3?? Caribbetiti iounml of Science. Vol. 44, No. 2, 215-222, 2008 Copyright 2008 College of Arts and Sciaices University of Puerto Rico, Mayag?ez New evidence of Ara autochthones from an archeological site in Puerto Rico: a valid species of West Indian macaw of unknown geographical origin (Aves: Psittacidae) STORES L. OLSON^'' AND EDGAR J. MA?Z L?PEZ,^ 'Department of Vertebrate Zoologu, National Museum of Natural History, Smithsonian Institution, P. O. Box 37012, Washington, D.C. 20013-7012, U.S.A. ^Centro de Estudios Avanzados de Puerto Rico y el Caribe, San Juan, Puerto Rico (USA), 00902-3970. *Corresponding author: olsons@si.edu. ABSTRACT.?The exinct macaw Ara autochthones, previously known only from a single bone from an archaeological site on St. Croix, Virgin Islands, is here identified from several associated bones from an archaeological site in south-central Puerto Rico. The species belongs to a distinctive intermediate size-class and was larger than the Cuban Macaw Ara tricolor. It is assumed to have been endemic to the West Indies, but prehistoric interisland transport of parrots by humans makes interpreting the natural distribution of the species impossible in the absence of fossils. Historical reports of macaws elsewhere in the West Indies are rendered dubious for the same reason. KEYWORDS.?Amazona, biogeography, extinction, human transport, parrots. INTRODUCTION The history and natural distribution of macaws {Ara) in the West Indies are clouded with uncertainties. The only speci- men evidence apart from, archeological re- mains is of the Cuban Macaw Ara tricolor, known from about nineteen skins and ex- tinct since about 1864 (Greenway 1958, Ol- son and Su?rez, in press). Contrary to pre- vious belief, there is no historical evidence for a macaw from Hispaniola (Olson 2005), the second largest of ?ie AntUlean islands. From Jamaica, Guadeloupe, Martinique, Dominica, and an unknown West Indian island, there are 18^*^ and 19*^^ century visi- tors' accounts of various macaws to which no fewer than seven scientific names have been applied {Clark 1905a,b; 1908; Roths- child 1905, 1907a,b), although these are all rightly to be considered entirely hypotheti- cal (Prestwich 1970), In addition, Fisher and Warr (2003) discovered and reproduced a previously unknown painting of a macaw supposedly from Jamaica dating from about 1765. From an archeological deposit on St. Croix in the Virgin Islands, Wetmore (1937) described and named a new species of ma- caw as Ara autochthones, based on a single tibiotarsus of an immature bird. Nothing further regarding this species has turned up in the 60 years since it was described and there has been no further evaluation of the species, which has been mentioned oc- casionally in various checklists and compi- lations; e.g. the curious statement by Prestwich (1970: 199) that: "Nothing ap- pears to have been recorded concerning this rather primitive macaw." flere we report on several associated skeletal elements from another archeologi- cal site in central Puerto Rico that we refer to Ara autochthones. These confirm the va- Kdity of the species and provide proof of the existence of a second species of macaw endemic to the West Indies. Information that we supplied concerning this material formed the basis for Wiley et al. (2004:96) reporting "Ara unknown sp." from Puerto Rico. Unfortunately, as the with archeologi- cal specimens of parrots from elsewhere in the West Indies reported by Williams and Steadman (2001), and in the absence of a fossil record, it is not possible to determine on which island this species of macaw 215 216 S. L. OLSON AND E. J. MA?Z L?PEZ originally evolved because of the potential for extensive trade in parrots among Am- erindians of the Antilles. MATERIALS AND METHODS Comparative material examined.?Skel- etons: Anodorhynchus hyacinthinus MHNT 1045, 1057, 1064, 1496, 1693, 1695, USNM 291249, 319969, 345230, 345854; A. leari FMNH 337716, 337860, 379161, MHNT 1540, 1547; Ara ambiguus LSUM2 90381; USNM 224811; A. ararauna MHNT 242,983, 1165, 1604, USNM 19355, 49891, 223952, 223993, 318791, 322286, 322337, 345207, 345848, 345849, 428243, 489411, 498698, 502499, 502500; A. auricolUs USNM 345846, 345847, 345851, 345852; A. chloropterus MHNT 825, 1653, USNM 225132, 226876, 345850,490125; A. couloni FMNH 291744; A. glaucogularis FMNH 337727, LSUMZ 168622; A. rmcao MHNT 753, USNM 18508, 18988, 226164, 288772, 290508, 321173, 321981, 322058, 322212, 430513, 430516, 431614, 502497,502498; A. manilatus USNM 344700, 345853, 621711, 621949, 622388; A. maracan? FMNH 337756, 390830, 398918, USNM 320003, 344670; A. militaris USNM 288554, 288605, 344772, 344848; A. nobilis USNM 344080, 344081, 502284, 502503, 622355; A. ruhrogenys FMNH 291402, 291404, 337744, MHNT 1812; A. severus FMNH 104484, 290489, 337748, MHNT 388, USNM 19115, 502504; Cyanopsitta spixii MHNT 820, USNM 346722. Measurements were also taken from X-radiographs of two mounted specimens of Ara tricolor USNM 135137, 171767, Qualitative comparisons were made with skeletons of Ara glaucogu- laris i"Ara caninde" auct. FMNH 337727), Anodorhynchus leari (FMNH 337716), and Amazona imperialis (USNM 318792, USNM 321883). RESULTS Genus Ara Lac?p?de, 1799 The new archaeological material is refer- able to Ara, rather than Amazona, the only other genus of large parrots in the West Indies, by the following characters: cora- coid more elongate with relatively nar- rower shaft, ventral lip of glenoid facet more protrudent; hum?rus with ectepicon- dylar process and attachment of pronator brevis situated decidedly more proximad; carpometacarpus proportionately much longer, process of alular metacarpal not curved proximad; femur with head propor- tionately larger; tibiotarsus very distinctive in having the inner cnemial crest more pointed and extending farther proximad, internal condyle much narrower. Although we have followed David and Gosselin (2002) in treating the generic name Ara as masculine, we do not endorse splitting the genus into three by resurrecting the names PrimoUus and Orthopsittaca (Tavares et al. 2006). Recognition of monophyly of the true macaws would be better served by in- cluding Cyanopsitta in Ara as it has long been delimited. Ara autochthones Wetmore, 1937 Holotype.?USNM 483530, left tibiotar- sus; vertebrate paleontological collections (formerly USNM 343033 in the bird collec- tions). Collected in kitchen midden depos- its from Concordia, southwestern St. Crobc, Virgin Islands, in 1934 by L. J. Kom (Wet- more 1937). Referred material.?USNM 448344 verte- brate paleontological collections: left cora- coid lacking a portion of the head, proximal and distal ends of left hum?rus, proximal end of right radius, left carpometacarpus lacking minor metacarpal, left femur lack- ing distal end, right tibiotarsus lacking ex- ternal part of proximal articular surface, proximal fragment and worn distal portion of left tibiotarsus, fragment of shaft (hu- m?rus?), unidentified fragm^ent (perhaps not avian). These bones are evidently a? from a single individual. Locality and age.?Collected by Maiz dur- ing an excavation conducted in March and April 1987 at the Hern?ndez Col?n (PO-13) archaeological site. The site, UTM E 755665/N1998980, represents an inland Saladoid/Ostionoid pre-Columbian Indian village of approximately 15,000 m^. It is lo- cated on the eastern bank of the CerriUos- Bucan? River, south central Puerto Rico, NE of the city of Ponce, Barrio Cerrillos (18? ARA AUTOCHTHONES IN PUERTO RICO 217 04' 05" N; 66? 35' 09 W). It lies at 76m amsl, 13,5 river km from the Caribbean Sea. Physiographically, the Hern?ndez Col?n site is situated in an alluvial terrace within the Semiarid Southern Foothills of Puerto Rico. Ten 2 X 1 m stratigraphie pits were ex- cavated after mapping the site. The pottery seriation and two radiocarbon dates re- vealed a multi-component site, with a local sequence of three archaeological phases: Pomarrosa Phase, Cerrillos Phase, and Ma- ragiiez Phase. The Pomarrosa phase is sty- listically related to the Hacienda Grande ce- ramic style {ca. 200 B.C.-400 A.D.) as defined for Puerto Rico by Alegr?a (1965) and Rouse & Alegr?a (1990). The Cerrillos and Marag?ez phases are in turn related to the Cuevas (400-600 A.D.) and Early Osti- ones (600-900 A.D.) styles as defined by Rouse (1952, 1992). The Hacienda Grande style is included within the Cedrosan Sala- doid subseries of the Saladoid series and corresponds with the first horticultural and ceramics groups that migrated to Puerto Rico from northeastern South America {Rouse, 1992). All cultural and faunal re- mains were collected using three gauges of screens: 6 mm (1/4 inch), 3 mm (1/8 inch) and 1.5 mm (1/16 inch). This archaeologi- cal recovery technique provided abundant zooarchaeological remains. The macaw bones came from pit 8, level 30-40 cm be- low surface (cmbs), from a midden deposit that is located at the base and beginning of the Pomarrosa phase (Ma?z L?pez 2002) within the local sequence. The begirming of the Pomarrosa phase at the Hern?ndez Col?n site is dated at ca. A.D. 300, based on a charcoal sample measurement (2 sigma- cal. 420-870 A.D.-Beta 23902). Measurements (mm) of referred material.? Coracoid: estimated length from head to in- ternal comer of sternal facet 43.5 (42.5 as preserved), length from base of procora- coid process to internal comer of sternal facet 26,5, length and width of glenoid facet 10.5 X 6.1, width and depth of shaft at mid- point 5.2 X 4.4, width of sternal facet 10,5. Hum?rus: proximal width 19.9, depth through external tuberosity 12.8, depth of head 6.2, estimated distal width 15.5, height and width of radial condyle 7.4 x 5.0. Carpometacarpus: length 55.8, proxi- mal depth 14.3, width of trochlea 5.7, width and depth of shaft at midpoint 4,5 x 5.0. Radius: greatest proximal diameter 6.0. Fe- mur: estimated length 51.5, proximal width 12.5, depth through trochanter 8.2, depth of head 6.2, width and depth of shaft at mid- point 4.9 X 5.4. Tibiotarsus: length from proximal articulating surface 74.5, length from distal end of fibuiar crest to external condyle 47.2, depth through inner cnemial crest 11,4, width and depth of shaft at mid- point 5.2 X 4.0, distal width 10.1. Comparisons.?The referred tibiotarsus is essentially identical in size with the holo- type of Ara autochthones and the referred material from Puerto Rico is therefore iden- tified as that species. As Wetmore (1937) noted, the holotype is from a juvenile indi- vidual, so the new material is all the more important for establishing the nature of the species. In size, most living species of ma- caws fall into two separate clusters repre- senting large species and smaller species (Table 1). Ara autochthones is distinct in be- ing intermediate between these two clus- ters. Only Ara glaucogularis and Anodorhyn- chus leari (and presumably the very closely related A. glaucus, which may be only sub- TABLE 1. Length measurements (range and mean in mm) of macaw bones {Anodorhynchus, Ara, Cyanop- sitta). Species are arranged by decreasing mean of the length of the tibiotarsus. Species n Carpometacarpus ; Tibiotarsus An. hyacinlhinus 10 61.3-73.2 (67.9) 87.2-99.1 (93.6) A. ambiguus 2 66.2-69.8 (68.0) 88.6-91.9 (90.2) A. chloropterus 6 63.5-67.9 (65,9) 85.3-89,5 (87.6) A. macao 15 59.3-68,1 (63,5) 79,7-89.2 (85.0) A. militaris 4 61.0-65.6 (63.9) 80.4-85.2 (82.8) A. ararauna 19 58.0-71.3 (63.9) 77.6-87.9 (82.7) An. leari 5 56.3-60.3 (58.0) 77.0-82.6 (79.6) A. autochthones 1 55.8 74.5 A. glaucogularis 2 53.4-57.6 (55.5) 67.8-70.7(69.3) A. rubrogenys 4 46.9^9,9 (48.3) 66.3-68.1 (67.0) A. tricolor 2 42.6-45.2 (43,9) 63.5-64.7 (64.1) A. severus 6 39.3-45.6 (41.31 56.2-61.6 (58.5) C. spixii 2 43.0-43.3 (43.1) 54.4-56.2 (55.3) A. couloni 1 39.2 54.2 A. manilatus 5 40.4-44.5 (41,1) 50.3-55.4 (53.0) A. auricollis 4 35,5-36,4 (35,7) 48,6-51.6(50.5) A. maracan? 5 33,8-37.8 (35.7) 47.6-52.5 (50.4) A. nobilis 5 27.1-28.5 (27.9) 40.3-43.4 (41,9) 218 S. L. OLSON AND E. J. MAlZ L?PEZ specifically distinct [Alvarenga 2007]) are similar in size. Although these species are very far removed geographically from the West Indies, it is still useful to make quali- tative comparisons of them with Ara au- tochthones to reduce the influence of pos- sible size-related differences. Compared with Ara glaucogularis and Anodorhynchus leari, in the hum?rus of Ara autochthones the pectoral attachment is less excavated and the capital groove is wider; the femur has a more massive head and in posterior view is more excavated under the head, neck, and trochanter?the more ro- bust shaft agrees with that in Ara and dif- fers from that of Anodorhynchus; the tibio- tarsus is more robust with the distal extremity more flared. In length the cora- coid, carpometacarpus, and femur are smaller than in either species although the tibiotarsus is longer than in Ara glaucogula- ris but shorter than in Anodorhynchus leari. Additional archeological specimens attribut- ed to Ara.?Two other specimens reported as macaws are known from West Indian archeological sites, A nearly complete cora- coid from Montserrat was stated to be "smaller than in Ara ararauna but larger than in A. severa or A. mantlata, although closer in size to the last two" {Williams and Steadman 2001: 180). Examination of this specimen (UF 4416) shows it to have the more pointed head of Ara versus that of Amazona, The head is broken in the speci- men of Ara autochthones but the bone from Montserrat is slightly smaller {head to in- ternal distal angle 41.6 vs. ca. 43.6 mm). This might be within the range of variation in A. autochthones, or possibly even that of A. tricolor. Its identity will have to remain imcertain in the absence of more material but at this point it cannot be used to estab- lish the existence of a third species of ma- caw in the West Indies. An ulna from an archeological site on Marie Galante was thought by Williams and Steadman (2001) likely to be referable to the hypothetical species Ara guadeloupen- sis Clark (1905a). We examined this speci- men (UF archeological collections Folle Anse 68, Box 68-10), which is lacking the proximal articulation. The distal end is somewhat worn and abraded so that such features as may possibly be diagnostic in this area have been altered to varying de- grees. There does seem to be a notch be- tween the internal condyle and carpal tu- bercle, however, which is like Amazona and unlike Ara. In size and robustness, the specimen is perfectly intermediate between the two individuals of Amazona imperialis that we examined, and we consider that it is probably best referred to that species. Amazona imperialis is probably the same as A. viol?cea, which is known only from de- scriptions of birds from Guadeloupe. From the same site on Marie Galante as the ulna, Williams and Steadman referred a tibiotar- sus to A viol?cea on the basis of its similarity to A. imperialis. DISCUSSION Ara autochthones was a decidedly larger bird than the Cuban A. tricolor (Table 1). It cannot be referred to any other known spe- cies of macaw and must be considered a valid, extinct species. Although Amerindi- ans could have transported macaws from the mainland to the West indies, it is far less likely that a mainland species would have been driven to extinction than a spe- cies restricted to one or more islands in the Antilles. Therefore, we assume that Ara au- tochthones was a West Indian endemic. The specific name autochthones was prob- ably one of the worst possible choices for this bird, as it is unlikely that it ever oc- curred naturally on St. Croix. We may question even whether it occurred natu- rally on Puerto Rico. Although fossil re- mains of both Amazona and Aratinga have been recovered from a number of pre- human sites in Puerto Rico (Olson, unpub- lished data), no fossils of Ara have yet been recovered. This is, however, not at all con- clusive, as macaws are imlikely to occur in cave deposits on an island where the only known cave-inhabiting predator was a rela- tively small bam owl (Tyto). The only fos- sils of macaws f oiuid to date in a paleonto- logical context in the West Indies are three bones of Ara tricolor, two of which were found in aquatic depositional environ- ments (Wetmore 1928, Olson and Su?rez in press). ARA AUTOCHTHONES IN PUERTO RICO 219 The indigenous natives of the West In- dies were excellent boatsmen and engaged in extensive interisland trade involving various commodities including parrots and feathers (Sauer 1966; Rouse 1986, 1992; Keegan 1992). That parrots were important in their culture was apparent from the out- set of interaction with Europeans. When Christopher Columbus first landed in the New World in 1492, somewhere in the Ba- hamas, the inhabitants "brought to us: par- rots, balls of cotton thread, (wooden) spears, and many other things" (Tyler 1998: 38). "In the early years [of Spanish coloni- zation] parrots were frequently mentioned in the islands as well as on Tierra Firme, by the Old World names of papagayo and perico, and the long-tailed brilliantly col- ored ones [macaws] by the Arawak name guacamayo. Along with doves and pigeons, they were appreciated as food by Spaniards as well as by natives, and were very abun- dant" (Sauer 1966: 184). Various other animals, both indigenous and otherwise, were moved about by Am- erindians, doubtless as living individuals, and were reared in captivity in places where some species did not occur naturally (Wing 2001). Agoutis (Dasyprocta) were transported from South America through the Lesser Antilles at least as far north as St. Kitts and St. Eustatius (Wing 1989). The large rodent Capromys pilorides, native to Cuba, was transported from that island to Hispaniola, where remains were found in an archeological context (Rimoli 1974). An- other large rodent, Isolobodon portoricensis, which occurs in pre-human contexts only in Hispaniola, was transported from there to Mona Island, Puerto Rico, and the Virgin Islands, including St. Croix, in all of which places it is found only m midden deposits (Olson and Pregill 1982, Woods 1989, Frank and Benson 1998). The extinct flightless rail Nesotrochis debooyi occurs both in paleonto- logical and archeological contexts in Puerto Rico and has been found in middens in the Virgin Islands, including St. Croix (Olson and Pregill 1982) and isolated Mona Island (Olson unpublished). St. Croix is separated from the Puerto Rican Bank by a deep oce- anic trench and hence was never connected to other islands (Heatwole and Mackenzie FIG. 1. Comparison of appendicular bones of ma- caws: in each group of three. Ara glaucogularis FMNH 337727 is on the left, Anodorhynchus leari FMNH 337716 is on the right, and Ara autochthones USNM 448344 (archeological specimen from Puerto Rico) is in the middle; h = Ara autochthones USNM 483530 holo- typical left tibiotarsus (juvenile). A, left humeri in an- conal view; B, left carpometacarpi in internal view; C, left femora in anterior view; D, right tibiotarsi in an- terior view. Scale = 2 cm. 1967, Pregill 1981). If it had had a native species of flightless rail, one would not ex- pect it to be identical to N. debooyi, so that human transport and captive rearing of that species is a logical conclusion. In northern Mexico and the American southwest, in areas where they do not oc- cur naturally, macaws (mostly Ara macao) were an important item of commerce and ritual among Amerindians and large cap- tive breeding facilities were maintained to sustain ritual sacrifices (Hargrave 1970, Minnis et al. 1993, Creel and McKusick 1994). Being the largest and most colorful of the highly esteemed parrots of the West In- dies, it is altogether probable that macaws were likewise items of great prestige and value in the Antilles and would doubtless have been traded far and wide. Europeans likewise valued macaws and would have carried them between islands and the mainland from the beginning of commerce with the New World. Therefore, 18'^ and 19* century accounts of macaws in Jamaica 220 S. L. OLSON AND E. J. MAlZ L?PEZ FIG. 2. Comparison of appendicular bones of ma- caws: in each group of three. Ara glaiicogularis FMNH 337727 is on the left, Anodorhynchus lean FMNH 337716 is on the right, and Ara autochthones USNM 448344 (archeological specimen from Puerto Rico) is in the middle; h = Ara autochthones USNM 483530, holo- typical left tibiotarsus (juvenile). A, left humeri in pal- mar view; B, left carpometacarpi in external view; C, left femora in posterior view; D, right tibiotarsi in pos- terior view. Scale = 2 cm. and the Lesser Antilles could be based on species originating almost anywhere in the Caribbean region, including the mainland. The macaw illustrated by Fisher and Warr (2003: 156) from several volumes of paintings by a Lt. L. J. Robins entitled The Natural History of Jamaica and dated to 1765, is stylized to a greater or lesser degree, but the overall plumage pattern is very similar to that of the Cuban Macaw Ara tricolor and it may well represent an example of that species that had been taken to Jamaica from Cuba. Although human-caused extinctions on islands has in many cases been rampant (e.g. Olson and James 1982, Steadman 2006), we very much doubt that "each Greater Antillean and Lesser Antillean is- land once sustained one or two indigenous if not endemic species of Ara" (Williams and Steadman 2001: 176). There is no cred- ible evidence in support of such a belief and the only relatively concrete evidence now available?the archeological record? FIG. 3. Comparison of coracoids of macaws (upper row, dorsal view; lower row, ventral view): A, Ara glaucogularis FMNH 337727; B, Ara autochthones USNM 448344 (archeological specimen from Puerto Rico); C, Ara sp. UF 4416 archeological specimen from Mont- serrat (image has been reversed to facihtate compari- son); D, Anodorhynchus lean FMNH 337716. Scale = 2 cm. suggests just the opposite?that there may have been only one other macaw in the An- tilles apart from the Cuban Macaw Ara tri- color. If so, the Antillean island most likely to have harbored a macaw larger than Ara tricolor would be Hispaniola, based on land area alone. It may be significant that if ma- caws were once present on Hispaniola then they seem to have disappeared before the first Spanish chroniclers began document- ing the fauna of the island in the 16* cen- tury (Olson 2005). To paint a completely hypothetical pic- ture, a Hispaniolan macaw may have been so precious a commodity that every pos- sible nest was sought to procure the young ARA AUTOCHTHONES IN PUERTO RICO 221 for trade. By such a means a large and com- paratively uncommon species might con- ceivably be reduced to extinction in the wild over the nearly two millennia that Amerindians occupied Hispaniola. The birds could have been maintained in cap- tivity in native villages through many other parts of the Antilles, but with the collapse of Taino culture shortly following the ar- rival of Europeans (Rouse 1992), the macaw would then have become extinct. Acknowledgments.?Measurements were obtained from specimens in the National Museum of Natural History, Smithsonian Instihition, Washington, DC (USNM); Mu- seu de Historia Natural de Taubat?, Brazil (MHNT, Herculano Alvarenga); Museum of Natural History Louisiana State Univer- sity, Baton Rouge (LSU, Steven W. Cardiff), Field Museum, Chicago (FMNH, Marcelo Stucchi). We are also grateful for the loan of critical skeletons from FMNH. X-radio- graphs of Am tricolor were made by Jeremy Jacobs and measurements calculated from them by Brian Schmidt, who also prepared the figures. We also thank David W. Stead- man for lending archeological specimens from the Florida Museum of Natural His- tory (UF), Gainesville, and for his com- ments on an early version of the manu- script. Ricardo Alegr?a and Francisco Moscoso provided bibliographical assis- tance. LITERATURE CITED Alegr?a, R. E. 1965. On Puerto Rican archaeology. American Antiquify 21:113-131. Alvarenga, H. 2007. Anodorhynchus glaucus e A. lean: osteolog?a, registros f?sseis e antiga distribui?ao geogr?fica. Revista Brasileira de Omitohgia 15: 427- 432. Clark, A. H. 1905a. The Lesser Antillean macaws. Auk 22:266-267. Clark, A. H. 1905b. The Greater Antillean macaws. Auk 22:345-348. Clark, A. H. 1908. The macaw of Dominica. Auk 25: 309-311. Creel, D., and C. McKusick. 1994. Prehistoric macaws and parrots in the Mimbres area. New Mexico. American Antiquity 59:510-524. David, N., and M. Gosselin. 2002. The grammatical gender of avian genera. Bulletin of the British Orni- thologists' Club 122: 257-281. Fisher, C. T., and F. E. Warr. 2003. Museums on paper: library & manuscript resources. Pp. 136-164 in N. J. Collar, C. T. Fisher, and C. J. Feare, eds. Why mu- seums matter: avian archives in an age of extinc- tion. Bulletin of the British Ornithologists' Club 123A, Frank, E. F., and R. Benson. 199S. Vertebrate paleon- tology of Isla de Mona, Puerto Rico. Journal of Cave and Karst Studies 60:103-106. Greenway, J. C, Jr. 1958. Extinct and vanishing birds of the world. New York, American Committee for In- ternational Wild Life Protection, Special Publica- tion No. 13. Hargrave, L. L. 1970. Mexican macaws. Comparative osteology and survey of remains from the south- west. Anthropological Papers of the University of Ari- zona 20:1-67. Heatwole, H., and F. Mackenzie. 1967. Herpetogeog- raphy of Puerto Rico. IV. Paleogeography, faunal similarity and endemism. Evolution 21:429-438. Keegan, W. 1992. The people luho discovered Columbus: the -prehistory of the Bahamas. Gainesville, University Press of Florida, Ma?z L?pez, E. J. 2002. El Sitio Hern?ndez Col?n: Ac- tividades subsistenciales de los antiguos habitantes del Valle del Rio Cerrillos-Bucan?, Ponce, Puerto Rico. Unpublished master's thesis. San Juan, Puer- to Rico, Centro de Estudios Avanzados de Puerto Rico y el Caribe. Minnis, P. E., M. E. Whalen, J. H. Kelley, and J. D. Stewart. 1993. Prehistoric macaw breeding in the North American Southwest. American Antiquity 58: 270-276. Olson, S. L. 2005. Refutation of the historical evidence for a Hispaniolan macaw (Aves: Psittacidae: Ara). Caribbean journal of Science 49 (2):319-323. Olson, S. L., and H. F. James. 1982. Fossil birds from the Hawaiian Islands: evidence for wholesale ex- tinction by man before Western contact. Science 217:633-635. Olson, S. L., and G. K. PregiU. 1982. A brief introduc- tion to the paleontology of Bahaman vertebrates. Pp. 1-7 in S. L. Olson, ed. Fossil vertebrates from the Bahamas. Smithsonian Contributions to Paleobiol- ogyiS. Olson, S. L., and W. Su?rez. In press. A fossil cranium of the Cuban Macaw Ara tricolor (Aves: Psittacidae) from Villa Clara Province, Cuba. Caribbean Journal of Science. PregiU, G. 1981. Late Pleistocene herpetofaunas from Puerto Rico. University of Kansas Museum of Natural History Miscellaneous Publication 71:1-72. Prestwich, A. A. 1970. Extinct, vanishing, and hypo- thetical parrots. AvicuUural Magazine 76:198-204. Rimoli, R. 0.1974. Prueba zooarqueologica de transito del hombre precolombino de Cuba hacia la His- paniola. Revista Dominicana de Antropolog?a e Histo- ria 4:27-28. Rothschild, W. 1905. [Notes on extinct parrots from the West Indies]. Bulletin of the British Ornitholo- gists' Club 16:13-15. Rothschild, W. 1907a. On extinct and vanishing birds. 222 S. L. OLSON AND E. J. MA?Z L?PEZ Proceedings of the Forth International Ornithological Congress London june 1905:191-217. Rothschild, W. 1907b. Extinct birds. London, Hutchin- son & Co. Rouse, L 1952. Porto Rican Prehistory. In Scientific Sur- vey of Porto Rico and the Virgin ?shinds, 18, pts 3^. New York, New York Academy of Science, Rouse, L 1986. Migrations in prehistory. New Haven, Yale University Press. Rouse, I. 1992. The Tainos. Rise and decline of the people who greeted Columbus. New Haven, Yale University Press. Rouse, I., and Alegr?a, R. E. (1990). Excavations at Maria de la Cruz Cave and Hacienda Grande Vil- lage Site, Loiza, Puerto Rico. Yak University Publi- cations in Anthropology 80. Sauer, C. O. 1966. The early Spanish Main. Berkeley, University of California Press. Steadman, D. W. 2006. Extinction & biogeography of tropical Pacific birds. Chicago, University of Chicago Press. Tavares, E. S., A. J. Baker, S. L. Pereira, and C. Y. Miyaki. 2006. Phylogenetic relationships and his- torical biogeography of neotropical parrots (Psit- taciformes: Psittacidae: Arini) inferrred from mito- chondrial and nuclear DNA sequences. Systematic Biology 55: 454-470. Tyler, S. L. 1988. Two worlds. The indian encounter with the European 1492-1509. Salt Lake City, University of Utah Press. Wetmore, A. 1928. Bones of birds from the Ciego Mon- tero deposit of Cuba. American Museum Novitates 301:1-5. Wetmore, A. 1937. Ancient records of birds from the island of St. Croix with observations on extinct and living birds of Puerto Rico, journal of Agriculture of the University of Puerto Rico 21:5-15. Wiley, J. W., R. S. Gnam, S. E. Koenig, A. Domelly, X. Calvez, P. E. Bradley, T. White. M. Zamore, P. R. Reillo, and D. Anthony. 2004. Status and conserva- tion of the family Psittacidae in the West Indies. journal of Caribbean Ornithology 17: 94-154. Williams, M. L, and D. W. Steadman. 2001. The his- toric and prehistoric distribution of parrots (Psitta- cidae) in the West Indies. Pp.175-198 in C. A. Woods and F. E. Sergile, eds. Biogeography of the West Indies. 2""* ed. Boca Raton, Florida, CRC Press. Wing, E. S. 1989. Human exploitation of animal re- sources in the Caribbean. Pp. 137-152 in C. A. Woods, ed. Biogeography of the West Indies, Past, Pre- sent and Future. Gainesville, Florida, Sandhill Crane Press. Wing, E. S. 2001. The sustainability of resources used by native Americans on four Caribbean islands. International Journal of Osteoarchaeology 11:112-126. Woods, C. A. 1989. Biogeography of West Indian ro- dents. Pp. 741-798 in C. A. Woods, ed. Biogeography of the West Indies, Past, Present and Future. Gaines- ville, Florida, Sandhill Crane Press.