INFERENCES FROM CULTURAL DIFFUSION TO TOWER HILL, JAMAICA, AND CUPERCOY BAY, ST. MARTIN. Ripley P. and Adelaide K. Bullen

INFERENCES FROM CULTURAL DIFFUSION TO TOWER HILL, JAMAICA, AND CUPERCOY BAY, ST. MARTIN Ripley P. and Adelaide K. Bullen In April 96, R. P. Bullen tested the Tower Hill site on Jamaica and later the same month we both worked at the Cupecoy Bay site on St. Martin. A report on the former was sent in 965 to The Institute of Jamaica but for a variety of reasons has never been published. That covering Cupecoy Bay was sent the same year to Nieuwe West-Indische Gids and published the following year (Bullen and Bullen 966). There are certain modal similarities between the pottery from these two sites in spite of the great distance between them. Some of these traits are also found in the Dominican Republic and on Puerto Rico, where we have good chronological information; some appear, but in lessened quantities, substantially further south in the Lesser Antilles. It seems important to bring these traits and their implications to the attention of students of the Lesser Antilles. Also of particular interest to this Congress, is a unifacial, percussion flaked, scraper-like tool from Cupecoy Bay which resembles products of the Casimira chert industry of Hispaniola. It seems to be unique in the Lesser Antilles. Tower Hill, Jamaica The Tower Hill site is located about a mile east of August Town, Jamaica. To reach it, one has to ford the Hope River and then to climb to an altitude of about 000 feet. The site, on a shoulder of Dallas Mountain, has a southern exposure and is protected against easterly storms by higher land. Behind the site, in a low part of the mountain' s shoulder, is a considerable area of sandy soil suitable for agriculture. The air-line distance to the ocean is two miles but if the valley of the Hope River is followed, the journey becomes nearly four miles in length. As the depth of aboriginal debris on the more-or-less level top of the site seemed slight, the test trench was located a short distance down the eastern slope where the cultural deposit was 8 inches thick. Excavation was by three arbitrary 6- inch levels which were kept parallel to the surface. The deposit consisted of a typical slope midden composed of clayey dirt, rocks, sea shells, food bones, and sherds of Amerindian pottery. At a depth of a foot we came down upon a hearth, feet across filled with ashes. Underneath the midden was sterile, reddish, clayey dirt. Typical sherds from this test are illustrated in Figure. Pottery from Tower Hill is thin (-7 mm), reddish-brown or blackish-red in color, tempered with medium coarse grit, and poorly finished with scraped and semi-burnished surfaces. Eight sherds, including the neck of a water bottle (Fig., J6), are brown-yellow in color and tempered with very fine grit. Workmanship may universally be referred to as poor. Rims are dominantly incurving or inturned. Flatish bases without feet or annular sup- ' ports prevail. Boat-shaped vessels (Fig., C4-5) are present. Decoration seems limited to folded lips (Fig., A-6), small lugs located on vessels sides (Fig., Bl-4) and grooves or gashes at shoulders (Fig., A5-U) except for the modeling prelintat 48

BULLEN AND BULLEN 49 the ends of boat-shaped containers (Fig., C4-5), double crest lugs at rims (Fig., A), and an appliquéd arm (Fig., B7). Red painting is completely absent. Pottery from Tower Hill may be divided into two main series or styles. The first is red-brown in color, has fairly well smoothed (relatively speaking) surfaces, exhibits traces of burnishing, and includes all the vessel sherds with folded lips or rims (Fig. ). While lacking incision, sherds of this group closely resemble those of the Meillac series of Haiti (Rouse 94) and the Dominican Republic (Florida State Museum, type collections). The other group, black-brown in color, exhibits very poor-- sometimes tacky--surfaces and includes all the vessels with large or heavy side lugs (Fig., B-C) and all boat-shaped containers. Surfaces of these sherds closely resemble some illustrated by Rouse (95: PI. 4, g-h) for the Santa Elena style of Puerto Rico. Sherds with gashes are intermediate between these two groups. Griddles, of which three types are present, seem relatively thick (5- mm). One, with a raised edge at its periphery (Fig., CI), would seem to be similar to the usual Caribbean griddle. The second, undoubtedly the rarest here, is the "pancake" type (Fig., C). The third is supplied with a raised edge located a substantial distance in from its periphery (Fig., C). This more complicated form we have not found in the Virgin Islands (Bullen 96) nor on Grenada (Bullen 964) or St. Vincent (Bullen and Bullen 97) but it is known for the Leeward Islands further to the north. We have been informed by C. Bernard Lewis, the late Robert R. Howard, and Ronald L. Vanderwal--all familiar with the archaeology of Jamaica--that the ceramic specimens from Tower Hill differ in no significant way from those found at the White Marl site located on the main road west of Kingston, Jamaica. Certainly the Tower Hill sherds fall within Howard' s (965) definition of White Marl ceramics. The White Marl site is radiocarbon dated from A.D. 870 to A. D. 490 with the main occupation A. D. 50-70 (Vanderwall 968) and, hence, was occupied for several centuries. The absence at Tower Hill of handles, incised animalistic lugs, or alternating oblique parallel lines incised above shoulders--all present at White Marl (Howard 965: Fig., d-g)--may indicate the Tower Hill site was either early or late in the White Marl cultural continuum or presents cultural differences at the tribal or clan level. In an attempt to see if any chronological suggestions are present in the data from Tower Hill, the pottery has been analyzed by ceramic modes and the results presented in Table. Examination of this table indicates that side lugs have a relatively deep, while folded lips and gashes at shoulders have a relatively shallow provenience, suggesting these traits to be relatively early and late respectively. Looking at Table from the viewpoint of the Lesser Antilles offers interesting correlations and speculations. Small lugs on vessel walls are a feature of the pre-a.d. 700 Caribbean Saladoid-related ceramics (Rouse 964) as found in the Cuevas style of Puerto Rico (Rouse 95: PI., d.), at Coral Bay on St. John island (Bullen 96: PI., e, g), and at Pearls on Grenada (Bullen 964: PI. 8) but not, to our knowledge, in the ceramics of later periods although large, gashed, side lugs are present in the Santa Elena style of Puerto Rico (Rouse 95; PI. 4, d) and known for the Petite Rivière Salée site on Martinique (collection J. Petitjean Roget). The fragment of a water bottle rim (Fig., B7), also in the lowest zone at Tower Hill, matches extremely closely a similar bottlëtsherd (Bullen 964: PI., ) from the Pearls site on Grenada. Could these traits be holdovers from the early Cuevas period of Puerto Rico although a Cuevas ce-

50 CULTURAL DIFFUSION ramie complex is not found west of Puerto Rico? The double crest or eared rim lug (Fig., A) is found in the Santa Elena of Puerto Rico (Rouse 95; PI. 4, i), on San Salvador in the Bahamas (Hoffman 970: 00), and in the related ceramics of St. John island in the Virgin Islands (Bullen 96: PI. 7, d). The typical scraped surface of Tower Hill pottery is similar to that of the Botany Series of the Virgin Islands (Bullen 96:6; PI., ; PI., _f) and reminiscent of the careful burnishing of the Caliviny Series of the southern Windward Islands (Bullen 964:48). Folded rims (or lips) are present but not common in Puerto Rico (Rouse 95: Fig. ), nor can we point to any from the Virgin Islands or the Windwards; but they are present in Cuba (Alvarez 956;Lam. 54, 5-6), in Hispaniola (Veloz, Ortega and Pina 97: Lam. 9), in the Bahamas (Hoffman 970; Fig. 9, d) and at the Cupecoy Bay site on St. Martin to be discussed shortly. Folded rims and gashes at sides, similar to Figure, A6-7, can be duplicated in the Dominican Republic (Kreiger 9: PI., middle) and gashes at sides are present in Cuba (Alvarez 956: Lam. 54, 4). Presence of gashed side lugs at the Petite Reviere Salée site on Martinique has been noted above. Folded rims are found in the newly delineated Cayo pottery of St. Vincent (Kirby 974). Table - Vertical Distribution of Ceramic Modes at Tower Hill, Jamaica Modes Depths in 0.6 6. - inches -8 Totals Double rim lug Grooved lip Folded rims Flat lip Simple rounded lips 4 7 7 7 75 Folded rim and gashes at side *Gashes at shoulders *Gashes at side, lug at shoulder *Lugs at sides of vessels Boat-shaped, modeled ends 6 4 0 5 *Neck of water bottle Appliquéd arm at side of rim Plain, fine temper Plain, medium coarse temper Griddle sherds 6 68 0 4 8 67 7 69 Totals 4 88 94 56 #Note that folded rims and gashes on sides or at shoulders have a relatively high provenience while side lugs (and water bottle sherds) are relatively deep.

i *> Fig.. Sherds, stone specimens, Tower Hill, J B A-, inward bent rim; -, 5-6, outward fol rim; 4, inward folded rim; 5, deep grooves a shoulder; 6-8, small side lugs with gashes o lugs and shoulder lines; 9-, gashes on sho - ers (9, on applique at shoulder);, deeply grooved lip;, double eared rim lug; B-, double side lug; -4, large side lugs; 5, two side lugs over support holes; 6, rim of water bottle; 7, large appliquée! (handle?), erode sherd with two, long and narrow, pinched side lugs; C-, griddle with raised lip;, panc type griddle;, griddle with raised edge an inch from outer lip; 4-5, modeled ends of bo shaped vessels (4 has flat outward slanting lips, 5 has two support holes); D-, pitted ha merstone;, sharpening stone for ax or celt. (Specimens about one fifth size)

5 CULTURAL DIFFUSION Fig.. Rim sherds from Cupecoy Bay, St. Martin a, thick rounded; b, flanged; c-d, inward folded; e, thickened, out turned; f, rim modified subtype 5; g, Pearls red-painted lip; h, side grooved; i, punctated lip. Fig.. Plate, body, and griddle sherds, Cupecoy Bay, St. Martin a-c, rim forms from plates; d, punctated appliqued strip; e, red painted; f, interior black painted; g-h, Suazey Scratched;_i-m, griddle rims.

BULLEN AND BULLEN 5 An effect similar to that of folded rims was produced at Magens Bay on St. Thomas by incising a line a short distance below the lip (Bullen 96: PI., d). This motif was in the highest zone at Magens Bay. It can also be cited for St. Vincent (Bullen and Bullen 97: PI. XXXVIII, a). An incised folded lip (Bullen 96: PI. 9, a_) was found on a sherd from the highest zone of the Francis Bay site on St. John. These traits are all relatively late in their respective areas. There is also a similarity in the boat-shaped vessels from Tower Hill (Fig., 4) and those from the Virgin Islands (Bullen 96: PI., c). This last similarity does not seem to be so marked further to the southeast but we believe it to be present on Antigua and it certainly is reminiscent of the Diamant Hammock shaped containers of the Lesser Antilles (Bullen and Bullen 97: 40, Pis. IV, _f, j). It is evident from the above brief survey that the pottery found at Tower Hill is related to that found in Cuba, the Dominican Republic, Puerto Rico, the Bahamas, the Virgin Islands, and to a lesser degree the Lesser Antilles. These relationships seem in all cases to refer to middle or late as opposed to early ceramic times or sometime after A. D. 800-900, the beginning of the Santa Elena period in Puerto Rico (Rouse 96). Stone artifacts from the test at Tower Hill include a pitted hammerstone from the highest zone, another (Fig., Dl) from the middle zone, three other hammerstones equally distributed vertically, and five chert flakes and three sandstone sharpening stones (Fig., D) from the two lower zones. These sharpening stones are interesting in that they have multiple, wide but shallow, sharpening grooves on both upper and lower surfaces. Our surface collection produced two fragments of polished celts made of a gray or greenish gray rock which could have been sharpened on these stones. Food remains in the midden consisted of marine shells and the bones of animals. Shellfish included oysters, three kinds of clams, cassis sp., Strumbus pugilis, Fasciolaria tulipa, murex, two kinds of snails, Vasum muricatum Born (?), and Area sp. This suggests the wide range of marine shellfish brought to the site for food. Animal bones have been identified by Elizabeth S. Wing, Assistant Curator of Zooarchaeology, Florida State Museum, with the support of the Caribbean Research Program of the University of Florida. She lists hutia (Geocapromys brownii), heron (Ardeidae), purple gallinule (Porphyrula martinica), sea turtle (Cheloniidae), sand or mâcherai or nurse shark (Lamnoidei), cat or requiem or hammerhead shark (Scyliorhiniodei), Nassau grouper (Epinephelus cf. striatus), snapper (Lutjanus sp.), barracula (Sphyraena sp. ), and sheep or goat. The minimum number of individuals represented was nine for hutia, two for barracuda, and one each for the rest. The sheep or goat undoubtedly was related to a recent house on the site. Dr. Wing commented that although the site was located inland, the inhabitants evidently depended heavily on sea food for protein. Her findings are in agreement with those given above for shellfish. Possibly the most interesting feature about the Tower Hill site is that it is at an altitude of about 000 feet above sea level. That Indians would bring fish, shellfish, and other food from the sea some miles away up an extremely steep trail indicates a decided preference on the part of someone to live on Tower Hill. That it was also a good location for horticulture does not, in itself, seem a sufficient reason for the selection of this location for a dwelling place. Only two considerations seem adequate: protection against human enemies or freedom from pests, hurricane tides, and heat. It should be remembered that the White Marl site, the largest known for Jamaica, is

54 CULTURAL DIFFUSION also located on a high ridge. Mr. Lewis tells us that the location of Indian sites at fairly high altitudes is common on Jamaica and we have seen such sites on St. Croix. It is hard to believe the inhabitants of the White Marl site, the largest on the island, had to live on a ridge for protection. It seems more likely Jamaicans of that period preferred to live in a place relatively cool and free of mosquitos and were willing to pay for this comfort by fairly long treks to the Caribbean Sea. Cupecoy Bay, St. Martin The Cupecoy Bay site is located behind a sandy beach near the western end of the southwest shore of St. Martin. To the rear is an arm of Simson Lagoon while on both sides are rocky hills. We made an extensive surface collection, dug a series of small tests which proved the site had been extensively damaged archaeologically by cultivation, and excavated a 5- by 0-foot stratigraphie test in April 96 (Bullen and Bullen 966). Typical specimens are presented in Figures -4 and the pottery summarized in Table. Pottery from Cupecoy Bay is similar to that from Tower Hill in that it is blackish, gray, buff, or reddish in color and tempered with medium-sized grit while surfaces are scraped and semi-burnished but seldom really smooth. As at Tower Hill bases are flat. Workmanship, however, is much superior at Cupecoy Bay and sherds are thicker on the average. As was true for Tower Hill, Cupecoy Bay sherds can be divided into two major groups. One is redder in color and better made with finer temper, smoother surfaces, and thinner walls. Except for the thinner walls, this group resembles the first Tower Hill group and like it contains sherds with folded rims (Fig., c-d). These sherds-- both in surface appearance and in folded rims--closely resemble some subsumed by Rouse (95: PI. 7, b_, d, g) under the Esperanza style of Puerto Rico. The other group is darker in color, coarser in texture, and poorer in workmanship. Sherds of this group are apt to be from casuela-shaped vessels. As a group they are reminiscent of the Botany-Bordeaux ceramic series of the Virgin Islands (Bullen 96:6-). "Thickened lips, flatly beveled inwards, or bevels, " but without folds, said by Rouse (95: 4) to be diagnostic of Ostiones, are also present. These are reminiscent of Caliviny style lips as found in the Windward Islands. At Cupecoy Bay, two rim sherds (Figs., _f, 4, h) have rim modifications like those found in the Windward Islands and designated as Caliviny Rim Modified, subtype 5 (Bullen and Bullen 97:4). Two examples of punctated rims (Fig., i) were found at Cupecoy Bay but gashes, like those at Tower Hill, were completely absent. Rims with small punctations (not finger indented) are also found in Jamaica (Vanderwal 968: Fig., p-q_) but not at Tower Hill, the Bahamas (MacLaury 970:P. VII, k) and the Virgin Islands (Bullen 96: Pi. IX, m). This is another Meillac ceramic Trait (Rouse 94: PI. ; Veloz, et al, 97: Lam. 9). Also present at Cupecoy Bay were six typical examples of Suazey Scratched sherds (Fig., g-h). Except for one sherd out of many thousands from St. Croix, examined by us at the Danish National Museum in Copenhagen, the island of St. Martin is the furthest northwestern provenience for this pottery type. A few are also known for Saba and St. Eustatius (de Josselin de Jong 947:, Pis. Ill, 8; IV,, 5-6- IX -) as well as Antigua (Hoffman 970: Fig., ). it is interesting to^otelhatboth '

BULLEN AND BULLEN 55 Table - Sherds from Cupecoy Bay, St. Martin Typology Surface Depths 0-6 in inches 6- Totals *Suazey Scratched *Caliviny Rim Modified, subtype 5 5 6 Cupecoy Series Plain Plain, fine Wide Handled punctated appliquéd punctated lip plate form, red-painted 0 9 8 8 99 7 7 Red Painted thick, rounded rim flanged rim *inward folded rim out-turned rim side-grooved rim 56 4 0 89 6 7 griddle sherds 0 57 98 Pearls Series Plain Red Painted red-painted lip black-painted interior 4 Totals 9 55 977 ^Presence of folded rims and relatively high provenience of Suazey Scratched and Caliviny Rim Modified, subtype 5, should be noted. Suazey Scratched and Caliviny Rim Modified, found together in presumed Carib sites in the Windward Islands, had relatively high proveniences at Cupecoy Bay (Table ). The few Pearls Series sherds hint at a much older occupation during Modified Saladoid times. Other specimens include an Oliva shell bead, three clam shells with rubbed edges, a perforated sherd disc or spindle whorl, a thick and crudely chipped side scraper of chert (Fig. 4, d), fragments of crude unfinished adze- or celt-like tools (Fig. 4, e-f), and nearly thirty smallish chert and quartz chips and wornout cores. The chips and small cores may relate to manioc graters, made by forcing chips into a wooden board, or may be debitage from the manufacture of the larger stone tools. Also found was part of a large sharpening stone (Fig. 4, _i) with a wide, shallow, grinding groove.

56 CULTURAL DIFFUSION Fig. 4. Miscellaneous specimens from Cupecoy Bay, St. Martin a, Oliva shell bead; b, clam shell with rubbed edge; c, perforated sherd or spindle whorl; d., unique scraping plane of a very pale greenish grey stone; e-f, chipped pole ends of broken, unfinished axes or celts, grey green stone; g, flat vessel base; h, rim modified subtype 5 sherd; i, sharpening or grinding stone. Scale in inches. One of the unique features of the Cupecoy Bay collection were seven broken ax- or celt-like tools (Fig. 4, je_-_f) which, presumably, broke during manufacture. They were made of a gray-green chert-like rock which has a coarse crystalline-like structure and does not chip very neatly. A complete specimen, collected in 957 by Dorothy and John Kuer and deposited by them in the United States National Museum, is a percussion chipped celt blank about inch thick and by 6 inches in width and length. Probably these blanks were to be eventually pecked and ground to form celts or axes, possibly of the petaloid form. The last stone specimen (Fig. 4, d) is by far the most unusual. The illustration presents the top view. The bottom is smooth, curves longitudinally, and is unworked except for a few use chips along part of one edge. The material is a fine grained, very pale gray-green chert with fine crossed cracks and inclusions of other materials. It is much finer grained than that of the ax blanks discussed in the previous paragraph and does produce concoidal fractures. This tool was made by percussion flaking with no marginal retouch. The top has been battered so that it resembles, in that respect, a "backed" blade.

BULLEN AND BULLEN 57 Typologically, it is a crude, unifacial, side scraper. If it represents a finished tool, it was probably used as a hand held scraping tool for smoothing wood. Alternatively it could have been hafted to form a pick-like tool. It is probable this tool antedates everything else at Cupecoy Bay. Its closest affinity in the Antilles is with the Casimira complete of the Dominican Republic (Cruxent and Rouse 969:50-5). Its material is not like that found in Antigua or Florida. In Florida and southeastern United States, such tools would be considered to belong to the middle or early Archaic period and to date to approximately 000 to 6000 B.C. Discussion In previous reports (Bullen 964:6; Bullen and Bullen 97:6-64, 66) we have traced the presence of Ostionoid traits in the Lesser Antilles and have pointed out the similarities between Ostionoid ceramic traits (Rouse 95: PI. ) and those of the Caliviny Series (excepting polychrome containers). Examples include double, cylindrical, horn-like handles (Bullen 964:49; Bullen and Bullen 97:4, rim modified subtype ) found on Puerto Rico (Rouse 95: PI., k) ; St. Lucia (Bullen and Bullen 970: Fig. 7, i-k); Mayreau in the Grenadines (Bullen and Bullen 97: PI. VIII, c), and Grenada (Bullen 964: PI. XVII, ). Mario Sanoja (personal communication, Chicago, August 97) advises that the Mayreau variety is also present in eastern Venezuela. Notched or incised appliquëd strips--sesmoidal or otherwise considered Ostiones or later in Puerto Rico (Rouse 95: PI., I) and also found in the Bahamas (Hoffman 970: Fig. 9, FA in Hispaniola (Rouse 94: Pis. 0-; Veloz, et al: 97: Lam. ) and reappear in Surinam (Geijskes 964: Fig. ). While we consider these as emanating from Puerto Rico there are cogent suggestions (Geijskes 964: Figs. -; Lathrop 970: PI. ; Evans, Meggers and Cruxent 959: Fig. and photographs of sherds not included in Figure ; Cruxent and Rouse 959, Vol. : Pis. 8, -;,, 76, 9, 4) that some of these motifs may have diffused in the opposite direction, i. e. from northeastern South America north up the Lesser Antilles. We concluded that influences, presumedly carried by Arawak speakers, brought these traits from Puerto Rico to arrive in St. Lucia, St. Vincent, and Grenada around or before A. D. 000. They may have brought with them small, simple, three-pointed stone zemis and red painted nostril bowls. The later are known for Grenada (Bullen 964: PI. XXII, ), the Grenadines (Bullen and Bullen 97: 9), and St. Lucia, as well as Puerto Rico (Oliver 97:-) and the Dominican Republic (Ortega and Pina 97:0). The former are well known in Guadeloupe, Martinique and St. Lucia, and, are also present on Carriacou in the Grenadines (Bullen and Bullen 97: PI. XI, a) while their furthest south proveniences are Mt. Irvine on Tobago and a site in the Caroni swamp on Trinidad (Thomas Cambridge, personal communication; examined in his collection in I966). Both are associated in Puerto Rico with late Ostionoid ceramics. However, small three-pointed zemis in shell if not in stone--have earlier dates in the Lesser Antilles (Clerc 97:80-8), and, hence, may have diffused in the opposite direction, i.e. from Guadeloupe to Puerto Rico. Both three-pointed stones and nostril bowls are ceremonial items. Their introduction to or development in the Lesser Antilles implies the simultaneous introduction of a new religion or, at least, of new cults. This certainly weakened, if it did not replace, the previously held cosmology of the inhabitants of the Lesser Antilles. To some

58 CULTURAL DIFFUSION extent this weakened or at least changed the then indigenous culture and prepared the way for the Caribs. Only one, possibly ceremonial, pottery type was held over to be still found in Caliviny-Suazey deposits. This is an open, tray-like, container decorated on the inside with zoned bichrome painting in which different colored areas are separated by incision. This has in the past been considered to be a variant of St. Lucia Zoned Incised, but it does not have some traits implicit in the definition of that type while it occurs consistently in significant quantities at such late sites as Lavoutte on St. Lucia (Bullen and Bullen 970: Fig., _i). This Ostionoid expansion is evidenced in Jamaica by plain, appliquéd sigmoid curves (Vanderwal 968: Fig., o) and by the earliest pottery there, known locally as "Little River" (Howard 965:5) and in the Bahamas by appliquéd sigmoid decoration (Rouse 95: PI., _; Hoffman 970: Figs. 8-9, collections at New World Museum, Watling Island). The later, assigned by Rouse to a late Ostiones time period, is also present in the Virgin Islands (Bullen 96: PI. Ill, m). Gashed appliquéd sigmoid curves are also known for both Jamaica (Vanderwal 968: Fig., d_) and Cuba (Alvarez 956: Lam. 55:). Data from Tower Hill, Jamaica, and Cupecoy Bay, St. Martin, indicate these radiating influences from Puerto Rico continued during the succeeding (Puerto Rican) Santa Elena and Esperanza periods. Such influences are clearly evident in the ceramics of Antigua as well as the Virgin Islands (Bullen 96; Pis. III-VII), the Bahamas (Hoffman 970: Fig, 9, g), and Saba (de Josselin de Jong 947; PI. IV, J_l-J, }]_-}8)- For the last island, Saba, de Josselin de Jong (947: Fig. 0) illustrates a casuela vessel sherd with parallel line incision forming a "V" motif on the neck that closely resembles Botany Incised Casuela sherds from Cinnamon Bay on the island of St. John (Bullen 96: Pis. IX, q; X, ). Reminiscent sherds are found in the Esperanza style of Puerto Rico (Rouse 95; PI. 7). Further south these traits are absent or extremely rare. In other words the Bahamas, Virgin Islands, and Greater Antilles including Jamaica and eastern Cuba continued to form an interacting culture area in post-ostiones times but it was cut off from the Lesser Antilles shortly after the start in Puerto Rico of Santa Elena style ceramics. We submit that this cut off was a result of the Carib take over of the Lesser Antilles. The Puerto Rican dates for late Ostiones and for Santa Elena, A.D. 050 and A. D. 890-0 (Rouse 96) interdigitate well with those from the Windward Islands where Ostionoid influences are dated to around A.D. 000 and Carib ascendancy completed by A.D. 00 (Bullen and Bullen 97:5). The Cayo pottery of St. Vincent (Kirby 974) should belong between these dates. Also of interest to the comparative student is the remarkable collection illustrated by Revert (949: Pis. XXI-XXII) from the Anse Belleville site in extreme northwest Martinique, thoughtfully brought to our attention by Jacques Petitjean Roget. Pictured is a carved bone vomit stick, a 47 cm diameter 68. 5 cm high circular bowl with an appliquéd head surrounded by appliquéd arms, a short hammock shaped vessel with a large face modeled on the side, and a unique human face adorno. Clearly these items seem to belong to the Greater Antilles while the accompanying sherds are described as footed bowls with circular feet, vessels of large dimensions, griddles, plate boat-shaped containers, and much red paint as well as some fine paste ceramics. Did Amerindians of the Geater Antilles have a trading post or outlying colony at

BULLEN AND BULLEN 59 Anse Belleville or do the Greater Antillean specimens represent booty from a Carib raid into the Greater Antilles? In closing we wish to point out that the percussion-chipped scrapping tool from Cupecoy Bay documents a "stepping stone" or way station between the newly discovered preceramic of Antigua and similar manifestations in the Greater Antilles. References Cited Alvarez Conde, José 956 Arqueología Indo-Cubana. Publicaciones de la Junta Nacional de Arqueología y Etnología. La Habana. Bullen, Adelaide K. and Ripley P. 970 The Lavoutte site, St. Lucia: A Carib ceremonial center. Proceedings of the Third International Congress for the Study of the Pre-Columbian Cultures of the Lesser Antilles, pp. 6-86. Gainesville. Bullen, Ripley P. 96 Ceramic periods of St. Thomas and St. John islands, Virgin Islands. Wm. L. Bryant Foundation, American Studies, no. 4. Orlando. I964 The archaeology of Grenada, West Indies. Contributions of the Florida State Museum, Social Sciences, no.. Gainesville. Bullen, Ripley P. and Adelaide K. I966 Three Indian sites on St. Martin. Nieuwe West-Indische Gids, vol. 45, nos. -, pp. 7-47. Martinus Nijhoff. ' s-gravenhage. I97 Clerc, Edgar I974 Archaeological investigations on St. Vincent and the Grenadines, West Indies. Wm. L. Bryant Foundation, American Studies, no. 8. Orlando. Les trois-pointes des sites précolombiens de la cote nord-est de la Grande-Terre de la Guadeloupe. Proceedings of the Fourth International Congress for the Study of the Pre-Columbian Cultures of the Lesser Antilles, pp. 7-8. Gainesville. Cruxent, José M., and Irving Rouse I969 Early man in the West Indies. Scientific American, vol., no. 5, pp. 4-5. De Josselin de Jong, J. P. B. I947 Archeological material from Saba and St. Eustatius, Lesser Antilles. Mededelinger van het Rijkmuseum voor Volkenkunde, no.. Leiden. Geyskes, D. C. I964 Arawaks in.the prehistory of Surinam. Premier Congres International d'etudes des Civilisations Précolombiennes des Petites Antilles, pp. 57-78. Fort-de-France.

60 CULTURAL DIFFUSION Hoffman, Charles A., Jr. 970a The Palmetto Grove site on San Salvador, Bahamas. Contributions of the Florida State Museum, Social Sciences, no. 6, pp. -6. Gainesville. 970b Implications from the Mill Reef, Antigua; Sugar Factory, St. Kitts; and Palmetto Grove, San Salvador, sites. Proceedings of the Third International Congress for the Study of the Pre-Columbian Cultures of the Lesser Antilles, pp. 95-06. Gainesville. Howard, Robert R. 965 New perspectives on Jamaican archaeology. American Antiquity, vol., no., pt., pp. 50-55. Salt Lake City. Kirby, I. A., Earle 974 The Cayo Pottery of St. Vincent. Proceedings of the Fifth International Congress for the Study of the Pre-Columbian Cultures of the Lesser Antilles, pp. 6-64. Gainesville. Krieger, Herbert W. 9 Aboriginal Indian pottery of the Dominican Republic. Bulletin of the United States National Museum, no. 56. Washington] MacLaury, James C. 970 Archaeological investigations on Cat Island, Bahamas. Contributions of the Florida State Museum, Social Sciences, no. 6, pp. 7-50. Oliver, Andres 97 Vasija inhalatoria aladoide del Bajo Igneri (Haicienda Grande). Boletín Informativo, vol., no., pp.. Fundación Arqueología Antropología e Histórica de Puerto Rico. Santurce. Ortega, Elpidio, and Plinio Pina 97 Un vaso inhalador de la colección del museo Nacional, Republic Dominicana. Boletin del Museo del Hombre Dominicano, no., pp. 8-. Revert, Eugene 949 Etude Summaire du Materiel Retrouve a l'anse Belleville. La Martinique. Etude Géographique et Humaine. Nouvelles Editions Latines. Paris. Rouse, Irving 94 Culture of the Ft. Liberté Region, Haiti. Yale University Publications in Anthropology, no. 4. New Haven. 95 Porto Rican prehistory: Introduction; excavations in the west and north. Scientific Survey of Porto Rico and the Virgin Islands, vol. 8, pt., pp. 07-460. New York. 96 Final technical report NSF-G4069: dating of Caribbean cultures. New Haven. Mimeographed. 964 Prehistory in the West Indies. Science, vol. 44, no. 68, pp. 499-5. Washington. Vanderwal, Ronald L. 968 The prehistory of Jamaica: A ceramic study. Unpublished M. A. Thesis. University of Wisconsin-Milwaukee. Milwaukee. Veloz Maggiolo, Marcio, Elpidio Ortega, and Plinio Pina 97 Fechas de radiocarbon para el periodo ceramista en la República Dominicana. Buletin del Museo del Hombre Dominicano, pp. 8-98. Santo Domingo.