade. (4 inclies.) 

tensions at one end. The type is char- 
acteristic of the igland of Guadeloupe, the two specimens here figured, 
plate 72, A, B, from the Guesde collection indicate, but a single 
specimen from St. Vincent, not figured, has resemblances to them 
and approaches in form an ax with meat-knife blade closely ap- 
proaching another group. 


There are several stones in the (Juesde collection in the Berlin 
Museum which are characteristic. These stones (pi. 73, A, B, C) 
have an oval or ovate form, with the polar diameter slightly less 
than the equatorial, which sometimes varies, being longer or shorter 
according to the points of measurement. 

The simplest of these (pi. 73, .4) have a shallow pit at one pole 
surrounded by a groove and a somewhat deeper furrow around 

■Op. cit., p. 787, flg. 104. 



A , 5 inches; S, 10 inches. 







B, 3.5 inches. 




the equator. In another figure (pi. 73. (') the simplest form of per- 
forated stone is allied to the engraved stones above mentioned, but 
is perforated by a beveled hole with a furrow at one side extending 
to the border, which is destitute of a marginal groove. 

In plate 73, B, the shape from above is oval, the longer diameter 
being double the shorter. This sjjecimen has on the upper side, turned 
to the observer, two grooves at right angles to each other, the longer 
bifurcating at each end and continuing into scrolls on each side and 
not united with the shorter groove. This object is now in the Berlin 
Museum and foiinerh' belonged to M. Guesde. 

In the specimen illustrated in plate 74. //, we have the highest form 
of the type of perforated stones. The perforation is not beveled and 
the marginal furrow is very deep. 

We find in the Lesser Antilles several perforated stones, some of 
which may be called rings. Scjme of these may be likened to the 
well-known stone collars of Porto Rico. 

The late Prof. Mason's reference^* to one of these specimens is 
short and differs from his figure : "A stone ring of great asymmetry. 
This perforation was probably made by pecking, its faces being 
rubbed down afterwards. Dimensions, 5.2 by 4.7 inches." 

One of the most instructive of the holed stones in the Berlin col- 
lection is one not heretofore figured, shown in the accompanjing 
figure (pi. 75, A). It suggests an unformed implement or itlol. 
Appended to one side there is a projection that reminds one of the 
knot of a small stone collar. This specimen, now in the Berlin 
Museum, belonged to the Guesde collection, and so far as known is 
unique. Its use is unknown. 

A ring (pi. 83, A) from Carriacou in the Heye collection has some 
points in common with that last mentioned, but is made of clay and 
has the representation of the furrow on one face instead of on the 
periphery. The edge of the perforation is in this case rounded in- 
stead of beveled or at right angles. 

The stone ring, plate 75, 5, is beautifully made, and resembles 
somewhat stone rings from the Totonac region in Mexico. 

Plates 74, 6', and 75, C, are mortars, the former boatshaped. 


Among the enigmas in the Guesde collection in the Berlin Museum 
there are several specimens (pi. 76, .4, B, C, D) the use of which is 
problematical, and so far as form goes, they belong to none of the 
types thus far described. 

"Op. cit., p. 817. 


One of these from Guadeloupe is shown in figures 24 and 25. Two 
distinct i-egions may be recognized in the object as seen fi'om above — 
one a circular body, the other an extension on one side forming a 

The body has a slightly curved upper surface and a flat underside, 
the latter so smooth that the function of a grinding implement is 
readily suggested. The smooth, flat surface is continued without 
break into the lower side of the handle, which is also polished, evi- 
dently by long-continued rubbing. 

Although the upper surface of both body and handle is slightly 
convex, the two curves are continuous and bounded by a shallow 
furrow, a short distance from the margin. 

The handle of this j^roblematical implement (fig. 25) is crossed by 
10 parallel straight grooves connecting the marginal furrow. At the 
point from which the handle arises from the body of the implement 
the marginal furrow is enlarged on each side into a deeper circular 
depression, suggesting places where tlie fingers might grasp the ob- 
ject, the palm of the hand resting on the handle. Held in that way, 
the grooves on the handle might enable one who used the implement 
to get a firmer hold in moving it back and forth as a grinding im- 

The only known reference to the object is found in Prof. Mason's 
account of the (niesde collection," and is as follows : 

" A very highly polished implement of dark brown color, and pre- 
senting one of those enigmatic forms that are ever springing upon 
us in the West Indian area. The general outline is that of a ladle. 
Upon the reverse the face is flat, but the broad portion of the obverse 
is slightly concave [sic] and bordered by a molding which is carried 
nearly to the narrow portion. The latter is Ungulate in form and 
has 10 concentric ridges terminating in the border, which is fluted 
externally. There is no duplicate of this form. Length, 12^ inches." 

There are many stone objects in the Guesde collection the use of 
which it is difficult to determine. One of these is shown in plate 79, 
Z>, E. This object is flat, triangular in shape, with two sides slightly 
curved and rounded, the remaining side square with two ears or 
rounded extensions imparting an irregular heart-shape to the object 
when seen from one face. A lateral view shows that the object is 
slightly warped at the point and obscurely convex on one surface, 
flat or concave on the other, but that the warped point prevents the 
stone from being used as a grinder, as neither surface could be made 
to fit flatly on another surface, which a grinding implement would 
imply. The object is smooth, well made, and evidently adapted 

« Op. clt.. fig. 171, p. 815. 


to a purpose now unknown. Its form is. however, so far as known, 
unique, although we have certain approximations, as the circular 
stone obtained by the author from Archbishop IMeriiio and figured 
in his Aborigines of Porto Rico."' 

Fig. 24. — Problematical stone. Guadeloupe. Berlin collection. 
(12.25 inches.) 

No implement is more enigmatical than that shown in the accom- 
panying illustration (pi. 76, A, B) from the Berlin Museum. At 
first sight we might suppose this to represent a new form of ax, and 
the curved cutting edge at one end seems to support such an iden- 

<" Twenty-fifth Ann. Kept. Bur. Amer. Elthn., pi. rrxlii, d. 



[ETH. ANN. 3i 

tification. but the break in one side (pi. 76. B) is unlike any feature 
in the many specimens thus far described. It may. however, be iden- 
tified as an unfinished ax. or one the outlines of which have been 
roughly worked out and then laid away to be polished and finished 
at some later date. 

Many similar instances of unfinished implements are found in 

caches, which lead to the surmise that the 
habit of partially finishing implements 
and putting them aside was not uncom- 
mon in the prehistoric West Indies. 

The object illustrated in plate 76, C, D, 
is from the Guesde collection in the Berlin 
Museum. This implement might more 
logically be classified as a grooved stone 
ax, having a well-marked furrow with en- 
circling ridges on each side. On the blade 
side of the groove this ridge is slightly 
larger than that on the head. 

The beautiful finish of this specimen 
^' (pi. 74, B) did not fail to impress the 
ti late Prof. Mason, and he gives a side 
■g view illustrating the groove on the rim. 
g As the perforation is not countersunk in 
^ this specimen, the author is inclined to 
2 believe that the two views (Mason's figs. 
^ 185-1 and 185-2) ascribed to the same 
I. specimen realh' belong to two different 
^ objects. Prof. Mason's reference °^ to 
g this specimen is as follows: 

"A beautifully finished stone pulley 
[pi. 74, B]. ' The points to be noticed are 
the nearly circular outline, the counter- 
sink perforation, the curved slope of the 
sides, and the groove in the circumference. 
This last feature is unknown to the author 
of these notes in any other stone imple- 
ment. The edge view is enlarged to ex- 
hibit the gi-oove. Diameter, l/^ inches." 
Guadeloupe has furnished several pre- 
historic mortars which as a rule have 


the same form as those from the St. Vincent area and are 
almost identical with others in various islands from Trinidad to 
Cuba. This uniformity is to be expected on account of the simplicity 

« Op. cit., p. 819. 


of these implements. Pestles from different islands differ more than 
mortars, their handles admitting of different ornamentation. 

The simplest form of Guadeloupe mortar '" is a slab of stone with 
a shallow curved depression on onfe side. Examples of this type are 
numerous, and, as they possess no features chai-acteristic of typical 
areas, they need not be considered. 

In the mortar (pi. 74, C) next to be considered, the longer axis 
has been elongated to such an extent as to give the object a 
boat shape. This specimen, now in the Berlin Museum, is said to 
have come from one of the Lesser Antilles, but from which is not 
definitely Icnown. In all these objects the cavity is much smoother 
than the outside surface and shows by the way it is worn signs of 
long-continued use. The size of tiiese mortars jjoints to their use 
in grinding of some cereal, as maize, rather than roots or fruits. 
They do not greatly differ from the forms of mortars found on the 
mainland of South America, where, however, flat slabs of rock were 
often used in grinding corn. It is, of course, not necessary to sup- 
pose that mortars from the Lesser Antilles were used for maize, even 
if their shapes are similar to those where corn was a food cereal. 

In a specimen (pi. 7.5, C) in the Berlin Museum we have a true 
mortar which has a spheroidal form, with a circular cavity of about 
half its longer axis on one side. The side view of this specimen 
differs from that above mentioned in the thickness of its walls, but 
especially in the character of the depression. 

The accompanying representations (figs. 26, 27) show a unique stone 
object from the Guesde collection in the Berlin Museum, concerning 
the use of which several suggestions have been made, but which has 
never been satisfactorily identified. When seen from above tlie form 
reminds one of the skeleton frame of certain cup-shaped sponges in 
which we have a hemispherical knob or attachment to the bottom of 
the sea from which arise walls flaring outward, leaving a deep con- 
cavity and wide-open mouth. The interior wall is smooth, the lips 
rounded, and the outer surface crossed by radiating grooves alternat- 
ing with elevations extending from the knob-like base near which 
they are deepest, becoming shallower as thej' approach the margin. 
Provisionally this object is identified as a mortar. 

The specimen is thus referred to by the late Prof. Mason ^ in his 
account of the Guesde collection: "An unique specimen of light- 
brown color and quite rough. It is hollow like a mortar, but the 
most remarkable feature about it is the series of flutings on tlie 
surface. M. Guesde is of the opinion that it was rather a cover for 
something than a grinding stone. In deference to this opinion it is 
drawn with the broad part downward. Height, 6^ inches." 

" Sovernl of the Guesde mortars have been described by the late Prof. Mason. 
iQp. cit., p. 814. 



[ETH. ANN. 34 


The simplest form of pestle from Guadeloupe is an enlarsed stone 
of a form to fit the hands, but with no attempt at shaping or super- 


Fig. 26. — Unidentified stone object resembling a mortar. (6.5 inches. 

Fig. 27. — Unidentified stone object resembling a mortar. (6.5 inches.) 

ficial ornamentation. Objects of this kind are numerous on all the 
West Indian islands, but the fact that they show no evidence of hav- 
ing been artificially worked has led to their having been overlooked 
by collectors. 








^- rl'- .-i^-'^S** 

•■■■■■■■-■,■'.■ .•,•-•'," ■-•■li-V -. ■-=e;-''i.(* 



C, 1.5 inches; D, 1.19 inches. 


One of the simplest forms of these inipkniients is shown in the fol- 
lowing series of Guadeloupe pestles (pi. 77, A, B). Here we have a 
spherical water-worn nodule evidenth- gathered from some river bed. 
showing depressions on two opposite sides in which the thumb and 
forefinger conveniently fit. This stone might have servetl for two 
jiurposes; it may have been a pounding and a rubbing implement, 
the depressions or pits serving simply to secure a good hold while in 
use. These handholds are generally situated at the ends of a line 
passing through the middle of the object. This object, now in the 
Berlin Musenm, is from the Guesde collection and is described by 

Conical Stones 

Among the peculiar forms of stone objects from the St. Vincent 
area may be mentioned certain conical specimens, in some instances 
like the fourth type of three-pointed stones from Porto Rico. These 
are more like amulets than idols and were possibly used as amulets. 
A conical stone specimen of this type is figured and described in the 
author's Aborigines of Porto Rico.- The general form of these 
stones, as their name indicates, is conical, their longer diameter being 
about twice that of the shorter. When placed on one side, which 
forms a natural flat base, their height is about the same as their 
length. The one feature besides shape that all these objects have in 
common is the groove girting the whole circumference and following 
the rim of the base. 

The several variations in shape of these cones are mainly in the 
breadth of this groove and modifications in the form of the cone, 
some having broad, others narrow grooves about the margin of 
the base, while a number have a serrated apex. In a few a 
ridge separated by deep grooves occurs on the sides of the cone. 
These conical objects (pi. 77. C) are not always made of stone, u 
few specimens existing in local collections being of shell and at 
least one of bone. Xone of these show any indications of a head 
or face cut upon them and are devoid of superficial decoration, aside 
from that above mentioned on the apex. It has been suggested 
that these stone or shell olijects were f(jrmerly attached to a handle 
and served as weapons, but the small size and fragile nature of 
those made of bone and shell would prohibit their use as striking 

The accomj^anying figure {p\. 77, D) shows one of these stones, 
the height of which is somewhat less than the greater diameter, 
as indicated in the figui •. As is common in the majority of speci- 
mens, the curve of the jasal groove is more gradual on the lower 

' Twenty-fifth Ann. Kept. Bu- .\mer. Ethn., p. 98 ; pi. xxiii, I. 


si lie than on tlie upper, which fact, so far as it goes, is opposed to 
the theory that it was lashed to a foreign body. This specimen is 
from the Heye Museum. 

The specimens (pi. 78^ A, B, C) are pestles so fashioned that we 
begin to have u ditferentiation of a disk or enlarged base in order to 
secure a greater grinding surface, and a handle ending in a knob. 
When the disk or base is seen from the side, it appears to be convex, 
flattened at the upper surface. Similar forms (pi. 78, />, E) re- 
sembling meat choppers in profile have been mistaken for weapons 
by Prof. Mason in his account of the Guesde specimens, but the 
indication is that these, like the one considered, should be called 
pestles or grinders. This form of pestle' appears to have been very 
abundant in (iuadeloujDe. 

In another specimen, also from the Guesde collection, the basal 
part or disk is globular when seen from the side and is surrounded 
by an enlargement which may have been carved into a rude figure, 
a character verj' pronounced in Santo Domingo pestles. 

The implement shown in plate 79, .4, in the Berlin Museum, is one 
of the characteristic forms of pestle in the Guesde collection. In this 
si^ecimen the base is massive, and when seen in profile it will be no- 
ticed that the outer upper edge is slightly curved inward, a feature 
that becomes very jDronounced in several specimens. The large size 
of this object suggests that it was used for bruising roots rather than 
for grinding grains, a function which it may share with several other 

The striking feature of the next specimen, also from the Guesde 
collection in Berlin, as shown in plate 79, 5, is the almost abnormal 
enlargement of both base and head, imparting to it a spool shape 
when seen in profile. The twaends of a dumb-bell object have their 
edges turned toward each other so that in profile they have crescentic 
outlines most pronounced in the basal region. This implement, like 
the preceding, was probably used in bruising roots rather than in 
pounding grains of corn. 

Plate 79, C\ is an ax of the characteristic St. Vincent type, with 
groove and extensions on the apex. 

The flattened form of grinders in which the base is extended on 
each side into horns is shown in the next specimen (pi. 79, F). also 
from the Berlin Museum. The object here takes the form of a chop- 
per with blunt edge; the handle is short with termination enlarged. 
This type of pestle is represented by several specimens in the (xuesde 
collection, all of which have as the common feature the extension of 
the base on each side and flat opposite faces. 



-1, 6.25 inches; JS, 5..3 iaches; t\ (i. inches; D, inches E, G inches. 



A , i<.2 inches; B, 4.SH iin.-lifb;; />, '>.2 imlu's; /:, 0.2 inches. 



A , 5.3 inches; B, 6.25 iJiches; C, 3..i inches; D, i.2o inches; E, 4.4 inches. 




.1 , o.s inches; B, 15.5 inches; C, 5.6 inches; E, 5.25 inches. 


A spool-shapetl grinder (pi. 80, .1) is tigured :uid described l)y 
Prof. Mason ^ as follows: "A gi'ooved implement of light-brown 
color. It is introduced here to follow figure 119 on account of sim- 
ilarity in groove. The ax function is lost in that of the smoother or 
rubber. There is a great deal of nice work on this example : indeed, 
as a work of art it is nearly faultless. The furrows of the sides con- 
tinued across the liottom of tlie shaft or neck below give a pleasing 
im^jression. Length. 6^^ inches; width of lower blade, 4i inches." 
• A similar specimen (pi. 80, D) was thus described by Prof. Mason,* 
who regarded it as a hammer. There are several reasons which lead 
the author to class this rather as a pestle, for the term hammer implies 
a handle, which is hardly necessary in the interpretation of this 
object. " 

'•A bell-shaped hanuner of blackish color. The very large curved 
base is not unknown to hammers or pestles outside of the West 
Indies. Several of nearly the same shape may be seen from the 
Haida Indians in the National Museum. The offset on the rim 
below at the base of the neck is unique. (See Timehri, III, pi. 10, 
fig. 19.) Height. 54 inches." 

This implement (pi. 80. E) is one of the few rubbing stones in 
which the handle is somewhat asymmetrical, and while it has some 
resemblances to a pestle, it has none to a rubber or grinder. As 
stated above, like the Kskimo knife, it would "fit the artisan's hand." 
It is figured and described by the late Prof. Mason *" in the accom- 
IDanying quotation. It will be seen on consulting Ids figure that there 
are certain differences between the two which are important, although 
not great enough to lead to any modification in interpretation. 

"A carved rubbing stone, of brown color. The slanting column 
and much-curved base, as well as the lateral flutings, extending 
everywhere except along the bottom, are noteworthy features. The 
Eskimo of Xorton Sound and northward excel in fashioning ivory 
scraper handles to fit the artisan's hand. At Sitka the Tlilinkit In- 
dians also cut out the upper portion of the stone hand-maul to fit 
the hand. Length, 4/;^ inches." 

Plate 81, 4, represents an implement in the Guesde collection in 
Berlin which may have l)een used as a wedge, although it has resem- 
blances to an ax. It has a form not unlike some of the rubbing 

The specimen (pi. 81, 5) is one of the elbow-shaped, asymmet- 
rical forms of grinders, having an extension on one side of the 
base and an elongated handle slightly enlarged at one end. This 
object is in the Berlin Museum, and apparently from the Guesde 
collection, although not figured by Prof. Mason. 

» Op. clt, p. 794, flg. 120. ' Op. cit., p. 804. " Op. cit., p. 807, fig. ITyl . 


The specimen (pi. 81, C) is a unique grinder or pestle, the handle 
of which tapers to a point which is gracefully bent over so that it 
almost touches the flattened top of the base. This object has been 
placed by some authors among problematical stones from the Lesser 
Antilles, but its general resemblance to pestles or pounding imple- 
ments is close enough to lead to its association with that type. As 
an example of prehistoric stone cutting it is unsurpassed in the 
West Indies. 

The unique character and admirable workmanship of this speci- 
men has already been commented upon by the late Prof. Ma- 
son,= whose reference to it is as follows: "The object is entirely 
unique, and indeed outlandish to the Antilles. It is admirable 
in workmanship and has been preserved without a scratch. The 
material is mottled green and brown. It would not be difficult 
to guess, granting this to be genuine, that the process of stone 
carving went on after 1193, the year in which Columbus discovered 
Guadeloupe, and that some ingenious lapidary had undertaken to 
imitate a hook in the tackle. Tliere is nothing improbable in this, 
for the Haida slate carvers to-day imitate steamers and other inven- 
tions of the whites in making their curious pipes. Height, 5^% 

This is as remarkable a specimen as it is rare, so far as known, 
unique and worthy to stand as the best known example of Antillean 
stone working. 

Of the most beautifully made of all the grinders two are repre- 
sented in plate 81, /?, E, of specimens in the Berlin Museum formerly 
in the (iuesde collection. These wonderful examples of stone work- 
ing have the bases prolonged into curved horns at opposite ends of 
their longest diameters. The handles taper uniformly to a head 
slightly enlarged and convex above. The shortest diameter of the 
base is slightly greater than the lower portion of the handle. 

This is certainly a remarkable specimen of Stone Age work, quite 
equal in technique to stone collars or zemis from the Greater Antilles. 


The island of St. Christopher, or, as it is commonly called, St. 
Kitts, was known to the Carib as Luiwa, the fertile land, on account 
of its great fertility. The wealth of archeological evidences indi- 
cates that it was once inhabited by a large aboriginal population 
devoted to agricultural pursuits. 

The archeology of St. Kitts and Nevis" has attracted the attention 
of local students, and is well considered in an instructive article by 

■^Op. cit., p. 809. 

* From the absence of aborigrinal place names of Indian derivation in either .St. Kitts or 
the neighlioring island, Nevis, it appears that the Indians were early exterminated on 
these islands. 


Dr. C. AV. Branch.' Since the publication of that paper Mr. E. Con- 
nell, an old resident of St. Kitts, who furnished much of the data 
to Dr. Branch, has added to his collection many prehistoric objects, 
as yet undescribed, which greatly enlarge our knowledge of the 
archeology of these islands. There are several specimens from St. 
Kitts in the Heye collection and in the insular Carnegie Library 
which are not incorporated in Dr. Branch's excellent article.^ 

The author visited several mounds and other antiquities of this 
island to determine the typical sites where stone and other imple- 
ments were found and visits were made to numerous pictographs 
referred to by Dr. Branch.* He spent some time in studying the 
many objects lately acquired by Mr. Connell and examined another 
small undescribed collection. 

Evidences of the prehistoric character of the culture in St. Kitts 
are afforded by mounds or middens, pictographs, and the very con- 
siderable collection of aboriginal objects in public and private hands, 
all of which indicate a culture (piite different from that uf Porto 
Eico or St. Vincent. As the antiquities from St. Kitts and Xevis are 
practically identical, it is supposed that they belong to the same 
culture area which may likewise include some of the neighboring 
islands, which have been little investigated. 


The middens on St. Kitts visited by the author are as follows: 
(1) West Farm or Two Mile Cut; (2) Stone Fort; (3) Wingfield 

In general character the St. Kitts middens do not differ from 
those of St. Vincent. They are ordinarily situated in cultivated 
fields and their presence can be detected only by fragments of pot- 
tery strewn over the surface of the soil, or, as at AVest Farm, where 
roads have been cut, laying bare a section of the mound. Speci- 
mens of marine shells invariably occur in the sites of these middens 
and deep below the surface as shown by cross sections, but not in 
sufficient quantities to lead us to designate them as shell heaps. 

' Aboriginal Antiquities of St. Kitts and Nevis. Amer. Anthrop., n. s., vol. ix, no. 2, 
pp. 315-333. 

® The majority of the objects of Dr. Branch's collection are now in the National 
Museum at Washington, to which institution he presented them when his article was 

" In addition to his article above referred to, Dr. Branch has published in Nature, 
vol. Liii, p. 580, Apr. 23, 1S96. the following : " Last year in St. Kitts, in a cjiff fresh cut 
by a wash, a gentleman found what were apparently the contents of a Carib grave — 
fragments of pottery, two complete utensils, and pieces of human bones. * • • This is 
the first discovery, so far as I can ascertain, of either bones or pottery in the Leeward 
islands. * * * Since then, however. I have found a kitchen midden, and procured 
plenty of small fragments, along with crab "claws, broken shells, fishbones, etc." 

'"These three localities .ind others are mentioned by Dr. Branch in his paper previously 
quoted. There are also middens at Indian Camp, Cayon, and Brighton Estates that were 
not visited by the author. 



There are several instructive pictographs of aboriginal make in 
the island of St. Kitts, one of the Ijest preserved of which occurs 
on a bowlder in a field at the Wingfield estate. There are others 
on a bowlder on the West Farm estate at Harts Bay, and still others 
in the gorge at Bloody River, near Stone Fort. Many of these pic- 
tographs are simply scratched in the surface of the soft tufaceous 
rock, and are in some instances undoubtedly helped out by modern 
visitors. The pictograph reported from Millets Bay by Dr. Branch 
is on a slab of stone which has been removed by Mr. Connell and' 
placed in his collection. This stone was once used by women as a 

Altar Stone 

Baird" thus describes a singular stone from St. Kitts: "Among 
the memorabilia of St. Kitts I find in my notebook honorable 
mention made of a somewhat singular stone which is to be seen 
almost on the very summit of a remarkable and singularly beauti- 
ful hill, called by the more appropriate than euphonious name 
of Monkey Hill, which hill may be said to form the southern ter- 
mination of the range which traverses the island. Monkey Hill is 
in itself a verdant object, with green and consequently beautiful 
cane fields and brakes extending to its very base; and on the sum- 
mit of it stands tlie large stone referred to, in form and shape some- 
thing like a cradle and having part of the top hollowed out so as 
to give countenance to the legend that it was used by the fierce 
Caribs for the inundation and burning of their human sacrifices." 

The Connell Collection 

The best collection of archeological objects from St. Kitts belongs 
to Mr. Connell, engineer of the central sugar factory, who has 
assiduonslj' collected antiquities from this island for many years. It 
is fortunate that the majority of objects found at St. Kitts drift into 
his hands and have augmented his collection, so that it has now be- 
come the largest in the island and one of the best in the West Indies, 
for he liberally exhibits it to all visitors and allows archeologists 
to study the objects it contains. His collection, like all others from 
the island of St. Kitts, is rich in grinding stones and as a rule poor 
in axes. There are very few winged headed axes like those so com- 
mon in St. Vincent, and a few petaloids, but no three-pointed stones, 
elbow stones, or stone collars. One of the rare forms of implements 
is shown in plate 82, A, B. It is exceptional in having a rectangular 

^* Impressions of the West Indies, pp. 67-68. 


projection on one edge. Another specimen {C\ I>) has projections 
on both edges, the latter resembling a stone object figured by Mason 
from Guadeloupe. Both are rare in West Indian collections. 

The grinder (pi. 82, E) is from St. James Parish, Xevis, and 
belongs to one of the type forms of grinders from St. Kitts. Its 
form is better suited to a rubbing stone than to a pestle such as occurs 
in Santo Domingo."" 

In the Connell collection is a remarkable specimen of stone carv- 
ing from the island, Nevis, similar in technique to a " pillar stone," 
now in the British Museum, described and figured by Joyce, but 
originally from Nevis. This remarkable specimen is a torso, of which 
two human legs and the lower part of the abdomen are carved in relief 
on the surface of a soft stone slab. The feet have been broken off at 
the ankles, and the remainder is more or less mutilated. The knees 
are extended, with several folds of skin indicating muscles on the 
inside of the legs, similar to the specimen in the British Museum. 
The sex is realistically indicated. The carving of the lower abdomen 
and limbs is of a high order of excellence. A remarkable character- 
istic of this specimen, a feature which it shares with the pillar stone 
described by Joyce, is a Medusa-like head, with curled hair cut in 
relief, between the extended knees. The eyes and nose of this head 
are represented, though considerably battered, and there is a kind of 
chaplet on the forehead. Grooves and rows of holes on the cheeks 
and chin suggest that feathers, or some other decoration, were in- 
serted in these holes. The hair is curly, like that of an African 
rather than an Indian, and is represented in relief by scrolls on the 
crown and the ears. The collector of this specimen records that there 
was formerly another fragment of this carving which has disap- 
peared, and may be that described by Joyce in the British Museum. 
The art of these two specimens is practically identical, and if they 
are pillar stones or examples of West Indian prehistoric art, as they 
seem to be, they present one of the best examples of Antillean stone- 
work known to the author. The supposition that they were pillar 
stones seems logical, in which case they may have been set up and 
treated as idols. The treatment of the hair is not typically Ameri- 
can, and it has been suggested that they may have been brought to 
this island from the Old World. 

One of the remarkable and exceptional forms of stone objects 
in the Connell collection, figured by Dr. Branch and represented in 
plate 82, A', is an elongated oval ring, the perforation being in the 

^^ .Toycp, Prehistoric antiquities from ttie Antilles, in the Britlsli Museum. Journ. 
Roy. Anthrop. Inst., vol. xxxFii, pi. Lii, flg. 3. 1917. 

160658°— 34 ETH— 22 11 


form of a slit with beveled sides. One end of this is enlarged into a 
knob, but the surface is not decorated. While it is barely possible 
that this object is a St. Kitts representation of the Porto Rican stone 
collars and may have been used in similar rites, there is no likeness 
to them in the superficial decoration. This is the only specimen of 
its kind that the author has ever seen in the Lesser Antilles. 


The grinders from St. Kitts ma}' be divided into the following 
types: (1) Conical with circular base and pointed apex, base curved 
or flat: (2) conical with oval base and pointed apex, base convex or 
flat; (3) frustum of cone with narrowing sides, top slightly en- 
larged; (4) base circular extending beyond handle, which is conical 
or rounded at apex; (5) base oval, handle narrower than diameter 
of base, rounded apex; (6) apex enlarged into knob, handle slender, 
base larger than apical knob; (7) aj^ex curved to one side, base larger 
than handle; j(8) spool-shaped, handle deeply cut; (9) flat, with 
handle decorated with parallel lines. 

Xone of the many grinding stones in collections from St. Kitts have 
the pestle form with a human, bird, or animal head cut on the end 
of the handle, which is in marked contrast with the known specimens 
of pestles from Santo Domingo and Porto Eico. 

Most of the above types of St. Kitts grinders also occur in the 
Guesde collection from Guadeloupe, and a few specimens are repre- 
sented elsewhere, but nowhere is there a relatively greater number of 
grinders than in St. Kitts and Guadeloupe. One or two conical grind- 
ers resembling those of St. Kitts have been collected in Trinidad. 
Similar implements in the Guesde collection have been described by 
Prof. Mason as axes. There are several stone objects shaped like 
pencils, one of which is not much larger than one's little finger. 
These problematical stones taper uniformly to a point. 

Two forms of grinding stones represented in plate 82, F, G. recall 
forms from Guadeloupe. The former is from Dayfords; the latter 
from the Hermitage. 


Implements and other objects of shell are abundant in the Con- 
nell collection. The lip of the shell. Cassis tuherosus, was often the 
material used for this purpose. Celts, axes, conoids, and raspers 
made of shell are common, but among these specimens tiie Barbados 
shoeiiorn-shaped shell implement is wanting. One type of shell 
implement has the end flat and beveled, and the corrugations of the 



A , 7 irifhes; B, 7 inches; C, i inches; D, 4 inches; F. 4.19 inches; H, 1.56 inches; /. 2 inches; J, 3 inches- 

K, 5.5 inches; L, 2.94 inches. 




B, 2 iDChes. 



B, 2.5 inches: C, :i.44 inches; D, 3 inches: /, 1.13 inches: J, 2.2.5 inches: A", 2.19 inches; L, 1.5 inches. 



A, 1.75 inches; B, 2.06 inches; C, S.06 inches; F, 2 inches; G, 3 inches; //, 3 inches; /, 5.25 Inches. 


lip along the side emphasized by deepening the grooves between them. 
Numerous implements have almost the same shape as petaloid or 
almond-shaped celts. The ordinary Scandinavian type with a blunt 
head is common. 

Cylinders made of shell, clay (pi. 82, /, J), and stone are found in 
the Connell collection. A shell cylinder shown has its surface in- 
cised with dots and lines forming a design (pi. S3. B) of unknown 
meaning. The same design appears on cylinders of clay (pi. 82, H), 
a broken fragment of which is in the Connell collection. 

Plate S3, ^1, represents a ring made of burnt claj'. 

The shell ornament (pi. 84, A) from Stone Fort is rectangular in 
shape and has three parallel marks connected with a transverse 
groove incised at each end. This unique form has been provisionally 
interpreted as an ornament, but its true meaning may later be shown 
to be a much different character. 

The shell object (pi. 8-4. B) is provisionally called a spoon on 
accoimt of the concavity on one side. It is convex on one side, con- 
cave on the opposite, and perforated at each end. 

The shell object shown in plate 84, (', is of unknown use, but the 
indications are that it was used as an ornament. Other similar forms 
Avith additional perforations also occur in the Connell collection. 

Ornaments of shell and stone of an elongated pendant form are 
niunerous. Some of the varieties of these are shown in plate 84, 
£', F. Figure D has a circular form and is perforated. In one 
or two instances these have a groove cut around one end. 

The object, plate 84, G, shown laterally and from above, calls to 
mind forms of labrets. It is ovate with medial groove, smooth, and 
of relatively snuill size. 

The conical objects like the accompanying figures (pis. 77, 6*, Z?, 
84, H^I^L) are of unknown meaning. These are specimens of stone 
and shell, one of the latter (/) having a pointed crest, which is ser- 


The pottery of St. Kitts is among the finest in the Lesser An- 
tilles and is commonly red or red and white, generally with incised 
decorations. The Connell collection contains several rare fonns never 
before figured. 

There is a marked resemblance between the St. Kitts pottery and 
that from the St. Vincent-Grenada area, but there is sufficient indi- 
viduality to indicate that the St. Kitts pottery belongs to a subarea 
allied in some particulars to the ceramics of the Greater Antilles. 

^ A specimen of the same type made of stone is figured in the author's Aborigines of 
Porto Rico. Twenty-fifth Ann. Bur. Amer. Ethn., pi. xxiii, I. 


Numerous heads made of earthenware, evidently the handles of 
bowls or jars, occur in the Council collection. That figured here 
(pi. 84, </, A) recalls Grenada ware, the mouth being repi'esented by 
an elevation, the eyes by a dot surrounded by a circle. There is an 
elevated knob on the forehead and a pair of similar knobs on each . 
side of the face. 

One of the most instructive specimens of St. Kitts pottery in the 
Connell collection is a broken jar of bright red ware with superficial, 
indistinct, white figures. This rare specimen (pi. 85, ^i) was found 
on the Brighton estate, Cayon, St. Kitts, and although partly broken 
is one of the most remarkable pieces of pottery from this island. 
The specimen evidently formerlj' had two handles opposite each 
other on the rim, one of which is now broken. The remaining han- 
dle, an animal head, is attached to the rim of the vessel, with mouth 
over the edge and large goggle eyes. From the head there extends 
a slim rounded handle attached midway in the curved side of the 
upper part of the bowl. 

The bowl is mounted on a base flaring below and of somewhat less 
diameter above. The ware is thick. Possibly this was a mortuary 
vessel. This specimen is the best known to me from the island of 
St. Kitts. 

The bowl figured in plate 85, B, found in an excavation in the 
road of the Cunningham estate, is practically of the same pattern as 
the " monkey vases " made by modern negroes of Nevis and sold in 
St. Kitts. It is of rough gray ware and has a snout with handles 
on each side. The base is flat, mouth somewhat constricted, snout 
protuberant, body of the vase enlarged at the equator and sloping to 
base and mouth. 

A platter of thick red ware (pi. 85, C) with a prolongation of 
the rim on one side, decorated on the interior with a double scroll, 
is fragmentary, but enough remains to indicate the general form. 
The bottom is flat, without basal ring, and the lip is slightly re- 
curved. This specimen was found at the botanical station on the 
island of Nevis. 

The pot here shown (pi. 85, D) was dug out of a grave at the Mills 
estate, and is a fine example of an amphora, few of which occur in 
the Imown collections. The lower part tapers uniformly to a flat 
circular base from an angular middle ridge, which narrows to the 
orifice and forms the lip. The two slender handles are attached to 
this part of the vessel. 

The comparatively large food bowl of reddish color shown in 
plate 85, E, has a wide flaring orifice slightly turned back. The 
base of this pot is circular, flat, and its sides slightly bulge midway 


between base and orifice. This bowl was found upside down about 
2 feet below the surface, covering a vertebra of a fish, 11 Hint 
scrapers, and 3 small chisels. 


There are two amulets in the Connell collection representing the 
frog, one of which, shown in plate 85, F, recalls the Prague speci- 
men described on page 234. This specimen is made of a green 
.stone-like jadeite and is perforated through the head. The raised 
ridge on the groove that separates the head from the neck is not seen 
in the cast of the Prague specimen and the eyes are wanting in the 
Connell object. 

An examination of the different forms of amulets that have been 
figured shows no fetish in form of a frog, but as the technique is 
purely Antillean there appears to be no doubt that this specimen 
came from one of the West Indies. From its resemblance to speci- 
mens from Haiti. Santo Domingo, and Cuba the author supposes 
that it came from the Greater Antilles. 

There is in the Public Library collection at St. Kitts a goblet- 
shaped mortar with two handles j^rojecting. one from each side. 
This specimen is unique, but mortars in which the base is not dif- 
ferentiated from the body and in which the handles are absent are 
not rare. 

A flat rectangular stone, with four short stumpj' legs, and a depres- 
sion as if due to grinding on the top, may have been used as a metate 
in grinding maize, yuca root, or cassava bread. This specimen also 
reminds one of seats (duhos) found in Porto Rico and some of the 
Greater Antilles, especially Jamaica. 

An exceptional worked stone object (pi. 85. G. H) in the St. Kitts 
Public ^luseum collection has a spherical or ovate shape with a de- 
pression on one side, imparting to it the appearance of a small 
mortar; but on one side of the wall of this concavity there is a 
projection resembling a handle extending into the depression, which 
is hollowed on each side as here shown. This unique specimen recalls 
somewhat an artificially worked stone in the Connell collection. 

Plate 86, A^ represents a shell object of unknown use. unique in 
West Indian collections, and B^ a shell disk which is perforated. 
C is the spire of a shell perforatect and apparently used as a tinkler. 
D represents the lip of a conch shell with ridges artificially inten- 
sified. The object shown from above, from below, and in section in 
E is made of bone, but its use is not known. F shows two imple- 
ments, possibly made of human bone. 


Plate 87, ^1, is a pestle or grinder of unusual form and use, while 
^ is a crescentic stone gorget perforated midway in its length. 
C and D are two views of a clay object with a resemblance to a 
human head. The remaining figures, A', F ^ G^ //, are made of stone, 
but their use is unknown. 

The few objects from Nevis in the Connell and other collections 
are exceptionally fine, and certain of them so closely resemble those 
found in St. Kitts that we may suppose the prehistoric people of 
this island and those of St. Kitts, only a few miles away, were iden- 
tical in culture. The indications are that there was a considerable 
population on this island in prehistoric times. 


A few pieces of potterj- from St. Croix belong to the Porto Rican 
area, and a few stone collars and three-pointed stones characteristic 
of prehistoric Borinquen have been found on this island. The objects 
from this island are almost identical with those from the other 
Danish Islands, San Juan and St. Thomas. Separated by a wide, 
deep channel from the St. Kitts group and geologically different 
from other Danish Islands, St. Croix shows recent volcanic action. 
The rock formation of the two ends of St. Croix are geologically a 
bluish slate, the area between being of a calcareous formation." 

St. Croix is called Ay- Ay in several of the early accounts of the 
island and is considered Carib, but the stone implements found in 
its territory are so like those of Porto Rico and the typical forms 
so different from the St. Vincent and St. Kitts area that it is con- 
sidered a member of the Borinquen group, and not as belonging to 
the Lesser Antilles, culturally speaking. 

Several private collections made on this island were examined by 
the author, who himself gathered about 100 specimens of petaloid 
celts. In the collection of the governor of the Danish West Indies at 
St. Thomas there are a number of prehistoric objects from St. Croix. 
It appears that the prehistoric population of St. Croix was fairly 
large, judging from the number of known mounds and middens. 
Early historical accounts of the island would lead me to suspect that 
it was smaller than the archeological evidence indicates. 

The inhabitants of St. Croix were valiant, as was evident to Co- 
lumt)us. who had an encounter on his second voyage with the aborigi- 
nes, in which one of his crew was mortally wounded by a poisoned 
arrow.'* The place where this encounter occurred was probably 

1' For g(X)logy seo Quin. Tlie Building of an IsLirid. New York, 1007. 
»* Tlie arrows of tlie natives were apparently tipped witli shai-p sticks or teetli, not 
with stone points. 




.4, ^.liiiRbes; B, 4 inches: C, l..'i iuches; D, .'> niches; E, :t.l:i inches; F. 7.63inclics. 



A , 7 inches; B, 5.5 inches; C, :i.5 inches; D. i-H iiiclies; K, .') inches; F, 4.n>. inches; (J, :>:!:• inclies; //, 5.5 inches. 


near the Salt Eiver settlement, about 7 miles from Cliristiansted. 
Here there was a landing place for canoes. The hostility of the na- 
tives led to the island being designated as a Carib island. 

The Salt River settlement has in the past yielded. many specimens 
of aboriginal artifacts. In recent times it has been examined by 
Dr. Christian Branch and Mr. A. Pinart, who obtained a small 
collection of jjottery fragments and stone objects; a few specimens 
were also collected by the author from a low bluff eroded by the sea 
at this place. The largest collection from this site was made by the 
late Theodoor de Booy.^^ 

The prevalence of petaloids and axes over grinding implements in 
the collections at St. Croix suggests warlike rather than peaceful 
pursuits; and as the island is climatically drj' and not very fertile, 
it may be supposed that agriculture was not as common among the 
natives as in well-watered fertile islands like St. Kitts. Historical 
accounts often speak of the Carib of this island, who made raids 
from it on Porto Rico and other islands of the Greater Antilles. 

Although St. Croix is distinctly called a Carib island, of the 
1,000 specimens from that island examined by the author, not one 
of the wing-shaped axes of St. Vincent was found, from which fact 
it is supposed, on archeological grounds, that the culture of the 
people who originally inhabited St. Croix was different from those 
of St. Vincent. 

Salt River Midden 

The Salt River midden, tlie most important known prehistoric vil- 
lage site in St. Croix, is situated on a slightly elevated point of land 
on the right hand entering the river, and covers a considerable ex- 
tent of the shore at that point ; but as the whole mound is now cov- 
ered with bushes which cover the surface much of its extent is con- 
cealed. The sea has made deep inroads on the edge of the bank, baring 
layers of shells and debris which are nowhere more than 3 feet thick. 
The author has been informed that a number of objects of stone, 
pottery, and other artifacts have been gathered along the beach, and 
that Dr. Christian Branch discovered human skulls in the lower 
layer, but on my visit a few fragments of coarse pottery, broken 
shells, with a few shell chisels, were all that was obtained.^'' 

There are one or two other shell mounds along the coast, one of 
which is an obscure midden near the landing at Christiansted, but it 
is much concealed by modern debris and rank vegetation, and quite 

'^ rirfe Booy. Archeology of the Virgin Islands, Indian Notes and Monographs, vol. 1, 
No. 1, 1919. 

'"This midden was partly excavated by Dr. Branch, and earlier by M. A. Pinart. 
Others have dug in it for supposed buried treasure. 


limited in size. The majority of objects now in collections from St. 
Croix were picked up on the surface, having been brought to light 
in the process of cultivation of the cane fields; but they are said to be 
most abundant on land bordering Salt River. 


The collections, like that obtained by the author, are largely made 
uj) of i^etaloid celts, which varj- from almond to spear shaped forms. 
One or two show shallow encircling grooves or notches on the sides 
for hafting and others are like edged tools of triangular form. The 
form of one stone implement is oblong, flat on one side and rounded 
on the other, with both ends sharpened to cutting edges. A large 
number of implements showing recent breakage were also collected. 
There is a general uniformity in the shape of St. Croix implements. 
the majority being like those of Porto Eico. No grinding tools ap- 
pear in the collections made by the author, but a mortar and pestle 
were seen in a i^rivate home. 

Among so many specimens of almond and leaf shaped implements 
which constitute the majority of objects in collections from St. 
Croix, it was a genuine surprise to find a stone implement which 
showed undoubted influence of Porto Eico. This was a fragment 
of a stone collar, or, to be more accurate, a fragment of the decorated 
panel, which had evidently been put to secondary use as a pestle. 
The surface was decorated with chevron lines, recalling the orna- 
mentation of certain collars from Porto Eico. 

The Carib at the beginning of the historic epoch had submerged 
the Tainan culture of the Virgin Islands and introduced manj^ ob- 
jects peculiar to them, but the earlier culture of St. Croix was 
essentially like that of Porto Eico. 

It is certainly significant that while the majority of stone imple- 
ments from St. Kitts showed marks of having been used for grinders, 
out of about 500 specimens examined by the author from St. Croix 
there was not found a single grinding pestle, and but one small mor- 
tar, too minute to be used for anything but pigment. Although 
situated about 100 miles from Porto Eico, no three-pointed stones 
or collars have yet been found in St. Croix. The collection owned 
by the governor of the Danish West Indies contains, in addition to 
many petaloids of almond shape, one or two other forms, as a 
paddle-like stone that recalls one figured by the author '• and a long 
needle stone pointed at both ends. Both of these are like the Porto 
Eican types. 


The Porto Eico area is separated from the Lesser Antilles by the 
Anegada Passage. The prehistoric culture which extends from this 

" Aborlginos of Porto Rieo, Twenty-fifth Ann. Rept. Bur. Amer. Ethn., pi. xxiii, /. 


passage to Cuba, called Tainan, is afjricnltnral. Geographically it 
includes not onlj^ the large islands, Porto Eico, Santo Domingo, 
Haiti, and eastern Cuba, but also the Danish Islands,^* St. Thomas, 
Santa Cruz, St. John, and a few small islands along the coasts of 
those above mentioned. The general features of this cultural area — 
the highest developed in the West Indies — have been outlined by the 
author in his "Aborigines of Porto Rico and neighboring Islands." "" 
It is essentialh' imlike any of the subgroups of the Lesser Antilles 
and has marked differences from Jamaica and the Bahamas. 

Here developed in prehistoric times the highest culture of the 
Indian race in the West Indies, and. although Arawak, it was related 
to that of the Carib. who had made settlements at certain points on 
the coast, but had not been able to'submerge the preexisting Arawalc 
or overlay the existing culture with their own. 

The Porto Rican cultural center is distinguished by the presence 
of three-pointed idols made of stone, stone collars of unique form, 
elbow stones, and wooden arid stone seats or duhos. The pottery is 
among the finest in the West Indies. Effigi' forms with raised heads 
for handles, incised rectangular lines, with enlargements at their 
.extremities, and encircling lines not joined predominate. In Santo 
Domingo the vessels are generally flask-shaped, often decorated with 
human figures in relief. 

The majority of the stone implements have a petaloid form, 
pointed at one end and flattened at the opposite, not grooved, but 
sometimes with a face engraved on one side. 

The prehistoric inhabitants of the Porto Rican area had cultural 
relations with Central America, but not close enough to indicate 
either an identity or intimate relationship. The decoration of 
stone collars recalls that of the stone yokes of the Totonacs, but the 
differences are so great that, although these resemblances have been 
repeatedly pointed out, the variations in details are important. 
There are also anomalies in the distribution of the stone collars 
which are difficult to explain on the supposition that they are related 
to stone. yokes. Cuba, especially the western extremity which ap- 
proaches very near Yucatan, has yielded no stone collars, and they 
are likewise absent in Jamaica, which lies between the mainland 
and Haiti. No stone collars have been found in the Totonac region 
and no stone j'okes in Porto Rico.^* 

^^ Mr. De Booy. who has lately madp a ronsiderable collection of prehistoric objects 
from shell heaps in the Vir^n Islands, questions the close relations here suggested, and 
may have good grounds for his doubts. The objects belonging to the historic epoch are 
more like Carib ; those of the prehistoric more Tainan or Arawak, like those from Porto Rico. 

"■" Twenty-fifth Ann. Rept. Bur. Amer. Ethn. 

" A stone ring in the Dehesa collection from Vera Cruz is unlike in all but form .1 
Porto Rican stone collar. 


Ca\'es, Shell Heaps, and Ball Courts 

Archeological objects have been found in many localities in Porto 
Eico, but they are especially abundant in the caves with which the 
island is honeycombed, in shell heai)S. and inclosures of rectangu- 
lar shape called batey, cercados de los Indios, or juegos de bola. All 
these sites are numerous and have been repeatedly referred to by pre- 
vious writers. 

The surface characteristics of these places are indicated in Aborigi- 
nes of Porto Rico."" where those of this island were first described 
in English, although known for many previous years through the 
descriptions of Dr. A. Stahl."" The vicinity of the ball courts to 
dwellings is mentioned by Oviedo and several other authors. 

In his pioneer reconnoissance of Porto Eico in 1902—1: the author 
had not the time or means to engage in prolonged intensive work 
of excavation of caves, shell heaps, and ball courts ( juegos de bola) .-"" 
There is much work to be done in this direction and a fair begin- 
ning has already been made. The opportunities are very great. 
Sites of prehistoric settlements are many, and those of historic char- 
acter can easily be identified. As in all the West Indies, the arche- 
ologist has barely begun his work, and much remains to be done 
before the stoi-y of the culture of the Tainan race can be adequately 
made out. One of the most promising islands awaiting the spade of 
the archeologist is Haiti, and it is to be hoped that ere many years 
the antiquities of this island may be explored. 

At the time of my visit a few desultory excavations had been 
made by local students in caves and shell heaps, but it remained for 
an expedition from the New York Academy of Sciences, under the 
direction of Prof. Boas, to pay especial attention to the subject. The 
results of extended excavations have not been fully published, al- 
though notices of the work have appeared in some of our journals. 

There have been several collectors of archeological material in 
Porto Eico since my Aborigines of Porto Eico was written, among 
whom should be mentioned Mr. and Mrs. Lathrop. who have added 
several unique specimens obtained by excavations in shell heaps and 
by purchase to the Peabody Museum at Cambridge. 

Although some light has been thrown on the so-called ball courts 
from an archeological point of view by the excavations of J. A. 
Mason and Haeberlin, they have added no new interpretation to the 
discussion bv Oviedo and "Stahl or to that in the author's Aborigines 

itio Twenty-flfth Ann. Rept. Bur. Amer. ElUn. 

=» Los Indios Borinquenos. Puerto Rico, ISSn. 

»° In Oviedo tliese sites are called h.itey. hut country people in I'orio Uico speak of 
them as Juegos de bola (game of ball). The old name Is still applied to open places 
before large buildings. 


of Porto Rico. The excavations have revealed the form and size of 
these inclosures and uncovered posts indicating buildings mentioned 
by Oviedo. 

The caves and shell heaps have likewise been investigated, but up 
to this time, with the exception of articles by Messrs. Aitken,-' 
Mason,-^ and Dr. Haeberlin,--' the results are not known to the author. 
Several unique objects have been excavated; others have been pur- 
chased from native collectors. 

' The most important result of this work in caves was the discovery 
of bones of extinct animals, and others belonging to modern periods, 
as sunilar bones occur also in shell heaps. The aborigines of the 
West Indies probably used many animals for food, thereby hastening 
their extinction. 

In both Porto Eico and Espanola we find evidences of a cave people 
and settlements of Carib — a vigorous nomadic stock — but the caves 
liad practically become burial places, or ceremonial in their use. Evi- 
dences of the use of caves for these purposes occur all over Porto 
Rico, and, although they have not been thoroughly investigated, 
enough material has been taken from them to show that the cave 
inhabitants had many points in common. 

There is considerable similarity in the artifacts from Porto Rico 
and Santo Domingo, and many differences between them and those 
from the Lesser Antilles, so that as a culture area the Greater An- 
tilles are well differentiated from the Lesser. 

Archeological Specimens 

The specimens in the Heye collection from Porto Rico add much 
to our Iviiowledge of the prehistoric period in Borinquen life, not- 
withstanding there are many duplicates of those alream' elsewhere 

The Heye collection contains several new forms of zemis, stone 
implements, pottery, wooden objects, and those made of bone and 
shell. The stone implements are naturally the most numerous and 
will be first considered. 

The type form of celt from Porto Rico is the petaloid, which, 
wherever found, has the same form and finely polished surface, 
as contrasted with the rough surface of the ax (fig. 28) or paddle-like 
stones of the Lesser Antilles.^* 

=> Porto Rican Burial Caves. Proc. 19th Int. Cong. Amer., pp. 224-228, 1917. 

== Excavation of a new .nrrhooloKioal site in Porto Rico. Proc. 19th Int. Cong. Amer., 
pp. 220-223, 1917. 

''Archeological work in Porto Rico. .\ni. Anthrop., n. s., vol. xix, pp. 214-238, 1917. 

^ For a iliscussion of distriliution of the petaloid and ax form of implement see 
Fewlies. Aboriginal Culture in the Lesser Antilles, Bull. Amer. Geog. Sec, vol. xLvi, no. 9, 






The pottery found by the author in Cueva de los Golondrinos, near 
Manati. fra<rments of which are figured in Aborigines oi Porto Rico, 
is coarse, not unlike that from the Lesser Antilles, and very much 
more like, that from Barbados. As was pointed out. there are like- 
nesses to pottery found in shell heaps of Porto Rico and in other places 
of the island. It was not possible to distinguish this pottery from 
sea caves from that found in open juegos de bola, as no extended col- 
lections had been made from the ball courts, and so far as could be 
judged this difficulty still confronts us. The pottery from the shell 
heaps is now better known than it was a dozen years ago, and it 
Mould appear that the ceramics from shell deposits is quite unlike 
that from some of the caves, as. for instance, the Cueva de los 

Golondrinos on the north shore. It 
is, however, difficult to make this 
statement comprehensive, as it ap- 
pears that caves were used for burial 
purposes, esf)ecialh' in the uplands, 
by a people more recent than those, 
if any, who lived in caves. 

" The present inhabitants of His- 
2^aniola still find the figures of Zemes 
in several j^arts of the island, and it 
is b}' this sign they know where In- 
dian towns formerh' stood, as well 
as bj' certain heaps of shells found 
under ground ; the Indians having 
been ver}' fond of shellfish; and as 
often as this happens, very curious 
discoveries are to be made, bj^ con- 
tinuing to dig a little, in the neigh- 
bourhood of such heaps; for here are generally to be found 
everj'thing this j^eople used; such as earthen vessels, flat earthen 
plates for baking cassava bread, hatchets, and those little plates 
of gold they used to hang to their nostrils, and sometimes to 
their ears; but above all, a considerable quantit}^ of Zemes of eveiy 
f omi.'''' -= 



28. — Outline of ax, shown ifrom 
edge an^side. (3.5 inches.) 

The majority of the stone axes from the Porto Rican area belong 
to a type which, from its resemblance to the petal of a flower, has re- 
ceived the name petaloid. A much closer resemblance in form, how- 
ever, is that to an almond nut, from which fact these implements 

^Jefferys. Katural and Civil History of the French Dominions in North and South 
America, pt. 2, p. 15. j^ 




are often called almond celts. The petaloid form occurs in great 
abundance in Porto Eico, but it is found in all the islands from 
Trinidad to the Bahamas. 

The size of the petaloid celt varies, some of the specimens being 
too small for use as implements, while others are very large. The 
latter may be ceremonial oljjects; the former mascots or charm stones. 

The mode of uniting the petaloid to its handle is indicated by the 
several specimens described by Joyce, Mason, and the author. In 
these specimens the handle and blade are both made of| one stone, 
wliich has suggested the name monolithic petaloid celt. 

Fio. -9. — Brukeu niuiK'litbic celt. (6 inches.) 

Monolithic Petaloid Celts 

There are two monolithic celts (figs. 29, 30) in the Heye collec- 
tion, from both of which fragments have been broken. There are, 
however, a few undescribed specimens of tliis form in otlier museums. 

The specimens of monolithic celts from the mounds of the United 
States, as those discovered by Mr. C. B. Moore at Moundville and 
in Jones' Antiquities of the Southern Indians, are particularly in- 
teresting on account of their variety and Antillean affinities. Those, 
from Central America are also instructive, but as yet no one has 
compared them to those from the West Indian region. 

One of the most common differences between petaloids is the 
variation in the relation between length and breadth or thick- 
ness. Plate 88, .1, shows one of the most unusual modifications in the 
otherwise regular form of these petaloids. The point in this sped- 



[BTH. ANN. 34 

Fig. 30. — Broken monolithic celt. 

men is less acute than is ordinarily the case and the sides enlarge 
as thej' recede from it, forming a cap, which appears to fit over the 
point of the celt as indicated. A cross section of the length of this 
modification of this form of petaloid is evident in the accompanying 

The most important differences in the petaloids are not so much 

in shape as in the form and na- 
ture of the heads, bodies, or faces 
engraved upon them. A special 
group has been made to contain 
petaloids with engraved surfaces, 
called ceremonial celts, several un- 
described S23ecimen.s of which have 
been studied by the author. An- 
other kind of petaloid celt, also 
elevated to a special group called 
monolithic petaloids, has a handle 
and ax in one piece. The celt pre- 
serves its almond shape, but the 
handle may be variously mollified. 
Several specimens of this group, one of which is in the Heye collec- 
tion, have been studied by the author. 

Engraved Cbxts 

The following is quoted at length from the author's account " of 
these implements in the Heye Museum. 

The majority of stone objects known from the larger islands 
are finely polished, while those from the Lesser Antilles are of 
a different foim, with a rough surface. The former are called 
celts; the latter are commonly known as axes. The peculiari- 
ties of these objects found in the AVest Indies indicate that these 
islands formed a sharply defined culture area in prehistoric times. 
Their technique suggests an occupation by man for a consider- 
able period, for it takes many years to develop the culture that 
they express. We find, moreover, that the geographical distribu- 
tion of the two different types of these objects can best be ex- 
plained on the supposition that they belong to two radically differ- 
ent culture regions which can be readily distinguished. These 
two subcultural centers, geographically speaking, are the Greater 
and the Lesser Antilles — the former characterized by the smoothly 
polished celt; the latter by the rough ax, having an enlarged, well- 
developed poll, differentiated from the blade by encircling grooves 

M Engraved Celts from the Antilles. Contributions from the Heye Museum, vol. il, 
no. 3, 1915. 


or marginal notches for the hafting of handles. "While the celt 
has no indication of a head, and the extremity of the blade tapers 
into a point, the head of the majority of axes from the Lesser 
Antilles is often broader or larger than the blade, and is variously 
modified into projections, horns, or bifurcations, very rarely tak- 
ing the form of an animal's head. 

The type form of celt from the Greater Antilles has an almond 
or petal shape, which has suggested the name " petaloid." Although 
the petaloid celt occurs in greater or less abundance " in all the 
AVest Indies, this form is particularly characteristic of the Greater 
Antilles. Implements of this type are generally made of stone, but 
when proper stone does not occur another material, as shell, may also 
be used for the purpose. 

No implement of the polished stone epoch in America can sur- 
pass the i^etaloid in tlie regularity of its form or the beauty cf its 
superficial finish. There is considerable variation in their size, but 
only a slight modification in tlieir outlines, as seen in profile, the 
general almond or petaloid form being constant. 

A cross section of one of these implements, taken midway in 
length, is as a rule oval ; but a few are rectangular, the angles being 
rounded. The greatest breadth of a typical petaloid is near the 
middle. These celts have blunt edges and taper to a point, being 
destitute of grooves.^' 

The petaloid type passes almost imperceptibly into the ax form, 
or those with one end modified into a head and without the pointed 

The petaloid celt figures extensively in the folklore of the modern 
islanders. In Porto Rico and other Spanish West Indies they are 
usually called " piedras del rayo," or thunder-stones, by which name 
thej' are commonly designated in the English islands. They are 
suijposed by the country- people to be endowed with magic powers, 
and are regarded as efficacious in healing diseases. They are like- 
wise supposed to protect the natives fi-om lightning, being fre- 
quently deposited for that purpose under the thatch forming the 
roof of their cabins. In St. Vincent and some islands they are 
placed in earthen jars to keep the drinking water pure and cool.^* 

^ Thp petaloid type occurs as far south as Trinidad, its northern extension being the 
Bahamas. About 90 per cent of all stone implements from the Greater Antilles are 
petaloid celts, and an equal percentage from the Lesser Antilles are axes. 

* It is sometimes stated that grooved axes do not occur in the West Indies, but In the 
Heye collection we find them well represented from St. Vincent and other Lesser Antilles. 
The author has seen very few of this type from the Greater .\ntilles. 

^ There exists considerable follclore and many superstitions connected with the^e 
stones. It is sometimes stated that they are found in trees strucli by lightning, while 
others declare they penetrate the earth and come to the surface in seven years. A true 
thunder-stone may, according to some informants, be determined by binding a thread about 
it and applying a lighted match. If the thread burns the stone is genuine. Several 
specimens bear superficial marlis of having been tested this way. 


They are conspicuous in the equipment of the African voodoo or 
obia men, who are said to employ them in some of their rites. 

The writer has excavated several of these implements in Porto 
Eico caves, shell heaps, or middens, but the majority, counting into 
hundreds, were purchased from the country people, who, finding 
t'hem in the soil while cultivating their " canucos," or small fai-ms, 
preserve them as curiosities for purposes above mentioned. Those 
purchased by the author were, as a rule, slightly nicked by the 
finders, their points or edges being broken under the impression 
that they contain "electricity.'" 

Specimens of Antillean petaloid celts with figures or faces cut 
upon their surfaces are rare. These probably were carried by the 
pointed extremity.^" The most highly ornamented bear a morpho- 
logical likeness to idols, and their forms imply more than the term 
" decorated celt '' would indicate, for some of these are practically 
figurines rather than celts. The step, however, from the incised celt to 
the idol is here, as elsewhere with man in lower stages, so slight that 
nothing can be gained by ascribing one use or name to the engraved 
petaloid and the other to the joetaloid in the form of a figure. ^^ 

The form and symbolism of the petaloids with life figures engraved 
upon them have led the author to the belief that these celts were never 
furnished with handles, but were used sjinbolically as insignia of 
rank, and carried directly in the hand by the pointed ends or inserted 
in staves or wooden sticks for the same purpose. 

Engraved celts have been collected in the Bahamas and the Greater 
Antilles, but have never been found in the Lesser Antilles. The fol- 
lowing list of specimens, named from the museums in which they 
now are, contains the more important variations in this type of celt : 

1. Berlin Museum, No. 1. 

2. Rae specimen. 

3. Berlin Museum, No. 2. 

4. British Museum (4 specimens). 

5. United States National Museum. 

6. Heye Museum. 

7. Museum of the University of Havana. 

8. Copenhagen Museum. 

9. Heye Museum (stone "dirk"). 

10. Blackmore Museum (stone "dirk"). 

""Thp term decorated celt or ceUs with incised decorations might suffice to designate 
the simpler forms, but this name Is inadequate to apply to the highly developed 

*• The monolithic stone celts, or those In which blade and handle are formed of one 
stone, are supposed to have been ceremonial in nature or tn have l)een used in religious 
practices. Incidentally they indicate the way a wooden handle was attached to a smooth 
undecorated celt, and may throw some light on the probable way the celt with a figure 
engraved on one side was used. 


An examination an<l comparison of the above specimens show that 
■while they have considerable variation in form there is a sufficiently 
well marked resemblance to refer all to the same type. In its sim- 
plest form the incised petaloid celt can hardly be distinguished from 
an undecorated petaloid, the main difference being the human face 
cut upon its surface. 

1. Berlin Museum, Xo. 1.^- — This specimen (pi. 88, B) is one of 
the simple forms of the type of ceremonial petaloid celts. Its thick- 
ness is slight, compared with the breadth, the lateral as well as the 
terminal edges being sharp. Its main character is the incised 
circle on one side, inclosing other circles indicating eyes and mouth. 
Below this face, the arms with elbows upward are also indistinctly 
represented by grooves. 

The s]5ecimen came to the Berlin Museum from Dr. Crrosser, Ger^ 
man consul at Plaisance, Haiti. It measures 13^ inches long by 3J 
inches broad and 2 inches thick. 

2. liae specimen. — Mr. Theodoor de Booy has called my attention 
to a similar, almost identical, specimen which was found on a farm 
in Xew Providence. Bahamas, and is now owned by Mr. C. S. Eae, of 
Nassau. ]\Ir. De Booy believes that there is a mention of this speci- 
men in some work on the Bahamas, but says he has never seen it 
figured. To supply this want he has sent me drawings from the 
specimen itself with the size indicated. Its greatest length is lOi 
inches; width, ^i inches; thickness. If inches. The face of this celt 
is 2 by If inches. The circle represents a face, and is confined to one 
side, the opposite surface being smooth. It seems probable that this 
specimen was brought to the Bahamas from the neighboring islands, 
and its close similarity to the one from Haiti above mentioned shows 
a great probability that it was derived from that island. 

3. Berlin Museum., No. 2. — An incised celt (pi. 88, C) in the same 
museum, said to have come from the island of St. Thomas, is one of 
the most beautiful forms of ceremonial petaloid celts known to the 
author. Its technique recalls stone objects from Santo Domingo, 
Porto Rico, and Haiti. "^ The outward form of this petaloid seen in 
profile is well preserved, the figure engraved upon it being well cut 
and unmistakably human. It measures 8 inches in length by 3 inches 
in width. 

One of the illustrations shows this specimen from the face: the 
other from the back.^^ The details of both surfaces are so well 

*' It is with ffri'at pleasure that the author here acknowledges his thanl?s to the 
authorities of the Berlin Museum fiir Vollierkunde for permission to publish figures of 
these petaloids. 

^ Probably it came originally from one of these islands, the locality St. Thomas being 
wrongly ascribed to it. 

" These figures and the preceding were made by W. von den Steinen, of the Berlin 

160658°— 34 ETH— 21' 12 


worked out in the illustration that there is no need of an elaborate 
description. The style of the fitrui-e is thoroughly Antillean throufrh- 
out. and comparable with that so constantly found hi engraved 
figures of human beings from these islands, especially those cut in 
low relief on stone, shell, or bone. This specimen may be regarded 
as an intermediate form between a smooth celt with a face incised 
on the surface and that from the Royal Museum in Copenhagen, to 
be considered later, where the celt form has almost completely 
disappeared. * 

4. British Miosev/m (four specimens). — Four specimens'*" in the 
British Museum preserve a petaloid form, but are not so symmetrical 
nor so well made as some others. They are, however, especially in- 
•structive, since they show an elongation of the pointed end in the 

form of a handle, while the cutting edge still survives above the 
head. The position of these celts in a hypothetical series illus- 
trating the modifications in the type would be nearer the simplest 
than the most complicated forms. This shows connections with 
another series, for from them one can readily pass into a group 
of globular stone objects or heads, with handles — a form without 
indications of a cutting edge and showing no affinity with celts. 

5. National Museum. — The type specimen of engraved or ceremo- 
nial celts, first published in the author's work on the Aborigines 
of Porto Eico,'^ was one of the most instructive specimens of Arch- 
bishop Meriiio's collection and was collected in Santo Domingo. 
The general characteristics, as shown in the figure referred to, are 
a sharpened edge extending over an oval head, with a pointed poll 
below, imparting to it the true petaloid fonn. The hands with their 
fingers drawn up under the chin and the low projection rising be- 
tween them are feeble attempts to represent either the body or its 
appendages. On its reverse side the surface of the implement is 
plain, slightly curved, as is true of all celts on which face or head 
is represented. Tliere can hardly be any good reason to believe 
that this celt ever had a handle, as no signs of such an attachment 
are to be seen, and even if there were the presence of a handle would 
conceal a part of the figure it bears. The specimen is of a greenish 
stone and measures 14 inches in length. 

6. H eye Museum. — Mr. Theodoor de Booy collected in the Bahamas 
a broken ceremonial petaloid celt, which is now in the Heye Mu- 
seum. The edge of this celt and the surface bearing the figure are 
very much mutilated, but enough remains of the body to enable us 
to show its form ^'' and verify his identification. 

"» These have been well figured and described by Joyce, in Journ. Roy, Anthr. Inst., vol. 
XXXVII, pi. I.v, figs. 1-4. 

^ Twenty-fifth Ann. Rept. Bur. Amer. Ethn., pi. xv. 

^This specimen is figured l)y Mr. de Booy in his article on Lucayan Artifiicts in .\mer. 
Antbrop., n. s,, xv, No. 1, p. G. 




A, 2.88 inches; £, 13.5 inches; C, 8.23 inches. 

















T. Museum of the University of Havana. — Aside from a brief field 
note made by the author, there is little information concerning a 
ceremonial petaloid now preserved in the University of Havana. This 
specimen was thus referred to in the author's article on the Pre- 
historic Culture of Cuba f "Among the objects seen in these two 
collections [Academia de Ciencias, and University at Vedado] are 
10 petaloid celts in the Academy museum and about double that 
number at the University. One of those in the latter collection has 
a stone hahdle like those obtained by me in 1903 in Santo Domingo. 
There is also a celt with a face cut on one side — evidently a cere- 
monial celt like one in Ai-chbishop Merino's collection." 

8. Copenhagen Museum. — There is a remarkably engraved petaloid 
celt in the Ethnological Museum of Copenhagen, Denmark. This 
specimen (pi. 89, A,B,C) is of a hard black stone, well made, with its 
surface smoothly polished. It is said to have come originally from 
Santo Domingo, and to have been added to the museum in the year 
1861. Through the kindness of Prof. Sophus MuUer, director of the 
museum, the author has been able to take the photographs from 
which the accompanying illustrations were made. The only other 
figure of this object is published by Eudolph Cronau in his book on 

Certain implements ascribed to prehistoric Santo Domingo find a 
fitting place in our classifications near the ceremonial petaloid celts. 
Tlieir form is somewhat different, but can readily be reduced to 
the same type. The main difference appears to be the absence 
of a cutting edge, the pointed end being prolonged into a long 
pointed blade. These objects with figures carved on their handles 
and long blades suggest stone dirks. Two of these are known to 
the author. 

9. Heye Museum (Stone "Dirk"). — On his expedition to Santo 
Domingo, in the interest of the Heye Museum, Mr. Theodoor de 
Booy collected in that Eepublic a stone dirk (pi. 90, .i, B, C) which, 
while allied to some of the incised ceremonial celts, still shows a form 
quite unlike any previously described. The specimen was formerly 
in the collection of the late Seiior Jose Gabriel Garcia, of Santo 
Domingo City. Its general form, as shown in a photograph 
taken by Mr. Theodoor de Booy, from side, front, and rear, re- 
sembles a dirk, the handle of which is modified into a rude figure of a 
human being, and the pointed end prolonged into a blade. It 
measures 8 inches in length. 

^ Amer. Anthrop., n. s.. vol. vi. No. 5. p. 594. 

» Amerika, vol. i, p. 357. Cronau labels his figure of this oljject, a " hand weapon," 
an identiflcatlon that is not far from correct. 


10. Blachmore Musetim (Stone "Dirk").— Mr. E. T. Stevens'" 
has figured another stone dirk which also came from Santo Domingo. 
This specimen shows signs of secondary work, and the " tool marks " 
upon it, according to Stevens, " have been removed by subsequent 
liolishing." The same author likewise calls attention to a " sketch 
of a somewhat similar weapon " engraved upon a map of Santo 
. Domingo, published in 1731.^" This specimen is described as having 
been found in an Indian sepulcher. 

In considering the morphology of these dirks we may theoretically 
suppose that they present the most highly specialized form of incised 
or ceremonial petaloid celts, but it may be that tl.e part interpreted 
as a blade was iised as a handle by which the object was carried in 
the hand. This handle may have been inserted into a staif, or set in 
a stand, or even planted in the ground. The use of this object is 
problematical, but there is little evidence that it was ever tied to 
a handle midway in its length. 

While the celts above described have in one or two cases lost the 
original form of this implement, their relation to an undecorated 
petaloid is evident and the main characters of the type are preserved. 
The modifications are not difficult to follow. 

The cultural differences in the prehistoric aborigines of the Greater 
and Lesser Antilles are nowhere better shown than in the character- 
istic types of implements. Stone implements of the former islands 
are celts without grooves for hafting, while axes with marginal 
notches, enlarged heads, or encircling grooves are characteristic of 
the latter islands. The celts of the former islands recall Central 
American or North American forms; the axes found in the Lesser 
Antilles are more South American. 

The type called celts, with smooth or decorated surfaces, described 
in the preceding pages, is not found in the chain of islands from 
the Danish Islands south from Anegada Passage to South America ; 
neither have similar petaloids yet been reported from the adjacent 
continent. It may therefore be regarded as characteristic of the' 
Greater Antilles, as are also elbow stones, stone collars, and tri- 
pointed zemis or stone idols. The technique of these objects from the 
West Indies is superior to those found in the southeastern area of the 
United States, but not better than those from the coast of Central 

"" According to Mr. Stevens fFlint Chips, p. 22fi) : " Two similar specimens are In the 
Christy collection ; another is in the collection of Mr. Ilnrtrter M. Wostropp, of Rookhurst, 
Cork ; and another, said to have been found at Ait'ui'perse, near Riom, Puy de Dome, 
France, is in the Clermont Museum." 

*"The following footnote reference to this map appears: "The map is entitled, ' L'Isle 
Espagnole sous le nom Indien d'llayti, ou conime elle ^toit possc-dcc par sea habitants 
naturels lors de la decouverte, avcc les premiers Efublissemens des Espagnols. Par le 
Sr. D'Anville, Geography Ord. du Roi. May, 1731.' The figure of the Implement has been 
copied in the 'Trans. Amer. Ethuo. Soc.,' vol. ill, part I, fig. la." 


America — a fact which, so far as it goes, points to ethnic kinship 
with the latter rather than with the former area. It can also be 
shown that the stone axes of the Lesser Antilles, like the ceremonial 
petaloids above described, bear incised decorations on one side, al- 
though so far as known these incised figures represent geometrical 
designs and not human faces or bodies. 

In connection with these ceremonial celts should be mentioned 
another tyjie reported from Cuba by Senor Andres Poey. who in 
1855 read before the American Ethnological Society a paper entitled 
" Cuban Antiquities ; a Brief Description of Some Relics found in 
the Island of Cuba." According to Dr. Brinton this paper was not 
published in English, but Senor J. L. Garcia printed a Spanish trans- 
lation of it in Volume TV of his " Revista de la Habana." This appar- 
ent contradiction is explained when we know that the edition of the 
publication in which Poey's article appeared was burned, but was 
subsequently reprinted in facsimile in 1909 by the American Eth- 
nological Society.''^ 

This object (pi. 93, ^4), according to Poey, was found by "D. 
JVIiguel Eodriguez-Eerrer in the eastern department of Cuba at a 
place called the Junco, in the jurisdiction of Baracoa. in the interior 
of a wood, at a dejith of 3 feet below the surface of the ground." 

The original account as represented is as follows:^- " [Fig. 2] is 
a correct representation of the second relic and is one-fourth of the 
size of the original. All the figures seen on one side are exactly 
reproduced on the other. These are so admirably executed that I 
am inclined to think they must have been done with a mold; the 
particular reasons for which conclusion are: 1st, That the measure- 
ments in lioth sides are so exactly alike that it seems almost impossible 
to have regulated the work merely by the eye or even by a compass; 
2d, all the figures are executed on both sides in alto-relievo ; 3d, the 
outlines of the figures are perfectly smooth. The stone, which is very 
hard and of a brownish red color, had originally a thick coat of 
varnish and was neatly polished, as is easily seen in those parts where 
the varnish has not been destroyed by friction. A, B is a vein of 
quartz which passes through the stone at an equal distance from the 
circumference. There is a slight groove cut all around it. The stone 
gradually dimini.shes in thickness from the center to the circum- 
ference. It is difficult to conceive what operations can have been 
performed with this implement unless we suppose it to be an ax. 
Were we to look for any animal representation that of a fish would 
most probably be the true one." *'' 

" Trans. Amer. Ethn. Soc. vol. iii, pt. 1, pp. 183-202. 
" Ibid., pp. 187-188. 

'= This object is also mentioned without figuii' in my worlj ■• Prehistoric Culture of 

Cuba," Amer. Anthrop., n. s., vol. vi, p. 589. 


As may be seen in plate 93, ^1, from a photograph by Senor Na- 
cisso Sentenach, procured in Madrid by Prof. Saville, this speci- 
men has an ax form •with a somewhat broken edge and a head in 
relief on the opposite end or poll. Its general form is oval, the 
surfaces plain, slightly convex, its margins rounded. The thickness 
is much less than the length or breadth. 

The unique characteristic of this ax is the head on the poll and its 
relative position resi^ecting the longer axis. The form of the nose, 
mouth, ej'es, and forehead suggests the head of a human being and 
the appendages represented in low relief on each side suggest fingers 
or flippers at the extremities of the arms. The boundary between 
body and head is a curved line near which on the under surface back 
of the chin are indications of folds. The straight line extending 
across the head and body passing through the eyes diagonally is the 
quartz vein in the stone to which reference is made by Poey. The 
resemblance to a fish is not striking. 

In comparing this blade with the ceremonial celts mentioned above, 
it will be noticed that in the latter the head or figure is engraved 
on one side, and the median line of the engraved figure corresponds 
with that of the middle line of the side of the ax, while the middle 
facial line of the Cuban specimen is at right angles to the middle line 
of the blade. This is the only known ax which has the butt modified 
into a head, although there is something similar in monolithic axes 
from Santo Domingo and Haiti. 

■ In the Heye collection there is a fractured ceremonial celt col- 
lected by De Booy in the Bahamas which connects ceremonial celts 
and engraved stones with heads or human figures which have lost 
their likeness to celts. This specimen preserves a likeness to deco- 
rated petaloids as well as to stone images where the petaloid form is 
lost, and may be said to connect these two types and possibly suggest 
a meaning for both. It was found in a negro cabin in Mariguana 
Island. Bahamas.^* and is thus referred to by De Booy, in an 
article. " Lucayan Artifacts from the Bahamas " : *^ 

"Although in fragmentary condition, this object shows clearly 
what the original outlines must have been, and it may be included 
among the best examples of prehistoric stonework from the Ba- 
hamas * * * The celt is petaloid and is made of a green, slate- 
like stone, possibly of volcanic origin." 

Mr. De Booy adds this important information about the object: 
"The figure on the celt is shown in a seated posture and is carved 
in low relief. The knees and arms point inward and the hands rest 

"The character o.f the stone of which It is made shows that It did not come from a 
coral island like Mariguana, and the resemblances in technique and culture ally it to 
Haitian forms. 

•'' Amer. Anthrop., n. s. vol. xv, p. C. 





> \ 


A,B, C, s inches. 







■p7 'Hjb ^^WM^^ 






1 T ' i^'^^TTliW^^^r 









under the chin. The body itself is not shown. The fingers and toes 
are represented by shallow grooves. The head is indicated by a 
carved circle, of which, owing to the fact that the top of the celt 
is missing, not more than half can be seen. However, the right ear 
is still shown outside the circle. The eyes and mouth are cut in 
intaglio ; the nose and the right eyebrow ai'e in low relief." ^"^ 

The general style of the stone cutting is very similar to that of 
an idol now in the Trocadero Museum, Paris, although the disposi- 
tion of the limbs is different and the latter has a large umbilicus, 
which is missing in the Bahama specimen. The outlines of tlie faces 
of both are similar and the details of the carving of the nose almost 

Mr. De Booj^, in the author's opinion, has correctly identified this 
specimen as a ceremonial celt, but it has certain features that impart 
to it an interest ajjart from its resemblance to an engraved petaloid. 
One of these features is the manner in which the hands are brought to 
the body, as the grooves representing fingers are longitudinal instead 
of horizontal. 


The passage from petaloid celts with human faces or heads en- 
graved upon them to stone heads that have lost all resemblance to 
celts bifurcates into two directions, one of which leads to stone heads 
with a prolongation on the back of the head, either above or below or 
in both directions; the other to stones with graven heads or faces, 
but with no projections. The former are called in the author's 
Aborigines of Porto Rico stone heads ; the latter, stone masks. 

In neither of these are the limbs or body represented with the 
heads, which separates them from another group in which portions 
of the body or limbs or both are represented. In both also the sem- 
blance to a ceremonial celt is lost by the absence of any representation 
of the point or cutting edge. In the other group we have simply a 
stone nodule with i-epresentations of a human head or other parts of 
the body carved on its surface. These forms have no visible projec- 
tions by which they could have been lashed to a staff or which may 
have served as a handle. 

Figures of various forms of stone heads with projections for lash- 
ing will be found in the article on the Aborigines of Porto Eico, and 
in Mr. Joyce's paper already mentioned,*^ where they are called stone 
heads. Heads or faces (pi. 91, A) cut on the surfaces of stones with- 
out projections by which they may be attached to a staff are also 
figured in my article under the name "stone masks" — a term open 

" Anier. Anthrop., n. s. vol. xv, pp. 6, 7. 

*' Prehistoric antiquities from the Antilles, in the British Museum. Journ. Roy. Antbr. 
Inst., vol. XXXVII, pp. 402^19. 


to some objection. In none of these are there representations of 
body 01' legs carved in stone with the head. 

Stone Heads 

One of the most interesting foi-ms of stone heads in the Heye col- 
lection is shown in plate 91, B^ and represented from the side in plate 
91, C. This head is globular, with two projections on the back, one 
above on the crown of the head and the other below under the chin. 
The cheeks are marked by dei^ressions. one on each, and above each 
depression there is a ridge, which becomes so pronounced in the fol- 
lowing specimens from the Pinart collection in the Trocadero Mu- 
seum that they seem to show the bones of the jaws or to resemble 
skulls, suggesting a death god. 

An illustration showing one of these heads with pronounced like- 
ness to a human ci'anium is shown in plate 92, ^1, from the Troca- 
dero Museum. This object, the gift of Mr. Schochler to the museum, 
was found in Porto Eico, and measures 0.145 by 0.09 m. It has a pro- 
longation in the place of a neck by which it was attached to some 
foreign body. 

The head shown in plate 92, 5, is cut in low relief on a stone 
nodule 0.12 by 0.09 m., and has visible projections by which it could 
have been attached to a foreign object. The nose, lips, and eyebrows 
are cut in low relief, the eyes and the mouth being indicated by shal- 
low depressions. 

The stone head shown in plate 92, 7>, is said to have come from 
Martinique, although it recalls in general features those from Santo 
Domingo. This specimen has been so mounted as to resemble a 
bust, and has marked European technique with the exception of the 
earrings, which show decided Antillean features. This bust is in 
the Trocadero Museum, Paris, and measures 0.20 by 0.18 m. I have 
many doubts of its prehistoric character, but have introduced it 
here to show the manner in which tlic ear ornaments were set in 
the lower lobe of the ear. It is mounted in a way different from 
that implied in the forms of other specimens, where the neck was 
probably lashed to a stave or staff. 

Stone Nodi' and Masks 

A stone face in the Heye collection represented in plate 94, C, has 
all the essential features of these objects as figured in the author's 
Aborigines of Porto Rico. 

The surface on which the face is cut is flat, the opposite convex, 
without prolongation or groove for attachment. The features that 
distinguish this mask from others are the two triangular engraved 
figures below the mouth which probably represent the limbs or body. 


Plate 94, 5, also in the Heye collection, represents one of the sim- 
plest forms of stone masks. It has an oval form, with eyes and mouth 
represented as shallow depressions, while nose and lips are in relief. 
It has all the features found in the typical stone masks, but is dis- 
tinguished from the majority by the two round symmetrical protu- 
berances, one on each cheek. In some of the engraved stones that 
have lost all resemblance to masks, homologues of these protuberances 
alone remain, and in one or two instances there are three of these 
projections, one on each cheek and one in the middle of the fore- 

The nodule type is differentiated from stone heads and stone masks 
because body and limbs, one or all, are graven on them. There sur- 
vives in them no indication of the point or cutting edge of a ceremo- 
nial celt and no projection by which they can be bound to other ob- 
jects. To all intents they resemble idols except that they are destitute 
of any flat base by which they may be made to stand upright. They 
show no evidence of having been attached to a wooden base or a pup- 
pet made of fabrics which has been suggested for the head. This 
type was imknown when the memoir. Aborigines of Porto Rico, 
was written, and so far as known has not been differentiated from the 
other types. 

There are in the Berlin Museum three specimens which can be re- 
ferred to this type, the essentials of which will appear in the follow- 
ing description : 

The first of these (pi. 93, B, C), from Cape Haitien, has an almost 
globular form, more like an ovate spheroid, in whicli the organs of 
the face and arms are cut in low relief on one side. The eyes, repre- 
sented by simiDle slits, are cut obliquely and are surrounded by a 
raised side. The lips are large and the nose more or less broken. 
Just below the lower liji we find representations of the fingers, four 
in number, connected by a ridge that surrounds and incloses the 
face. This ridge probably represents the arms, the elbows being 
situated just below the hands. 

The nodule figured (pi. 94, .4) is from Les Cayes, Haiti, and is 
mucii more complicated in its carving, because both arms and legs 
are represented. In this specimen the chin is pointed, and the face 
has the eyes, nose, and teeth carved in relief. The nose has an in- 
verted T shape. The forehead is low and, like the chin, indicated 
by a curved line. An exceptional feature of the face is the two 
parallel vertical markings on each cheek. These recall the grooves 
on the cheek of an etched stone from' Nipe Bay, Cuba, elsewhere 
figured. *° 

" Twenty-fifth .\nn. Rept. Bur. Amer. Ethn., pi. isxu, b, and Amer. Anthrop., n. s. vol. 
TI, no. 5, pi. xviii, 1. 



[ETH. ANN. 31 

Ridges representing the arms on each side of the stone extend back- 
ward, forming shoulder blades, and lengthwise for upper arms to the 
elbows, ending in hands a short distance below the chin. The legs 
are made in low relief extending to the Icnees. 

A flat form of stone nodule is shown in plate 93, Z), E. This speci- 
men has the hands turned upward and the raised ridges representing 
the arms inclose the face. The eyebrows arch over the eyes and are 

continued into the nose, 
which bifurcates above 
tlie teeth, inclosing the 
mouth and forming the 
lower lip. The teeth 
are indicated by parallel 
ridges alternating with 

This specimen from 
Gonaives, Haiti, which 
is now, like the two pre- 
ceding, in the Berlin 
Museum fiir Volker- 
kunde, was collected by 
Dr. Grosser, German 
consul in Haiti. 

The stone object shown 
in figure 31, purchased 
in Paris by Mr. Heye, 
is cylindrical in shape, 
with a human face cut on one end, and with arms in low relief. The 
remarkable feature about this object is the knobs, 15 in number, ar- 
ranged with some regularity, covering the sides and back of the body. 
The lower part of the body and the legs of this specimen are miss- 
ing, probably broken. The specimen is ~i\ inches long and about 3| 
inches in diameter at the level of the chin. 

In various collections studied by the author there occurs another 
type of Antillean stone handiwork, the technique of which resembles 
that on the ceremonial petaloid celt, the stone head, and the mask, but 
in which there is no projection that might serve as a handle. These are 
oval or spherical stone balls with face, body, limbs, or complete human 
forms cut in relief on their surfaces. They recall the ceremonial peta- 
loid celts of the British Museum, figured by Joyce and referred to 
above. If, for instance, the handle of the latter were reduced in length 
and the head made more globular, we would have the ceremonial ball 
referred to. 

Fig. 31. — Stone idol, a, Front view. 
(7.25 inches.) 

b. Side view. 





There are several semicircular stones in the Heye collection, one 
of which, shown from front and side (fig. 32), is particularly well 

The distinctive features of this type are the semicircular form, 
flat on one side, convex on the other, and the two ear-like lappets, 
sometimes modified into heads, one on each side of the, by 
which the stone was lashed to a foreign object. In the stone heads 
these cleats for lashings appear above the head and below the chin, 
forming a neck, whereas in the semicircular type the ear-like lap- 
pets are at right angles to the axis of the face when such exists. 

Fig. 32. — Semicircular stone with face, from front and side. (7.31 inches.) 

The face on this specimen is indicated by a well-made circle, 
within which are two circular pits of equal size, side, by side, for the 
eyes, and a third, oral in form, for the mouth, all three being sur- 
rounded by ridges. 


There are no more characteristic objects from the West Indies 
than the so-called stone collars, which are prehistoric in origin, and 
of unknown use. These objects appear to be confined to Porto 
Eico and Haiti, although specimens in some of the European mu- 
seums are labeled from the Lesser Antilles. The Heye collection 
contains several stone collars, one of which (pi. 95, A, B) is unique. 

The stone collars of Porto Rico and Santo Domingo have a variety 
of forms which athuit of a classification. There are two great types 
known from characteristics that are evident— the massive and the 
slender ovate. Both massive and slender collars are secondarily 
divided into two groups, right and left handed, accordingly as the 
knob is on the right or left hand side as the collar is placed in a 
natural position.^" 

^ The " natural position " would seem to be in case of the slender ovate with the 
pointed pole uppermost, the panels facing the observer. In the massive collars, when 
there is no pointed pole, the situation of the panels determines the upper pole. 



[ETH. ANN. 34 

The specimen of a massive collar that is represented in plate 95, 
A, B, was purchased by the author ^^ from Seiior Seiyo, of Arecibo. 
Porto Eico, and is one of the most remarkable of all. This sj^ecimen 
is evidently a connecting link between the massive collars and the 
slender ovate, the symbolism on it belonging to the massive variety. 
There is a remote likeness in this to the paneling of the slender ovate 
variety, but no other collar has a head carved in such high relief. 
Several massive stone collars have been figured by archeologists. 
They have the general features shown in plate 97, D,^^ from Paris, 
which is one of the heaviest of its group. Although massive collars 
have sometimes been interpreted as unfinished specimens of the slender 
ovate, there is every evidence that the massive type is wholly dis- 
tinct and its symbolism should 
be interpreted from its own char- 
acters. We are forced to the 
conclusion, if the slender ovate 
is regarded as a finished form 
of the massive, that having cut 
one form of decoration for the 
massive they changed it later, in 
making that of the slender ovate, 
to another type of symbol. 

The various parts of a slender 
ovate collar are designated in 
figure 33 for reference and are 
known as follows: b, boss; dp, 
decorated jDanel; dpr/, decorated 
panel groove; dpT), decorated 
l^anel border; dphp, decorated 
panel border pot ; p, jirojection ; 
upp, undecorated panel pit; upff, 

Fig. 33. — Si-hi'inatlc view of stone collar. 

«, shoulder; sJi, shoulder bowl; 
undecorated panel groove. 

The corresponding parts of a massive collar are designated in the 
accompanying cut by the same letters. There are intermediate forms 
connecting the two groups, and the evidences api^ear strong that both 
massive and slender collars had the same use. 

The collar represented in plate 96, ^, shows the marginal border 
of the panel highly decorated, while the projection modified into a 
head recalls one with the snake's head in the museum at Bremen fig- 
ured by the author. 

'i Thi.s lioautiful sppcimen was first figured by the .author in his- articlo on " Porto 
Rican Elt^ow-Stones in the Heje Museum." The author's figure Is reproduced by Joyce, 
" Central American and West Indian Archaeology," pi. xix. 

"The author is indebted to Prof, M. H. Saville for the photograph here reproduced. 




;?■■;'■ M^v55j''S' '-"'"'' " ■^®|i%-'^''' 


\V-; ^^^^^f:^?^-- :': f / 
"■ ■ •^- ■■■-■■■ • ,,l?" 







20. i by 16 inches. 







C, Is inches; />, 17 incties. 









I> c 



The specimen, jjlate 97, ^1, from a photograph procured by Prof. 
Saville from Dauberton, Paris, shows modifications in the edge of the 
undecorated panel border, somewhat different from the hist, although 
comparable with it. The parts reiu'esented are a median head with 
lateral appendages, recalling the more extensive design on the elabo- 
rate form of the massive collar (pi. 95, ^1). The slender ovate collar 
figured in jjlate 97, B, is in the Trocadero Museum, Paris, and is 
exceptional in the relatively large development of the boss. There is 
also a remote resemblance in the projection to a head, which assumes 
the reptilian form in the specimen from the Ihemen Museum. Inci- 
dentally attention may be called to the neat method of installation 
which is worthy of adoption in other museums. 

In another specimen from the same museum shown in plate 97, C\ 
the boss is not as prominent, and the region of the undecorated panel 
is more massive. The shoulder band is broad, the projection not 
being visible in the view here given. The photograph (pi. 97, D) 
here reproduced was likewise made by Dauberton, and procured for 
the author by Prof. M. H. .Saville. 

One of the most instructive stone collars known to the author 
was described by him in his article on "A Prehistoric Stone Collar 
from Porto Rico."^^ The knob of this collar is modified into the 
head of a serpent or some reptile. In this article the author shows 
that the knobs of several collars may represent the heads of some 
reptilian form and if that conclusion has any important significance 
in a determination of the identify of the type we may adopt the 
serpent theory, or that the stone collar represents a serpent idol. In 
this article the homologies of other parts of stone collars are con- 
sidered to such an extent that the paper is here quoted at length. 

Attention was first called to a stone collar with knob modified 
into a snake's head in the following lines of an article on Porto Eico 
Stone Collars and Trii^ointed Idols :^* " Sometimes the projection is 
ferruled. often with jiits like eyes, and in one collar the prominence 
is said to have the form of a snake's head." To this is added the 
following note : " This specimen is owned by Mr. Leopold B. Strube, 
of Arecibo, who has sent the author a drawing which shows the 
knob in the form of a snake's head.'' This reference was later 
quoted in the writer's memoir on the Aborigines of Porto Eico.^^ 

On a recent visit to Europe the author examined the specimen, 
now in Bremen,^'' and made the drawings reproduced in figures 

^ Amer. Anthrop., n. .s., vol. XVI, pp. 319-330. 

" Smithsonian Miscellaneous Collections, vol. xi.vii, pt. 2, 1904. 

"'' Twenty-flfth Ann. Rept. Bur. Amer. Ethn. 

^ The author acknowledges with pleasure his indebtedness to I>r. Johannes Weissenborn. 
curator of ethnology in the Stadliche Museum, Bremen, for the opportunity of studying 
this instructive specimen. 



[ETH. ANN. 31 

34-37. A glance at the first of these shows that it belongs to the 
type called by the late Prof. O. T. Mason " the " right-handed va- 
riety of the slender oval group." 

This collar is made of a hard, light-gray andesite or diorite, with 
surface fairly smooth but not finely polished. Its general form is 
not unlike other examples of the slender ovate type. The special 
differences are found in the ornamentation of the decorated panel 
border and the modification of the projection or knob into an 

animal head. It measures 15 and 11 
inches in greater and lesser diam- 
eters, respectivel}'. 

The undecorated panel shows no 
exceptional features, except that 
the rim is pinched midway of its 
length into a triangular projection, 
as shown in figure 34, but which 
could be better seen from one side. 
A slightly raised band extends 
around the collar, just below the so- 
called boss or elbow, joining the 
upper and the lower margin of the 
decorated panel border. As will 
be pointed out i^resently, the head 
carved on the panel border is very 
Avell made and instructive. 

Lateral and dorsal representations 
of the knob modified into a head are 
shown in figures 35 and 36. 
Before the author had examined the Strube specimen he was of 
the impression, from sketches of the objects, one of which was kindly 
sent to him several years ago by Herr Strube, and the other by Prof. 
W. H. Holmes, who saw the specimen in the Bremen Museum, that 
the head replacing the projection or knob represents that of a serpent, 
but he is now able to point out a more striking resemblance to the head 
of some otlier reptile, a conclusion reached mainly from comparative 
studies of similar heads found in some of the three-pointed stone idols 
of the first type, figured elsewhere. ^^ 

The three-pointed idols with heads like those of the Strube collar 
also possess legs, which would prohibit their identification as serpent 
idols and would weigh against acceptance of the opinion that the 
head on the collar represents a snake, were it not for the fact that 
primitive man is not always consistent in fashioning his images; 

"Latimer Collection of Antiquities from Porto Rico in the National lluseum at Wash- 
ington, Smithsonian Report for 1876. 

™.\borigines of Porto Kico, Twenty-fifth Ann. Kept. Bur. Amer. Ethn., pis. xxxix. a, 
a' ; XLi, b, c; xmi, a, b; XLiii, a, a'. 

Fig. 34.- 

-Tbe Strube stone collar (Bre- 
men Museum). 




35. — Lateral view of " knob " of the Strube stone 
collar (Bremen Museum). 

hence the heads of both, even when furnished with limbs, may repre- 
sent some serpent monster, the iguana, or a reptile with the body and 
apjjendages of a turtle. 

The modification in the projection in this collar, although less usual 
than other features, is not more instructive than the unitjue figures 
graven on the border 
of the decorated pan- 
els. The surface of 
the panel is not excep- 
tionally ornamented, 
but its border is sculp- 
tured into the form of 
a head with lateral ap- 
pendages much better 
made than is generally 
the case. 

The appearance of the head and legs on the j)anel border of this 
specimen (fig. 37) are as exceptional in form as the knol;). tor. unlike 
the heads cut on the panels of other slender oval collars, the head 
of this specimen is in high relief. The relation of the head to the 
collar is here exactly reversed, as compared with that of almost all 
other collars, for the forehead adjoins the panel instead of being 
turned away from the decorated panel. The two lateral appendages 
extending along the border on the sides of this head are readily com- 
parable with similar figures, in the same position, on other collars. 
A representation of the head and appendages as seen from below 
shows that the lower jaw is pointed and triangular. 

The form of the decorated panel border of the Strube collar (fig. 
37) bears directly on our interpretation of this feature in other collars 
and sheds light on the meaning of certain conventionalized figures on 
other speciniens in which the head form is not so evident as in this 

specimen, as may aj)- 
pear from the follow- 
ing comi)arisons. 

The decoration on 
the panel borders of 
different stone collars' 
falls naturally into a 
series jjassing from 
realistic to convention- 
alized figures, shown 
in tlie accompanying figures. In order to interpret these decorations 
we may pass from the most complicated to the simplest form. 

Commencing with the form shown in figure 38, representing a 
specimen now in the Heye collection, we have a massive collar with 
a head cut in high relief on the surface of one side. This head (/<) 

Fig. 36.- 

-Dorsal view of " knob " of the Strube stone 
collar (Bremen Museum). 



[ETH. ANN. 34 

resembles those constantly found engraved on stone images or 
modeled in terra cotta for han^lles of Antillean bowls or vases. It 
represents a being wearing a kind of Phrygian cap, with mouth lialf 
open, large eyes, and other features recalling a turtle or some reptile. 
Two anns («, «), with the elljows bent and showing the palms of 
the hands and the fingers, are well represented and rise from ujider 
(,. . , the chin. The hands 

i ,' /■ pa. fc. 

' appear to hold' up 

rings cut at the sides 
of the head, which 
they touch on each 
side, and are inter- 
preted as representa- 
tions of ears or ear or- 

FiG. 37. — Doooratod panel and panel border of the .Strnbe naments. These rings 
stone collar (Bremen Museum). ^.^^^jj ^j^^ j^^gj. ^^^^^ 

of the ears in certain stone yokes found in Vera Cruz and other 
Mexican States. The umbilicus appears on the body just below the 
chin, and on each side are rectangular carvings {d, p), supposed to 
represent otlier parts of the body. 

In the collar of the Heye collection there is practically no sepa- 
ration of the panel border and the panel, or rather the former has 
extended over the latter, which remains as a rectangular design 
(d, p) filling the areas on each side of the anterior appendages (a) 
and below the problematical lateral extensions (pa). 

Extending on each side of these rings on the upper margin of 
the collar there is an interesting conventional figure in relief, unlike 
a leg or any other part of the body,"but which is seen constantly 
in modified form in 
other collars. In a 
general way this dec- 
oration (pa) consists 
of a distal portion, 
which is more or less 
angular and of cubi- 
cal form with a me- 
dian pit (h). and a 
proximal region con- 
nected by means of a knee-shaped relief figure (y), with the head 
and all other portions of the design. The parts represented in this 
carving are the head, forearms, ear lobe or ornament of the ear, 
and a knee-like problematic body. Every organ except the last 
can be reatlily identified, but in order to determine the meaning of 
the knee-like member we must consider similar relief designs on 
collars in other collections. 

Fig. 38. — Panel of stone collar (Heye MuJt>um). 




The ornate desifni on the panel border of the Strube specimen 
in Bremen naturally next claims our attention. In considering 
this example (fig. 37) it will be noticed at once that the mouth, eyes, 
and all other parts of the face are reversed when compared with 
the head of the collar in the Heye collection (fig. 3s). This is due 
to the fact that its left side represents the right side of the Heye 
collar, as will be seen when these collars are laid with the decorative 
panels uppermost for comparison, in which case the lower jaw in 
the former is naturally below, while in the latter it is above, a re- 
versal caused by one of these collars being right-handed while the 
other is left-handed. This does not prevent a comparison of similar 
parts in the ornamen- 
tation of the collar, 
but it must be borne 
in mind that they are 
in reversed positions. 

y\'e fail to discover 
on sides of the head 
of the Bremen collar 
any indication of 
those rings or ear or- 
naments in relief that 


Fig. 39. — Panel of stone collar (Latimer collection). 

are so conspicuous in the Heye Museum specimen. There are likewise 
no homologues of arms and hands below the chin, but the lateral 
figures carved in low relief on each side are represented in somewhat 
modified form. Here occur representations of a joint {j) and the 
terminal circle with a deep pit (i), leading us to consider them the 
same organs. The panel is distinct from its border and has no sign 
of legs. 

Passing to a consideration of a collar figured by Prof. Mason and 
said to be from Guadeloupe, we discover on the decorated panel 
border a still greater simplification of the head which here (fig. 39) 
appears as a circle (h), with eyes and mouth represented by shallow 
j)its. The problematical lateral organs (pa) have here become sim- 
ple scrolls, with a pit (b) in the middle of the distal end, a conven- 
tionalization which is paralleled by that shown in another design 
on the panel margin of a collar from the Latimer collection figured 
by Mason,^" where the lateral appendages (pa) are reduced to 
scrolls, although the joint is still angular. 

A similar decorated panel is found in one of the collars of the 
Latimer collection (fig. 40). 

"Op. clt. 

-34 ETH— 22- 




[ETH. AXN. 34 

Fig. 40. — Panel of stone collar (Latimer collection). 

AVe pass now in our comparison to a collar (fig. 41) in wliich the 
face on the panel border is divided medially into two parts, and the 

remainder of the 
'"'• ■*■ figures, especially 

the lateral scrolls, 
have undergone a 
strange elongation. 
The simple jiits rep- 
resenting eyes still 
remain, and each of 
the halves of the 
former head is con- 
tinued into an extension curved into a scroll in which the only recog- 
nizable feature is the jointed organ. 

Another variation in the figure on the decorated panel border 
(fig. 42) occurs in another of the Latimer collars. The vertical 
division between the eyes separating the face into halves has not 
extended wholly across the head, and the forehead here remains un- 
divided. The scroll- 
like lateral append- 
ages (pa) that make 
up the remainder of 
the figure of the dec- 
orated panel border 
have no exceptional 

In still another 

collar of the Lati- I^'« 41.-Panel of stone collar (Br«nen Museum). 

mer collection, the conventionalization of the panel border figure has 
proceeded so far that the resemblance to a head with lateral append- 
ages is completely lost. Here we have simply two scrolls with one 
extremity of each approximated and tlieir distal ends widely sepa- 
rated and extended. 

In another collar of the Latimer collection the decoration of the 

panel has been subject- 
ed to further modifica- 
tion in form, the panel 
figure taking the form 
of two rectangles rep- 
resenting the half- 
circles of the divided 
face, each bearing a pit 

Fic. 42.— Panel of stone collai- (Latimer collection). rej^resentin"' ail e\e. 

Tlie elbow-like scrolls are present with their terminal dots rising 
one on each side of the rectangle representing half of the face. 

A. e. 




Fig. 43. — Panel of stone collar (Latimer collection). 

Any resemblance of the panel decoration shown in fifjure 43 to 
a human head with lateral appendages has whollj' vanished. Here 
the decoi-ated pa'nel 
border takes the form 
of a narrow rectangu- 
lar figure with round- 
ed ends, slightly 
curved upward and 
crossed at regular in- 
tervals bj' three paii"s 
of bars. In each of 
the intervals there is a 
small pit, two of which (e, e) represent all that of the eyes, 
and two (h. h) those constants at the extremities of the scroll-like ap- 
pendages that exist in 
the figure of the compli- 
cated panel border. 

There remain other 
designs on panel bor- 
ders, one (fig. 44"°) of a 
collar in the Trocadero 
Museum at Paris, and 
the other (fig. 4.'5) in the 
Latimer collection. The 
outlines of these show 
important mod ifica- 
tions, but these also in 
reality teach the same 
morphology as the 
preceding, viz, that fig- 
ures on the decorated 
panel borders are simjily 
highly conventionalized 
heads with extended lat- 
eral appendages. 

There is one feature 
lacking in the figures 
last mentioned that 
should be explained. 
Since the pits which 
represent the eyes, as we have pointed out, are here absent, it might be 
supposed that the conventionalized head is also wanting; but if we 
compare them with the underside of the figure cut in the panel of the 
Bremen collar (fig. 46), the reason for this lack is apparent. All of 

-Sloue collar sbowiny unique decuiaieJ panel 
border (Trocadero Museum). 

'" The author is indebted to Prof. M. H. Saville for this illustration. 



[ETII. ANX. 34 

Fig. 45. 

-Decorated panel and panel border of stone collar 
(Latimer collection). 

these represent the underside of the lower jaw, not the upper part of 
the head where eyes, mouth, and nose are present. 

From the comparative data given above we are able to say that 
Avhere\er we have figures cut on decorated panel borders they prob- 
ably represent a head, bod}', arms, or legs, often highly convention- 
alized and sometimes lost. As the arms or forelegs appear in the 
more completely represented form, figure 38, accompanied with the 
problematical lateral scrolls, we can not regard these scrolls as dupli- 
cate arms or fore 
limbs; if they are 
appendages they 
must be posterior 
limbs or legs. The 
posterior append- 
ages in all these 
instances have 
been brought for- 
ward into the same 
plane as that in 
w h i c h the head 
and anterior legs lie, and by this contortion have lost all likeness to 

This interi^retation of the ornamentation of the decorated panel 
border of the stone collar reduces it to a figure of the same general 
character, but it takes no account of certain figures on the surface 
of the panel itself. The figures engraved on this area are sufficiently 
distinctive to bear certain resemblances whose meaning is doubtful 
The decorated panels of several stone collars (figs. 39, 41, 42, 
43) bear an incised ring or circle, sometimes with and sometimes 
without a central 
pit. On each side 
of this circle there 
are constantly 
represented well- 
made figures, of 
unknown signifi- 
cance, that have 
certain common 

resemblances in all specimens in which they occur. It may be 
assumed, but without positive proof, that these figures repre- 
sent parts of the body; for example, the circle, which so often 
appears in Antillean art, represents the umbilicus, while the in- 
cised geometrical lines on each side of it resemble figures of legs 
or arms. 

Fig. 46. — Underside of decorated panel of tlie Strube stone 


In several of the decorated panels we find this circle doubled; 
or these duplicated circles may be connected or modified in siicli a 
way as to appear as spirals;"' or at times parallel lines may extend 
from the circles. The figures on the decorated panels of several 
collars consist of geometric parallel lines arranged in squares and 
chevrons, a form of decoration sometimes found on panels of massive 
collars. These are regarded as decorations of the body of the animal 
or the human form represented. 

The main difference supposed to exist between the Bremen collar 
and other examples of its kind would seem to be the modification 
of the projection or knob into an animal head, and yet when we 
examine a series of collars we find several specimens in which the 
projection is carved in such a way as to suggest the conventional 
head of some animal. 

Many massive stone collars "- and some of the slender ovate ^' 
varieties have two " knobs," one of which projects on each side of a 
binding band or shoulder band filling the interval between them. 
In one instance the two ends are not united by a band but are hooked 

No decorations appear in any of these double knobs, and all are 

without eyes or other indication of the presence of a head, which is 

likewise true of those examples in which the projections do not rise 

above the surface of the collar, although a remnant of the shoulder 


When the projection bears any design, it is commonly flattened, 
with a pit on each side. Another form of simple flattened knob, hav- 
ing circles on each side and parallel lines between them, is found on 
the second Bremen specimen. In an example in the Heye collection, 
where the jjrojection is not very prominent, it is marked I)}- a single 
transverse and .several parallel grooves, recalling the parallel lines 
between the pits in an und'cscribed collar in the Bremen Museum. 

The simplest interi^retation of these variations in the so-called pro- 
jection or knob of a stone collar would be that, like that of the Strube 
specimen, it represents a highly conventionalized head, and that the 
accomiDanying pits or circles are eyes. 

Although several forms of stone collars have been added to those 
known to the author when he published his account of the Aborigines 
of Porto Eico, and one or two new theories concerning their use have 

"" This iform suggests the ornamentation of a fragment of a specimen of doubtful 
relation in the Stahl collection, now in the American Museum of Natural History New 

" Aborigines of Porto Eico, Twenty-fifth Ann. Kept. Bur. .\mer. Ethn.. pis. Ixiii, 
Ixiv, Ixv. 

" Prehistoric Antiquities from the .Antilles in the British Museum, Journ. Antbr. Instl- 
tulc, vol. XXXVII, pi. xl. 1907. 

"" .\boriaines of Porto Rico, Twenty-fifth Ann. Rept. Bur. .\mer. Ethn., pi. Ixv, /, 


been brought forward, we are still somewhat perplexed as to what may 
be regarded as their true use. A few of the more I'easonable theories 
are mentioned below : 

1. Insignia of office, worn on tlie person. 

2. Sacrificial objects. 

3. Idols for animal worship — serpents, lizards. 

4. Idols for tree worship, especially yuca. 

5. To assist childbir*^h — representations of female organ of 


6. Collar for men or women dragging canoes. 

In the following pages the author will return to the interpretation 
of the Antillean stone collars in connection with elbow and three- 
pointed stones, but he will here state that he inclines to combine the 
third and fourth theories mentioned aliove as the nearest approach 
to a correct interpretation of tiie stone collars. The beneficent super- 
natural being of the Porto Ricans and Haitians was probably the 
Yuca god or the S\<j supernatural who brought life to the food plant, 
yuca, and, as occurs on the neighboring continent, was represented 
by a mythic snake or dragon. Stone collars represent this god of 
serpent form but also with human features carved upon them. An 
midoubted serpent made of wood "'''' iuis the same form of head as the 
knob of the collar above figured. 


In an article entitled '' Porto Eican Elbow Stones in the Hej'e 
Museum, with Discussion of Similar Objects Elsewhere,''"* the author 
has published the following account of these instructive objects: 

"Many prehistoric stone objects found in Porto Rico have taxed 
the ability of archeologists to explain and have furnished the theorist 
with abundant material for speculation. Among these may be men- 
tioned three-pointed idols, both with and without animal or human 
heads. Other forms, from their resemblance to horse collars, ha\e 
from the first been designated as collars or collar stones. Those 
prehistoric Porto Rican stone objects that, from their shape, are 
called elbow stones, are the least known and apparently one of the 
most enigmatical types. 

" Elbow stones resemble, in general form, fragments of broken 
collars, but a detailed study of various elbow stones and comparison 
with stone collars, rather than bearing out this seeming resemblance, 
tends to show that they form types distinguished by highly special- 
ized characters. 

"'" Aborigines of Porto Rico, Twi'iity-flfth Ann. Ropt. Bur. Amor. Ethn.. pi. xc. 6. 
I" American Anthropologist, n. s., vol. xv, pp. 435-459. Reprinted as Cont. Heye Mus., 
uo. 4. 


'• The elbow stone type of objects is represented by 12 specimens 
in tlie archeolojrieal collections studied by the author. Objects of 
this type are therefore less numerous than the collars, of which there 
are about 100 in different collections. Elbow stones have not been 
found in Cuba, Jamaica, or the Lesser Antilles, and have never been 
reported from the American mainland. Their distribution in the 
West Indies corresponds closely with that of stone collars and three- 
pointed stones, which are practically confined to Porto Eico, His- 
paniola (Haiti and Santo Domingo), and possibly eastern Cuba. 
The author is of course aware that stone collars and three-pointed 
stones have been recorded from certain of the Lesser Antilles, but 
their number, or rather their relative proportion to other jjrehistoric 
objects from the same islands, is so small that he is inclined to ques- 
tion the recorded provenance of these specimens. Thus the late Pro- 
fessor O. T. Mason described and figured a single collar in the Guesde 
collection from Guadeloupe, and M. Alphonse Pinart ascribed an- 
other specimen of this type to the same island. Among several hun- 
dred stone objects from St. Vincent the author has not seen a single 
collar or three-pointed stone, and he therefore reasonably suspects 
that the locality of the single broken fragment of the latter type 
ascribed to St. Vincent by Mr. Joyce is doubtful. It is the writer's 
belief that these objects are not indigenous to the Lesser Antilles. 
With a collar in the British Museum described by Joyce and said to 
have Ijeen found in St. Thomas, Danish West Indies, the case is 
somewhat different. St. Thomas, St. Croix, and neighboring islands 
belong to the same prehistoric culture area as Porto Eico. hence stone 
collars may rightly be expected in them; in fact, a fragment of a 
collar undoubtedly found in St. Croix is now in the Nordby collec- 
tion at Christiansted. the cliief city of that island. 

"As the author intends to point out elsewhere that the localization 
of characteristic stone objects determines certain archeological areas, 
he will now only briefly mention the existence of several well-defined 
prehistoric Antillean culture areas. The majority of stone objects 
from the St. Vincent-Grenada area are radically different from those 
of St. Kitts. and these in turn differ from those of the Barbados 
area. Stone collars, elbow stones, and three-pointed stones are pecu- 
liar to the Porto Rico-Haiti culture area, and when found elsewhere 
in the West Indies are believed to have been introduced. * * * 

" For convenience of study the two arms of an elliow stone [pi. 98. 
B] may be designated as right and left {RA, LA), and their point of 
junction the anffle or elbow. One of the arms is either decorated 
or has a panel : the ends of both may be fluted, while their general 
form tapers more or less uniformly. One or both arms may have 
a groove on the outside called the sulcus (.s). which, when situated 
on the paneled arm. extends lengthwise from the panel border to 


the end of the arm. A cross section of an elbow stone near the boss 
following the elbow band is, as a rule, about the same as that near 
the pointed pole of an ovate slender collar. The surface of an elbow 
stone, esjDecially the boss, is generally rough, but several examples 
have the remaining parts finely polished. 

" So close are the general likenesses between the boss and the arms 
or shoulders of collars and elbow stones that an identification of the 
latter with broken collars is most natural. In order to explain minor 
differences in the two types, it is held by some of those who entertain 
this opinion that a broken collar has been subsequently fashioned 
into an elbow stone and its surface redecorated to fit it for secondary 
use. • So radically different, however, are the carvings and symbols 
on the surfaces of these two types of objects that this conclusion 
seems imreasonable. 

" Other archeologists believe that the elbow stone is a fragment 
of a type of collar differently ornamented from any that have been 
found entire. The resemblances are believed by "them to be close 
enough to indicate identity and the differences are looked upon as 
special rather than as general characters. 

" The belief that the elbow stone belongs to a distinct type is far 
from the thought that there is any utilitarian difference between 
the two classes of objects. All indications tend to show a like use, 
and that if we could satisfactorily explain the meaning of one type 
we should be in a fair way to interpret the other. It is, in fact, pri- 
marily to shed some light on the significance of the stone collar that 
the author presents the following results of his comparative studies: 

" Elbow stones, like stone collars, may be divided into right- 
handed and left-handed, or right-armed and left-armed, according 
to the position of the decorated arm. AVhen an elbow stone is placed 
so that the panel will show, this feature will be seen either on the 
right or the left hand, thus determining the designations ' right- 
handed ' and ' left-handed ' elbow stones. The significance of the 
difference in this feature is not known; it may mean nothing, but it 
would appear that its very occurrence in both collars and elbow 
stones has some important bearing on the function of the objects. 

" The style of ornamentation furnislies data for a classification 
of elbow stones on other grounds. Two distinct varieties of these 
objects can be readily recognized accordingly as a head, face, or 
body is sculptured on the outer surface of one of the arms. This 
sculpture, when it appears, is generally in low relief, and always 
represents human features, never those of an animal. In elbow 
stones on which such a sculptured figure does not appear there is 
always a panel with a shallow, oval, concave pit hollowed in the 
middle, in which is sometimes a secondary depression, as shown in 
figure ."50 [see pi. 99, B], The aim bearing this panel with its pit cor- 






A, .5.25 inches; B, 9.0.3 inches. 



J , i;.siiu-hes; C, C, 8 inches. 


responds with that on which, in decorated elbow stones, is cut a human 
head or bodj'. A homologue of this plain panel (identical with the 
undecorated jsanel of a stone collar) does not occur on those elbow 
stones in which carved heads or faces are found : consequently it is 
supposed that the decorated panel of the stone collar is not repre- 
sented bj' a sculptured head in elbow stones. 

" In those specimens of elbow stones in which a face is sculptured 
on one arm it will be noticed that the middle line of the face or head 
is placed longitudinally and not transversely to the axis — always 
lengthwise of the arm, never crossing it. The position of these fig- 
ures on known elbow stones differs radically from that of the heads 
on panels of stone collars, for in the latter the middle line of the 
face is at right angles to the panel. The figure on a collar is situated 
generally on the border of the decorated panel, and is small and in 
low relief; but in an undescribed collar in the Hej'e Museum [pi. 
95], which is unique in this respect, the head rises above the 
surface. An examination of this collar shows that in general form 
it belongs to the massive stone collar group, while the decoration is 
more like that of the slender oval collar; but the head cut on the 
panel is so different from any j'et described that it can hardly be as- 
signed to the latter group. It is therefore regarded as a connecting 
form having affinities witli both massive and slender oval collar 

" It is instructive and may be significant tliat tlie faces on all the 
elbow stones are anthi-opoid; and the same is true also of the stone 
collars, the heads on all of which have human features. The syra- 
bolism of the spirit depicted represents a human, not an animal, 

" Description of Elbow Stones 

"The following classification includes the known elbow stones in 
various collections, designated by the name of the owner, the col- 
lector, or the museum in which they are deposited : 

".-1. ^\'itll jure cut in relief on one arm 

a. Face on the rifrht arm. 

1. lladi-id spet-imeu (fig. 48). 

2. Heye Museum specimeu (pi. 98, A). 

3. Latimer specimen, National Museum (tig. 49). 

b. Face on the left arm. 

1. American Museum specimen. 

2. American Museum specimen. 

3. Pinart specimen. 

4. Heye Museum specimeu ipl. 98, B). 


"B. Anil nithout (ace, but icitl( ijanel 

a. Panel on the right arm. 

1. American JIuseum specimen. 

2. American Museum .specimen. 

3. American JIuseum specimen. 

4. National JIu.seum specimen (tig. 47). 

b. Panel on the left arm. 

1. Heye Museum specimen (pi. 09, B). 

'■('. Ellioic Htuiic of doubtful type 


" n. Face on tlie ripht arm 

"1. Madrid specimen. — The most perfect and elaborately decorated 
of all these objects is an elbow stone in the Museo Arfiiieologico of 

Fig. 47. — Elbow-stone In tin' I'niti'd States National .Museum 

Madrid, which has face, arms, and legs sculptured on one arm. This 
specimen has been figured by Neumann and several other writers, but 




as it is almost unknown to archeolofrists a new illustration [fi<r. 48] 
indicating the variations in the decorations of these objects, is here 

" Fi'om an inspection of the figure it appears that both arms of 
this beautiful specimen, unlike those of most elbow stones, are un- 
broken. The right arm shows the longitudinal groove (sulcus) com- 
mon to these objects, extending from the lower margin of the panel 
to the extremity of the arm. The surface is almost wholly occupied 
by the figure sculptured upon it, the head, arms, legs, and horseshoe- 
shaped headband or fillet being in relief. This fillet, which is of 
about the same breadth through- 
out, is decorated with a number 
of incised pits, one of which is 
placed medially over the fore- 
head. The fillet ends on each 
side of the face, near the cheeks, 
where there are depressions ap- 
parentlj" representing ears. This 
headband recalls those found on 
heads of three-pointed stones, 
with which it is seemingly 
homologous. The two ends of 
the fillet merge into the shoul- 
ders of the figure and continue 
to form the arms. The fore- 
arms are folded on the breast, 
as is common in Antillean ob- 
jects of art in stone and shell, 
and the fingers are rudely rep- 
resented by grooves. Near the 
wrists, a short distance from the 
fingers, there is a slight projec- 
tion on each arm, which recalls the protuberances commonly repre- 
sented on the ankles of Antillean figures. Relatively the body is 
abnormally small or inadequately represented, the space between chin 
and legs being so restricted that not even the umbilicus, so constantly 
found in stone images from Porto Rico, is represented. The soles 
of the feet are turned upward in an extraordinary way and the toes 
are folded back, a common feature in Antillean idols. The mouth 
is large, nose broad, cheeks prominent, the whole recalling faces on 
three-pointed stones. 

"2. Heije Museum specimen. — The second specimen of elbow stone 
(pi. 98, A, A') with a face on the right arm is less elaborately sculp- 
tured than the Madrid example, the arms and body not being 
represented. The right limb is apparently broken off just below 

-Elbow-stone, M-idrid 
(12.62 inches.) 



the carved face, so that there is nothing on this arm correspond- 
ing to a ferruled end. On the forehead of the figure may be seen 
a triangular area in which is a central pit. The head is fringed 
by a fillet less elaborately made than that of the Madrid speci- 
men. The end of the small arm appears to have been broken, 
there being no sign of fluting, although it shows indications of 
a sulcus. On the outer side of the small arm, near the angle, 
there are two series of parallel lines, or chevrons, cut in the sur- 
face, recalling the decoration of a massive collar elsewhere figured. 

" In order to compare this elbow stone with certain stone heads 
figured by the author in his Aborigines of Porto Eico (pis. li, 
Lir, Liii), we may suppose that the two arms are much reduced 
in length, as in plate lii here referred to, and the face cut in 
high relief instead of being low or flat. A still further' reduction 
in the homologues of the arms appears in certain stone heads and 
in stone disks with faces illustrated in the plates mentioned, in 
some instances all traces of the arms having disappeared. The 
stone head shown in plate liv, ff, a' has the neck developed into 
a short handle, giving the appearance of a baton and recalling 
certain ceremonial celts. The objects called 'stone heads' in the 
author's work above cited so clo.sely resemble three-pointed stones 
that they may be allied to the third type of zemis, in which the 
conoid projection is modified into a head. A like parallel occurs 
in the first type of three-pointed stones, tlie heads of which recall 
those of men, lizards, and birds. The few known specimens of the 
second type have human faces. 

" The figures representing lizards in both the first and the third 
type of three-pointed stones are characterized by elongated snouts, 
eyes, and two pits, representing nostrils, placed near the extremity 
of the ujDper lip. The human faces of the first type generally 
have the ornamented fillet reaching from ear to ear, which is never 
represented in reptilian three-pointed stones of the first type, but 
is present in reptile figures in the third type. Ears appear in 
human but never in bird or reptilian forms. In place of a depres- 
sion or pit in the median line of the headband, the reptilian figures 
of the third type have a device consisting of a low convex pro- 
jection and pit of the fii'st form. This last-mentioned feature is 
sometimes situated in a fold extending downward over the fore- 
head, suggesting a frontal ornament. 

";3. Lathier sper'nneit. — T^his elbow stone [fig. 49] was first fig- 
ured by Prof. O. T. Mason, who regarded it as a part of a collar, 
and afterwards by the author, who founded the tyiie now known as 
elbow stones upon its characteristics. Although the form of the 
Latimer elbow stone is somewhat aberrant in several particulars, it 
presents the distinctive features of the type. Its arms are ap- 




parently imbroken at their extremities, and the face is cut on the 
right limb. Instead of the encircling grooves on the arm bearing 
the face, the arm is perforated near its end, where it is crossed by a 
single transverse groove supposed to serve the same purpose as the 
grooves in the fluted specimens above considered; in other words, 
for attachment to a staff or some other object. The oval face, eyes, 
nose, and mouth are typical of Antillean art. The headband has 
a pit medially placed above the forehead and is ornamented by a 
series of parallel incised lines. The slightly protruding ears at 
the termini of the headband have large circular pits. The shorter 
arm has a shallow longitudinal groove (sulcus?) and obscure elbow 

I'lii. 4'j. — Elbow-Stone in the Lalimer collection. Bide and front views. 
(Length 7 J inches.) 

" /;. I'ucc on the left arm 

" 1. American Museum specimen. — Among the elbow stones in the 
American Museum of Natural History there is an instructive speci- 
men in which an arm is ornamented with a human face in relief, por- 
tions of the body, and anterior appendages; the legs are drawn to- 
gether and merge into a beaded end with longitudinal sulcus and 
accompanying encircling grooves. The face sculptured on this speci- 
men is oval, the cheeks are prominent, the eyes and mouth circular. 
The ear pits are prominent, and the fillet or headband bears a medial 
circular protuberance with its accompanying pit. The arms are bent ; 
the legs are separated above by a space in which is a triangular 
depression. The umbilicus is indicated by a circular design. The 
shorter arm is girt by parallel grooves and tapers to a rounded 

" 2. American Museum specimen. — In the same museum there is 
a second specimen of elbow stone, on the left arm of which is carved 
a rude face. This example is broken on one edge. It has no grooved 


arms, but in place of them is a perforation near the end of one arm, 
as in the Latimer specimen. The sulcus is absent. 

"3. Pinart specimen. — The Pinart elbow stone, said to have been 
at one time in the Trocadero Museum, Paris, belongs to that group 
in which the left arm is the larger and bears an oval face, wliich 
lias large open mouth, prominent ears, and headband, with a circular 
pit over the ft)rehead. Representations of arms, legs, and uml)ilicus 
are present; the legs are separated liy a triangular depression as in 
n former specimen. In tlie figure given by Pinart there are indica- 
tions of the grooves or furroivs of the terminal ends of botli arms, 
liut iis his illustration is imperfect this feature is difficult to de- 
termine satisfactorily. 


"a. Pant! on tlir ririlit arm 

" There are five specimens of elbow stones with flat panels instead 
of figui'es on the arms. Three of these are in the American Museum 
of Natural History, New York; a single specimen of the same 
type is in the National Museum collection; and there is one in the 
Heye Museum. All, except the last, are right-handed. 

" These objects are simpler in form than those of the previous 
group, otherwise they are of the same general character. Each has 
a sulcus on the surface of one arm, which, however, is without encir- 
cling arm grooves. Although the panel pit, a constant featui'e of the 
panel, is about uniform in position, it Aaries in shape and size in the 
several specimens. From its general shape and simplicity it would 
appear that tlie panel in these specimens served as a base to whicli 
another object, possibly a stone head, was attached. 

" The specimen in the National INIuseum is said to have been col- 
lected at Vieques Island, a new locality from which elbow stones 
have been recorded. The paneled arm of this specimen is long and 
slender, the other limb short and grooved, but with a well-marked 
sulcus not shown in the figure. 

" b. I'aiicl III! tlie left arm 

" 1. TIeye Museum specimen. — Tlie Icft-lianded elbow stone [pi. 90, 
5] in the Heye Museum is a fine specimen, surpassing the others 
of the same group in foi-m and superficial polish. Its left arm ends 
in a series of fluted joints, but is without a sulcus; the right arm is 
short, with an encircling groove. The panel is almost wholly oc- 
cujiied by an elongate oval depression, in whicli is a second oval pit, 
the surfaces of both of which are smooth. Tlie panel is suri'ounded 
by a polished border slightly raised and evenly rounded. 



" There are one or two aberrant specimens that are doubtfully 
identified as elbow stones. In discussing the Guesde collection Prof. 
Mason figured and described an unusual object from Punto Duo ( ?), 
allied to elbow stones but of highly aberrant form, as follows : 

" ' Fig. 195. An ornamental piece, of bluish green color. It is rare lu form, 
but not absolutely unique. In the American Museum at New York is a similar 
specimen. The cbamferiuf: and fluting are gracefully blendej. The left-hand 
extremity is perforated for suspension. Length of long limb, S inches: of short 
limb, 5A inches." 

"The differences between this specimen [pi. 99, (\ T'] and the 
typical elbow stones lie mainly in the 'chamfering.' nevertheless it 
shows certain characters pecidiar to elbow stones. The ' similar speci- 
men ' in the American Museum, referred to by Mason, is possibly 
one of those above mentioned under group B. 

" The Guesde stone is exceptional in several particulars. The 
figure shows no indication of a panel or a head, and the sidcus like- 
wise is missing. On account of the absence of the panel it is diffi- 
cult to tell whether it belongs to the right-arm or tlie left-arm group. 
The fluting on the longer arm reminds one of the specimen in the 
National Museum at AVashington, and the perforated shorter arm 
is like that of the Latimer specimen. The grooves of the smaller 
arm extend halfway round the arm. while those of the longer arm 
girt it entirel}'. 

" Morphology and Interpketation 

"The many similarities between three-pointed zemis, elbow stones, 
and stone collars would seem to indicate a corresponding similarity 
in use, consequently any light on the morphology of one would aid 
in the interpretation of the other types. The author believes that 
the life figures on these three types of objects are symbolic repre- 
sentations of zemis, or spirits, which were worshiped by the pre- 
historic Porto Ricans. They were idols, and bore the name of the 
particular spirit represented (as well as the general designation 
'zemi') — a usage common to primitive religions. 

"A consideration of the differences in form, or the morphology, 
of these objects is desirable before the many theories as to their use 
can be intelligently discussed. 

"A forward step in the interpretation of the morphology of 
stone collars was taken by Mr. Josiah Cato and later by Prof. Mason, 
who recognized that the ' shoulder ridge ' faintly resembles a lashing 
of the two ends of a hoop. 

■'An important suggestion has been made by Mr. T. A. Joyce ***" that 
the Antillean stone collar is a copy of an archaic zemi made of 

•""Prehistoric Antiquities from the Antilles, in the British Museum, Jour. Royal Anthr. 
Institute, vol. xxxvilj pp. 402-419, pi. xlviii-lvi, 1907. 


branches of a tree bent into a hoop and fastened at their ends. He 
was the first to associate tlie stone collar with ' tree worship - — 
an important advance in the solution of the enigma. Mr. Joyce 
described a stone collar in the British Museum in which there is no 
shoulder ridge, but what appear to be the two ends of branches 
' overlapped ' and ' hooked together ' at the point where the shoulder 
ridge ordinarily is found. This led him to suggest that in studying 
a stone collar we must ' retranslate ' it to its wooden prototype, and 
recognize that the juncture of the ends in this case, and perhaps in 
all, was effected as follows : ' When the limbs of the fork [of a tree] 
were trimmed, the stump of a small subsidiary branch growing in a 
convenient jiosition toward the end of each, was left projecting; the. 
longer limb was bent round, and the projection toward its termina- 
tion was hooked round the i^rojection on the shorter limb; the ad- 
dition of a cotton bandage would hide the joint and make all 
secure' (p. 410). 

" ' It is perfectly obvious,' Mr. Joyce says, ' that these collars were 
constructed originally of wood. A young tree was selected and cut 
off immediately below a fork ; the two ends of the fork were trimmed 
into unequal lengths, the longer bent round so as to overlap the 
shorter, and the two fastened together bj' a band of cotton smilar to 
the leg bandages worn by the natives.' He also states (p. 410) : ' Start- 
ing with the supposition that they were originally constructed of 
wood (which seems to me to be almost certain), it seems possible that 
a clew might be found in the prevalence of tree worship in the 

"'The heavy collars,' continues Mr. Joyce, ' which appear to have 
been formed of a single and comparatively stout stem bent into a hoop 
and the ends secured by a bandage may represent a semi made origi- 
nally from the straight trunk of a tree without a fork.' Thus a second 
and important step in the interpretation of the meaning of the collar 
was taken by Joyce in the recognition of the collar as a zemi con- 
nected with tree worship, its original prototype being made of wood, 
the stone form being a more lasting one, but one in which certain 
characters of the wooden or archaic form still persisted. 

" In Ramon Pane's account of how Antillean wooden zemis were 
made, as directed by a tree spirit, we have evidence of tree worship 
in Haiti; but the testimony afforded by this account is too meager to 
prove that when the tree referred to by the Catalan father was felled 
it was made into the form of a collar. The author suspects that the 
idol referred to by Eamon Pane represented the Yuca Spirit,"^ but 
this suspicion is still subject to proof. 

"The Tuca splrU or Tocahu (Yocabana) Maorocon, Mahouya, Huracan, or Great 
Serpent, whose idol JIarolo was one of the two stone images In the Cave of the Sun, 
worshipped for rain and blessings. 


■■ In Antillean, as in other tree worship, it was the spirit of tlie 
tree that was the object of adoration, and that worship was more or 
less connected with the material benefits desired — generally the food 
that the tree j'ielded. The deitj' that controlled the manioc (yuca), 
or the Yuca Si:)irit, Yucayu, was worshipped for temporal benefits, 
the wooden idol being the visible, material symbol. 

" In differentiating the elbow stone from the collar as a distinct 
type, it has been shown above that the position of the heads of both 
relative to the axis does not coincide, since one is transverse to the 
axis, the other longitudinal. In one case the object must be placed 
vertically, in the other horizontally, in order to bring the face into 
a normal position — a difference in position that remains to be satis- 
factorily explained. If, however, the elbow stone was carried, it may 
be that one arm only of the elbow stone was attached to a staff and 
the object carried ujji-ight, while the collar was laid horizontally 
when in use, bringing the head into the same relative position. 

"It is evident tliat tiie furrow, or sulcus (s), is an important 
feature in the morphology of elbow stones. This groove, non- 
existent in the collars, may have been cut in the surface of the elbow 
stone for the insertion of a rod or staff', to which it may have been 
lashed with cords held in place by the grooves girdling the arm. It 
is not always limited to one arm, but is sometimes found on both 
arms, and it would appear that occasionally either two sticks were 
attached to the stone, one at each end, or the two ends of the same 
stick were bound to the arms, in which latter case the stick would 
have to be bent into a hoop resembling in shape a stone collar, part 
wood, part stone, the elbow being of the latter material. 

" The attachment of an elbow stone to a rod or staff was probably 
by means of vegetal fibers. In some instances this was unnecessary, 
since there was sometimes a depression in the end of each arm, as 
in an elbow stone reputed to be owned by Senor Balbas, of Porto Rico, 
to which reference has been made elsewhere, but which the author 
has not examined. In this case it appears as if there are depres- 
sions in which the sticks were possibly inserted, rather than lashed 
to the stone. 

" The theory that the extremity of a staff was laid in the sulcus 
and lashed to the elbow stone would preserve the normal position of 
the face carved on the panel if held vei'tically. If carried by means 
of this staff, the face cut on the arm would be upright or in a nat- 
ural position. Some of the elbow stones may have been carried in 
the hand without an attached staff, thus accounting for the absence 
of a sulcus. * * * 

160658°— 34 ETH— 22 14 


" An examination of certain celts, clubs, and other stone artifacts 
leads to the belief that the prehistoric Antilleans had many kinds 
of objects which they carried in their hands on ceremonial or other 
occasions. Several of the almond-shaped or petaloid celts with heads 
or human figures cut on the sides have their pointed ends prolonged 
into a handle, and even those without such a prolongation can hardly 
be supposed to have been hafted, as in such a case much of the 
design cut upon them would have been concealed. Many of the 
beautiful axes for which the island of St. Vincent is famous were 
too bulky to be carried in war and too dull to be used as cutting 
implements. They may have been carried by chiefs on ceremonial 
occasions as badges or insignia of office. 

"A remarkable stone object [pi. 99, ^1] in the Heye collection 
has the appearance of having been used as a baton, but its form 
is different from that of any yet described, and would suggest that 
it was carried in the hand, but it may have been inserted into a 
wooden staff. One end of this object is enlarged, with the surface 
cut into a definite form, while the other end tapers imiformly, pro- 
viding the handle, possibly for attachment to a rod. The figure 
on the larger end has a median crest or ridge extending over the 
extremity, on each side of which is a prominence, the arrangement 
recalling the crest and eyes of some highly conventionalized animal. 
The crest or ridge is found on examination to be double and to 
extend round the larger end, the two parts coalescing at one end 
and uniting by a transverse band on the other. On the sides of 
this median crest are the protuberances, each with a circular pit 
and extension from the margin. The only object known to the 
writer that approaches in form the stone referred to is one made 
of burnt clay found in Barbados, many miles away. This specimen 
also has an enlargement representing a head at one end and tapers 
uniformly to the other extremity in the form of a handle. The 
Barbados object also has a crest extending along the middle of the 
enlarged part and ending aliruptly near a hole which may be likened 
to a mouth; on each side of this elevation there are pits that may be 
regarded as eyes. The ridge or crest suggests a distorted nose, or 
the beak of a bird, a suggestion that would seem to comport with 
the parts on the enlarged end of the stone baton above described. 
The double median fold and lateral elevations with pits represent, 
beak and eyes." 

Another stone object (pi. 100, A, B) from Guadeloupe, descril)ed 
by Prof. Mason, evidently belongs to the same type as the stone cere- 
monial baton, or some form of badge mounted on a staff. 


The stone object (pi. 100, C\ D) sliuwn from side and top reminds 
one of the bird-shaped ceremonial liaton. This object was evidently 
so fashioned as to be held in the hand by the pointed end, the carved 
portion bein<r held aloft. Its shape is similar to the bird-formed 
baton from Arecibo, Porto Rico, above mentioned. The Heye speci- 
men is distin<ruished from the latter l)y a double row of rounded pro- 
jections, regularly arranged one row on each side of a median 
groove, running along the enlarged end of the baton. It is probable 
that it was carried in the hand. The specimen is (H inches long by 
4 inches wide at its broadest end, tapering to a point at the opposite 

Elongated stone ceremonial batons with figures cut on one end and 
enlargements at the other have l)een reported from several West 
Indian islands, but the author knows no specimen of quite the same 
form as that figured in plate 100. i\ D. 


The group of stone objects known as three-pointed stones, also 
called zemis and mammiform stones, is confined to Porto Rico and 
Santo Domingo. These objects may be divided into the following 
groups: (1) Those with head on the anterior end; (2) those with 
head on the anterior side of the conoid projection; (3) those with 
conoid projection modified into a head : (4) those without head or 
face cut upon them.'^" The Heye collection contains one specimen 
with features of both the first and second groups, and one referred 
to the fourth group, which in place of an engraved head has an 
incised circle on both anterior and posterior ends. 

First Type ok Threk-I'Ointku .Stones 

It is possible to still further divide the first group of three-pointed 
stones into four subdivisions: (1) with human heads (fig. 50) ; 
(2) those with reptilian or mammalian heads; (3) those with bird 
heads; (4) heads of nondescript animals. A single specimen of the 
first subgroup has a head on the posterior as well as the anterior 
end. but as a rule the posterior end of stones of this type has a pair 
of lirtibs cut upon it. A specimen (pi. 101, ^1) from the Heye Museum 
has a sujDerficial feature on the cone, hitherto undescribed. Two cir- 
cular pits or depressions occur on each side of this projection, each 
surrounded on the lower half h\ an incised line, which is connected 
by an incised line with the edge of the base."' 

"An aberrant form of the fourth subfrroup occurs also in the St. Vincent-Orenada 
region, but with thi<! exception — three pointed stone idols are not found outside the Santo 
Domingo-Porto Rico area. 

"A similar pair of pits is figured in another specimen in Aborigines of Porto Rico, 
Twenty-fifth .\nn. Hept. Bur. Amer. Ethn., pis. xxxvii, a; xxxix, c ; xLi, b; and xLiii, c. 



[ETn. ANN. 3* 

A second specimen (pi. 101, B) in the Heye Museum is closely 
allied to that group of three-pointed stones with a representation of a 
human head on the anterior end, and shows a pair of pits on each side 
of the middle line of the conoid projection near the apex. These 
pits, unlike those of the last specimen, are without encircling lines, 
and are destitute of the lines on the side of the specimen, plate 101, A. 
The mouth is represented widely opened, recalling that of a fish. 
The form of the legs is quite exceptional, being unlike those of the 
majority' of three-pointed stones of the subgroup to which it be- 

The specimen (pi. 101, C) belongs to the same group as those 
already considered, but differs from them in wanting two pits on the 
side of the conoid projection; the ear is rudely engraved, lips not 
being well represented. Its legs are drawn up ; eyes barely outlined. 
The frontal fillet on the forehead is smooth and undecorated. 

This specimen (pi. 101, D) in the Museum fiir Volkerkunde in 
Berlin, has the typical human form of a three-pointed stone of 

the first tyi^e. The frontal fillet is not 
clearly indicated, and the legs or posterior 
appendages are rudel}' cut. The apex of 
the conical projection leans slightly for- 
ward, the body being short and thick. 

The specimen (pi. 102, .1) differs from 
all others thus far considered in the form 
of its posterior appendages, which are 
folded backward and notched at the ends. 
The frontal fillet is not differentiated 
from the eyebrows, the nose being some- 
what broken and the ears being obscurely 
represented. The form of the head sepa- 
rates this specimen from any in the group 
of three-pointed stones with human heads 
to which it belongs. It is rudely carved, 
its legs being in low relief and the ears not even indicated. This speci- 
men has the appearance of never having been finished, notwith- 
standing which it shows unmistakably marked features of the first 

A three-pointed stone shown in plate 102, 5, has its mouth wide 
open, like one of the specimens above represented. The body is 
very long as compared with its height, and the apex of the conical 
projection does not bend forward. The head suggests a human being, 
its fillet being without engraved decoration. The ears are indicated 
bv incised circles and triangles. 


50. — Three-pointed stone, 
first type. 


Side view. 

Lateral view. 


A, B, U.5 inches: C, D, G.5 inches. 



A , in inches; C, 10 inches; D, 10.7.'. imhes. 








The next specimen to be considered, shown in plate 103, A, is in 
the Berlin Museum. It shows all the essential characteristic features 
of the first group, the fillet on the head being somewhat more elab- 
orately decorated than is the case in the majority of these objects. 
Its nose is broken, but the chin is quite protuberant. The legs are 
slim and bifurcated at the extremity, suggesting webbed feet, as in 
the figure taken from the posterior end. 

The head of plate 103, B, represents that of a human being, while 
C of the same plate is reptilian. The crest on the head of the latter 
is unique in thi-ee-pointed idols. 

Mr. Theodoor de Booy, of the Heye Museum, collected in Santo 
Domingo a three-pointed stone of the first type, which is different 
from those yet described. This specimen is made of a brown stone 
resembling fossil wood. Morphologically it belongs to the second * 
group of the first type, or those with a reptilian head, and is now in 
the Heye Museum. The exceptional feature of this specimen is its 
rounded base, which curves upward around the edge. There is a 
circular de2>i"ession in the micklle of this base, situated about equal 
distance from anterior and posterior ends. 

Another specimen belonging to the first tj' pe of three-pointed stone, 
also collected by Mr. de Booy in Santo Domingo, differs from the 
preceding in the position of the mouth, which instead of extending 
longitudinally in the axis of the base is vertical to it. Perhaps the 
nearest approach to this is the form figured on plate xl of my Abo- 
rigines of Porto Eico. The stone of which this specimen is made 
has a whitish color, and its surface is quite rough. 

Three imperfectly described specimens (pi. 104, B, C, D) of 
three-pointed stones in the Madrid collection, all of the first type, 
are worthy of notice. One. of these {C) had the anterior ends cut 
in the form of tlie head of a bird with engraved wings on the sides. 
Figure B has a highly decorated fillet, engraved lines of which alter- 
nate with three pits, each surrounded by a ring. One of these pits is 
situated on the median line, the other two laterally. 

Plate 101, C. in the Madrid Museum, represents a zemi in the form 
of a bird. When seen from above the bill appears to be upward. The 
head is distinct from the body, but appears to be swollen just back 
of the lieak, the relative position of which to other organs is 
unique, the mouth generally pointing forward and not upward as 
is here the case. The apex of the conical projection is modified into 
a knob tipped slightly forward. The posterior end of the idol 
shows no indication of feet, legs, or other organs. This is one of 
the most exceptional forms of known three-pointed idols and differs 
radically from that of the other known bird forms. 

A most remarkable three-pointed stone from the Trocadero 
Museum, shown in plate 105, A, is unique in having a head 


carved on the posterior as well as on the anterior extremity. 
This specimen also bears exceptional incised markings on the 
anterior and posterior slope of the conical j^rocess. They are 
evidently decorative and have grooves extending from the apex 
a little to one side of the median line of the cone. From these 
arise 7 short parallel lines on one side and 10 on the opposite. These 
lateral grooves arising on the same side of the longitudinal line ap- 
parently have no reference to the position of the conoid projection 
between them. They extend toward, but do not join, other longi- 
tudinsil scratches not very clearly indicated. If the whole figure 
could be made out we would probably find it to consist of two ir- 
regular rectangular designs crowned by a number of parallel lines, 
the two figures separated at their narrow sides by the conical eleva- 
»tion, but what was intended to be represented by these areas and 
parallel lines is not wholly clear. 

The most exceptional feature of this three-pointed idol is its bi- 
cephalism, a head being found on each end. The structure of the 
forehead is also exceptional, for it would appear that representations 
of a pair of limbs had been crowded into the area just above the 
eyes on each side of the circular pit which marks the middle of the 

If we interpret the parts back of the eyes as arms or legs it 
would appear that the figure is kneeling, the knees being pointed 
forward and ending just back of the nose in claws that embrace the 
circular figure in the middle line; but it should be mentioned that 
the representations of claws, fingers, or toes are indistinct and the 
appendages are somewhat problematical. Back of these so-called 
folded leglike appendages there exists a rounded ridge on the side 
of a groove, plainly showing indications of a place of attachment by 
a cord that may have bound the idol to a foreign object. 

The three-pointed idol shown in plate 10.5, 5, belongs to the first 
type, or those with head on the anterior and legs on the posterior 
point; but although its form is somewhat different from any pre- 
viously known, the essential structures are not very clear. The eyes 
ai^pear to liulge from the sides of the head, while the nostrils are 
represented by pits mounted on papillae like those of reptiles. The 
head is not relatively as long as that of the majority of three-pointed 
stones representing reptiles, but the mouth is large and extends back- 
ward rather than transversely, as is usually the case in stone rep- 
- resentations of these animals. 

The essential feature of the three-pointed stone, plate 105, C, also 
from the Trocadero Museum (Pinart collection), is the two pairs 
of circular pits, surrounded by ridges in high relief, engraved on 
the sides of the conoid projection. These appear to be separated 
from each other by another ridge extending from the point of the 













conoid process to the margins of the base. A corresponding ridge 
likewise extends lengthwise of the three-pointed stone from the apex 
of the conoid to the base. 

The three-pointed stone represented in plate 106, A, has a wide 
oi^en mouth unlike any other specimen of this type, but recalls in 
other respects the first group of the first type. 

Plate 106, B, from the Heye collection, represents a zemi of the first 
group of the first type, and plate 106, C, from the same collection be- 
longs to the same type. This specimen is also distinguished by two 
elevations in the frontal fillet, and the characteristic shape of the 
ear, by which it can be sej^arated from the majority of three-pointed 
stones with human heads and those where the mouth is distinctly 
anthropoid. This specimen, unlike other representations of the 
group to which it belongs, has toes at the ends of the legs. The apex 
of the conoid projection is broken, but shows evidence that formerly 
it inclined slightly forward. 

There is a good example of the second group of the first type in 
the Heye collection. The head of this specimen (pi. 106, D) re- 
sembles that of a lizard. 

There is a three-pointed stone in the Berlin Museum fiir Volker- 
kunde that has features of the head like a reptile, but differs from 
this animal in certain well-marked characteristics. Along the median 
line of the head of this specimen (pi. 10", A ) there is an elevated ridge 
on the side of which are parallel markings recalling heads of rep- 
tiles. The nostrils, which are constant features of the thi-ee-pointed 
stones of the first class representing reptiles, are indicated by two 
pits situated slightly behind the tip of the snout. These are not, as 
is usually the case, surrounded by a ridge or moimted on an elevation. 
No ears, fillet, or nose were made out in the specimen. 

The Heye collection has a specimen of tlu-ee-pointed stones (pi. 107, 
B) with characteristics of tlie fii-st and second types and seems to be a 
connecting form. This remarkable specimen is not only exceptional 
in having a head carved on the anterior surface of the conical projec- 
tion, but the conical projection is deeply incised, the position of the 
ear being indicated by an incised triangle. The posterior extremity 
is also quite exceptional and unlike three-pointed stones of the first 
and second types. 

The three-pointed stone, plate 107, C, belongs to the Heye Museum 
and has likenesses to a bird, but likewise resembles a turtle. Gen- 
erally in other bird zemis wings are represented. 

The three-pointed stone shown in plate 108, A, was collected by 
Mr. de Booy in Santo Domingo. It belongs to the first type and 
shows relations, in the form of the head, to those placed in the rep- 
tilian group. The object is made of a brown stone, said to be fossil 
wood, which effervesces with acid. 


Pei'haps the most exceptional feature of this object is the curved 
base, instead of flat, ending on each side in the pointed rim shown 
on the side. This base has a median circular depression about mid- 
waj^ in its length. The lower lip extends forward and is curved 
upward, the mouth impression being a deep-cut groove. 

The specimen shown in plate 108, B, belongs to the first type of 
three-pointed stones, and is remarkable in the position and shape of the 
mouth, which is turned upward instead of being horizontal, as is 
usually the case. The specimen was collected by Mr. de Booy in 
Santo Domingo. 

Second Type of Three-Pointed Stones 

The only specimen of the second type of three-pointed stones, or 
that with a face on the side of the conical projection, in the Heye 
collection is the form, plate 108, C, which appears never to have 
been finished. The position of the eyes is indicated by a depres- 
sion, and there is a slight bridgelike elevation representing lips. 
The cone is low. its apex bent forward. 

Specimens of these stones are not very conunon, and as a ride come 
from Santo Domingo. 

Although pointed stones of the second type are so rare in Antil- 
lean collections, Mr. de Booy has added another to those already 
known. This specimen, like most of the others, was collected in 
Santo Domingo and resemljles that figured on plate xLvir in Abo- 
rigines of Porto Eico. Like this specimen it has a pit on the sur- 
face of the conical projection, opposite the eyes, surrounded by an 
incised ring, around which are three lines of a triangular incised 
figure. In the published figure the ear has the form of the figure 6, 
but in this .specimen the ear is a circle. The lower jaw of the speci- 
men collected by Mr. de Booy is more pronounced than in any of 
the three-pointed stones of the second type yet described. 

In a collection of Porto Eican antiquities presented to the National 
Museum by Miss B. A. Gould and described by the author"^ there 
are several instructive three-pointed stones — one of the first, two of 
the second, and one of the fourth type — which so far as known are 
unique. They are referred to as follows in the article above quoted : 

" Specimens of the type in which a face is carved on one side of the 
conoid projection, or between its apex and the anterior projection, are 
much less abundant than those of the first type in Porto Eican col- 
lections. Only five zemis of this kind are described in the author's 
memoir, and the majority of these came from Santo Domingo.'* 

" Further notes on the Archeology of Porto Rico, Amer. Anthrop., n. s., vol. x, No. 4, 

'° The .second t.vpe was originally distingni.shed from the flrst in the memoir above men- 
tioned. The author has seen no representations of this type in other publications. A 
Rimilar method of representing Joints liy incised circles is found in many Central 
American figures, and in Mexican bas-reliefs. 



A , B. C, 10 inches. 



,4 , .3..0 inches. 




A , S.ol inches: iJ, y.o inches: C, s inches; D, 3.75 inches. 











B - «'.■*- * 













a, b, 4.13 inches. 


There is one additional specimen of the same type in the collection 
here considered [pi. 108, D], and Seilor Grullon has sent the author 
a photograph of still another [pi. 109, A, B], now in the Santiago 
Museum. These seven known specimens fall logically into two 
groui3s — three having limbs carved in relief on the sides, and four 
without any sign of appendage. One of the three-pointed stones 
here described belongs to the latter, the other to the former gi'oup. 

" The specimen of the second type, sent by Miss Gould, came from 
Aguas Buenas, Porto Rico, and according to its label was found in a 
cave."' It is instructive in several particulars, not the least being its 
geographical localit}', indicating that the type is Porto Rican as 
well as Dominican. This specimen has legs cut in low relief on the 
sides of the conoid projection. These appendages rise from the back 
and extend to the anterior projection, where they terminate in feet 
which are brought together below the mouth. Round depressions, 
or pits, are found near the position of the joints, and just below the 
apex of the conoid jDrojection is a small lateral depression. Grooves 
worn in the base of the conoid projection seem to indicate that the 
object was lashed to some foreign body. The face of this idol is 
without nose, while lips and ears, which ordinarily are prominent in 
the type, are inconspicuous. The tip of the posterior projection is 
considerably battered, but striaj in the stone at this point would 
appear to have been intended for feet. The specimen measures 6 
inches in length by 4 inches in height. 

" The author's attention has been called by Seiior Grullon to an- 
other fine and instructive specimen of this type from Santo Do- 
mingo [pi. 109]. It resembles that figured in plate xlv, figures 
h, h', of the authors memoir on the Aborigines of Porto Rico, but 
unlike that specimen has incised scrolls around a circle on the back 
like the object represented in plate xlvii of the same paper. 

" Unlike the one last mentioned, this zemi has no indication of 
legs or other appendage on the side of the conoid projections; but 
the ears are elaborately cut in relief, the mouth is large, the lips are 
rather narrow, the eyebrows flattened, and the nose is prominent. 
The ferrule back of the head, which possibly indicates a neckband, is 

" Fourth Type of Theee-Pointed Zemis 

"The fourth type of three-pointed zemis includes all those which 
are destitute of head on either the anterior projection or the conoid 
prominence, and have no indication of a face on any part of the 
object. The specimens of this type vary considerably in general 

™ Miss Gould has kindly furnished a photograph of the exact point in the cave where 
she was informed this specimen was found. 


form, most of them having the anterior and posterior projections 
blunt and rounded, the cone being of limited heiglit. The best figure 
of this variety can be seen in plate l, e, of the author's memoir on 
the Aborigines of Porto Eico. Another subdivision of the type 
[pi. 110, A, B] has more pointed anterior and posterior pi-ojec- 
tions, the surface lying between the anterior projection and the apex 
of the cone being slightly concave, while that portion which extends 
between the posterior projection and the apex is slightly convex. 
There are sometimes pronounced lateral ridges that extend from the 
ajjex of the cone to the edge of the base. 

" In the third subdivision of the type the conoid projection is 
slender, while in the fourth the cone seems to rise out of a depression 
surrounded by a slightly elevated lip. The first two subdivisions 
of tills type have been figured elsewhere (op. cit., pi. l) ; the second 
two, here distinguished from the others for the first time, have not 
hitherto been illustrated. They will be considered in turn, beginning 
with the one last mentioned. 

"An instructive new form of three-pointed zemis, to which the 
author's attention was called by Seiior Grullon, is pi-ovisionally 
placed in the fourth subdivision of the fourth type, from which 
it differs in having an elevated fold or raised ridge inclosing 
a depression, out of which rises the conoid projection. Although 
the general appearance of this stone has suggested phallicism, the 
author would not so interpret it. This is the only specimen of this 
form thus far described.'"''' 

"Another three-pointed zemi from Santiago has the conoid pro- 
jection quite slender, more so than that of any other specimen. Its 
apex tips slightly forward toward the anterior end of the zemi. A 
photograph of this idol was sent to the author by Seiior Grullon. 
This specimen belongs to the third subdivision of the fourth type." 

Four examples of the fourth type of tripointed stones are here 
described for the first time. Two of these (pi. 110, A, B) are in the 
Berlin Museum and one in the Heye collection. 

The fourth type of tripointed stones is easy to distinguish, from 
the fact that while it shows the three points — anterior, posterior, and 
conical projections — it has no representative of human or animal head 
carved on it, although not without geometrical designs, as parallel 

Plate 111, A, represents a specimen which shows grooves near the 
anterior and posterior ends for lashing to some foreign object. The 
cone is girt by jDarallel lines, and the base is slightly concave. 

Plate 111, B, represents one of the Heye specimens, which differs 
from others in having vertical lines near the apex of the conical pro- 

70o Figured in Furtlier Notes on tlie Arclieology of Porto Rico, Amer. Antlirop., vol. %, 




ViG. 51. — Three-point- 
ed stone with face 
on anterior end. 

jection. These lines are also represented in the second Berlin speci 
men (pi. 110, ^4) , in which the base has a central groove with a well 
marked rounded edge on each side. The object shown in plate 111, C 
represents the passage from this type of three- 
pointed stone to the first type. 

Other specimens of the three-pointed stones of 
the fourth t3'pe were collected bj' Mr. de Booy 
in Santo Domingo. They do not differ essen- 
tially from those already figured ; certain of them 
have rough surfaces and show indications of hav- 
ing been used for purposes different from that for 
which they were originally made. 

The fourth type of three-pointed stones i.s a 
large and comprehensive one containing several 
different forms, as may be seen by an examination of tlie objects fig- 
ured in mj- Aborigines of Porto Rico. It will probably be necessary 
later to divide this tj'j^e into several subgroups, and it is possible that 
.specimens now referred to the type have different uses. 

Figure 51 presents features of a three-pointed stone of the fourth 
type so far as general form is concei'ned, but with a face cut 
on the anterior end, consisting of a circle with dots for eyes. There 

is a similar face with like 
representations of ej-es and 
mouth on the posterior end 
in plate 111, C. This fea- 
ture calls to mind one of 
the three-pointed stones of 
the first group, already de- 
scribed (fig. 50). 

The si^ecimen shown in 
figure 52 belongs to the 
fourth tjqie, but has cer- 
tain peculiarities different 
from it, one of which is 
four jJarallel incised lines 
near the point of the cone, 
the meaning of which is 
not known. 

The third type of three-pointed stones, one of which is shown from 
front and side in plate 111, D, grades into stone heads. A well-made 
specimen (pi. Ill, E) in the Heye collection presents all the essen- 
tial features of the same group as described and figured in the author's 
article on the Aborigines of Porto Rico. A similar specimen belong- 
ing to the same group is also figured by Dr. Haeberlin." 


52. — Three-pointed stone of fourth type, from 
side and base. (5 inches.) 

" Archaeological work In Porto Eico, p. 234. 



[ETH. ANN. 34 

There are included in this group of stone heads objects (pi. ill. H) 
that might more properlj' be called three-pointed or triangular stones, 
and they grade into circular or oval disks upon 
which faces are indicated. 

Three-pointed Stone Used for Pestle 

The stone object shown in figure 54 shows 
marked evidences of a secondary use. It is evi- 
dently a three-pointed zemi of peculiar form, the 
anterior end of which is modified by evidences 
of hammering as if used as a i:)estle for bruising 
roots or grains. 

If we interpret the original use as the same as 
that of other three-pointed stones we have pe- 
culiarities not jjreviously recorded for this type 
of stone objects. On the supposition that it is the 
posterior end which has been modified or shows 
the effect of pounding, the centrally placed of the 
three knobs, as shown in the figures of the object 
from above, would represent the apex, while the 
two lateral knobs in the same figure represent excrescences that are 
jjresent in certain forms described in previous articles. There is a 
central circular depression surrounded by a raised rim midway in the 
length of the base. A^Tiile the median extension 
shows no evidences of eyes the whole ajipearance 
of this specimen recalls the bird group of the 
fii-st type of three-pointed stone zemis. The in- 
clination of this projection forward would indi- 
cate that the right-hand end of the stone, which 
does not sliow the marks of hammering, could be 

Fig. 53 . — B a s c of 
three-poiated stone 
of [fourth typo, 
showing , longitudi- 
nal furrows. (6 

Fiu. 54. 

(in liody. 

^toiii' recalling thice-pointed idol, with supiTtii'ial knobs 
Front and side views. (4.75 inches.) 

interpreted as the anterior end of the figure or a handle by which it 
may have been carried as a baton. The general appearance of the 
object suggests a new type of three-pointed stones or an aberrant 
example of the first type. 


/<>■'■-■.■>. ■■ r ■ Is ■; .- .•■',■^ -v -^'>;•■■'w■■o',■-H* f'.wffiSK 




-4 , H iiu-hes; Bf 2.0 iiiclu'S. 


B ^ 


,1 , 2. .56 inches; B, 6 inches; C, 4.4 inches; D, 7 inches; E, (i.5 Inches. 



B, C, 10 inches. 




B, 16 inches. 



Several stone objects of spherical or oblong form are covered with 
projecting warts or wens, tlie signification of which is unknown. 

A few of these, resembling idols, but showing marks of having 
been used as pounding implements, were purchased in Paris by Mr. 
Heye. The stone idol shown in figure 31 shares with the stone balls 
the problematical excrescences. 

One of the most in^ructive idols in the Heye collection from 
Santo Domingo obtained by Mr. Theodoor de Booy near Boca 
Chica, on the south coast of Santo Domingo, is said to have 
been found in a cave. This object, figured in plate 120, recalls 
in several respects an object figured by Pinart and is remarkable on 
account of the large size of the legs, as if aiiiicted with elephantiasis. 
How much of these legs represents a seat is not known, but the rest of 
the human anatomy is so well formed that one is tempted to interpret 
these appendages as part of the seat. The relief figures on the shoid- 
ders and that on the belt are instructive. The ornamentation of the 
back of the head is very similar to that frequently found on tri- 
pointed zemis. The hunchback is likewise not without a parallel 
in clay figures already elsewhere described by Pinart. 

The head has a close likeness to figure «, plate Ixxxii, in Aborigines 
of Porto Eico, where we also have the same elephantiasis in the 
lower legs. Particularly interesting is the head on the belt shown 
in profile, which may have been a buckle. The features of tech- 
nique are well brought out in the plate. 


A worked stone from Arecibo purchased by the author from Sefior 
Seiyo for the Heye collection has a bird form and is the only speci- 
men of this form known to the author, with the exception of one 
described by him in the Aborigines of Porto Eico (pi. lvi, fig. a, a'). 
This specimen has the different parts as legs and wings of a bird 
somewhat more elaborated than that in the National Museum. 

This specimen (pi. 112, A) represents a bird with legs drawn up 
below the breast. The wings are represented in the conventional 
form, the joints being indicated by depressions, while the surface is 
crossed by curved lines and triangular figures which may represent 
feathers. The use of this object is not known, nor can the specific 
bird represented be identified. 


The ordinary form of Antillean mortars and grinders, like those of 
other primitive races, is destitute of decoration and has a depression 
in the surface. It may be of stone or wood. The former was gen- 
erally used for grinding corn. Yuca roots were ground into meal 


for cassava bread on a flat board set with sharpened stones — a form 
still used by the Carib. 

In the collection of the United States National Museum there 
are several flat, rough stones, rectangular in shape, with rounded 
corners, that have been identified as grinding stones. These objects 
have a paired extension on one margin. They are supposed to have 
been used in grinding seeds or roots, and are not common. Similar 
flat stones with two projections occur in several European museums, 
the best of which known to the author are in the Trocadero Museum, 

A mortar jii-esented to tlie United States National Museum by 
Miss A. B. Gould has a concavity on each side, and a distinct groove 
extending around the body. Of all West Indian mortars seen by 
the author this is the most interesting and is of most exceptional 

There are, however, many other forms, not unlike those described 
by the autlior, some of which are represented in the Heye collection. 
Haeberlin figures a fragment of a clay mortar from Porto Rico. 

One of the stone mortars, closely resembling in form an earthen- 
ware vessel, is figured in plate 112, B, C. This utensil is ornamented 
with incised geometrical designs on the outer surface. The decora- 
tions cover also the bottom of the mortar and bowl and are quite 
different from the majority of geometric figures on earthenware 
objects. They have, however, this instructive feature, that most of 
the curved lines end in an enlargement — a common characteristic of 
incised decorations. 

It is difficult to distinguish an implement on which the root of the 
yuca, chocolate, or corn was ground from a seat (duho), and one or 
two of the grinders are identified by some writers as seats. There is 
reason to suspect these identifications, but not enough to prove their 
falsity. It is probable that some of the objects described as seats or 
Porto Rican duhos are in reality grinding stones, especially those of 
A'ery small size; but at present the author has not been able to differ- 
entiate the two, and it is possible that some of them were used for both 
purposes. In this connection an incident quoted from Du Tertre may 
be as instructive as it is amusing : " I recall," he writes (p. 43.3) , " that 
an Indian capitaine, wlio was quite newly clothed, was rebuked quite 
sharply by Madame Aubert, the wife of our governor in the island, 
for his having sat upon her bed (hamas) , which was of white fustian, 
where he had left a good part of his breeches (pantaloon legs). Then 
M. Aubert, lier husband, invited him to dinner. He had much diffi- 
culty in coming to a determination what to do, seeing in advance that 
he would redden the bench on which he would sit; but having cast 
his eyes on his plate, he imagined that this round article, which 
only needed three legs to make of it a stool, had been placed there 
to serve as a place to sit on, he took it and put it on the bench and 


sat on it; and seeing every one laughing at his action, he became 
angry and informed us by means of an interpreter that he did not 
know what posture to assume among the French, and that as long as 
he lived he would never return ! " 

Some of the chocolate slabs have short, stumpy legs; others are 
flat stones. These flat stones are, however, easily distinguished from 
those used in cooking cassava bread. 

According to Labat,'- " before the Europeans Drought the iron 
plates [for frying cassava cakes] they made their cassava on large 
flat, thin stones, which they adjusted for this purpose in decreasing 
the thickness. Many of these stones are found on the seashore. It 
is a kind of sandstone or i^ebble of the color of iron, ordinarily 2 or 3 
feet long and oval. They heated it for the purpose of removing more 
easily the fragments and reducing it to the desired shape. I saw one 
of these stones in 1701 at quay St. Louis, in the island of Santo 
Domingo, at the house of a man named Castras, manager of the house 
of the company of the Isle a Vache. It was 22 inches long and 14| 
inches wide and 3 inches thick. It was very even, and it would have 
been difficult to make it better with tools. In digging the ground 
they found it with some pieces of pottery and figures of grotesque 
shapes, which were supposed to be Indian idols, worshipped on the 
island when it was discovered by the Spanish." 

The specimen shown in plate 113, .4, is about twice as long as broad, 
rounded below, and mounted on four stumpy legs. It has- a head on 
one edge. This specimen was found near a 'town situated in the in- 
terior of Porto Rico not far from the military road from San Juan 
to Ponce. 

On certain grinders (pi. 113, f) the position of the heads, legs, 
and tails represented by simple extensions reminds one of a turtle 
form. This specimen, now in the Berlin Museum, was found at Cape 

The most instructive form of chocolate grinders (stone seats or 
" duhos ") in the Heye Museum is shown in figure 55. This beautiful 
specimen is made of a greenish stone with surface very smooth and 
polished on both the upper and lower surfaces. 

The head is a continuation of one side and closely resembles that 
of a turtle, having a blunt nose, mouth extending far backward, and 
CA^es obscurely indicated by elongated depressions in the side of the 
head. On each side of the neck above the forelegs are projections 
comparable with the fore flippers of a turtle, without any indication 
of claws or leg joints. The concave upper surface of the seat follows 
the line of the neck, narrowing as it approaches the back of the head. 

There are four short, stumpy round legs, one of which is broken, 
which support the body of the seat. The diminutive size of this 

" Nouveau Voyage aui Isles de r Amerlque, vol. i, pp. 400-110. 


specimen ■would seem to indicate that it was of a very different type 
from that used by the caciques for seats, and it may have been used 
for some other jDurpose. Unlike many other seats it has no back, 
and the smoothness of the upper surface implies that it was not used 
as a mortar. It may have been one of those seats on vphich idols were 
placed, or it may have been used for some houseliold god. Its re- 
semblance to some of the known stone seats has led me to place it 
among the duhos, several specimens of which are described in my 
Aborigines of Porto Rico. 

The most elaborate duho made of stone is that shown in plate 
113, Z?, in which we have the buck prolonged be3'ond the seat and 
modified into a human head. The remarkable things about this 
specimen are the arms and portions of the upper body engraved on 
the in.side surface of the back. 

Fie. 55. — Chocolate grinder shaped like a seat or duho. (9.31 inches.) 

The seat and its short legs are not greatly unlike the corresponding 
parts of other duhos. The anterior legs are much larger and thicker 
than the posterior, and between them is a projection comparable 
with the turtle head of the specimen just described. Neither of 
these parts is sufficiently well cut to clearly indicate that they were 
intended to represent any life forms. The posterior legs are longer 
than the anterior and are destitute of ornamentation. 

That neither of these represent any part or appendages of the body 
to which the head graven on the extremity of the back belongs is 
evident from the position and character of the arms engraved on 
the flat inner surface of the back of the duho. These arms are so 
represented that the hands are folded on the breast, the four fingers 
reaching almost to the neck. The upper arms are straight, the 
lower curved, a line connecting the elbows indicating the margin 
of the junction of the body and the seat. This point of union is also 
well marked on the rear side on a level with the elbows. The only 


part of the body that can be identified is a circle with central dot, 
probably the umbilicus, which for some unknown reason is so con- 
stant on figures of Antilleans. 

In examining the arrangement of parts in tliis duho, and especially 
the relation of the head and arms of the human figure which forms 
its back to that of the seat itself, it appears that the former is an ad- 
dition and not an essential part of the latter. In instances, however, 
where an animal head appears between the two anterior legs of the 
duho, these legs may be regarded as the fore limbs of the same 
animal. Such is probably true of this duho, but the head on the 
back has nothing to do with the rude anterior limbs here rejiresented. 

This stone seat, formerly in Paris, now in the Museum f iir Volker- 
kunde, Berlin, has already been described and figured by the late 
Prof. O. T. Mason in his account of the Guesde collection. The 
specimen is one of the best-made examples of these objects yet found. 

This specimen, according to the late Prof. Mason, was identified by 
M. Guesde as an " idol." Mason considered it a duho or chair, as 
shown in the following quotation : 

"A stone stool or chair of tlie variety mentioned and illustrated 
in the Smithsonian Report, 18TG, i^age 376. The material of those 
there described, however, is either sandstone or wood, and the device 
is some animal form. In M. Guesde's specimen the material is a dark 
brown volcanic stone, and the device is the human form. Moreover, 
the position is inverted. The man is lying on his back, with his feet 
drawn up to form the legs of the stool. His arms, without any at- 
tempt at accuracy of delineation, are doubled on his neck. The ej'es 
and mouth are like the same features in all aboriginal statuary, and 
beautiful shells were doubtless inserted in them. The ears have large 
openings in which were inserted plugs of wood, stone, shell, or 
feathers. The legs of the chair, just beneath the man's shoulders, are 
mere projections from the stone. The markings in the head and fore- 
head are quite tastefully designed. The back does not slope upward 
as much as in the Latimer specimens. In Dr. Liborio Lerda's Eldo- 
rado is figured a mummified human body seated on a stone stool in 
a cist. The figure in this paper and notes of im Thurn (Timehri, I, 
271) should be consulted. The impossibility of using such objects 
as mealing stones was pointed out by the author of these notes 10 
years ago, and im Tluirn adds the very pertinent argument that the 
ancient West Indians did not grind maize, subsisting mainly on 
cassava. Dr. Joseph Jones quotes Sheldon as saying: 'When a 
Carib died his body was placed in the grave in an attitude resembling 
that in which they crouched around the fire or the table when alive, 
with the elbows on the knees and the palms of the hands against the 
cheeks.' Length, 16 inches; width, 6i inches; height of head, Qi 
inches ; of feet, 2 to 3 inches." " 

" Mason, op. cit., p. 827. 

160658°— 34 ETH— 22 15 


By the author's interpretation the extended portion of this chair 
should not be confounded witli the horizontal portion or seat proper. 
We find the head, arms, and upjier bodj' of a human figure graven on 
the former, while an animal figure is represented in the latter. The 
anterior appendages or front legs of the seat have a rude head be- 
tween them, and the legs of the upright figure are not represented. 
The identification of the projection on the anterior border as a head 
conforms with other seats having a head carved in the same position. 

Labat describes how these grinders are used. Apparently, from 
his account, after the kernels of cacao have been roasted to a 
paste and cleaned of their skin in a wooden mortar, they are 
pounded on a stone to make the meal finer. "The stones," he 
writes, " used by them should be hard, a little porous in order that 
the fire placed below should heat them more readily: but they should 
not be liable to split, nor to calcine, and their grain should 
be sufficiently hard not to disintegrate, because it would spoil the 
taste. They should be polished with care, and cleaned, washed, and 
well rubbed immediately their use is finished. Ordinarilj' they are 
15 to 18 inches in width by 2| feet long. They are excavated their 
whole length, so that they are concave and are left 2 or 3 inches 
thick. They have at each extremity a foot about 4 inches square 
and 6 inches high to hold up the stone and raise it high enough from 
the ground to place fire under it." '* 

Labat, in speaking of the seats used by the Carib in bathing, 
hints at the form of the chocolate stones. He says (p. 109) : "The 
Caribs arise before day and a little before sunrise, withdrawing 
from the house for their necessities, which they never do near the 
houses, but in a place a little away, when they make a hole and 
subsecpiently cover it with earth. They immediately go into the sea 
to bathe if there is no river near; if there is, however, they do not 
go into the ocean. When they return they sit down on a little seat of 
one jriece of icood shaped like a chocolate si^ojie."'" 

The dead were buried in a contracted or sitting posture. Jefferys 
aays : ^^° " The corpse was not laid out horizontally, but seated on a 
little bench under a kind of wooden arch, to hinder the earth from 
falling in upon it. * * * But the bodies of the Caciques were 
not interred till they had been first well embowled and dried by fire." 


The pestles of the Porto Rico-Santo Domingo area are among the 
finest stone implements in the AA'^est Indies, and are readily dis- 
tinguished by the carvings on the head. In the Berlin Museum there 
is an undescribed pestle from St. Thomas (fig. 56) that approaches in 

"Labat, vol. vi. pp. 58, iJS. 

^^ Op, cit., p. 109. This would certainly imply that duhos and "chocolate stones" were 
practically the .same in some instances — a conclusion arrived ai l>y comparative studies. 
""French Dominions, pt. 2. p. 16. 




form the Porto Rico or Santo Domingo type. While the provenance 
of this specimen ma}' be doubtful, its presence in a Danish island 
would not be unexpected, if the Porto Rico-Santo Domingo culture 
areas embrace also the islands of St. Thomas, Santa Cruz, and 
San Juan. 

The St. Thomas pestle (pi. 114, A) has a well-developed disk, a 
thick handle, and a prominent ferrule, the tip of the handle being 
sculptured into a head with protruding lower jaw, sunken eyes, and 
prominent eyebrows marked with a fillet, which ends in slightly 
developed ears after arching over the 

As this specimen is the only one of 
its form described from St. Thomas, the 
author suspects that it was brought 
there from Santo Domingo. The Porto 
Rico pestles are generally destitute of 
carved heads on the handle. 

Plate 114, B, shows front and side 
views of a pestle from Haiti in the Ber- 
lin Museum. This specimen shows 
characteristics of the Santo Domingo 
type, the most marked of which is the 
presen?e of a ferrule at the 2^oint where 
the handle joins the enlarged lenticular 
base. The surface of the base is con- 
A-ex; the diameter of the handle at its 

union with the base is less than midway 
in its length. The end of the handle is 
enlarged, bearing a carved imitation of 
a human head, bod}', and retracted 

A pestle somewhat better sculptured 
than that last mentioned was found in 
Haiti and is now in the Berlin Museum. 
Its essential features are seen in plate 114, C. The resemblance of the 
head to that of a human being is better than that just described, and 
the limbs are more skillfully carvetl. Both of these pestles are in- 
ferior to some of those in the Merino collection, elsewhere de- 

Quite inferior in sculpture to the pestle in the Meriiio collection 
from Santo Domingo or those in the Berlin Museum, above described, 
is one in the Heye Museum, figured in plate 116, A. This specimen 
is somewhat smaller than the preceding, differing from it in the 

Fig. 00.— 

I'roblematical object 
likr' a pestle. (3.5 

'"Aborigines of Porto Eico, 25th Ann. Kept. Bur. Amer. Ethn. 



[ETH. A.NN. 3t 

absence of a ferrule and lens-shaped base. Its handle terminates in a 
head, from which project widely extended ears. The eyes are rep- 
resented but are not proininent. 

A glance at plate 115, B, shows a pestle more closely allied to 
those from Porto Eico than to those from Santo Domingo. The 
ferrule is absent and the disk thick, even massive. The head, as 
shown from the side, is separated by a deep groove from the handle. 

The typical form of pestles of the Porto Rico-Santo Domingo area, 
as repre&;ented by the preceding specimens, is easily distinguished 
from that of pestles found in the Lesser Antilles, one of which is 

Fig. 57. — Front and back views of head of an end of decayed pestle handle. 

shown in plate 115. C, />, from frcjnt and side. We have in these 
specimens no ferrule differentiating a lens from the handle, and, 
instead of the head being cut on one end of the handle, a face is en- 
graved on the side. 

The two heads, figure 57. represent front and back views of a 
broken end of a pestle in the Heye collection. The grooves incised on 
them form a strictly Antillean design, which, however, is better 
brought out in the form of the face, ears, and mouth. The object 
suggests Santo Domingan rather than Porto Kican art, the ends of 
the pestle handles from the former islands being much more elab- 
orately sculptured than those from the latter island. 



A , 4 iuches; if, 4.13 inches; C, 4.25 inches. 




























The fine pestle shown in figure 58 closely resembles those from Santo 
Domingo. It is elsewhere described and figured in tlie article, " Fur- 
ther notes on the archeology of Porto Eico,'""' 
and is now in the United States Xational Museum. 

This specimen was presented to the Smithsonian 
Institution bj- Seiior Don Juan Cabezas, of Caro- 
lina, Porto Eico, and, according to its label, was 
plowed up near his estate. It is made of a hard, 
smoothly polished stone, and is one of the finest 
examples of pestles from Porto Eico. Tlie handle 
is elongated, slightlj' ta^jering, with a well-carved 
head at one end and a lens with slightly chipped 
periphery at the other. This pestle, unfortu- 
nately, has been broken at the neck. The handle 
has no ferrule. The lips, nose, eyes, and ears are 
well carved in high relief. Each side of an eleva- 
tion on the crown of the head Iieare a ring-like 
protuberance unlike anything in other described 
pestles from this region. This specimen measures 
7J inches in length. 

The lobate stone shown in figure 59 belongs to 
a type the use of which is unknown. It has cer- 
tain relationshijas to the three-pointed type of 
idol, but its form is quite different, and it may 
have been used as a rubber. 

This object came from Haiti, according to the label in the jMuseum 
fiir Volkerkunde, where the specimen is now on oxliiliition. I have 

Fig. 58. — Stone pestle 
with heiid on end of 
thi> liandle. 

Fig. 59. — Problematical stone implement, proi 
ably when in use lashed to a wooden handl 
(4.5 inches.) 

personally placed it among the grinding implements, not, however, 
without some misgivings, as it seems to be unique among prehistoric 
Porto Eican or Antillean objects. 

' Amer. Anthrop., n. s., vol. x, No. 4. 1908. 



The aborigines of both the Greater and the Lesser Antilles prized 
many kinds of ornaments to decorate their heads and bodies. Tliey 
are described by contemporaries as a cleanly people, bathing often, 
and are even said to have built their habitations near the water so 
they could take frequent baths. They painted, stained, and tattooed 
their bodies and faces with pigments in elaborate designs and varie- 
gated colors. These designs were said to represent in some instances 
zemis. possibly totems. 

The early authors frequently speak also of stone pendants, shell 
or stone necklaces, gold earrings and gorgets, and other ornaments 
which they found on their bodies or in.serted in nose and ears. Speci- 
mens of these objects in the Heye collection have a great variety of 
form. These are generally made of stone, but are probably of the 
same form as those made of gold. 

Several ornaments here described have a crescentic form that sug- 
gests the " caracolis " referred to by Labat '' and others. The ac- 
count by Labat is instructive, although none of the prehistoric orna- 
ments were made of other metals than gold. 

"The ornaments most valued by them are caracoli, which are cer- 
tain plates of metal purer than lirass and less valuable than silver. 
It has the property of neither tarnishing nor rusting. This causes the 
savages to hold it in great estimation. Only the chiefs or their chil- ' 
dren wear them. It has been thought that the caracoli came from 
the island of Hispaniola, otherwise called St. Domingo, but the 
savages assert to the contrary and say they trade for them with their 
enemies, who call them alouagues through certain understandings 
they have among them who make presents to those from whom they 
I'eceive things. To know whence these Alouagues are obtained is a 
difficulty. They say the gods whom they adore, who make their home 
in frowning rocks, in inaccessible mountains, give them to them so 
that they may have greater reverence for this .sovereignty. If true 
I believe, however, that it may be that the devil abuses the feeble 
minds of these ignorant ones by this artifice. However that may be, 
these caracoli are very rare amongst them, and are brought from the 

"They are of different sizes; the largest are twice as large as a 
piastre. They have the form of a crescent. They wear them at the 
neck incased in wood. They wear bracelets of white beads, not at 
the wrist, but on the arm near the shoulder. They also wear them 
on the legs in place of garters. The women dress their hair like the 
men do but do not ever use feathers stuck in the hair and never 
wear a crown. They color themselves with arnotto as the men do, 
also wear bracelets as they do. but at the wrist and not on the upper 

'" NoHveau Voyage aux Isles de I'Amerique. Paris, 1742. 


arm. They carry necklaces of different kinds of stones, such as crys- 
tal, amber, greenstone, and beads. I have seen them with over six 
pounds hung at the neck. In their assemblies they wear belts of 
plaited cotton and chains of white beads. They hang to the different 
jjarts of this belt little bundles of six or seven chains of beads of a 
finger's length, and a large number of little bells, so as to make 
more noise in dancing. All the women and girls, excepting the slaves, 
wear from their earliest youth a certain half stocking wliich grasps 
the leg from the ankle to the calf of the leg, and another between the 
calf of the leg and the knee. At the top of the cotton stocking there is 
attached a kind of enlargement larger than a plate, plaited from 
reed and cotton, and a smaller one at the bottom than at the top, so 
that these two enlargements j^ress on the calf of the leg in such a 
manner and press out tlie calf that it looks like a Holland cheese 
between two plates. 

" The caracolis worn by the savages are made in the shape of cres- 
cents, according to the part of the body where worn. Ordinarily 
they are worn one at each ear. From one end of the horn to the 
other is about 24 inches. A little chain with a hook is held attached 
to the ear. Where they have no chains (for all do not have them) 
they are held by a cotton thread which is passed around the center 
of the crescent, of which the weight is like that of a piece of 15 sols 
(halfjiennies). They wear another of the same size at the space 
between the nose, which strikes on their mouth. The lower part of 
the underlip is pierced, where they attach a fourth caracoli which 
is a third larger than the preceding ones. Finally, they have a fifth 
one, which has an opening of 6 or 7 inches, which is incased in a little 
black wooden board centering in the crescent, falling on their breast, 
being attached at the neck by a cord. I leave it to be imagined 
what resplendence this gives to a man's head, and if it does not 
resemble a mule with his plates. 

" AVhen they are not wearing these caracolis they are careful to fill 
the holes in their ears, in the nose, and in the lips with little sticks to 
prevent them closing up. At such times they resemble hogs that have 
had pins to prevent them rooting up the ground. Sometimes they 
wear greenstones in the ears and in the lip, and when they have neither 
greenstones nor little sticks, nor caracolis, they put in them the 
feathers of parrots or red. blue, or yellow ("aras") paroquettes, 
which give them mustaches 10 or a dozen inches long on both sides of 
the mouth, both above and below, without counting that which they 
have in their ears, M^hich gives them the most pleasant countenance in 
the world." '' 

The Porto Eicans wore strings of gilded beads (fig. 60). crescentic 
stones, and earrings of shell or bone. The Heye collection, how- 

*= Op. cit., T. II, p. 85. 



[ETII. ANN. 34 

V ever, contains no specimen of a gold ornament; but pendants made 
of shell carved into amulets have heen described by one or fwo 
authors. Beads made of bone, shell, and stone are known from pic- 
tures of Porto Rican Indians in early writings. None of the gold 
beads of prehistoric times escaped the cupidity of Europeans, and 
those given them by the natives found their way into the melting 
pot. Beads made of stone, however, exist in several collections. 
Their form is cylindrical rather than globular, perforated, and often 
having a second hole at right angles for insertion of feathers. A 
necklace made of these beads had the appearance of a feather col- 
lar, as described by early authors. These stone beads are sometimes 
cut in the forms of animals or human beings. They were strung on 
a string side by side for a collar, similar to the necklaces figured by 

Fig. (JU. — Various furms of stone beads, plain anU decorated. 

Giglioli,*' Andree,'^ and other ethnologists. These ornaments must, 
howevef, be distinguished from amulets or zemis, worn on the fore- 
heads when they went into battle. 

In the Berlin Museum there are 40 shell sticks of brown and white 
color of a similar form. 

In several of the eai-ly accounts it is stated that every medicine 
man among the Carib carried a zemi, by which he was known. In 
some cases this was worn on a necklace, in other instances attached 
to his forehead, and still other forms were painted or tattooed on the 
body. These and other facts lead me to believe that in a way we 
may regard the zemi as a totemic sj'mbol, representing the divinized 
ancestor in much the same way as the Katcina among the Hopi. 


Tliere is in the Heye collection a unique stone object, plate 110, .1, 
of unknown use. from Santo Domingo, which is provisionally called 
a pendant and supposed to have been worn as an ornament. It is 
made of black stone and has the form of a cone pointed at one end 
and flattened at the opposite, or base, recalling a grinding stone. 
The conical end is, however, perforated and the hole beveleil on 

" Ropt. XVIth Int. Cong, of Anier., pt. 2, p. .310. 

" Baessler-Archiv, Band IV, pp. 31-32, Leipzig, 1914. 








A , 2.tBS iuches; B, 2.7 inehes; C, 2 inches: B, 3.1 inches; E, 2.0B inches; F, inches. 



-•i ,.i indies; /;, i-iilargcd; £, .i..'. inches; i^, 2.7.i inches; O, '^.4 inrhe.s. 


each side. An exceptional feature of this specimen is the four flat 
circuhir disks or knobs, one in each quadrant, attached to the sides. 
The bases of attaclinient of these disks adjoin, their diameter being 
about half the whole length of the object. This unique object is well 
polished and may be considered one of the finest specimens of Antillean 


One of the shell amulets in the Heye collection is the best made of 
these objects known to the author. It is represented in plate 116, Z?, 
and while in general features it resembles fetishes or amulets figured 
elsewhere,"^ it has certain features that are characteristic. The head 
of this amulet is well cut, showing mouth with rows of teeth, eyes, 
nose, and crest on the forehead. The forearms are fluted in such a 
way as to bring the hands below the chin with palms pointed out- 
ward. The body is elongated into a round shaft, which terminates 
in an angular, cubical enlargement, to which the posterior append- 
ages, the parts of which are not clearly indicated, are attached. The 
back of the head is perforated from side to side, on a level with the 
nose, indicating that the object was formerly suspended. The general 
form of this object and its perforations leads me to regard it a 
pendant worn with beads as a necldace about the neck. 

Several amulets (pi. 116, C) from Santo Domingo resemble those 
the author has already figured, but one or two are better examples 
and have characteristic features. The specimen shown fi"om face and 
back in plate 116, £', resembles that on plate lxxxvii of my Porto 
Rico memoir.^'' It has a perforation for suspension, which is shown 
in the figures. 

In another amulet (pi. 116. F) there is a disk-like addition to the 
head, recalling the tabla or table added to the heads of wooden 
images. This table addition appears to be tyjiical of several speci- 
mens, all from Santo Domingo. The ear projections are prominent 
features in several of these amulets. 

One of the most remarkable amvdets from Guadeloupe is now in 
the Vienna Museum, and was described by Prof. Franz Heger.'^" 

Tlie three views given in figure 61 show the form of this amulet 
from the front, back, and side. The exceptional feature is the exist- 
ence of four constrictions dividing the body of the specimen into five 
regions and imparting to it the form of a segmented animal like a 
worm or centipede. Each of the body segments has markings on the 
sides that might be mistaken for legs. They have the form of simple 
grooves, sometimes, as in the first segment, bifurcated. This like- 
ness to a centipede is enhanced by the form of a proboscis-like ap- 
pendage to the head and the form of eyes and mouth. The relative 

^Aborigines of Porto Rico, Twenty-flftli Ann. Kept. Bur. Amer. Ethn. 

^°" Aus den Sammliing<>n dfr anthropolojri^^cli-ethnosrrapliisclien Alitlieilun^ des k. k. 
naturliixtorisolicn Hofnuiseums in Wien, Mittlifil. dcr Anthrop Gesell. in Wien. Bd. ix, 
p. 132, pi. I, Wien, 1880. 



[ETH. ANN. 31 

position of the perforation by which this amulet was suspended is 
the same as those of other frontal amulets, which leads the author 
to include the specimen in this group. It is recorded that an amulet 
of a frog was found with a human skeleton on Guadeloupe. 

The museum of Prague, Bohemia, is said to have one of the 
finest amulets from the ^Vest Indies. The author knows of it from 
a cast in 'the Berlin Museum. This specimen (pi. 117. ^4) is made of 
a greenstone, probably jadeite, and has the form of a frog, the head 
and hind legs being cut in low relief. The perforations by which it 
was suspended appear in a view from the underside. The smooth 
surface on this side, combined with peculiar perforations so like 
those of other amulets, would indicate that it might have been worn 

on the forehead bj' warriors 
when they went into battle, 
as described by Gomara and 
&i] \ , I ill -- H Peter Martyr. 


The problematical object, 
plate 117, E, 2)reserved in the 
Berlin Museum, is made of 
bone, one end having a 
snake's head, the opposite ex- 
tremity being slightly flat- 
tened and bifurcated. A 
perforation through the mid- 
dle of the head would appear 
to indicate that it was worn 
on the body, possibly as an 
ornament suspended about 
the neck. The side view in- 
dicates that a section of 

r--^ I 


Fig. 61.— Amulet in Vienna Museum. Shown it is rOUnd or OVal, and that 

the flattened spatulate ex- 
from the body and head. 

from front, back, and side. (4.3S inches.) 

tremity is separated by a shoulder 
It is not unlikely that this specimen may have been used as a 
spatula in modeling, the e.xtremity ornamented with a head being 
held in the hand for that purpose, and the flattened end applied to 
the soft clay. 

Plate 117, F, shows an unusual spoon-like object made of bone, a 
unique form in the Heye collection. One surface is convex, the con- 
cavity on the opposite side or bowl of the specimen being decorated 
with incised lines on one end. Further resemblance to a spoon is 
lost from the absence of a handle, the end near the bowl on which 
the ornamentation appears being cut off sharply and replaced by a 
smooth surface. 


A bone swallow stick was found by the late Theodoor de Booj- in 
the Virgin Islands, and is now in the Museum of the American Indian 
(Heye Foundation). The eyes and mouth are inlaid with shell. 


The shell mask (pi. 117, C) in the Heye collection is ahnost iden- 
tical with that figured by the author in his Aborigines of Porto Eico. 
but is much better carved. It was purchased by the author from 
Senor Seiyo, of Arecibo. As in the Merino specimen, there are two 
holes near the rim for suspension by a band. These holes alternate 
with round elevations. There is also a perforation in the chin— a 
feature absent in the Merino specimen. On each cheek of the Heve 
specimen there is an engi-aved circle, recalling the elevations or wens 
on cheeks of the stone head in the Berlin Museum previou.sly referred 
to. The curved incised line on the forehead of the Heye"^ specimen 
ends in three extensions, recalling a hand with fingers. The anoma- 
lous position of these appendages, which do not occur in the Merino 
.specimens, is an objection to our identification of this as a fore limb. 

The Heye collection has an imitation of teeth made of shell. This 
ol)ject (pi. 117. B) was probably formerly inlaid in a wooden figure 
of some "West Indian god and represented its teeth. It has been 
suggested that these imitation teeth were used as an amulet, either 
suspended ar.iund the neck or carried with other fetishes in a special 
sack for tliat purpose.*" 

The object, the carved surface of which is shown in plate 117, D, is 
made of shell and is perforated in the middle. 


Disks made of clay or burnt earthenware often have incised figures 
upon them, suggesting that they were used as stamps for pottery or 
fabrics. These disks also, in some instances, bear knobs forming 
handles fashioned in the form of animals. One of these stamps, with 
the Iniob broken hut the incised legs of an animal still shown, is fig- 
ured in my memoir on the Aborigines of Porto Eico and Nei'ghbor- 
mg Islands." The incised designs on the surface of this disk rep- 
resent the legs of a frog, turtle, or other animal. The body antHiead, 
formerly in relief, have been broken off, but the author has seen speci- 
mens from Santo Domingo in which the whole body of a frog is rep- 
resented on one of these stamps. 

Clay Ctlindebs 

In essential features the clay cylinder (pi. 117. G) is unlike other 
objects of the same type el sewhere described; the figure incised on its 

I^^r't'-^l^i' "hj.'c( was tied to a belt in front, as recorded in older histories. 
"^ Twenty.fitth Ann. Rept. Bur. Amer. Ethn., pi. lxxxti. 



[ETH. ANN. 34 

surface is characteristic. This cylinder was probably used as a 
roller for stamping decorations on jjottery or fabrics, since iden- 
tical designs to those upon its surface are on the surface of 
earthenware vessels. Similar cylinders have been collected on 
other islands of the Santo Domingo-Porto Eico area, and have 
been reported likewise from the Lesser Antilles, where, however, they 
are generally replaced by disks with a knob or handle in the middle. 
These latter apparently serve the same i^urpose as the cylinders, al- 
though that has not yet been definitely determined. Another use to 
which these cylinders and circular stamps may have been put is sug- 
gested by a studj' of certain 
tribes of Venezuela and Guiana 
who are said to use similar ob- 
jects in stamping patterns on 
woven fabrics. Woven prehis- 
toric Antillean fabrics are rare, 
but a few show evidences of repe- 
tition of the same designs as if 
stamj^ed with cylindrical i-ollers 
or flat tablets. 

Circular stamps with central 
knobs are not rare in some of 
the Antillean shell mounds and 
village trash heaps. Some of 
these from Santo Domingo 
caves, as shown in the collections 
obtained by De Booy, are orna- 
mented with figures in relief. 
Other stamps have slender han- 
dles projecting from the middle 
of each end of the cylinder, 
at each end (fig. 62), imj^art- 
ing an ovate form to the object. One of these is elsewhere figured.*' 

Fig. tj:;. 

-Clay stamp or die with incised 

These cylinders sometimes taper 


The pottery from the Haiti-Porto Eican region is rather coarse 
as compared with that of the Lesser Antilles, especially Trinidad and 
St. Kitts. It is, as a rule, thick, its surface unpainted, but decorated 
M-ith incised lines, ridges, and raised figures that appear to have 
been added after the rest of the bowl had been fashioned. The in- 
cised lines are generally rectilinear, but are sometimes spiral, in the 
former case being arranged in triangular and in the latter in cir- 
cular figures. The spaces about the center, which is occupied by a 

» Xotis on Archaeology of Porto Rico. Amer. .\nthrop., n. s., vol. x, no. 4, 1008. 


circle or dots, are well filled. Straijrht lines often end in pits, which 
may Ije sejjarated from the ends of the lines. They were apparentlj^ 
made with a pointed hn2:)lement. 

Food bowls are not uncommon, their decoration being commonly 
a modification of lugs, or handles, into heads, which when broken 
off are often called idols by the country people. The margins of 
the openings are not turned outward or upward, but are without 
modifications. There are rarely necks to the bowls or feet or bases 
for them to stand upon. 

The earliest described whole piece of i:)ottery from Porto Rico 
was figured in 1907 in the article on the Aborigines of Porto Rico. 
Se^-eral additional specimens are now known in different collections. 
As shown elsewhere, there are two distinct types of Porto Rican pot- 
tery — one from caves (Cueva de los Golondrinos, not far from 
Arecibo) ; the other from the shell heaps and ball courts (batey). 
The vessel ^^^ from the mound near the ball court at Utuado is of 
rough ware, the handles slender, not as broad as those from the caves- 
If the reputed provenance of specimens can be relied upon, the pot- 
tery from inland caves differs from that of the caves alongshore — 
a condition that might be expected if caves were at times inhabited 
and later used for mortuary purposes. The pottery from Porto 
Rico is quite different from that found in middens of the Lesser 
Antilles; the incised or relief decorations are not the same. It is, 
in fact, possible to differentiate art designs from the different islands, 
although there is a general similarity in pottery forms froin Trinidad 
to Cuba, indicating a lilce cultural condition of the prehistoric inhab- 
itants, the modifications being largely due to environmental condi- 
tions, character of clay, and pigments used in decorating. 

The most abundant fragments of pottery from Porto Rico and 
Santo Domingo are lugs or handles of bowls and vases made in the 
form of human and animal heads. To consider all these fragments 
in any collection would be a great undertaking, and many such heads 
have been figured by students of collections from these islands. 

A\Tiile the pottery of Porto Rico and Santo Domingo in general 
characters resembles that of the other West Indies, it can be readily 
distinguished from that of the St. Kitts, St. Vincent, Barbados, and 
Trinidad areas. Many of the specimens in the Heye collection are 
fragmentary and are more or less worn on the surface by use, but as 
a rule they exhibit no evidence of a gloss which they formerly had. 
Even these fragments indicate that the ancient islanders were excel- 
lent potters, showing superior art and workmanship. As in the other 
islands, handles of vessels in the form of clay heads predominate, in- 
dicating that effigy bowls, vases, and jars were very numerous. 

*'" Aborigines of Porto Rico, Twenty-flfth Ann. Kept. Bur. Amer. Ethn., pi. Lxxvii, c; 
tile cliaracter of cave pottery is sliown in pi. Lxxin. 


The flat open-mouthed bowl (pi. 118, A) is destitute of lugs or 
relief figures and has the base convex, the outer surface being slightly 
inclined inward. This upper region is decorated with incised pat- 
terns, in which circles, parallel lines, crescents, and triangles may be 
easily recognized. Some of these lines are enlarged at their extremi- 
ties into pits, and the decoration is also helped out by sunilar pits, 
free from the end of the incised lines. The figures engraved on the 
upper zone of the bowl recall those on the decorated panels of certain 
stone collars from Porto Ilico and possibly are the same symbols. 

One of the best objects made of burnt clay from Santo Domingo 
in the Heye collection is shown in plate 118. B. This specimen is in 
fact one of the most beautiful examples of Antillean pottery j-et col- 
lected. Its general form is oval, the opening being about half the 
longer axis, the rim or lip of the orifice without decoration. Al- 
though the surface is worn, it shows evidence of decoration with in- 
cised lines which girt the vessel or are arranged in such a way as to 
indicate organs of the animal intended by the effigy. The head stands 
out in bold relief from the body of the vessel, from which it is sepa- 
rated by a well-developed neck. This head recalls in some features 
those of certain three-pointed idols already elsewhere figured. The 
representation of the fore limbs consists of four short parallel in- 
cised lines situated on each side not far from the junction of the 
neck and bowl. There is a triangular figure on each side, but not 
elevated above the surface of the jars. A similar triangular figure 
occupies a like relationship to each of the posterior appendages. The 
hind legs are represented by two club-shaped appendages which arise 
on each side near the pole of the bowl, opposite the head. These 
posterior limbs are represented in such a manner that the soles of 
their feet face outward toward the observer — a common feature in 
Antillean art. The sole of the foot is triangular in shape and has 
a slightly curved border and indications of four toes. 

The leg itself is short and stumpy, without joints. Its general 
form and situation as regards the remainder of the effigy recalls the 
flippers of a turtle or leg of a lizard — an identification not opposed 
by the general shape of the^ead and the existence of nostrils, so 
common in clay representations of reptilian heads. The tail is elon- 
gated, round, and enlarged at the extremity into a button, which is 
further marked with parallel incisions above. Around the enlarged 
base of its attachment there are two ferrules or raised sides separated 
by circular grooves, from the middle of which the tail seems to 
emerge. The existence of this tail favors a reptilian or turtle identi- 
fication of the vase. It is a feature unknown in effigy forms of bowls 
and other potterj^ objects. 

The bowl shown in plate 118, C, belongs to the Museum fiir ^'6lker- 
kunde in Berlin, to which it was brought from Santo Domingo. It 



A, B,W inches. 



ii,5.4inches; C, fiinches. 




has a globular form with orifice wide open and lugs in relief in the 
decorated zone. These vertical handles are perforated, but as the 
openings are too small for insertion of the fingers it would seem that 
the bowl was once suspended by strings in these lugs. The decora- 
tion is confined to the upper half and consists of alternating recti- 
linear parallel lines, four of which occur in the horizontal and three 
in the vertical areas. The lip opening is slightly bent outward. 

A small bowl (pi. 119, B) in the Heye collection is decorated with 
two heads, placed opposite each other, in relief, recalling some of 
those from the Erin Bay midden in Trinidad, British West Indies. 
It has a convex base, and the opening is only slightly less than the 
diameter of the bowl. The surface decoration consists of parallel 
grooves, whose extremities are separated by the heads in relief or 
lines at right angles to the same. There are several pits near the 
ends of the rectilinear grooves — a feature characteristic of pottery 
from Santo Domingo and Porto Eico. The heads are attached in 
liigh relief on the two opposite sides of the bowl, the rim flares out- 
w^ird, extending over a deep groove or neck that separates two areas 
of incised lines that form the main decoration of the outer surface of 
the vessel. The eyes and ears are represented by circular rings in 
low relief. The back of the head corresponds in curvature with the 
inner surface of the bowl. 

Tlie bowl shown in plate 119, C. from the Heye collection from 
Santo Domingo, has sloping sides, with a flat base. The exterior 
bends inward to form an undecorated lip, which also serves as the 
place of attachment of two heads, placed opposite each other, pro- 
jecting slightly outward and upward from the margin of the opening 
and forming lugs or handles. The eyes and other organs of the head 
take the form of dumb-bell appendages to the rim and are situated 
one on each, side of the hantUe. 

In an account of his trip to Santo Domingo Mr. De Booj- added 
several instructive forms of pottery to those already known from 
that island, thus increasing our knowledge of the ceramics of 
the Porto Rico-Santo Domingo area. Several of these water jars 
have nozzles ornamented with human faces and other organs as 
elsewhere figured and described.*' The nozzles of these specimens 
resemble those on vases in the United States National Museum pur- 
chased from Archbishop Merino and figured in my Aborigines of 
Porto Rico. 

The suggestion that these flask-shaped necks contained charcoal 
and were used for filtering water is unsatisfactory, but they may have 
been used in filtering the fermenfed juice fi'om some plant, as the 
yuca, a theory that can hardly be looked upon as more than a 

'" Aborigines of Porto Rico, TweDty-flfth Ann. Rept. Bur. AmpT. Ethn., pi. Ixxx, a, a', a". 


The additions made to the Heye Museum by ISIr. De Booy are 
very important, as they, belong to a few unrecorded types. The 
number of whole pieces from Santo Domingo in the Heye JNIuseum is 
greater than in any other West Indian collection. Many have a 
gi'eat value, because they were found in caves, the floors of which 
are now submerged. 

One of the most unique forms of Porto Eican pottery yet unde- 
scribed is a bowl in the iluseimi fiir Volkerkunde, in Berlin, said to 
have been found in one of the caves on the left side of the road from 
Arecibo to Utuado. This bowl has a plain surface without relief 
decorations, its upjDer zone having two encircling incised grooves 
inclosing incised figures, placed horizontally and vertically, as shown 
in figure 63. 

In figure A, plate 119, from a specimen in the Heye collection, we 
have a very elaborate neck, to "which is added heads in relief, all 
that remains of a large vase, the body of which has been lost. Its 
style is Santo Domingan ratlier than Porto Eican. The well- 
preserved specimens of an exceptional type of pottery from pre- 
historic Porto Eico figured by Dr. Haeberlin should be mentioned. 
One of these specimens is particularly instructive as having a raised 
base, which feature, although known for many yeare ^'' from the 
Lesser Antilles, has not previously been recorded from Porto Eico, 
the bowls and vases hitherto described from that island having a 
rounded or flattened base. 


The island of Cuba, the largest in the Greater Antilles, has up to 
a few years since yielded a smaller number of archeological remains 
than Porto Rico, Jamaica, or Haiti. This is due to want of explora- 
tion, as shown by the lai'ge collections made by Mr. M. R. Harrington, 
of the Museum of the Ame'rican Indian (Heye Foundation), who has 
made extensive collections in the island since 1914 and is preparing a 
memoir on this subject. 

The eastern end of Cuba shows evidences of a higher culture than 
the western, as if affected more by influences from Haiti and Porto 
Rico. This is especially seen in jjottery obtained from caves of the 
eastern end. Apparently the western extremity, toward Yucatan, 
which lies nearer to the continent than any of the Greater Antilles, 
was inhabited by a race of low comparative culture who spoke a 
different language from the inhabitants of the eastern extremity. 

""One of these bowls with liasal ring from Trinidad is figured in Aborigines of Porto 
Itiee, Twenty-fifth Ann. Rept., Bur. Amer. Etlm.. pi. Lxxxv. 




The archeology of Cuba has been studied by several scientific men. 
and we have valuable contributions by Morales, Dr. Montane."^ Poey. 
Carlos Torre, Brinton, and others. The author published a few 
years ago the following article which adds a few new observations 

Fig. fi3. — Bowl with iniised decoratiou. (9 inches.) 

on a subject which ijromises great results after systematic explora- 

o^lhe article " L'Homme de Sancti Spiritus (He de Cubaj" (Extrait du Compte Rendu 
dn xiii Congres international d'AnthropoIogie et d'.Archeologie prehistoriques, Session de 
Monaca, 190t;i contain.s valuable information on the archeological olijects in the Uni- 
versity of Ilabana. For the Indians of Cuba see Culin, in Bull. Free Mus. of Sci. and 
Art, vol. Ill, no. 4. 1902. 

1606.58°— 34 ETH— 22 16 

242 ISLAXn CVLTrKE area of AMERICA [eth. ANN. 34 

Prehistoric Cultire ok Cuba "- 

Although the early Spanish writers ascribed to Cuba a large abo- 
riginal population, thej' recorded very little regarding racial differ- 
ences of natives in different parts of the island. The majority, con- 
sidering the inhabitants as homogeneous in culture, paid little atten- 
tion to variations in language or to diversity in mode of life, while 
later authors, who are few in number, have added little to earlier ac- 
counts. Archeological investigations, to which we must now look for 
more light on this subject, have thus far been limited, and our 
museums are very poor in prehistoric Cuban objects. Few speci- 
mens are known to have been found in the province of Pinar del Kio, 
or the western end of the island, and local collectors are unanimous 
in saying that all the aboriginal objects they possess came from the 
eastern extremity. This limitation is significant, especially when we 
consider that Yucatan, where the natives attained high culture, is 
such a short distance from the western end of Cuba, and that it was 
from the Cubans that tlie Spaniards first, heard of the highly de- 
veloped Indians of Mexico. The present paper, based on studies and 
collections made during a brief visit to Cuba in 1904, suggests an 
explanation for this paucity of prehistoric objects and the limitation 
of the localities from which those known have been obtained. 

I A study of the available evidence, both documentary and arche- 
ological, shows that the aboriginal culture of Cuba differed in differ- 
ent ijarts of the island. Some of the inhabitants reached a compara- 
tively high degree of culture development, others were rude savages. 
The former had polished stone implements and knew how to make 
the fertile soil yield their food supply, but the latter were naked 
cave dwellers, who gathered for food roots or tropical fruits that 
grow spontaneously in the rich soil of the island. There were also 

_*fishermen, who subsisted on a natural supply of the products of the 
sea when their habitat made it possible. Contact with people of 
high culture had raised them somewhat above the dwellers in the 
mountains to whom they were related. 

Columbus commented on the resemblance of the aborigines of Cuba 
to those of the Bahamas, regarding tliein the same in language and 
customs; but this supposed identification was true only in a very gen- 
eral way. The diary of the first voyage of the discoverer, as found 

I in the writings of Las Casas, affords no direct evidence of a more 

' primitive race in Cuba, although it suggests the theory that such a 
people existed. 

Historians do not agree as to the first landfall of Columbus in Cuba, 
but no one doubts that it was somewhere on the northern shore of 
wliat is now Santiago Province. At whatever point he landed he 

"Reprinted from Amer. Anthrop., n. s., vol. 6, pp. 585-59S, 1904. 


found the natives living in houses, making use of hammocks of cotton 
and palm fibers, and possessing stone idols and carved wooden masks. 
Cohmibus learned from them of a ruler, whom he called king, of a 
country to the south, which was rich in gold. Xothing is said in his 
diary of the natives to the west of the landfall, but he sailed westward 
a few leagues along the northern shore without finding people worthy 
of special mention. Later, turning Iiack, he rounded Cape Maysi 
and e.xamined a section of the southern coast, but was not attracted 
farther toward the west. On this side of Cuba he again heard of 
the wealth of the Indians of the south. The implication is that the 
people of eastern Cuba Iniew of the Haitians and recognized that their 
culture was superior to that of the western end of their own island. 
They held out no inducement to Columbus to extend his explorations 
westward, as we might susj^ect they would have done had there been a 
superior race in that end of the island. 

The great Genoese returned to Cuba on his second voyage and 
explored the entire southern shore. Bernaldez, to whom we owe 
an account of this visit, scarcely mentions the Indians in this part 
of the island, although he describes the Jamaicans in .some detail, 
regarding them a highly developed race. Many native fishermen 
were seen along the shore, but they were evidently lower in develop- 
ment than the Jamaicans, whose canoes (according to Bernaldez) 
were painted, better made, and more luxuriantly ornamented than 
those of the Cubans. 

Numerous references might be quoted from the writings of those 
who followed Columbus, showing that the prehistoric customs and 
languages of the natives of the eastern and western ends of the island 
were not the same. In the judgment of many of the Spanish con- 
querors, among whom Diego Velazquez may especially be mentioned, 
the natives of Cuba were more susceptible to Christianity than the 
other West Indians, but they say that this docility was not true of 
all the Cubans, some being less tractable than others. The extreme 
western end of Cuba was said to have been inhabited by barbarous 
Indians similar to those living in Guacaj'arima,"^ the Province at the 
western end of Haiti. The Spanish writers declare that these natives 
could net speak ; by which is probably meant that their language was 
different from that of any other Indians of these islands. Bachillery 
Morales says that the Guanahatebeyes (Guanacahibes), who lived in 
the interior of Cuba, were savages who did not treat with the other 
Indians. He adds that they lived in caves, which they left only to 
go fishing, and quotes from older writers''^ that there were other 

™ A town on the island of Trinidad, where survivors o( the Indians still live, is called 
Arima. There is another Trinidad village crilled Naparima. 
" Cuba Primitlva, p. 2S0. 


Indians called Zibnneyes, a tribe that included the inhabitants of the 
islands off the northern and southern coasts, called the Gardens of 
the King and Queen, who were enslaved by the other natives. 

According to La Torre "^ the Indians of Cuba form one of the 
natural groups of the Tainos and are generally Icnown by the name 
Siboneyes. They inhabit, he says, the whole island and have the same 
customs, although in certain parts of Cuba there are backward tribes, 
as the Guanacahibes of Cape San Antonio. The original authority 
for these statements is found in the Munoz collection, and reads as 
follows r'*" 

" Lo mismo podra hacerse con los indios de los Jardines del Rey e 
de la Eeina, que son muchos islotes de indios que no suelen comer sine 
pescado solo. E estos se les dura menos trabajo, pues no estan acos- 
tumbrados sino a pescar. lo niismo se entiende para unos indios al 
Cabo de Cuba, los cuales son salvajes que en ninguna cosa tratan con 
los de la Isla, ni tienen casas, sino estan en cuevas continuo, sino es 
cuando salen a pescar; Guanahatabeyes otros hay que se Uaman 
Cibuneyes, que los indios de la misma Isla tienen por servientes e 
casi son ansi todos los de los jardines." 

Diego Velazquez, the conqueror, wrote''' to the King of Spain in 
1514 that there were two provinces in the western part of Culja. 
and that one of these was called Guaniquanico. the other Guanahata- 
bibes. The latter was situated at the western extremity, where the 
natives lived as savages, having neither houses nor farms, subsist- 
ing on game captured in the mountains or on turtles and fishes. 
Peter Mai-tyr d'Anghera says that the inhabitants of the Haitian 
Province of Guacayarima, to which these Indians are said to have 
been allied, lived in caves and subsisted on forest fruits. 

Gomara'^ mentions the fact that the inhabitants of different j^arts 
of Cuba have different languages, and says that both men and 
women wear little clothing. He thus writes of a peculiar custom 
which they practiced in their nuptials : 

" Si el Xovio es CaciQue tcilos lo.s Caciques (■(iniliidiidos (lueniien cnii la Xovia, 
priiiiero que no el ; si iiiercadnr, los iiiercailiires ; i si lahrador, el Senor o algun 


The earliest contribution to the archeology of Cuba we owe to 
Senor Andres Poey, who in 1855 read before the American Ethno- 
logical Society a paper entitled "Cuban Antiquities: A Brief De- 
scription of some Eelics Found in the Island of Cuba." Although 

"5 Manual 6 Guia para los exfimencs dp los Maeatros y Maestras, p. 45. 

^"^ Vol. LXV. See also Ferrpr, Naturaleza y Civiliaucion de Cuba. vol. ii, p. 142. 

■^ DocuniPntos infditos del Archive de Indias, vol. xi, pp. 224, 225. 

■" lUstoria, chap. 51, p. 4t. 


Brinton''^ says this paper was not published in English, Senor J. Q. 
(Jarcia. in 1855, edited what lie calls a Spanish translation of it in 
the fourth volume of his Eevista de la Habana. 

The figures accompanying this article include two stone images, 
a few clay heads copied from Charlevoix,^ and a stone pestle taken 
from Walton. = The stone images are from Cuba, but the pestlo 
and the clay heads came from Santo Domingo. The images more 
especially concern us in this article. One of these, called an idol, is 
made of a hard stone of reddish color, highly polished, with a head 
cut on one end. Poey believes it was originally covered with n 
varnish which has been worn off in exposed places. He is probably 
right in this conclusion, for remains of a resinous substance which 
once covered some of the three-pointed stone idols from Porto Eico 
still adhere to several specimens. This so-called idol lias the gen- 
eral form of a celt, although it differs in details from the cere- 
monial celts which have thus far been described as from the West 
Indies. It is now in the Archeological Aluseum at Madrid. There 
is no doubt that the other image, described and figured both by 
Poey and Ferrer,^ is an idol. The former likens its attitude to that 
of a dog resting on his hind jijarts, the forelegs crossed over the ab- 
dominal region. This specimen is now in the Univei-sity of Hal)ana, 
to which institution it was presented by Ferrer. 

The form of this idol is different from that of idols from Santo 
Domingo and Porto Eico, but its technicjue indicates an equally 
high development in stone working. 

In a brief article of four pages, Brinton, " without aiming at 
completeness," gives a review of the labors and results of students 
of the archeology of Cuba. He calls attention to some of the con- 
tributions of Poey, Ferrer, Garcia. Pi y Mai-gall, and others, and 
shows that the archeology of Cuba " has not been wholly neglected 
by intelligent Cubans, although it is true that there has been little 
serious investigation of the remains." He considers that " the most 
promising localities for research would seem to l)e the extreme 
eastern and western provinces, Santiago and Pinar del Eio. In the 
caves of the latter we should, if anywhere, find traces of the Mayan 
culture." ^ 

"» The Archfpology of Cuba, American Archseologist, vol. 2. No. 10, Oct., 1S98. 

* L'lnstoire de I'lsle Espagnole ou de S. Domingue, Paris, 1730. 

=* Present State of the Spanish Colonies. Including a Particular Account of Hispanola, 
London, ISIO. Mr. Walton finds in these pestles evidences among tJie Haitians of phallic 
worship like that of the Hindoos, and Poey devotes considerable space in his articles to a 
discussion of thi.s theory, which he supports. The comixirisons of this pestle to the yonl 
and lingam appear to me to l>e strained, esjiecially when we examine a series of tliese 
objects, some of which represent birds and other animals. 

= Congreso Internacional de .\mericanistas, ,\ctas de la Cuarta Reunifin, Madrid. ISSI, 
p. 245. 

* Brinton says that according to Ferrer there are caves along the Rio Cuyaguatege. in 
Pinar del Rio, in which the aborigines interred their dead. Archaeology of Cuba, Amer. 
Arclia;ol,, vol. Ii, no. 10, 1898, p. 256. 


According to Brinton.^ Senor Garcia gives in " one of the num- 
bers of the Eevista de la Habana an illustration of what is called 
a duchi, which is tlie common term in Cuba for the figures of stone 
or claj' attributed to the aborigines. This particular clAichi was a 
stone ring, with eyes and ears of gold, and was supposed to have 
been the seat or throne of a chief, but probably was a stone collar." 
The author has not been able to find this illustration in the Eevista 
de la Habana. although he has examined and copied Garcia's two 
articles which he claims to be translations of Poey's paper read before 
the American Ethnological Society, which have not been seen. 

Brinton's suggestion that this duchi was a stone collar does not 
appeal strongly to me, for the term duchi, duho, or dujo was given 
by the West Indians to native seats or stools in the form of ani- 
mals with eyes and ears of gold.'^ 

According to Bachiller y Morales,' D. Tomas Pio Betancourt, in 
his Historia de Puerto Principe, says that D. Pedro de Parrado y 
Pardo, in a book on the genealogy of families of Bayamo, written 
in 1775, gave the name duho to one of these seats, in possession of 
Dona Concepcion Guerra, that formerly belonged to the Cacique of 

■ I am unaware that the following statement hy Brinton^ has ever 
been verified : " I have also learned," he writes, " of a locality, which 
T will not now further specify, in central Cuba, a river valley, along 
which, from time to time, one meets grim faces carved from the 
natural rock, and sometimes monolithic statues, the work of the 
aborigines and believed to represent the guardian spirits of the 
river. This locality I hope to have visited by a competent pereon 
this winter." A verification of these statements and a description 
of these supposed '' monolithic statues," with figures of the same, 
would be an important contribution to Cuban archeologj'. It 
would also be interesting to know whether the river valley where 
they are reputed to have been found was in the eastern or the west- 
ern provinces of the island. 

At the Madrid session of the International Congress of Ameri- 
canists, in 1881,'^ Sefior Rodriguez-Ferrer read a paper in support of 
the theory that there was evidence of the existence, in prehistoric 
times, of Cuban aborigines different from those discovered by 
Columbus. The thesis is defended mainly by facts drawn from 
crania found in caves, but the two archeological sjjecimens which 

"Op. cit., p. 253. 

« So far as known, stono collars and three-pointed idols, which characterize Porto Rican 
aboriginal culture, have not l>een found in Cuba. 

'Cuba Primltiva, p. 268. 

' .\rcha>aloj;,v of Cuba, p. 2.15. 

^ Congreso Internacional de Aniericanistas : Actaa de la Cuarta Reunion, Madrid, 1881, 
vol. I, pp. 224-267. 


he elsewhere describes and figures are also brought to the support 
of this theory. There is nothing to show that this cave people dif- 
fered in any respect from those to whom early writers allude as 
living in the central and western parts of the island. All the 
evidence appears to support the tlieory that some of the natives of 
Cuba lived in caves at the time of the discovery, and the conclusion 
is natural that they were the lineal descendants of the oldest race 
which they resemble in bodily and cultural characters. 

Senor Rodriguez-Ferrer, in his valual)le work.'" referring to the 
letter of Las Casas and to other evidence published in the Docu- 
mentos Ineditos del Archivo de Indias (vol. viii, p. 34), points out 
certain differences in the culture of the natives in different parts of) 
the island, which are jiractically the same as those indicated byj 

An important addition to our knowledge of Cuban archeology 
was made by Don Eusebio Jimenez," who, in October, 1850, exca- 
vated some mounds in the central part of the eastern end of the 
island. According to J. de J. Q. Garcia these importaiit remains 
were found on the farm of D. Francisco Rodriguez, nearly 5 miles 
southwest of Moron. Various utensils and objects made of hard- 
wood, stone, and burnt clay were recovered from these mounds. The 
description which Garcia gives of the excavations leaves no doubt 
that these mounds, called canej^s. were aboriginal burial places, and 
they suggest the existence in the neighborhood of one of those dance 
places called cercados de los Indios, or juegos de bola, which occur 
in Santo Domingo and Porto Rico. One of the best known of these 
aboriginal iiulosures in Cuba is the so-called Pueblo Viejo, situated 
in the eastern end of the island, near Cape Maysi. Although this 
inclosure has been described by several writers, no one has yet 
called attention to its resemblance to the dance inclosures of the 
neighboring islands. 

It is evident from the contents of the numerous caves that have 
been excavated by Dr. Montane and others in Santiago and Puerto 
Principe Provinces that cave men lived in those provinces after the 
introduction of a higher culture from the neighboring islands. 

Although there is considerable literature on the somatology of 
the Cuban Indians, especially on crania found in caves, a considera- 
tion of this subject is foreign to the scope of the present article, 
which is devoted mainly to the consideration of evidences of the 
existence of a high and a low culture in Cuba at the time of its 
discovery. The crania found embedded in calcareous rock in caves 
near Cape Maysi and elsewhere on the eastern end of the island 

"I Naturalez.i y Civilizacion df la grandiosa Isla do Cuba, Parto Segunda — Civilizaclon, 
Madrid, 1887. pp. 142-144. 

"^ See El Perlodico de Puerto Principe, and Faro Industrial diarlo de la Marina. 


have been amply described by anatomists, antl are liighly instruc- 
tive in a consideration of the antiquity of man in Cuba, but I am 
not yet ready to express myself fully on their significance. The 
natural inference would be that these skulls support the theory of 
ancient cave man in Cuba, of whom the Guanahatabeyes were the 
survivors in the fifteenth centurj-; but West Indian caves were used 
as burial places after the discovery, and no one has yet satisfactorily 
shown any great difference in the crania embedded in rock from 
those found under usual conditions in the caves. ^- 

Dr. Enrique Gomez Pianos, in his valiuible work on prehistoric 
Cuba,^^ mentions several caves on the island from which human 
remains and pottery have been taken, and gives an interesting 
resume of Cuban archeology without adding much that is new to the 

Another work containing considerable material on Cuban eth- 
nology is that of Bachiller y Morales,^* a most valuable compilation, 
but very carelessly edited. It contains much information in regard 
to the aborigines of the Greater Antilles, but the title " Cul)a Primi- 
tiva" is somewhat misleading, for while it contains chapters on the 
subject of primitive Cuba, the larger part of the book deals with 
Haiti and Poi'to Rico. 


The distribution of polished stone objects in Cuba may be said 
to confirm the historical accounts of a difference in culture between 
the inlialiitants of the eastern and those of the western provinces. 
Those of Santiago resemble objects from Haiti and Porto Rico, but 
no similar implements are found in Pinar del Rio at the western end 
of the island. 

There are two collections of Indian objects in Habana which 
contain objects of interest to the archeologist. One of these, the 
smaller, is in the museum of the Academia de Ciencias, on Calle de 
Cuba ; the other is in the university near Vedado, a suburb of 
the city. Both collections are under the directorship of Dr. Luis 
Montane, who has conducted excavations in several caves of the 
island and has in prejoaration a memoir on the subject. The collec- 
tion at the university is particularly rich in crania from caves, and 

■^For an account of these remains see Anales de la .\cademia de Ciencias, vol. xxvii, 
Habana. 1800. 

'^ Prehistoria de Iji l.sia de <Mili;i,. Analc>s de la AcJidemia de Cienoias. vol. xxx\ii. 
Habana, August-December, 1900. 

'* Cuba Primitiva : Origen, Leoguas, Tradiciones e Historia do los Indios de las 
Antillas Mayores y las Lucayas, 2d edition, 1SS3. In his paper on the Archeology of 
Cuba (1898) Brlnton thus refers to this valuable book, fifteen year.s after its publication: 
" The announcement of it, which is before me, dated ' Havana, 1881,' states that it will 
discuss the antiquities of the island, and the traditions and languages of its early inhabi- 
tants. Whether it was published or not I have not learned." 


contains several interesting objects, descriptions and figures of 
which have not been published. Dr. ilontane has kindly shown 
me many jjhotographs and charts illustrating his explorations, and 
has courteously permitted me to photograph some of the more 
striking objects, including a stone collar from Porto Eico.'^ The 
majority of the archeological specimens came from the eastern end 
of the island and closely resemble in technique those from Porto 
Eico. Among the objects seen in these two collections are ID i>et- 
aloid celts in the academy museum and about double that number 
at the university. One of those in the latter collection has a stone 
handle like those obtained by me in 1903 in Santo Domingo. 
There is also a celt witli a face cut on one side — evidently a cere- 
monial celt like one in Archbishop Merino's collection. This like- 
wise is a product of Tainan culture, as is the stone pestle with a 
well-fashioned head on the eml of the handle. 

The three choicest specimens in Dr. Montane's collection are a 
wooden idol, a stone tui'tle, and a shell with a face cut on one side. 
The wooden idol has a perforation, as if for attachment to a staff, 
and may have been used in ceremonial dances like those of the 
Salivas and other Orinoco tribes described by (xumilla.^" The turtle 
of stone recalls one of wood collected by Ober in a cave in St. 
Vincent in 1878 and now in the Smithsonian collection, but, unlike 
it, the latter is not perforated for attachment. An account of these 
objects in the university museum, with localities and figures, would 
increase our knowledge of the archeology of Cuba. 

In the Santiago museum were two idols made of coral rock, one 
(fig. 64) of which, according to the label, is from Cueva de Boruga, 
near Baracoa, the other (fig. 65),^' which is smaller, from the Loma 
del Cavuco, Gibara. The former was loaned to Prof. W J McGeo, 
Ethnologist in Charge of the Bureau of American EthnologA-, by 
Seiior Quesada in 1900, at which time Mr. DeLancey Gill made 
front and profile photographs of the specimen. This idol represents 
a seated figure, with elbows on the knees and hands to the breast. 
Its whole appearance is different from that of any West Indian idol 
that I have ever seen.^' 

Tlie smaller idol, also of coral rock, shows the septa of indi- 
vidual coral animals scattered over the surface, and has the form 

'5 The idol presented by Senor Ferrer (flg. 66) is historically the most interesting in 
this collection. 

'• EI Orinoco, ilustrado y dHfendido, Madrid, 1745. 

'• Prehistoric Culture of Cuba, Amer, Anthrop., n. s. vol. 6, No. 5, 1904. 

'^ The exceptional form of this idol, when compared with those from Santo Domingo 
and Porto Rico, may lead some archeologists to doubt its authenticity. The form of the 
mouth, however, is almost identical with that of the head of a pestle from Santo 
Domingo, and the attitude recalls that of the wooden idol in the university museum at 



[ETH. ANN'. 31 

of a pestle, the arms beingj obscurely indicated and the legs being 
replaced by a base upon which it stands. 


Via. 04. — Idol of roial nick fi-om Citova lie Borii?a. P.araooa. Ciiha (Santiago Musrum). 

The idol shown in figure 6G is the same as that elsewhere mentioned 
as presented to the university museum by Seiior Ferrer and figured 
by him and by Seiior Andres Poey. 

The cei'emonial celt, also in the Santiago museum, has a rude head 
cut on one end and arms carved in low relief on the sides. This 




specimen is said to have been found at tlie Indian town of El Canev ; it 
belongs to the same type as the ceremonial celt described and figured by 
Poey, Ferrer, and others. Its general character allies it to stone prod- 
ucts of the Tainan culture of Santo Domingo and Porto Eico. 

The celts collected liy me in Cuba have the same forms as those 
from the other West Indian Islands, and are known to the country 
people by the same name. piecJras del rayo, or thunderbolts (fig. 67). 
They are petaloid in form, 
smoothly j^olished. and 
without grooves for haft- 
ing. As in Porto Eico, 
there is considerable folk- 
lore in Cuba connected 
with these implements. 
Twentj' petaloid celts were 
collected in the neighbor- 
hood of Santiago at El 
Cristo, El Caney, and the 
outskirts of the citj'. 

While in Santiago I pur- 
chased a small collection of 
Indian objects from Nipe 
Bay, on the northern coast 
of Cuba, which includes 
petaloid celts, fragments 
of pottery (fig. 68), a shell 
implement, and other ab- 
original objects. Among 
the last is a water-worn 
stone on which is cut in 
outline (more like a picto- 
graph than in relief) a 
human face with mouth, 
eyes, and what might have 
been intended for a nose 
(fig. 69). The specimen is unique in form, and although not flattened 
on one side, in certain particulai's it reminds one of the so-called stone 
masks of the ancient Porto Eicans. The chief characteristics of this 
outlined face are the oblique eyes and the three curved lines extending 
from their lower ends to the incised line which borders the face. Its 
use and significance are unknown to me, nor am I familiar with any 
similar specimen from the other West Indian Islands. It will be 
observed that this and the following specimens came from the eastern 
end of Cuba and can be referred to the Tainan culture. 


1. — Idol or pistil' I'lom Loma di'l Cayuco 
(Santiago Museum). 



ETH. ANN. 31 

In character the pottery from Cuba is practically the same as that 
from Porto Rico. The collection made by me consists wholly of 
frajrments of clay heads from bowls or vases. The specimens shown 
in figure C8 were obtained from Nipe Bay on the northern coast, 
but I have seen almost identical fragments from Pueblo Viejo/' the 
dance inclosure near Cape Maysi. 

The archeological evidences of a low culture stage in the western 
provinces of Cuba are thus far negative, for no objects which can be 

ascribed without question to the 
aborigines have yet been found 
in those parts. The known pol- 
ished stone imijlements, idols, 
and like objects from Cuba re- 
semble those characteristic of 
tlie Tainan culture, and are con- 
fined to the eastern end of Cuba. 
Naturalists have long recog- 
nized a marked difference in the 
fauna and flora of the two ends 
of Cuba. The prehistoric cul- 
ture of these two localities was 
also different. 


It appears from both histori- 
cal and archeological evidences, 
so far as now known, that the 
Tainan or Antillean culture 
which was found in eastern 
Cuba did not originate on that 
island, but was introduced from 
Porto Eico or Haiti, where it 
reached its highest develop- 
ment. The germ of this culture 
came to both these islands from 
South America, but had grown 
into a highlj' specialized foi-m 
in its insular home. There were minor differences in the different . 
islands — Cuba, Jamaica, Santo Domingo, and Porto Rico — but these 
differences were all modifications of the polished-stone age. 

There was considerable likeness in culture between the inhabit- 
ants of the keys of Florida and those of the Cuban coast and the 

'" This inclosurf has been described by Ferrer and others ; but one of the best .iccounts, 
and I he only one in English that is Isnown to me, is by Mr. Stewart Culin : The Indians of 
Cuba, Bulletin of the Free Museum of Science and Art, University of Pennsylvania, vol. 
Ill, no. 4, 1902. 

l'"ia. 00. — Stone 

idol (University Museum, 



small adjacent islands, due 
either to early contact of these 
two peoples or to migration 
from one to the other localitj- 
in limited nmnbers. The In- 
dian villages of Carahate (near 
the site of the modern Cuban 
town of Sagua la Grande) and 
Sabaneque (near Eemedios) 
were pile dwellings,^" not unlike 
those of the Indians now inhab- 
iting the delta of the Orinoco 
and the shores of Lake Mara- 
caibo in South America; but 
these adaptive conditions do not 
necessarily show kinshiij, and 
more probablj- were of inde- 
/ pendent origin. The resem- 
' blances between Floridian and 
Cuban coast peoples were due 
to contact and interchange of 

There were at least two dis- 
tinct stages of culture in abo- 
riginal Cuba. The natives in 
the first stage were savages with 
few arts, but those of the second 
stage were as highly developed 
as any of the West Indian abo- 
rigines. The one was an archaic 
survival, the other an intro- 
duced culture which originated 
outside the island. 

The people of the first stage 
were survivors of the earliest in- 
liabitants of the island, but they 
have left little to the archeolo- 
gist to indicate the status of 
their culture; nevertheless, it 
was evidently of a very low or- 
der. The natives of the Cuban 
coast and of the numerous small 
islands were fishermen. Their 

'' These houses built on piles were 
called barbacoas. The polygonal or cir- 
cular house with conical roof was known 
as a caney, and the quadrangular dwell- 
ing, with two-sided roof, a bohio or bnjio. 

Fig. 6i. — Petaloid celt (Santiago Museum). 



lETH. ANN. 31 

culture was higher than that or the others referred to, but the highest 
prehistoric cultui'e was confined mainly to the eastern jjrovinces, 
esjjecially Santiago, and was ap^Darently introduced from Haiti, 
where it reached a high development, although even in the mountains 
of that island there were survivors of the savage, or lower, culture 
stage which j^redominated in western Cuba. 

The Carib, who occupied the Lesser Antilles from Trinidad to 
Porto Rico, were the last of the several South American tribes 


68, — Fi-agmrnts of pottery from Nipe Bay (U. S. National 

which invaded the West Indies. This virile race at the time of the 
discovery had conquered and assimilated the original inhabitants of 
the Lesser Antilles and peopled them Svith a comiaosite people. The 
evidence that the Carib settled on the coast of Cuba is not decisive. 
They probably visited the island in their marauding expeditions, 
but they contributed little to the existing culture of Cuba or that of 
the neighboring peninsula of Florida. 




The Indians of Cuba, like those of Haiti, Porto Rico, and the 
Bahamas, were harassed by the Carib from the Lesser Antilles, but 
it is yet an open question whether tliese marauders had settled in 
any considerable numbers on the island at the time of the dis- 
covery. The inhabitants of the extreme eastern end of Cuba, like the 
Ciguayos, who occupied the region from Puerto Plata to Higuey, 
from exposure to the inroads of the Carib had become more war- 
like than the other people of Cuba, but this does not necessarily 
mean that they were Carib, as some writers appear to believe. 
The discovery of flattened skulls in caves near Cape Maysi, and 

their identity in this re- 
spect with deformed Carib 
crania from Guadeloupe, 
does not prove identity of 
race. According to Dr. 
Carlos de la Torre,-' the 
explorations of Sefior Mi- 
nfuel Rodriguez -Ferrer. 
Valdes Dominguez, Mon- 
tane, and himself tend to 
confirm the opinion of 
Rafinesque that the Caribs 
had settled south of Bara- 
coa. but the evidence pre- 
sented in support of this 
theory is not conclusive. 

The original coloniza- 
tion and prehistoric cul- 
ture of Cu^a must com- 
prehend three different 
conditions of aboriginal 
life, practically three dif- 
ferent peoples — viz, the 
primitive cave dwellers of the central region and western extremity 
of the island ; the fishermen living in pile dwellings in some places ; 
and the Tainans having the true Antillean stone-age culture. The 
derivation of the last-mentioned culture from Haiti and Porto Rico 
is reasonably certain. The comiection of the coast fishermen of Cuha 
with the shell heap and the key population of Florida was intimate, 
but it is still undetermined which was derived from the other. 

The origin of the cave dwellers and of the rude savage race of 
Cuba is the most difficult of all to determine. Their ancestors were 


Gil. — Stout Willi liKe t'loni Nipt* 
National Museum). 

Uay (I'. S. 

^Manual 6 Guia para los Ex&menes de los Maestros y Maestras, Habana, 1901, p. 45. 


the first colonists of the island, but we know little of their language, 
arts, names, and customs, and lack a basis for comparing tlieiu with 
peoples of North America or South America. It is probable that 
these people were lineal descendants of those whose semifossil skele- 
tons found in caves have excited so much interest. Xo evidence has 
yet been presented to prove that this race had vanished when Cuba 
was discovered by Columbus. 

The prehistoric objects from Cuba in the Heye collection have been 
greatly augmented in nuniliers since the author began his studies of 
the collection, and as these additions should be considered as a unit 
by their indefatigable collector, Mr. IM. R. Harrington, it would not 
be justice to him to anticipate any of his discoveries by further com- 
ment in this^ report. The author has, however, introduced the pre- 
ceding quotation from his account of the prehistory of Cuba, as it is 
a part of the work that has been done in preparation for this general 

In a new field of research like the Antillean. when energetic col- 
lectors are annually adding so much new material, a delay in prepa- 
ration of this report has rendered some of its results antiquated be- 
fore thej were 2)ublished. This is as it should be, and it is a great 
satisfaction to feel that a subject like Cijba, of which so little was 
known when the author began his studies of the West Indies, has 
I attracted so man}' other younger and better equipped archeologists. 
In antiquities from Cuba the Heye collection is one of the richest in 
the world, but there are many objects from that island in the Berlin 
Museum, collected by Bastian, and a very fine collection in the Uni- 
versity Museum of Halaana. 


The largest outlying island of Cuba is the Isle of Pines, which 
shows evidence of being inhabited in jjrehistoric times bj^ Indians, 
who have, unfortunately for science, left little to indicate their 
culture. We may suppose from this paucity of material that they 
were low in development and allied to the natives of the western 
end of Cuba, as no objects of Tainan culture have yet been found 
on the island. 

The author spent several weeks at the capital, Xueva Gerona, 
from which, as a base, he visited all parts of the island. He dis- 
covered a few Indian skeletons in one of the numerous caves (cuevas 
de los Indies) near that city and investigated the sunken pits called 
Indian cacimlias, but found no artifacts that seemed to him ade- 
quate to determine the relationship of the prehistoric inhabitants. 
There is no collection of Indian antiquities from this island in 
the University of Habana : and. although since the author's explora- 


tion a great many objects from Cuba have been added to the Heye 
Museum, little material has been obtained from the Isle of Pines. 

Among enigmatical structures of supposed Indian origin are the 
so-called cacimbas — an Indian word which occurs in place names in 
Venezuela and even on the west coast of Florida with the spelling 
'■ caximba." The Indian word means a pipe, but the structures to 
which it is applied are subterranean depressions with openings at 
the level of the ground shaped like buried ollas. They likewise have 
in a few instances lateral entrances, and are generally accompanied 
by small areas showing evidences of fire and much charcoal. They 
occur among the pine trees; and as tar and evidences of turpentine 
occur on their inner surfaces, they may be places where tar or other 
products of the pine were obtained for ships or canoes. They are 
widely distributed in the Isle of Pines, a few being situated only a 
.sliort distance from Nueva Gerona, the capital of the island. 

It seems probable that these cacimbas ascribed to the Indians of 
the Isle of Pines may have been constructed by the aborigines under 
direction of their Spanish masters.' Nothing of distinctly Indian 
culture was found in these subterranean depressions or near them, 
a fact that may have a bearing on the relative time when they were 
const I'ucted. 

In the neighborhood of Nueva Gerona there are numerous caves 
called ■' cuevas de los Indios." in the floor of which skeletons and frag- 
ments of the same, including a mutilated cranium, were found, but 
no accompanying artificial objects. These bones are of Indian 
origin, but whether they date back to prehistoric times or not it is, 
of course, impossible to say. It was a custom, not only in Cuba 
but also in all the AVest Indies, to bury the dead in caves, but ante- 
cedent to that custom we have good authority for the belief that 
caves were inhabited. Whether, therefore, the bones exhumed from 
the Cueva de los Indios, near Nueva Gerona, indicated this original 
cave population of the western extremity of Cuba, or burials in the 
historic epoch, no one can now tell. No pottery or mortuary objects 
of any kind which might have shed light on this question were 
found with the human i-emains. 


The antiquities of Jamaica are well known from the researches 
of Duerden,^- and many specimens from this island in the Heye 
collection, collected by Theodoor de Booy. have been described by 
'him in a paper in the American Anthropologist,-' leaving it un- 
necessary for me to consider that island in detail. 

2= Aboriginal Indian Remains in .Tamaica. Journ. .Tamaica Inst., vol. ii. pt. 4. 
=3 Certain Kitclien-middens in .Tamaion. Amer. Antlirop., n. s., vol. kv, pp. V25-4Z4. 
160658°— 34 ETH— 22 17 

258 ISLAND CULTURE AKEA OF AMEfilCA [eth. axx. 34 


The author made a special visit to the Great Cayman in order to 
investigate its archeological features, but found nothing of im- 
portance to satisfy him that it was peopled in prehistoric times. He 
collected four stone axes, which maj' have been brought by turtle 
fishers from Honduras or Jamaica, and discovered a cave from which 
fragments of pottery were said to have been collected years ago. 
He is convinced that if there were any aborigines on the island in pre- 
historic times they belonged to roaming, nomadic Carib, who landed 
there and remained only a few days. "No middens or shell heaps were 
called to his attention. Fragments of pottery occur in certain caves 
and a few celts were obtained from the natives. As some of these 
closely resemble those of the other islands they may have been 
brought there by visitors. 


In the precedinfT account of archeological material from different 
islands the plan has been to group them as far as possible on a geo- 
graphical basis. It is evident that there are great differences in the 
remains from different islands, and it is sought to account for these 
differences by minor variations in culture. A distinction in variety 
of cultures, probablj' in the beginning more marked, was more or 
less broken down by interchange of material cult objects before the 
advent of the Europeans, and while the problem is a very complex 
one a brief summary of differences in prehistoric pottery, imple- 
ments, and ornaments may be instructive. 


Although there is a general similarity in pottery from the West 
Indies, there are marked differences in the ceramics from different 
islands. In a general way it may be said that it indicates two great 
culture areas — that of the Greater Antilles (Haiti, Porto Eico, 
Cuba, and Jamaica) and that of the Lesser Antilles, the islands 
extending from the eastern end of Porto Rico to Trinidad. There 
are differences of note between pottery objects from Cuba, Jamaica, 
Haiti, and Porto Rico which would differentiate these as subculture 
areas of the Greater Antilles. The pottery from Haiti and Porto 
Rico is so close in likeness that these two islands are embraced in 
the same ceramic area, and this corresponds with what is known of the 
distribution of stone collars and three-pointed stones, which are char- 
acteristic of these two islands and not found in Cuba or Jamaica. 
It may be possible later to chronologically separate pottery from 
caves near the sea from that of caves in the upland of these islands, 
but at present this separation is conjectural. There are some diffei'- 
ences between the pottery of Haiti and Porto Rico and the other 
Greater Antilles, but these are not very great. Jamaican pottery is 
the most aberrant of the Greater Antilles. 

In the Lesser Antilles we also have a difference in ceramics from 
different islands. As a rule, the pottery from these islands is of 
better character than that of the Greater Antilles, perhaps the highest 



grade being from Trinidad, which indicates a special subculture area 
so far as pottery is concerned. St. Kitts pottery ranks higher tech- 
nicall)' and artistically. 

One imjiortant diiference between the pottery of the Lesser and 
the Greater Antilles is the presence of a superficial slip in the former. 
This is generally red in color, the paste is finer, and there is a supe- 
riority from an artistic point of view in the ornamentation. The 
incised decorations of pottery from the Lesser Antilles are rectilinear 
or curved, but the lines have not the terminal pits which occur pretty 
generally in pottery from the Greater Antilles, Porto Rico, Haiti, 
Cuba, and the Bahamas. The ornamentation of pottery from the 
Lesser Antilles is not as commonh' produced by applied relief figures, 
and the relative number of effigy vases is less than in the (ireater 
Antilles. The ring-shaped base is common in the Le.sser but rare in 
the Greater Antilles. As a rule, animal Iieads are more common 
in the former than the human heads so constant in Santo Domingo 

The pottery from Barbados is as a rule coarse, the relief decora- 
tions low and crude. It can hardh' be placed in the same category 
as the beautiful ware from Trinidad. Grenada ceramics resemble 
the Trinidad forms. 

The pottery from St. Vincent is quite distinct from that of St. 
Kitts, which includes also Nevis. It resemljles that from neighboring 
islands, as Carriacou, and marks a distinct ceramic area. 

The island of St. Kitts had a development in ceramics quite di.stinct 
from any other and shares with Trinidad the position of the highest 
development of the potterj^ technique in prehistoric West Indies. Ii 
is as a rule simpler in form and does not have the eomplexit}' of de- 
velopment of handles, lugs, and superficial additions so prominent 
in the pottery of the Greater Antilles. The forms are more graceful, 
and incised decorations are less frequent. Perhaps it is more closely 
related to pottery from Trinidad than to that from any other island, 
but it is distinctly unique. This is likewise in conformance with the 
character of stone implements, as notched axes, pestles, and orna 
ments. On St. Kitts pieces of pottery with incised decorations filled 
in with white pigments are found, imparting an appearance of 
painted ware — a condition not yet recorded from the other islands. 
As a rule the heads found in pottery from St. Kitts are not as gro- 
tesque but more realistic than those from the larger islands. 

Mr. Joyce " has pointed out the resemblance of some of the heads 
of Trinidad prehistoric pottery from Erin to that which occurs 
"throughout the liasins of the Aruka and Araau tributaries of the 
Barima River, not far from Morowhanna. The Aruka hills, iso- 

» Central American and West Indian Arclaaeology, p. 254. 


lated eminences, stand now in a tidal mangrove swamp, and were evi- 
dently at no very distant date actual islands." This significant rela- 
tionship points to a cultural connection of the prehistoric inhabitants 
of the strictly continental islands like Trinidad to those of the neigh- 
boring coast of South America, as shown also by De Booy's archeo- 
logical investigations on the island of Margarita. 

A word may be said in regard to the affinities of the pottery of all 
the Lesser Antilles and that of the Greater from the chronological 
point of view. The pottery from the former is often styled '" Carib 
pottery " ; that from the latter, " Tainan " or "Arawak." There may 
be little foundation for these designations. When Columbus dis- 
covered America he recognized the difference between the inhabitants 
of the Greater and the Lesser Antilles, and all the early writers have 
regarded the two as distinct people and commented on their customs. 
That the inhabitants of the Lesser Antilles were different from those 
of the Greater at the dawn of history should not lead to the state- 
ment that objects found on the two groups indicate the same dif- 
ference in culture. Many of the so-called prehistoric objects from 
the Lesser Antilles antedate the submergence of the people who made 
them by the Carib, and it is probable that the conquered people 
might have been more closely related to those of Porto Rico; in fact, 
might be called Tainan or peacefid agriculturists. That the conquer- 
ing race obtained by acculturation many features of culture of the 
conquered is true; in fact, we know it appropriated the wives of the 
Tainans. who transmitted a higher culture to the Carib of the 
islands. The pure Tainan culture of the Lesser Antilles was no doubt 
modified by mixture, but there is everj^ reason to believe that the 
antecedent form of life in the Lesser Antilles in ante-Carib days 
had many points of resemblance to that dominant in Porto Rico in 
the days of the advent of Europeans. 

Thus far little has been done in the determination of the relative 
age of the different kinds of pottery found in the West Indies and 
the comparative age of pottery from shell heaps, caves, and "juegos 
de bola." Some of the shell heaps offer good opportunities for 
stratigraphical studies, and caves should be excavated, paying at- 
tention to the relative depths at which different kinds of pottery 
occur. The provenience of many specimens of West Indian clay 
heads in our museums is doubtfid. and a more accurate knowledge 
of the localities from which they were taken is a necessary pre- 
requisite for exact generalizations on its distribution. The pottery 
from Cueva de los Golondrinos. on the north shore of Porto Rico, 
described and figured in plate Ixxiii of Aborigines of Porto Rico, is 
quite crude as compared with the best Porto Rican ceramics and 
would seem to have been the product of a people of different culture. 


The handles of bowls are destitute of heads or realistic decorations of 
animal form, and are vertical bands, pits with a raised ridge, or sim- 
ple lugs. Whether the caves in the interior of the island have similar 
l^ottery is not known, but there is everj' probability that the numer- 
ous excavations made in them since my visit will shed much light on 
this subject. It is very desirable that those having more material 
on cave and shell-heap ceramics should make thorough examinations 
of the character of potterj' from different parts of Porto Eico. 


The stone artifacts from the West Indies indicate a division into 
two large groups, corresponding roughly with the geographical 
divisions called the Greater and Lesser Antilles. The predominating 
type of stone implement of the Haiti-Porto Rico group is the celt of 
almond shape called the petaloid, which is also common in collec- 
tions from Cuba and Jamaica. The name conveys accurately the 
form of these implements. They rarely show any sign of a groove 
for hafting, have a pointed poll and a curved, often blunt, edge. The 
predominating cross-section of these petaloids is oval, but convex- 
concave outlines also exist, and some are slightly angular or rounded 
at the corners. Their surfaces are smooth, often highly polished, 
and are among the most beautiful known forms of stone implements. 
Outside the Greater Antilles area they are rarely found. 

There are several known examples of these petaloids with faces, 
heads, and human figures cut on the surface.-^ The mode of hafting 
of the petaloid is known from several sjiecimens. One specimen has 
been found with a wooden handle and in others the wooden handle 
is re2:)laced by stone, which is sometimes engraved on surface and 

The stone implements from the Lesser Antilles are quite different 
from those of the Greater Antilles, although the petaloid occurs in 
scanty numbers. The typical ax with blunt poll has been found on 
several islands, and many of have a shallow groove for hatting. 
A cross section of these is generally oval or square with rounded 
angles. The ax with two deep opposite angular marginaL notches, 
a strictly South American form, occurs in St. Kitts and Trinidad, 
but is not found in great numl)ers. It has not yet been recorded from 
Haiti and Porto Rico. The iroll of this type is flat, the cross section 
oval, and the thiclmess much less than the breadth. 

By far the most numerous form of ax from the Lesser Antilles 
is that found often in caches and abundantly figured in this article. 
The forms vary from a circular, oval, or almost square body with 
regiilar or asynmictrical outlines. The greatest modificatioii appears 

»Fewkes, Engraved Celts from the Antilles, 1915. 


in the poll, which is bifid, or crossed by vertical ridges modified into 
what looks like birds' heads, and various other grotesque figures. 
These have no well-marked groove for hafting and many are too 
large for practical use. The surface is rough; none have the polish 
of the petaloid. but many have incised figures cut on their surface and 
some are perforated, the perforation being confluent with the edge, 
making that region bifid. A very large number as well as a great 
variety of these axes have been found in St. Vincent, and the form 
occurs in other islands, but this ax is characteristic of the Lesser 

The theory that this form sometimes represents unfinished weapons 
on account of their occurring in caches is interesting, but has not yet 
been fully demonstrated. Their edges are not sharp enough to cut 
wood or other objects, and there is no evidence that thej' were hafted. 

The foi-ms of grinders or pestles distinguish at least two sub- 
culture areas of the Antilles. Thus the pestles from the Haiti-Santo 
Domingo area have elaborated heads, which are absent in those 
from Porto Eico. The pestles of St. Kitts are as a rule conical with- 
out decorations. The grooved pestles found in Guadeloupe are de- 
scribed by Prof. Mason as blades.^" There are about 65 pestles in the 
Guesde collection in the Berlin Museum. There is not a single pestle 
with decorated head from that island, but a very large number of 
grinding stones (or hammers). 

The most important difference between stone objects fi'om the 
Porto Eico-Santo Domingo area and the Lesser Antilles is the 
I^resence in the former of stone collars, elbow stones, and three- 
pointed stone idols, and their total absence in Cuba, Jamaica, and 
the Lesser Antilles. We find, however, certain crude forms of three- 
pointed idols in the latter islands, but the well-made, highly deco- 
rated si^ecimens are i^eculiar to Porto Eico and Haiti. These two 
types seem to have been autochthonous in the Haiti-Porto Eico area, 
and are alone sufficient to diffei'entiate this area from all other West 
Indian prehistoric cultures. 

The very natural desire to know the uses of these peculiar and 
exceptional objects can not yet be gratified. There are many theories 
of their use, and it is prettj- generally conceded that they are zemis 
or idols rather than badges of office or secular paraphernalia. Their 
elaborate symbolism and variety of forms make their interpretation 
difficult. Their relative abundance connects them intimately with 
some great desire of their owners and points to some widespread 
want in the Antillean mind. 

20 Among the grinders (pestles) in the Guesde collection figured hy Prof. Mason as 
■'blades" may be mentioned figs. 119. 120, 121, and 122. Nos. l.S.i, 136. 13T, 139, 140, 
141, 142. 143, 144, 145, 146, 147, 148, 149-1.17, 15.8, l.irt, and 100 are variously identified 
as hammers and stones for various uses. The majority are, however, rubbing stones or 


There was no material necessity greater in the life of an agri- 
cultural people than the fructification and growth of their food sup- 
ply. The West Indian culture was based on the growth of the yuca 
plant, from which cassava meal was obtained. It would be in accord- 
ance with what is known of other jarimitive religions of agricultural 
people for the Antilleans to center their religious ideas about their 
food plant, and naturally the production of yuca was the great need 
of the Antilleans. One of the two great gods of the Haitians, accord- 
ing to Pane, bears in its name the element yuca. Yucayu was the 
great Skj' god, whose name is varied, but under various designations 
controls the rain, the winds, and other phenomena of the sky. This 
is supposed to be the beneficent Sky god, and there was a stone 
image in the Sacred Cave of the Haitians called Maroio (Morahu), 
which from its holy character must have been one of the chief shrines. 
It is, perhaps, a coincidence that the Yucayu or Yuca god of the 
skj has for part of its name Guamaonoco, or Maorocon, in which 
we detect the element maroio of various spellings, as if the name of 
the idol and the name of the beneficent Yuca god were the same. 
They were at least similar conceptions, as both were appealed to for 
rain. There was another stone idol of equal sanctity to Moroio in 
the Cave of the Sun, the sanctuary of the Haitians. This idol was 
called by Pane, Bonael, and was also appealed to for rain — a func- 
tion of the highest deity in the Antillean pantheon. Bonael was 
supposed to be the idol of a being in the Sim, as Maroio was that of 
the Moon god. The sun and moon are also symbols of a higher god 
of the sky, the Yucayu or the god of winds and rain which the An- 
tillean farmers worshiped. 

Bonael (el signifying son; the son of Bona) was not the chief god, 
but was the son of another who might have been. Who, then, was 
Bona, his father ? Ma boya, the great ( ma) Boya, was the great god to 
which the Antilleans appealed for rain. He must have been a Sky god , 
and a great one at that. Boa is the serpent and Maboya is the great 
serpent. Guabansex was a woman to whom the Antilleans prayed for 
~Tk\n. Her two assistants were her agents in bringing rain, etc. Ma- 
boya and Guabansex have many points of similarity. In this ap- 
parently tortuous linguistic method the conclusion is arrived at that 
the great Sky god to whom the Antilleans prayed for success of their 
farming was a serpent god ; not a god of evil, as early authors stated 
in their interpretation of Antillean mythology, but a beneficent god, 
although at times a ruler of the hurricane. The great serpent might 
readily have been the god of the yuca. 

Did the Antillean represent this serpent in stone, as we are told he 
did in wood? iMr. Joyce has shown the relation of the collar stones 
to a bent tree, and the author has interpreted the designs on them as 


representing the idol of the yuca plant. Porto Rican stone collars 
have certain resemblances to serpents with tail bound to the neck. 
The collar stone thus becomes the idol of the great beneficent serpent, 
Hura kau, to whom the Antilleans prayed for rain for their yuca. 

The other important power worshiped by the Antilleans was 
Mamona, or Tierra the earth, Mother Earth, the female element that 
produces the yuca. A problematical stone idol of the Antillean that 
is not less numerous than the stone collar is the three-pointed stone. 
According to Pane three-pointed stones were worshiped " to make 
^le yuca grow.'" These are supposed ta represent clan tutelaries, 
children of the earth power. 


Although a considerable number of amulets, beads, and objects of 
personal adornment are known from the West Indies, our luiowledge 
of these is mainly confined to the Haiti-Porto Rican area. We can 
hardly expect any great difference in those _from different islands, 
since they are portable and no doubt were exchanged in trade even in 
prehistoric times. Gold objects traveled in the hands of traders from 
one end of the West Indies to another, and, although there is good 
evidence that many ornaments of this metal existed on the islands 
when discovered, no single specimen is now to be found in any of our 
museums. From contemporaneous descriptions of these objects that 
have come down to us there is little doubt that the Antillean gold- 
smiths were not less clever workmen in this line than their neighbors 
on the Continent. The avidity with which they exchanged their work 
for objects made of baser metal and the richness of the sources of 
gold in the Cibao Momitains would seem to indicate a considerable 
abundance of gold objects among the prehistoric Haitians, Porto 
Ricans, and Cubans, and possibly renewed search may yet bring some 
of these objects to light. 

There is little doubt that the aborigines of the (ireater Antilles had 
more gold objects than those of the Lesser Antilles. 


The foregoing data may be discussed from two points of view — 
first, the geographical distribution of the archeological material, and, 
second, its distribution in time.-' 

A study of the geographical distribution of archeological objects in 
the West Indies leads me to divide the area into three great divisions — 
the Greater Antilles, the Lesser Antilles, and the Bahamas. The 
characteristic orographical feature of the Greater Antilles, as com- 
pared with the other regions, is their size, their geological character, 
and the extension of their axis east and west. The four great islands, 
vmiversally called the Greater Antilles, including Porto Rico, Haiti, 
Jamaica, and Cuba, are tops of high mountain chains, reaching a 
moderate altitude, composed of sedimentary, azoic and volcanic rocks, 
in which the first predominate. An extension of this mountainous 
chain westward would unite it with the Mexican-Central American 
highlands, Jamaica being continued westward by a submarine bank 
Avhich rises out of the sea in the Caymans. Many geologists are in 
accord in the belief that formerly the Caribbean Sea was closed on 
the north by a continuous elevated region extending practically from 
the Yunque Mountain in eastern Porto Eico to the coast of Yucatan, 
and others find evidences of this former communication well marked 
in distribution of animals and plants. It is an interesting fact to note 
that cultural likenesses between the Greater Antilles and Central 
America are limited to the islands of Porto Rico, Haiti, and Jamaica, 
and the eastern end of Cuba. The western end of Cuba, where it 
approaches nearest to Yucatan, has not a close likeness, culturally 
speaking, to the mainland. As can be shown from the natui'e of the 
artifacts characteristic of the culture of man in this region of the 
Antilles, the indication of a derivation from Central America is not 
close. Evidently it occurred in very ancient times, before a char- 
acteristic Antillean culture had developed on the islands. The char- 
acteristic cult objects from the Antilles do not occur on the main- 
land l)Ut there is a likeness in fauna and flora. 

The geology of the Lesser Antilles is more closely allied to that 
of South America, all the West Indies, from the Virgin Islands 
to Trinidad and Barbados, having their longer axis north and south, 
forming a chain of islands, either submerged i^eaks or volcanoes, 
some of which are still active, or a submerged ridge of land extend- 
ing from South America, having all the geological characteristics 
of the adjoining region on the continent. The fauna and flora of 
the Lesser Antilles are distinctly South American, and the cultural 
relationship of the prehistoric inhalntants of these islands shows a 

''The relations of aboriginal culture and environment in the Lesser Antilles have been 
discussed in my article in Bull. Amer. Geog. See, vol. xlvi, no. 9, ]914 ; reprinted as 
Cont. flpye Mus., vol. i, no. 8. 



like kinship. It is not known whether the sejjaration of these islands 
was before or after man's advent. 

The Bahamas, forming the third geological area, are low coral ' 
islands, closely connected geological!}' with the adjacent peninsula 
of Florida, but not mountainous like the other two divisions. [ 
Culturally and biologically the Bahamas have relations on the one 
side to the adjacent regions of Xorth America and on the other to the 
neighboring islands, Cuba and Porto Eico. The cultural features 
follow in distribution the fauna and flora, as might be expected. 

P'rom the chronological point of view the evidence of the archeo- , 
logical objects above recorded indicates a longer occupation of all 
areas considered and a distinct cultural development in those areas. 
In other words, it appears that while there has been a jsrocess of 
acculturation going on throughout the West Indies there has been 
a specialized develoi^ment of culture in the different groups of 
islands. If we compare tlie antiquities of the Greater Antilles with 
those of the Lesser we readily note these imjiortant differences, 
which have been emphasized in previous contributions to the study 
of the aborigines of these islands. Two distinct cultures, one of 
.which may be called the Tainan, the other pre-Carib, are evident. 
The Tainan was that of an agricultural people, and we maj- conjec- 
ture that the ancestors of the Tainans settled the islands in very 
ancient times before specialized features which distinguish that cul- 
ture had been developed. This pre-Carib colonization formerly ex- 
tended over the whole Lesser Antilles and took place before written 
history began. It was reiDlaced by waves of Carib immigration from 
South America, being submerged in all the Lesser Antilles as far 
north as the Virgin Islands. These waves of Carib invasion, over- 
running these islands, radically changed the character of Antillean 
culture. The Carib killed the men and appropriated the women, the 
result being a mixed race, in which the arts of the preexisting Tainan 
people were appropriated. Evidence of this mixture may be found 
in the difference between the language of the men and the women, 
to which attention was called by Columbus and the early travelers; , 
the women, who were the slave wives of the Carib, i-etainine the t 
language of the conquered, the men speaking a tongue akin to the 
Carib on tlie Orinoco. Most of the objects called Carib, especially ; 
ceramics and basketry, which represent the prehistoric Lesser An- 
tilles, belonged to a submerged culture. Although represented in 
such abundance in archeological collections from the Lesser Antilles, 
they do not occur in the same abundance or have the same charac- 
teristics among the Carib on the continent of South America. They 
are characteristic of an epoch -'" previous to the submergence of the 
island people by the Carib. 

'~" So Ions as all the artif.Tcts found on an island aip arbitrarily assigned to the same 
culture epoch archeology will be unscientiflc. 


In the Bahamas the relationship of prehistoric aborigines with 
those of North America is shown most strikingly in the character 
of their pottei'y. The Bahamas also suffered from the incursions 
of the Carib, and their aboriginal culture shows strikingly the marks 
of this influence. 

From the data now in hand we can distinguish three cultural 
epochs in the West Indies. The earliest people were cave dwellers,^* 
a mode of life that had not wholly disappeared at the advent of Co- 
lumbus, who recorded the existence of cave dwellers in Cuba and the 
western extension of Haiti. This cave dwelling culture, or the earli- 
est known in the West Indies, was not generally dissimilar to what 
has been recorded in coastal South and Central America. It appar- 
ently extended throughout both the Greater and Lesser Antilles, 
but on account of the absence of caves it naturally did not exist in 
the Bahamas. It appears to have been allied to that of the coast 
of Florida. As the great shell heaps occurring on the islands wouhl 
indicate, and the pottery from these shell heaps emphasizes, the life 
of the caves is almost identical with that of the kitchen midden. 
The absence of fine stone objects separates the West Indian cave 
man from that of a following epoch, the agricultural West Indian, 
whose stonework reached a perfection not excelled elsewhere in the 
two Americas. The characteristic of tlie agricultural epoch, which 
was highest in development at the time the islands were discovered, 
is best known from the stone collars, stone idols, oi-naments, and 
various other artifacts which have been described in the preceding 
pages. While pottery was more highly developed in the Lesser An- 
tilles than in the Greater, the celts and objects made of stone of the 
Lesser Antilles were as a rule inferior to those of Porto Eico. This 
superiority in pottery found in Trinidad and neighboring islands 
is undoubtedly due to the vicinity of South America, the coarse pot- 
tery of Porto Rico not being highly influenced in that way. The 
archeological evidences of the third epoch, or that of the mixed race 
formed by an amalgamation of agricultural and Carib elements, 
appear to indicate a decline in the arts, as would naturally be ex- 
pected from the nature of the life of the inhabitants. All three 
states of culture — caveman, Tainan, and Carib— coexisted in the 
West Indies when discovered. The first mentioned had been driven 
to isolated, undesirable localities: the Tainan held the Greater 
Antilles, but had been sul)mergcd on the Lesser except in Trinidad : 
the Carib occupied the islands between Trinidad and Porto Eico 
and was slowly encroaching on the Greater Antilles when Columbus 
gave a new world to Castile and Leon. 

2* The difference between pottery found in tlie Cucva de los Golondrlnos and that from 
caves in the interior of the island of Pnrto Hico is easily recognized. 


Abbad y Lasierba. Fray Inigo. Historia geografica, civil y natural Je la isla de 

San Juan Bautista de Puerto-Kico. Nueva Edicion. Puerto-Rico, 18G6. 
AiTKEN, IJoBEKT. Porto Ricaii l)iirial caves. Proe. Nineteenth Int. Cone. Anier., 

Wasliinston. 1915. pp. 224-228. Wastiinston, 1917. 
Andkee, Richard. Seltene Ethnographica des stiidtischen Gewerbe-Museums 

zu Ulni. Baessler-Archiv. Band iv, pp. 29-38, Leipzig, 1914. 
BACHrLLEB Y MORALES, Antonio. Cuba Priniitiva. Habana, 18S3. 
Baird, [Robert]. Impressions and experiences of the West Indies and North 

America in 1849. Phila., 1850. 
Benzoni, Gibolamo. History of the New World. Tr. and ed. by W. H. Smyth. 

Hakluyt Soc. Pubs. [No. 21]. London, 18.57. 
Branch, C. W. Aboriginal antiquities of Saint Kitts and Nevis. Araer. An- 

throp., n. s., vol. ix, no. 2, pp. 31.5-333, 1907. 
Brett. W. H. The Indian Tribes of Guiana. London, 1868. 
Brintox. Daniel G. Arch.T?ology of Cuba. American Archseologist, vol. n. pt. 

10. pp. 253-256, Columbus, Ohio. 1S98. 
Charle:\-oix, Pierre F. X. de. Histoire de I'lsle Espagnole ou de S. Domingue. 

Amsterdam, 1733. 
Coll y Toste, Cayetano. Prehistoria de Puerto-Rico. San Juan, 1907. 
CoLLENs, J. H. A guide to Trinidad. A handbook for the use of tourists and 

visitors. 2nd ed. London, 1888. 
Cronau. Rudolf. Amerika. Die Geschichte seiner Entdeckung von der altesten 

bis auf die neueste Zeit. Vols. i-ii. Leipzig, 1892. 
CuuN, Stewart. The Indians of Cuba. Bull. Free Mus. Sci. and Art. Univ. 

Pa., vol. in, no. 4. pp. 18.5-225. Phila., 1902. 
D'Angheka, Peter Martyr. De Orbe Novo. Tr. from the latin with notes and 

Intioduction by Francis Augustus McNutt. Vols. i-ii. New York and Lou- 
don, 1912. 
Davies, John. The history of the Oaribby-Islands. London, 1666. 
De Bogy, Theodoor. Lucayan artifacts from the Bahamas. Amer. Anthrop., 

n. s., vol. XV, no. 1, pp. 1-7, 1913. 
Certain kitchen-middens in Jamaica. Amer. Anthrop., n. s., vol. xv, 

pp. 425-434, 1913. 
Certain archiEological investigations in Trinidad, British West Indies. 

Anier. Anthrop.. n. s., vol. xix, no. 4, pp. 471— J86, 1917. 
Archaeology of the Virgin Islands. Mus. Amer. Ind., Indian Notes and 

Monographs, vol. I, no. 1. New York, 1919. 
De Hostos. Adolfo. Prehistoric Porto Rican ceramics. Amer. Anthrop., n. s., 

vol. XXI, pp. 376-399, 1919. 
De la Borde. P6re. History of the origin, customs, religion, wars, and travels 

of the Caribs. Timehri : Journ. Roy. Agri. and Com. Soc. British Guiana, 

vol. v, pt. 2, Demerara, 1886. 
Duebden,,J. E. Aboriginal Indian remi^ins in .laniaica. .Tourn. Inst. Jamaica. 

vol. II, no. 4, Kingston, 1897. 
Dtr Tertre, .Iean Baptiste. Histoire genei-ale des Isles des Christi.iiihe, de la 

Guadeloupe, de la Martinique, et autres dans I'Amerique. Paris, 16.54. 



llDWAKDs, Bryan. The history, civil and coniinercial, of tlie British colonies in 
the West Indies. 4th ed. Vols, i-m and atlas. London, 1807. 

Ferrer. See Rodriguez-Ferrer. 

Fewkes, J. Walter. Prehistoric culture of Cuba. Amer. Anthrop., n. s., vol. vi, 
no. 5, pp. 585-598, 1904. 

— — — Aborigines of Porto Rico and neighboring islands. Twenty-fifth Ann. 
Rept. Bur. Amer. Ethn., pp. 3-220, Washington, 1907. 

Further notes (in the archeology of Porto Rico. Amer. Autbrop., n. s., 

vol. X, pp. 624-633, 19()S. 
Porto Rican elbow stones in the Heye Museum, with discussion of simi- 

lar objects elsewhere. Amer. Anthrop., n. s., vol xv, pp. 435— l."i9, 1913. Re- 
printed as Cont. Heye Mus., no. 4. 
Prehistoric objects from a shell heap at Erin Bay, Trinidad. Amer. 

Anthrop., n. s., vol. xvi, no. 2, pp. 200-220, 1914. 
Prehistoric stone collar from Porto Rico. Amer. Anthrop., n. s., vol. xvi, 

pp. 319-330, 1914. 
Relations of aboriginal culture and environment in the Lesser Antilles. 

Bull. Amer. Geog. Soc, vol. xlvi, no. 9, pp. 66i-678, 1914. Reprinted as Cont. 
Heye Mus., vol. i. no. 8. 
Prehistoric cultural centers in the West Indies. Journ. Wash. Acad. 

Sci., vol. V, no. 12, pp. 436-^43, 1915. 
Engraved celts from the Antilles. Cont. Heye Mus., vol. ii, no. 3, New 

York, 1915. 
Forte, Joseph. Note on Carib chisels. Journ. Anthrop. Inst. Great Britain, vol. 

XI, pp. 2-3, London, 1882. 
GiGLiOLi, Enrico H. Intorno a due rari cimeli precolonibiani dalle Antille, molto 

probabilmente da San Domingo, conservati nel Museo Etnogratico di Firenze. 

Proc. Int. Cong. Amer., Sixteenth sess., Wien, 1908, pt. 2, pp. 313-320, Wien, 

Gomez y Planos, Enrique. Prehistoria de la Isla de Cuba. Anales de la Acade- 

mia de Ciencias, vol. xsxvii, pp. 71-110, Habana, Aug.-Dec, 1900. 
GuMiLi-A, Joseph. El Orinoco -ilustrado y defendido. Vols i-ii. Madrid, 1745. 
Haeiierlin, Herman K. Some archieological work in Porto Rico. Amer. Anthrop., 

n. s., vol. XIX, no. 2, pp. 214-2.38. 1917. 
HucKERBY, Thomas. Petroglyphs of .'^t. Vincent, British West Indies. Amer. 

Anthrop., n. s., vol. x\^, no. 2, pp. 238-244, 1914. 
Hughes, Griffith. The natural history of Barbados. London, 1750. 
iM Thuen, Everard F. Among the Indians of Guiana. London, 1883. 
Jefferys, T. The natural and civil history of the French Dominions in North 

and South America. London, 1760. 
Joyce, Thomas A. Prehistoric antiquities from the Antilles, in the British 

Museum. Journ. Roy. Anthrop. Inst., vol. xxxvn, pp. 402-419, 1907. 

Central American and West Indian archaeology. London, 1916. 

[Labat, Jean Baptiste.] Nouveau voyage aux isles de I'Amerique. Vols, i-vi, 

Paris, 1722. 
Laet, Juan de. L'histoire du Nouveau Monde on description des Indes Occi- 

dentales. Leyde, 1640. 
La Torre y Huerta, Carlos de. Manual 6 Quia para los ExSmenes de los 

Maestros y Maestras. Habana, 1901. 
LiGON, Richard. A true and exact history of the island of Barbados. London, 

Martyr, Peteb. See D'Angheba, Peter Mabtye. 
Mason, J. Alden. Excavation of a new archeological sice In Porto Rico. Proc. 

Nineteenth Int. Cong. Amer., Washington, 1915, pp. 220-223, Washington, 1917. 


Mason, Otis T. The Guesde collection of antiquities in Pointe-a-Pitre, Guade- 
loupe, West Indies. SuiitUsou. liept. for lss4, pp. 731-837, Washington, 1885. 

The Latimer collection of antiquities from Porto Rico in the National 

museum, at Washington, D. C. Smithson. Rept. for 1876, pp. 372-393, Wash- 
ington, 1877. 

Montana, Luis. L'Homme de Sancti Spiritus. Compte Rendu XIII Congrds Int. 
il'Anth. et d'Arch. Prehist., Monaco, 1906. 

MuNOz, Juan Bautista. Historia del Nuevo Mundo. Madrid, 1793. 

OviEDO Y Valdez, Gonzalo Fernandez de. Historia general y natural d^ las 
Indias. Madrid, 1851. 

Peppee, George H. The Museum of the American Indian, Heye Foundation. 
Geog. Review, vol. ii, no. 6, pp. 401-418, New York, 1916. 

PoEV, Andres. Cuban antiquities. A brief description of some relics found in 
the island of Cuba. Trans. Amer. Ethi). Soc, vol. in, pt. 1, pp. 183-202, New 
York. 1853. Reprinted in facsimile, 1909. 

PoiNCY. Louis DE. See Rochefokt, Charles C. de. 

QuiN, John T. The building of an island, being a sketch of the geological struc- 
ture of the Danish West Indian island of St. Croix, or Santa Cruz. New 
York, 1907. 

Rochefort, Charles C. de. Histoire naturelle et morale des iles Antilles de 
I'Amerique. Roterdam, 166.5. [Also attributed to Louis de Poincy.] 

Rodriguez-Ferrer, Miguei.. De los terrfcohis cubanos con anterioridad a los 
que all! encontr6 Col6n, segfln puede inferirse de las antiguedades encontradas 
en esta Isla. Proc. Int. Cong. Amer., 4th sess., Madrid, 1881, vol. i, pp. 224- 
261, Madrid. 1882. 

Roth, H. Lino. Aborigines of Hispaniola. Journ. Anth. Inst. Gt. Brit, vol. 
XVI, pp. 247-286, London, 1887. 

Sheppard, Charles. An historical account of the island of St. Vincent. Lon- 
don, 1831. 

Stahl. Augustin. Los Indios Borinquenos. Puerto Rico, 1889. 

Stevens, Edward T. Flint chips. A guide to prehistoric archjeology. London. 

Walton, William. Present state of the Spanish colonies; including a particu- 
lar report of Hispaiiola. London, 1810. 


AcKLiN. number of specimens from- 50 

AcoMA Indians, visit of, to Wash- 

iuf^tnn . 



Indies 5''' •"i'?-'">8 

Hi-r Tainan culture. 

AouAS KCENAS. specimen from 217 

AiTKEN. Robert, article by 171 


Frachtenberg 21 

.\I>TAR STONE, St. KittS 1I>0 

Amecos. a Trinidad nation 64 

.\MEKicAN Museum, elbow stones 

m 205-206 

AMULETS 113-113, 233-234 

from Guadeloupe 136 

in Connell collection 165 

,'\\ecdotb from IKi Tertre 222 

Animal figuhes in .\ntillean art. 

i<ee Bird stones. Frog, Heads. 

Monkey. I'eccarv. 

Shark. Squirrel, Turtlr 
Apache Indians, visit of, to Wasli- 

ington 28 



Arawak, The — 

a stationary people 52 

artifacts of 5'"> 

as pottery makers •'»6. 77 

.as.sjisined to tlie Greater Antilles. 55 

islands inhabited by ."il.."i2 

natives of Trinidad classed as — 64 

pottery of 261 

submerged by Carib— 52,^61, 267, 268 
.S'w 'fainan. 
Architecture, Stone Age, of Amer- 
ica and Europe compared •"i4 

Arotle midden 92 

Assyrian sub.tects. painted on 

cotton 30 

.\WLS from Guadeloupe 135 

Axes — 

absent from Porto Rico collec- 
tions 131 

anchor 148 

a.s-ymmetrical • 102-104, 138-139 

ceremonial 133, 142 

characteristic of Lesser An- 
tilles 174. ISO 

eared 108-109. 139-143 

eared, where found 59, 93 

engraved 109-110, 144-147 

from Oominica 125 

from Great Cayman 258 

from Guadeloupe--. 132-133,138-143 

160658°— 34 ETH— 22 18 

Axes — Continued. 

from St. Lucia 131 

from St. Vincent region 94-104, 


from Tobago 78 

grooved 101-102, 145 

notched '•*■ '8 

"f unusual size 130 

perforated 147-148 

typical of Lesser Antilles 146. 202 

winged, from Tobago 78 

with animal head on poll .14n 

with caps 100-101 

with regular margins 138 

Av-Ay, name tor St. Croix 166 

ijACHiLiER y Morales, work of, on 

Cuban ethnology 24S 

Bahamas — 

cultural relations of 267 

number of specimens from ."lO 

Balsas, SbSor, elbow stone owned 

by 209 

Ball, stone, from Barbados 86 

Ball courts of Porto Rico 170-171 

B.vlliceaux, island of 89-90 

excavations in 10 

numtier of specimens from 49 

Ballou, How.^rd M., compilation of 

hililiography liy 25 

Banana Bat. mound at 89 

Barbacoas. name of pile dwellings-. 253 


aboriginal population of 78—79 

archeological work in , 11 

artifacts from 86-88 

(lepi.pulalion of S7 

early maps of 79 

English landing at SO 

number uf specimens froni 49 

Bathing customs of the Caribs.- 226. 230 

Batons — 

ceremonial 210-211 

I'lay, from Barbados 87 

from Dutch Guiana 132 

of wood 132 

Battowia. Indian caves of 89 

DE.4DS 231-232 

Bequia — 

kitchen middens of -^0 

number of specimens from 49 

Berlin Museum FtjR Volker 


celts in-- 177-17S 

Guesde collection in 128 

West Indian objects in 50.215,218 

Big Man, portrait made of 28 





Bird stones 221 

Boas, Fiiaxz — 

archeological work of 170 

editorial work of 20. 21 

excavations under supervision 

of 51 

paper by 26 

BO(:(iRAs.WAi,iJB\[AK,author of Chuk- 

chee grammar 20 

BoNAEi., an idol in tile Cave of the 

Sun 264 

Bone, objects of 75. 234. 235 

Howls — 

fragments of 70,71 

from Porto Rite area 237-240, 241 

from Trinidad 69 

Box, CLAY 70 

Bracelets of beads 230 

Branch, Dr. Christian, archeolo- 

gical work of 167 

Bremen MrsEUM. West Indian ob- 
jects in 50 

Brighton estate, rare specimen 

from 164 

Bbitish MtJSBDM, specimens in 178 

Burial customs 90, 225, 226 

Burial i'laces — 

caves as 57,61,248 

mounds as 247 

HusHNELL, D. I.. Jr., work of 22-23 

Byington, Cyrus, author of Choc 

taw dictionary 26 

Cabezas, Don Juan, specimen pre- 
sented to National Museum by 229 

Caches, valuables buried in 61-62 

Cacimbas of the Isle of Pines 256-257 

Calab.\siies. valuables hidden in 62 

Camp Fire girls, requests for infor- 
mation from 28 

Canoes, method of making 105 

Cape IIaitien, stone nodule from 185 

Caracolis, described by Labat 230-231 

Carib Indians — 

a nomadig people 52 

artifacts of 55 

assigned to the Lesser Antilles. 55 

Black, origin of 89 

culture epoch of 56 

inhabitants of Lei^ser Antilles 

submerged by 52. 88. 261 

invasion of Lesser Antilles 

by 254-255.207 

islands inhabited by 51, 52, S,s 

jnol pottery makers 56 

.settlements of, in Porto Rico_- 171 

still living in Dominica 124 

survival among, of older cus- 
toms aufl objects 52 

theory concerning 77-78 

Carib pottery, meaning of the term 261 

Carib stones, where found 93 

Carinrpagotos, a nation of Trinidad 64 

Carriacou — 

number of specimens from 49 

pottery of lis 

Cassava, legend concerning, in Do- 
minica 120-127 

Cassava bread — 

made from yuca 57 

stones for baking 223 . 

Casse-t15tes from Guadeloupe 134 

Catawba language studied by Dr. 

Michelson 17 

Cato, Mr. Josiah, stone collars in- 
terpreted by 207 

Cavki dwellers — 

culture of 56.268 

of Cuba 246-247, 255-256 

Cave dwellings of Barbados 87 

Caves — 

as burial places 57.61 

of Barbados 11. 82-83 

of Isle of Pines 237 

of Porto Rico 170-171 

of Ti-inidad 06 

Celts — 

ceremonial 176, 178, 251 

characteristic of Greater An- 

tUles 174, 180 

engraved 174—183 

from Barbados 86 

from Dominica 126,131 

from Guadeloupe 133—134 

from Mfirtini(iue 131 

from I'orto Rico 131 

from St. Lucia 131 

from Trinidad 74 

iiKinolithic. not found in Lesser 

Antilles 94 

monolithic, use of 176 

monolithic petaloid 173, 174 

of fossil .shell 134 

petaloid, called thunder stones. 175, 

petaloid. from Porto Rico_ 171, 172-183 

petaloid, from St. Croix 168 

petaloid, from Tobago 78 

petaloid, scarce in St. Vincent 94 

petaloid, typical of <iireater 

Antilles 59, 146, 262 

shell, where found 59 

Ceremonial objects. See Axes, 
Batons, Celts, Collars, Knives, 
Swallow sticks. Three-pointed 
STONES, Zbmis. 

Chambeklain, a. F., work of 21 

Cherokee Indians, work among, of 

James Mooney 13 

Chester, Greville J., specimens 

cidlected by 80 

Cheyenne Indians, visit of, to 

Washington 28 

Chip-chip shell heap, description 

of 66-67 

Chippewa Indians, visit of. to 

Washington 28 

Chippewa mcsic. paper on 26 

Chisels — 

fossil shell 80. 137 

stone 94-95, 135 

Chocolate grinders 223.224 

Sec Grinders. 
Choctaw language. Dictionary of 

the ' 20 




CnrKCHBE LANGUAGE, work on 20 

CiGUAYOS, a Cuban tribe 255 

Clay heads. See Pottbhy. 

CLtBs. See Batons. 

CocHiTi Indians, visit of, to Wash- 
ington 28 

Collars, stonej 187-198 

characteristic of Greater An- 
tilles 180 

compared with elbow stones 200- 

201, 207-209 
confined to Porto Rico and 

Haiti 187 

distribution of 199 

fragment of. found in St. Croix_ 168 

interpretation of form of 207-209 

parts of 188 

possible use of 198 

probable origin of 61 

two types of 187 

theories concerning use of_^ 263—265 
where found 60,61.169 

Collections, West Indian — 

Connell 159 

Dehesa 169 

Guesde 128-1.S7 

Heye 49-51 

in European Museums 50 

in Habana 248 

made by De Booy 49, 63 

Merino 227 

Taylor 11, 81. 86 

transferred to National Mu- 
seum 29 


excavations made by 66 

quoted on archeology of Trini- 
dad 68 

COLOK OF specimens - 132 


at Trinidad 6.3-64 

explorations of, in Cuba 242-243 

CONNBLL, Mk. E., acknowledgment 

to 11 

CONNBLL COLLECTION 87, 159, 160-166 

CONNOUN, number of specimens 

from 49 

COOKSEY, Rev. Mr., article by 80 

Coos langfagd. grammar of, by Dr. 

Frachtenberg 21 

Copenhagen Museum, West Indian 

object.s in 50, 179 

CouN, introduction of, into the West 

Indies 58 

Crania found in caves of Cuba 247-248 

Crbscextic implements. See 


Crooked Island, number of speci- 
mens from 50 

Caow Indians, visit of, to Washing- 
ton 28 


archeological remains of 240—256 

natives of. described by early 

writers 243-244 

number of specimens from 49 


Cuba — Continued. 

prehistoric culture of 242-244 

stages of culture in 240, 

242, 247, 253-254, 255 

western, destitute of pottery 60 

writers on archeoloiiy of 241, 245 

tery found in 172, 261-262 

CuLTi'RE CENTERS of the West 

Indies 58 

CcLTiRE epochs of the West Indies. 56,268 

CrxNiXGHAM ESTATE, Specimen from. 164 

Cylixdbbs — 

clay 235-236 

sheU 163 

Dakota Dictionary, necessity for 

revision of 21 

Daibebton. photographs made by 189 

Davies, quoted on aborigines of 

Dominica 126-127 

Paybwawaindung. portrait made of- 28 

Deaulb. Mr., acknowledgment to 66 

De Booy. Theodooh — 

drawings furnished by 177 

mention of 74 

specimen described by 182—183 

specimens collected by 49, 63, 

167, 213, 215, 
216. 219, 221 

Delaware LANorAOE, study of. by 

Dr. Michelson 16 

Densmore, Miss Frances — 

paper by 26 

work of 21 

Descent, matbilineai 88 

Dirks, stone, from Santo Do- 
mingo 179-180 

Dishes from Guadeloupe 135 

Disks — 

clay, used as stamps 235 

from Guadeloupe 136 

perforated 123 

stone 75 

Dominica — 

agricultural race in 126 

Carib still living in 124 

celt from 131 

culture area 123-128 

number of specimens from 49 

old negro culture in 124 

.sacred lake of 124 

DCCHI, native name for seat 246 

Duddeley. Sir Robert, journey of, 

through Trinidad 64 

DlHOS — 

from Battowia 89 

grinders mistaken for 222 

of stone 223-226 

Dunn. Jacob P., work of 24-25 

Eabrixgs, gold 2"ii 

East Caicos. number of specimens 

from 50 

Ediculb from Guadeloupe 136 

Efpigy bowls 68,70.237 

Elbow stones 198-200 

arms of 199-200 



Elbow stones — Continued. 

characteristic of Greater An- 
tilles INO 

compared with collars 200-201, 


distrihution o( 60,199 

theories concerning use of 263-265 

with face in relief 201. 202-20(1 

without face 202,206-207 

Eleuthera, number of specimens 

from 50 


from St. Vincent 115-118 

Sre Problematical objects. 
Ebin Bat, description of settlement 

at 65 

Erin Bat midden, excavations in 9 

ExCAv.\TiONS. Indian, in Barbados-- 83, 



enigmatical objects found at 115 

midden at 91 

Feathers, use of 230 

Fexner, Clarence N., collnborator_ 26 
Fetishes — 

■ shell, from Barbados 86 

stone 113-115 

FEWKE.S. J. Walter — 

papers b.v 25 

work of 8-12 

Finger ring, made from seed of 

gou^ou palm .' 75 

Fish idol, described b.v Poe.v 181 

Flask-shaped vessels 239 

Fletcher, Alice C. songs tran- 
scribed by 18 

Flint, knives of, found in Guade- 
loupe* l'^7 

Florida natives, resemblances of, 

to Cuban coast people 252—253 

Food bowls 237 

Forte, .Toseph, quoted on shell 

chisels 80 

Fowler, Hon. FI.. mention of 68 

Fox Indians — 

sacred packs of 16,30 

studi<'d b.v r>r. Michelson 15 

Fkachtenberg. Leo .T., work of 21 

Freire-Marrbco, Barbara, collabo- 
rator 2.-!, 27 

Frbshwatbr Bay, settlement at 82 

Frog, amulet representing 234 

Garcia, J, Ia., translation by. of 

paper by Poey 181 

(i.\TscHET, Mrs.' lioriSA H.. vocabu- 
laries restored by 31 

(Jbologv of the West Indies 266-26S 

fiiLBOA, Mfu-NT, caves of 8;! 

<!iLFiLL.4N, Rev. ,Tos. a., gift by, of 

Chippewa letters 31 

Gill, Db Lancet — 

specimen photographed by 249 

work of . 28 

Gold — 

objects of, in the West Indies,- 265 

ornaments of 230 


GoNAivEs, Haiti, specimen from 186 

(iori.n, Jliss B. A. — 

collection presented by 216 

mortar presented by 222 

Grand Caicos. number of specimens 

from 50 

Great Abaco, number of specimens 

from 50 

Great Catman, archeology of 258 

Great Ixagua, number of specimens 

from 50 

Grenada — 

I'ulture reliitionships of 88 

number of specimens from 49 

pottery from 118, 119, 121 

Griddles for cassava bread 119, 120 

Grinders — 

for chocolate 223, 224 

from Guadeloupe 15-1—158 

from Porto Rican area 221-224 

from St, Kitts 162 

from St. Vincent region 112-113 

identified as s<'ats 222 

See Mortars. Pestles. 

Grosser; I>r., .specimen collected by_ l86 

Grullon, Senor, mention of 217, 218 

Gcabansex, an Antillean ^loddess 264 


archeological specimens froin_ 128—158 

numtier of specimens from 49 

GiESDE, M„ quoted by Ma.son_ 60, 129-137 


described 120-138 

drawings of 137 

size of 137 


ethnological objects from 30 

paper on Indians of 26 

(iiRLEY, J, G., work of 25 

Haeberlin, II, K, — 

excavations made by 170 

specimen figured by 219 

IIaessler, Miss, work of 21 

Haiti, pestles from 227 

IIalbekt, II, S, — 

acknowledgment to 14 

editorial work of 26 

Hammers 154 

grooved, from St. Vincent re- 
gion 101-102 

Handbook of .\merican Indians, 

demand for 27 

Handles of vessels 71-74, 2.'i7 

Harpoon from Gaudeloupe 135 

Harrinoton, J. P. — 

collaborator 27 

I>aper by 26 

work of 23 

llAiiuiNGTON, M. R., collections made 

by 49, 240, 256 

Hatchets found in Hispaniola 172 

Havana, collections in 248 

Heads — 

animal 68. 72-74, 118-121 

as handles 69, 71-74 



Heads — Continued. Pase. 

grotesque human, on Porto Rico 

pottery 76 

peccary or wild hog Tl 

reptilian 69. 76 

stone . 183-184 

turtle 121 

HEr.ER. Fr.\xz. amulet described by. 233 

Hexdeeson, JuNils, collaborator 23,26 

Hewett, Edgar L.. collaborator 23. 26 

Hewitt. .T. N. B.. work of 17-18. SI 

Hbte. George O. — 

aclcnowledffment to 8 

collection of 49 

objects purchased by 221 

Hevb MrsEt'M — 

artifacts in. from West Indies. 49-51 
work under auspu-es of 8 

Hisi'ASioL.x. another name for 

Santo Domingo 131 

Hodge. F. W. — 

Bulletin editi'd by 25 

report of 5-32 

work of 7-8 

Hollow Houx Bear, portrait made 

of 28 

Holmes, W. H. — 

collaborator--^ 26 

sketch by 190 

work of 22 

Hooks from Guadeloupe 135 

Hrdlh^ka, AleS. bulletin by. in col- 
laboration with others J 26 

HrcKERBY. Rev. Thosias — 

collection of 88.90 

excavations made by 66 

mention of 115 

pendants collected by 122 

quoted on objects from St. Vin- 
cent 116 

specimens purchased by 50 

HrcHES. Rev. Griffith — 

article l)y 80 

Indian Castle described by 84—85 

Hi'RA KAr. the beneficent serpent 265 

HrTSON, Dr. .Tohx — 

collection of 86 

mention of 82 

Idols — 

from Cuba 245. 249-250. 251 

from (Juadt'loupe 136 

from Porto Rico 131.221 

from St. Vincent area 113—115 

in Heye cnlli^ctions 186,221 

in University Museum, Havana. 245, 
250, 252 
stone collars possibly used as._ 198 
three-pointed, with heads and 

legs 190 

wooden, in Montana's collec- 
tion 249 

lEiiiAX fTLTi'RE of the Lesser An- 
tilles 56 

ILU'STRATIONS. preparation of 28 

IMPI.E.MENTS, SHELL 59,78,79,80 



age of 55 

chipped, rare in West Indies 55 

classification of. from St. Vin- 
cent 93-94 

comparison of, from Old and 

New Worlds 54 

crescentic 107-108 

<lcvelopment of. in the West 

Indies 54 

fishtail, limited to volcanic 

area 61 

flaked, not made in Antilles. 1.30. 137 

from Dominica 125 

from St. Vincent region 104—105 

from Trinidad 74-75 

from West Irdies. two groups 

of 262 

winged. limited to volcanic 

area 61 

See AWLS, Axes, Celts. Chis- 
els, DtttKS, Grinders, Ham- 
mers. llARPOOX. K N I v B s , 
NEEDLES. Stamps. Tools. 
I'TExsiLs, Weapons. 

Incised stones 148-149 

IxiMAX Castle — 

described 83-84 

visited by-Dr. Fewkes 11 

IxDiAX excavations in Barbados 85-86 

Indian River, mound at 81-82 

Ikon Bear, portrait made of 28 

Isle i»k I'ines, cacimbas of 256—257 

IsLETA Indians, visit of, to Wash- 
ington 28 

.Tadbite. pendant of 75 

.lAMAICA — ■ 

antiquities of 257 

cave burials in . 61 

cultural relations of 61 

number of specimens from 49 

.Iimexez. Don Eusebio. mounds ex- 
cavated by 247 

,IONES. William, myths gathered by. 15. 16 
.lovcE. T. A., theory of. concerning 

stone collars 207 

KiOKAPOO MYTHS translated by Dr. 

Michelson 16 

Kiowa Indians, visit of, to Wash- 
ington 28 

Kitchen middens. See Middens. 

MoiNDS. Shell heaps. 
Knives — 

ceremonial or sacrificial 107— lOS 

c]f flaked flint 137 

Kxors. specimens showing 186,220 

Kroeber. a. L., to prepare volume 

on California Indians 24 

La Flesche. Francis — 

specimens collected by 30 

work of 18-20 

Languages, Indian, Handbook of 26 

Lathrop. Mr. and Mrs., collection 

made b.v 170 




Latimbu, O., collection presented by_ l^^l 
Latimeh specimen of elbow stones, 204—2(15 

Learv, Miss Ella, work of 2fl 

Les Caybs. Haiti, nodule from 1S5 

Lesser Antilles — 

comparison of pottery from 2r>r) 

inhabitants of, at time of Colum- 
bus 130 

invaded by the Carib 52. 88, 261 

theory of kinship of aborigines 

of - 126 

LiNcnsTics as an aid to archeol- 
ogy 61 

IjIttt.e Abaco, number of specimens 

from . 50 


LouBAT PRIZE awarded to Dr. Swan- 
ton 14 

Lubbock. John, mention of 1,30 

Mabova, a Sky god 264 

McGeb. W J. mention- of 249 

Madrid Mttsecm — 

elbow stone in 202-203 

specimens in ^3 

Mamoxa. the Earth Mother 265 

Mandan and IIidatsa music, study 

of, l)y Miss Densmore 22 

Manuscripts of the Bureau 30-31 

Mariguana — 

number of specimens from 50 

specimen found in 182 

Marl Hill, I'.arbados, collection ob- 
tained at 87 

Maroio, .stone image called 264 

Martin, Mrs. Ada, poltery presented 

by 30 

Martinique — 

culture relationship of 12s 

specimens from 128, 131 

Masks — 

shell 235 

stone 183, 184-185 

Mason, J. A., excavations made by__ 170 

Mason, Otis T. — 

Guesde collection described by_ 12!), 


on form of stone collars 207 

Maya hieroglyphs, work on, by Mr. 

Morley 23 

Mkdicind Crow, portrait made of 28 

Menzies, Mu., acknowledgment to 66 

Meri.^o collection, mention of 227 


paper by 25 

work of 15-17 


Argyle 92 

Oberland . 92 

of Harbados 81-82 

of St. Kitts 159 

of .St. Vincent 91-92 

Salt River 167 

Sec Moundsj- Shell heaps. 

Mills estate, .specimen from 164 

MiNDORO Island, photographs of na- 
tives of 29 

Monkey head of clay 68 

■' Monkey " teapot, from Barbados^ 80 

Monkey vases made by negroes 164 

Monolithic celts 173,174,176 

MoNTANf:. Dr. Luis, collections under 

directorship of 248 

MooNBY. .Tames, work of 12-13 

Mores, number of specimens from .^0 

Morgan, Mr., enigmatic objects 

found by 115 

Morley, .Sylvancs G.. memoir by.- 23 
Mortars — 

from Guadeloupe.,. 134-135. 152-154 

of Porto Rican area 221-226 

ordinary form of 221 

resembling a bat 131 

Trinidad 68 

Sec Grinders. Pestles. 
Motul-Maya Dictionary, photostat 

copy of 31 

Mounds — 

excavation of. in Cuba 247 

f?ee Middens. 

Mount Gilboa. caves of 11.83 

MuLLBR. Prop. Sophus. acknowledg- 
ment to 179 

MuNN AND Co.. gift by, to Bureau 29 

Munroe, Miss Helen, work of 27 

MuNSEE Indians, visited by Dr. 

Michelson 10 

Murie. .James, work of 24 

Music — 

Chippewa, papers on 26 

Siouan, study of 21-22 

Necki.acbis — 

from Porto Rico 131 

of shell or stone 230,231 

Needij:s, netting — 

from Guadeloupe 136 

from Porto Rico 131 

Nbgroes, influence on, of Indian cul- 
ture 12 

Nepoios, a nation of Trinidad 64 

Neumann, Felix, editorial work of- 25 

Nevis, island op — 

archeology of 158 

number of specimens from 4!) 

perforated stone from 12."i 

torso from 11 

New Providence, number of speci- 
mens from 50 

New York Academy op Science, 

archeological work of 51 

N'EwsAM. A., mention of 68 

Nichols, Dr.. collection loaned by 124 

NiPB Bay. objects from 251.254.255 

NoDiiLES. stone ; 184—186 

North Caicos, number of specimens 

from 50 

Oberland midden 02 

Obia men, influence of 124 

Omaha Indians, gaining dice of, 

presented to the Bureau 30 

Ornaments 2.30-232 

of gold 230.265 

of shell 163.232 

See Beads, Caracolis, Neck- 
laces, Pendants. 



Osage Indians — Page. 

specimens collected from 30 

studied by Mr. La Flest-he 18-20 

visit of, to Washington 28 

Paoet. John, mention of 80 

Pane. Ramon, account by, of inak- 

ing zemis 208 

Panels of stone collars 193-196 

Passamaquoddy Indians, visit of, to 

Washington 28 

Peccary, hbmd of, in clay 121 

Perforated stones 148-149 

Pendants 114 

found in St. Vincent area 122-123 

jadeite 75 

stone 232-233 

Pepper. Gojrgb H., quoted on speci- 
mens from St. Vincent 116 

Pektorated stones 148-149 

Pestles — 

comparison of, from diffi'rent 

areas 263 

found in Le.sser Antilles 228 

from Guadeloupe 134, 154-158 

from St. Vincent region 112—113 

from Trinidad 74-75 

of the Porto Rico area— 220. 226-229 

variations of 74 

.Sec Grinders. Mortars. 

Photographs of Indians — 

exhibition of 28 

requests for 28 

Pictographs — 

of Guadeloupe 130 

of St. Kitts 160 

of St. Vincent 92, 130 

PlEDRAS DEL RAYO 175-176.251 

Pile dwellings of CtJBA 253 

Pillar stones 161 

PiNART. A., collection made by 167 

PiNART elbow stone 206 

Place names, Indian — 

in Trinidad 65 

occurrence of 62 

Plana Cayo, number of specimens 

from 50 

Planos. Enrique Gomez, work of, 

on Cuba 248 

Plenty Coups, portrait made of 28 

Poey, Andres, on Cuban antiqui- 
ties 181, 244 


Guiana, excavations in 76-77 

PoNCA Indians, visit of, to Wash- 
ington 28 

Population. Indian, work of .Tames 

Mooney on 13 

Porto Rico — 

absence of axes in 131 

'cultural relations of inhabitants 

of 169 

culture area 168-240 

excavations in ."il 

highest culture developed in 169 

Latimer collection from 131 

number of specimens from 49 

Pottery — -Page. 

ago of 261 

as an indication of an agricul- 
tural people 77 

culture areas indicated by 259 

excavated in Trinidad 9 

fragments of, from Porto Rico- 131 

from Barbados 86 

from Cuba, character of ^ 252 

from Greater Antilles, affinities 

of 261 

from Greater Antilles, compari- 
son of 259.260 

from Greater and Lesser Antilles 

compared 260 

from Lesser Antilles, affinities 

of 261 

from Porto Rican area 236-240 

from St. Kitts 16.3-165. 260 

from St. Vincent area 118-122 

from Trinidad.- 60,67-74.75.76 

glazing of 68 

of Porto Rico 172 

Porto Rican, characteristics of- 58 

Porto Rican. quality of 169 

Stone Age, of Old and New 

Worlds compared 54 

Tainan 261 

Pottery bests 70, 86, 120. 121 

Pottery stamps. See Stamps. 
Prague, West Indian objects in mu- 

si'um at 50.234 

ProbleM-itical objects 149-152, 

154. 220. 227. 229 
from St. Vincent—- 110-111.115-118 

of bone 234 

Property of the Bureau 30 

Providence, number of specimens 

from 50 

Publications — 

correspondence relating to 27 

d'-tribution of 27 

Pueblo Indians, preparation of bib- 
liography of 7-8 

Pulley, stone, in Guesde collection. 152 

Quoits, played by Caribs 136 

Rae, C. S., celt owned by 177 

Ragged Island, number of speci- 
mens from 50 

Raleigh, Sir Walter, races of Trini- 
dad enumerated by 64 

Red Cloud, portrait made of 28 

Red FIawk. portrait made of 28 

Red Willow, Nebb., fragments of 

pottery from 30 

Reptiles, heads of, in pottery 69. 76 

Rings, stone — 

in Connell collection 161-162 

in Dehesa collection 169 

in Guesde collection 152 

in Heye collection 149 

Robbins, Wilfred William, col- 
laborator 23. 26, 27 

Robinson, Sir W., mention of 68 

Rodriguez-FerbeBj Miguel — 

mention of 247 




KiiDRKUEz-FEnEER, MIGUEL — Continued. 

paper read hy 240 

specimen figured by 245.250 

specimen found by 181 

EOTH. Walter B. — 

paper by J 26 

specimens collected by 30 

BOTAL, number of specimens from 50 

Saba, number of specimens from 49 

Sacred packs of Fox Indians — 

acquired for National Museum IG 

purchased by Dr. Michelson 30 

Sacrificial knives 107-108 

St. Christopher. See St. Kitts. 

St. Croix — 

aborigines of 166-167 

artifacts from 168 

culture of, like that of Porto 

Rico ' 168 

geology of 166 

specimens from ^ 16tj-168 

St. Kitts — 

archeological work in 11 

archeology of I08-I66 

number of specimens from 49 

pottery of 260 

St. Thomas — 

number of specimens from 49 

pestle from 227 

St. Vincent — 

number of specimens from 49 

pottery of 118-122 

specimens from 90-91 

visit to shell heaps of 10 

St. Vincent — Grenada cultur i; 

AREA 88-12.3 

Salivas, a nation of Trinidad 64 

SALV.AGAT, a nation of Trinidad 64 

San Juan Indians, visit of, to Wash- 
ington 28 

Santa Clara Indians, visit of, to 

Washington 28 

Santa Cruz — 

collection gathered at 12 

number of specimens from 49 

Santa Lucta — 

artifacts from 9.") 

celts and axes from 131 

number of specimens from 49 

Santo Domingo — 

celt from 131 

number of specimens from 49 

Saona, number of specimens from__ 49 


acknowledgment to 183 

photographs obtained by 50, 189 

ScHOMBURGK, Sir Robert, mention 

of 80 

School op Ameeican AECH.EOL0Gr, 

inve,stigations conducted by 23 

Seats, stone 223-226 

not found in Lesser Antilles 60 

Seito, SetSob, specimens purchased 

from 50-221 

Se! stones 187 


Seneca texts, edited by Mr. Hewitt. 17 

Serpent, Great, story of 127-128 

Serpent god of the Antilleans 264-265 

Shark, head of, as handle of 

ve.ssel • 73-74. Sij 

Shell, objects of — 

cone-shaped 86 

disks ; 80 

from Barbados 86 

from St. Kitt.s 162-163 

from St. Vincent 123 

implements 78, 79, 80 

mask 235 

pottery stamps : 86 

teeth 233 

Shell-heap people — 

customs of ."iii-d" 

probably cave dwellers 57 

Shell-heaps — 

excavations in 9, 62 

of British Guiana 70 

of Porto Rico 170-171 

on eastern coast of Trinidad 60 

iS'ee Middens, Mounds. 

Shiblds, used by natives of Trini- 
dad 6.". 

Shosho.nt Indians, visit of, to 

Washington . 28 

SiouAN MDSic, study of, by Miss 

Densmore ' 21-22 

Siorx Indians, visit of, to Wash- 
ington 2S 

SiusLAW language, grammar of, by 

Dr. Frachtenberg 21 

Sky God, worship of 128 

Slaughter, Mrs. Ella, work of 29 


eruption of 91,92 

mention of 110 

Squirkel, whistle in shape ot 68 

Stahl, ArousTiN, archeological work 

of -_- 170 

Stamps — 

clay 74, 123. 23.i 

shell 86 

Statuejs, monolithic, mentioned by 

Brinton 246 

STENHOtJSE, Walter J., work of 28 

Stevenson, Mrs. M. C. — 

I)aper by 20 

work ot 14-15 

Stillwbll. Miss Margaret B., 

transcription of manuscript by 24 

Stone .4ge, in Old and New Worlds. 53 

Stone Fort, midden at 159 

Stones, incised and perforated- 148-149 

Stones, problematicai 149-152 

tS'ce Problematical objects. 

Stonework — 

of Central America .SS 

of Mexico 58 

of the Greater Antilles 58 

Stribe. Leopold B., stone collar 

owned by 189.103 

Stubbs, midden at 91-92 




SuLcrs, an important feature of 

elbow stones 209 

Swallow stick of bone 235 

SWANTOX, John R., work of 13-14,20 

Tainan cuLTt'RE — 

of Porto Rico 56 

origin of 252 

submerged !>>• Carib 267, 208 

.^'er Agricultural epocb, Arawak. 

Taos Inihans, visit of. to Washing- 
ton 28 

Tattooing in the Antilles 230, 232 

Taylor collection, mention of- 11, 81. 86 

Tewa Indians — 

papers on 26, 27 

researches among, by Mrs. Ste- 
venson 14 

study of, by Mr. Harrington 23 

Three-pointed .stones — 

areas dififereDtiated by 2i"i,s 

dlstrihutinu of 199 

not found in L*esser Antilles 00 

of four types 211-220 

theories concerning use of 263—26.5 

used for pestle 220 

where found 61 

.See Zemis. 



TlMT'CUA LANGtJAGB. Work Oil. by 

Dr. Swanton 14 

Tobago — 

artifacts of T^ 

grinding stone from 125 

number of specimens from 49 

Tools. See Implements. Utensils. 

ToRTOLA, number of specimens from- 49 

Tree worship in the Antilles 208 

Trinidad — 

archeologioal work in 9 

culture area 62—7.S 

descriiition of 63 

named by Columbus O.'! 

natives of 03-<i5 

number of specimens from 49 

objects from, compared with 

others 7.5—77 

pottery of 60.07-74 

Trocadero Museum — 

specimen in 21." 

stone collar in 195 

Tsimshian mytholoot, paper on 20 

Tufa, volcanic, objects made 

from n.-j-lls 

Turtle — 

effigies 6S, 70, 72-7.'! 

heads, clay 121 

stone. In Montana's collection.- 249 

wooden, from Battowia 89 

wooden, from St. Vincent 249 

Two Moons, portrait made of 28 

Union Island, number of specimens 

from 49 

160658°— 34 ETH— 22 — -19 

U. S. Board ox Geographic Names. 
Smithsonian Institution repre- 
sented on 8 

U. S. N.iTiONAL Museum — 

collection presented to, by Miss 

Gould 216 

elbow stone in 202 

engi'aved celt in 178 

Latimer collection in 131 

graved celt in 179 

Utensils. See Bowls. Griddles, 
Grinders, Implements, Mor- 
tars, Pestles, Pottery, Seats, 
Vases — 

from Guadeloupe 135 

from Trinidad 68-09 

Victoria Insiitutb, pottery in 68 

Vienna Museum, West Indian ob- 
jects in 50 

Von den Steinen, Mr., illustra- 
tions made by 128, 137 

Walcott, Dr. Chas. D. — 

letter of transmittal to 2 

report of F. W. Hodge to 5-32 

Weapons. See CASSE^T£TES, Dirks, 

Weissenijorn, Dr. .Iohannes, ac- 
knowledgment to 189, 190 

West Caicos, number of specimens 

from 50 

West Farm, midden on 159 

West Indians, condition of, at time 

of Discovery .5.3 

West Indies — 

geological divisions of 59 

visit of Dr. Fewkes to 8 

Wichita Indians, visit of. to 

Washington 28 

Willis, Bailey, collaborator 26 

Winofield EST.iTE, midden on 159 

WiNSHip, George Parker, acknowl- 
edgment to 24 

Wright, F. E., collaborator 26 

Yaios, a nation of Trinidad 64 

Yokes, stone, of the Totonacs, 

mention of 109 


cassava bread prepared from 57 

cultivation of 57-58 

West Indian culture based on — 264 
YuCA Spirit, idol of 20,8-209'' 


name of Yuca Spirit 209 

the great Sky god 264 

Zemis — 

as a totemic symtwl 232 

carried by medicine men 2.32 

found in Ilispaniola 172 

tri-pointed. characteristic of 

Greater Antilles 180 

See Three-pointed stones. 

ZuSi Indians, paper on 26 



, ■■lililll 

3 9999 06313 743 2 

fcl % : ^JS^^-^^W-W^ 

^yt^ fi