sign in sign up
<<
Home , French Polynesia

A New Eastern Limit of the Pacific Flying Fox, Pteropus tonganus (Chiroptera: Pteropodidae), in Prehistoric : A Case of Possible Human Transport and Extirpation1

Marshall I. Weisler,2 Robert Bollt,3 and Amy Findlater4

Abstract: Five bones, representing one adult of the Pacific Flying Fox, Pteropus tonganus, were recovered from an archaeological site on (151 21 0 W, 22 270 S), , , making this the most eastern extension of the species. For the first time, flying fox bones from cultural deposits were directly dated by accelerator mass spectrometry, yielding an age of death be- tween A.D. 1064 and 1155. Their stratigraphic position in an Archaic period ar- chaeological site and the absence of bones in the late prehistoric to historic layers point to extirpation of the species. No flying fox bones were found in pre- human deposits and human transport of the species cannot be ruled out.

During the past few decades archaeo- 1994), and possibly low coral (Weisler zoology (the study of animal bones from 2001). In essence, no insular was archaeological sites) has made important con- immune from the effects of colonizing hu- tributions to island biogeography by recover- mans who cleared land for horticulture ing bones of birds previously unknown to (Kirch 1983, Athens and Ward 1993) and in- science (Steadman 1995), extending the range troduced commensal animals such as pigs, of extant species (Wragg 1995), and under- , chickens, and rats (Green and Weisler standing the dynamic relationships of people, 2004, Matisoo-Smith and Robins 2004) as fauna, and flora within Pacific island eco- well as vital cultigens, including medicinal systems during the late Pleistocene and and ornamental plants (Barrau 1963, Yen throughout the Holocene. It is now relatively 1973, Kirch 1982). commonplace to document human-caused In this paper we report what is now the avian extinctions on high volcanic islands most easterly occurrence of the Pacific Flying such as Hawai‘i (Olson and James 1982, Bur- Fox (Pteropus tonganus), recovered from ar- ney et al. 2001), the large continental islands chaeological deposits dating to as early as the of (Anderson 1989, Worthy eleventh century A.D. on Rurutu (151 21 0 1998), landforms in the southern W, 22 27 0 S), one of the seven Austral Is- Cooks (Steadman and Kirch 1990) and Hen- lands, located @470 km south of in derson, (Wragg and Weisler central East Polynesia (Figure 1). The sample size of five bones represents at least one adult individual. Their stratigraphic position in an 1 This project was supported by a National Science Archaic period archaeological site, and the Foundation doctoral dissertation improvement grant distance of Rurutu from documented popu- (No. BCS04-37581) to R.B. Manuscript accepted 3 Au- lations of flying foxes in the southern Cook gust 2005. Islands, @600 km to the west, may suggest 2 School of Social Science, University of Queensland, human introduction. No flying fox bones St. Lucia 4072, Queensland, . 3 Department of Anthropology, University of Hawai‘i were recovered in prehuman deposits, and at Ma¯noa, , Hawai‘i 96822. the small number of bones in the oldest cul- 4 Department of Anthropology, University of Otago, tural layer suggests that the Pteropus popula- Dunedin, New Zealand. tion was never very large. We also provide the first direct radiocarbon age determina- Pacific Science (2006), vol. 60, no. 3:403–411 tions on Pacific Flying Fox from prehistoric : 2006 by University of Hawai‘i Press contexts and discuss some probable causes of All rights reserved its extirpation on Rurutu.

403 404 PACIFIC SCIENCE . July 2006

Figure 1. Map showing the location of Rurutu in the Austral Islands. The archaeological excavations were conducted at Uramoa near a pass, just south of Peva on the east coast.

The Pacific Flying Fox is one of the most Ethnographic and Archaeological Evidence of the widespread Pteropus species and is currently Flying Fox in Polynesia found from the to the south- ern (Flannery 1995, Miller and The Pacific Flying Fox was clearly a food Wilson 1997). Martin (1968) documented that item, although it was considered sacred on flying foxes were taken to Tahiti in the mid- some islands. Beginning in Uvea, West Poly- nineteenth century, and Flannery (1995) nesia, the flying fox was sometimes captured believes that because carried fly- by shaking the branches from which it hangs ing foxes as pets it is possible that some pop- to sleep, although by 1932 it was only taken ulations result from introductions. This is with shotguns (Burrows 1937). During his merely the historic continuation of mid- fieldwork in during 1920–1921, Gif- Holocene introductions of animals by hu- ford (1929) recorded that the flying fox was mans in island that extended still considered a deity on Tongatapu. In Tu- species beyond their natural range (Flannery tuila, , Buck (1930) reported that flying et al. 1988). Our contention here is that P. foxes ( pe‘a) were caught at night in a fowling tonganus may have been transported from the net with a long handle as they flew low to the southern Cook Islands to the Australs some- ground; occasionally they were hunted with time during the eleventh century A.D. Direct bow and arrow. Polynesians built permanent transfer is unlikely because the first apprecia- platforms in the tops of tall trees in ble landfall from the southern Cooks is Rima- where, in the late afternoon, a snare was tara, some 150 km west of Rurutu, although used to trap a flying fox ( peka) that was used no records of flying foxes are known from as a decoy to attract other individuals that there. This opens up the possibility that pre- were caught in a net. Up to 200 could be cap- historic records of flying foxes may be discov- tured in one night (Loeb 1926). There is also ered in other islands in the Australs. a Niuean legend that describes how the flying Eastern Limit of Pteropus tonganus in Prehistoric Polynesia . Weisler et al. 405 fox carries her ‘‘children’’ clinging to her agree with Worthy’s conclusion that the breast instead of leaving them behind in the Niue bones are contemporaneous with pre- nest like a pigeon (Loeb 1926). Henry’s historic occupation but do not provide evi- (1928) major treatise on ancient Tahiti was dence for prehuman existence of fruit bats based on observations made by Reverend on the island, although it could well be an in- Orsmond in the early nineteenth century and digenous taxon based on its widespread oc- no mention was made of the flying fox, currence throughout the Southwest Pacific suggesting that it was of little importance, (T. Worthy, pers. comm., 2005). Three of extirpated, or perhaps never present prehis- the southern Cook Islands have prehistoric torically. The flying fox was, however, a records of P. tonganus (known locally as moa mid-nineteenth-century introduction (Martin kirikiri). From the west coast of Aitutaki, a 1968). The Pacific Flying Fox is thought to sandy archaeological deposit dating to <1000 be indigenous to the southern Cook Islands, B.P. contained a molar and wing phalanx where historically it is known from Mangaia (Steadman 1991). Walter (1998) found a sin- and Rarotonga (Hill 1979, Flannery 1995), gle flying fox mandible in an Archaic period which, until now, was thought to be the east- deposit on Ma‘uke, and the small quantity of ern limit of the species. F. Alan Seabrook, flying fox bone suggests that the taxon was who did extensive ethnographic fieldwork on extirpated soon after human occupation. The Rurutu in the 1930s, stated (1938), ‘‘The most extensive assemblage of Pacific Flying Cook Islands bat did not reach Rurutu.’’ By Fox bones from any prehistoric site in Poly- the early 1920s, there was no record of the nesia was recovered from a habitation rock- flying fox in , Austral Islands, when shelter on Mangaia from layers dating to the Aitken (1930) spent eight months inventory- eleventh to fourteenth centuries (Steadman ing the flora, fauna, and customs. and Kirch 1990). The frequency of the bones The published archaeological occurrence at the site diminishes over time, suggesting of Pacific Flying Fox in Polynesia is limited local depression of the population. There is to Tonga, Niue, and the southern Cook Is- no evidence of flying fox bones in the Cook lands. On ‘Eua, southern Tonga, Koopman Islands before human occupation, but, as on and Steadman (1995) recovered 17 bones of Niue, we would expect indigenous popu- P. tonganus (including a mandible, lower lations. These sites of the southern Cook Is- incisor, three radii, three metacarpals, seven lands are contemporaneous with one another, manal phalanges, a femur, and a tibia) from as well as with the site on Rurutu—all dating excavations at three limestone caves. Only to the Archaic period of East Polynesia, when one of these bones was reported as coming most insular extinctions took place. from prehuman deposits, and two specimens were from the cultural layer in ‘Anatu´ cave The Study Area dated to 570 G 70 to 2710 G 70 B.P. (Koop- man and Steadman 1995). An additional 17 The Austral Islands of French Polynesia in- specimens were mostly fragmentary and may clude the volcanic islands , Rurutu, represent either P. tonganus or P. samoensis Tubuai, , Rapa, and two uninhabited (Koopman and Steadman 1995). On the ma- islets, Maria (formerly Hull), an , and katea island of Niue, 11 bones and one tooth the small isolated rock spires of . of P. tonganus were recovered from surface The Australs have the smallest total landmass and excavated layers from one archaeological (144 km2) of any East Polynesian archipelago and one ‘‘non-cultural’’ context (Worthy et except the Pitcairn group (43 km2). Although al. 1998, Worthy et al. 2002). None of the high islands, their elevation is not sufficient bones or contexts was dated, but eight of the to produce orographic rain. The Australs are bones from surface and excavated deposits at the southernmost archipelago in French Poly- Ulupaka Cave, south entrance, are assumed nesia, lying across the . to be of archaeological origin (Worthy et al. The average yearly temperature is 23.1C with 1998; T. Worthy, pers. comm., 2005). We an average annual rainfall of 1,848 mm/yr 406 PACIFIC SCIENCE . July 2006

(ORSTOM 1993). Distances of approxi- in situ combustion features or carbonized mately 150–200 km separate the islands from wood dispersed throughout deposits has been each other. radiocarbon dated (e.g., Steadman and Kirch The nearest neighbors of Rurutu, our 1990). Because Pteropus roosts primarily in study island, are Tubuai, 225 km to the east, trees (Miller and Wilson 1997), it is possible and Rimatara, 150 km distant. Rurutu has a that bones could enter archaeological sites landmass of 32.3 km2 and a maximum eleva- through natural death and not as human tion of 389 m. It consists of a volcanic core, food refuse (Weisler and Gargett 1993). ringed by makatea limestone, which faces the Consequently, we selected the two radii frag- sea in escarpments up to 100 m high. These ments of P. tonganus for accelerator mass cliffs subdivide the island into valleys. The spectrometry (AMS) age determinations. The vegetation of Rurutu was exceptionally im- samples were processed at the National Ocean poverished compared with that of most other Sciences Accelerator Mass Spectrometry Polynesian high islands by the time of - Facility (NOSAMS), Woods Hole Oceano- pean contact in the late eighteenth century. graphic Institution, Woods Hole, Massachu- Its barren landscape prompted Joseph Banks setts. The Pacific Flying Fox bones were (1962) to write in 1769, ‘‘The Island to all ap- processed using organic combustion to pro- pearance that we saw was more barren than duce CO2. Collagen was extracted from each any thing we have seen in these seas.’’ The bone using the EDTA (ethylenediaminetetra- small coastal plains that border the island acetic acid) method. The collagen was then consist of sand and coral debris. Rurutu’s combusted and converted to graphite. fringing reef is generally small and straight with neither a nor a barrier reef. The results reef flats are for the most part 100 m wide but not uniform around the island (Pirazzoli and The material recovered from the marae ‘‘Ura- Salvat 1992). moa’’ site excavation includes typical Archaic East Polynesian artifacts such as untanged materials and methods adzes and one-piece pearl-shell fishhooks. The excavation yielded abundant faunal re- From May to August 2003, R.B. excavated an mains (mostly fish, but also turtle and pig, Archaic East Polynesian site in the sector of with lesser quantities of rat and bird) and Peva Rahi, Peva Valley, on the east coast of charcoal from earth ovens in intact strati- Rurutu (Figure 1). The excavation took place graphic layers representing two distinct cul- on the grounds of the marae (temple) ‘‘Ura- tural occupations. The lower layer D (@80– moa,’’ constructed atop a coastal sand dune, 110 cm below surface) represents a habitation on the terrain known as Onatietie. It is situ- site from the Archaic period (beginning as ated approximately 100 m from the coast in early as the eleventh century A.D. and con- an overgrown section of unused private land. tinuing until ca. A.D. 1400), and the upper The dune extends the entire 1-km length of layer A (0–20 cm below surface), associated Peva Iti. All sediments were dry-screened us- with the marae, is from the Classic/Early His- ing 1/8-inch (3.2-mm) mesh. Further details toric period (late eighteenth to early nine- of the excavations are presented in Bollt teenth century A.D. [Figure 2]). (2005). The Pacific Flying Fox bones are from units G10 and G11, layer D of the excavation site and consist of one proximal end and one Radiocarbon Dating distal end of radius that do not Pacific Flying Fox bones have never been di- fit together but have a combined length of rectly dated. In most examples, Pteropus has @150 mm, which is within the range of vari- been dated through its association with pre- ability of the forearm length range of P. ton- historic artifacts, food remains, and cooking ganus as recorded by Flannery (1995); one stones in cultural layers where charcoal from index claw; and one left mandible (Figure 3). Eastern Limit of Pteropus tonganus in Prehistoric Polynesia . Weisler et al. 407

At least one adult individual is represented. the Department of Anthropology, University Trevor Worthy identified the flying fox man- of Hawai‘i at Ma¯noa. dible, and R.B. identified four other bones Table 1 presents the results of the age- with a Pteropus reference specimen held at determination analysis. Samples were cali- brated with Reimer et al. (2004) for terrestrial samples after subtracting 10 yr from the mean for material (Stuiver et al. 1998). Because the two samples are likely from the same individual, we averaged the calibrated ages (Reimer et al. 2004), and there is a 74% probability that the age of death of the flying fox was between A.D. 1064 and 1155. However, it must be noted that these dates are approximately 200 yr older than the average of six charcoal dates (on unidenti- fied wood) from the same cultural layer (Bollt 2005). This discrepancy may be due to post- depositional contamination of the bone samples (Hedges and Van Klinken 1992). Considering all eight radiocarbon age deter- minations, however, the deposit is securely within the Archaic period, when similar extinction and extirpation events occurred throughout East Polynesia.

discussion and conclusions

Figure 2. A typical stratigraphic profile of the Marae The suggestion by Flannery (1995) that hu- Uramoa site showing prehistoric cultural layers A and mans may have been a dispersal agent for the D. The lowest cultural layer contained bones of the Pacific Flying Fox seems possible with the Pacific Flying Fox, Pteropus tonganus. G13 and G14 refer to excavation units. Designations such as 10YR3/1 refer identification of the five bones of this taxon to Munsell soil/sediment colors, generally from dark from the archaeological site at Peva, Rurutu, (upper layer) to lighter (lower layers). in the Austral Islands—now the most easterly

Figure 3. The left mandible of Pacific Flying Fox (Pteropus tonganus), seen in superior, lateral, and medial views. 408 PACIFIC SCIENCE . July 2006

TABLE 1 Radiocarbon Age Determinations for Pacific Flying Fox Bones from the Peva Site, Unit G10, Layer D

2s Cal Age Relative Area Lab. No. 14C Age yr B.P. 13C/12C Ranges A.D. under Distribution

NOSAMS48011 995 G 35 18.22 990–1048 0.50 1087–1122 0.38 1138–1150 0.11 NOSAMS48049 920 G 30 17.79 1028–1184 1.00

known extent of the species. Only one bone generally unsuitable habitat for flying foxes of P. tonganus has been recovered from pre- and may have been a natural barrier that pre- human contexts in the whole of Polynesia; vented the further eastern extension of the Koopman and Steadman (1995) reported species. the find from Tonga. The Pacific Flying Fox The five bones recovered from Rurutu has not been found in prehuman contexts are probably from the same adult individual, on Niue, the Cook Islands, or Rurutu— and direct radiocarbon age determinations currently the only other archaeological oc- yield a time of death during the mid-eleventh currences of the species in Polynesia. We to twelfth centuries, or perhaps to the four- agree with Flannery that human translocation teenth century if wood charcoal dates are a of the flying fox leading to establishment of more reliable indication. Despite intensive a thriving population seems unlikely (T. excavations covering 46.5 m2 of the site, no Flannery, pers. comm., 2005), especially other flying fox bones were recovered, sug- given the archaeological records for Ma‘uke gesting that this species was never common and Rurutu, where the flying fox was extir- on Rurutu during the Archaic period— pated soon after human occupation. Given assuming that it was a desirable food item. its biogeographic history, it is likely that the That no flying fox bones were recovered flying fox dispersed to the eastern margins of from a large excavated sample in the late its range unaided by humans, but we strongly prehistoric/Early Historic layer A suggests advocate direct dating of flying fox bones to that P. tonganus was extirpated—although ad- clearly document the occurrence of Pteropus ditional archaeological sites on Rurutu with before human settlement. similar stratigraphic evidence would provide The flying fox is not known from prehis- a clearer picture of the process of extirpation. toric deposits in the , yet fu- The paucity of bird bones in layer A also ture excavations using fine-mesh sieves may documents that other endemic species had recover bones of this species: the distance been drastically diminished by the fourteenth from the southern Cook Islands to the Aus- century A.D. trals and the high islands of the Societies Due to the relatively impoverished nature (with favorable habitats) is quite similar. The of the Rurutu vegetation by the late eigh- emerged only during the late Holo- teenth century, as reported by early European cene (Dickinson 2004), and, consequently, visitors, it is likely that deforestation played a flying fox could not use these islands as ‘‘step- major role in the extirpation of the flying fox ping stones’’ to (@1,400 km from on the island. With their low reproductive the Australs) or to the Marquesas (@1,800 rate, these animals are very sensitive to over- km), both of which do not have records of hunting, introduced predators and competi- flying fox. However, it is unlikely that flying tors (such as the Pacific rat, Rattus exulans), foxes could have made such long water cross- and, most important, habitat loss (Miller and ings. Composed of low coral atolls and one Wilson 1997). Rurutu is relatively low in alti- uplifted makatea island, the Tuamotus are tude for a high island, and the vast majority Eastern Limit of Pteropus tonganus in Prehistoric Polynesia . Weisler et al. 409 of its terrain is easily accessible on foot. On of Rurutu and specifically its , Fre´de´ric such an island, mass deforestation could have Rive´ta, as well as the site landowner, Fernand occurred quickly after human settlement—a Roomataaroa; his hosts on the island, Pierre scenario that was played out on many islands Atai, Ingride Drollet, and their sons Takiri across the Pacific (Rolett and Diamond 2004). and ; Rurutu’s Minister of Tourism, Although human predation is clearly a Yves Gentilhomme; the government of major reason for the disappearance of flying French Polynesia, its former Ministre de la foxes in Samoa (Craig et al. 1994) and pro- Culture, Louise Pelzter, and Henri Marchesi bably most oceanic islands, small insular pop- of the Service de la culture et du Patrimoine ulations are also very susceptible to abrupt and the Muse´e de Tahiti et ses Iles; and espe- landscape changes, whether human induced cially his assistant on Rurutu, . We (Kirch 1983) or the result of drought or hur- thank Trevor Worthy for identifying the fly- ricanes (Craig et al. 1994). These climatic ing fox mandible. We appreciate the helpful perturbations must be considered when de- comments made by Carl Christensen and Da- termining the causes of insular extirpations vid Steadman. Les O’Neill and Michael Has- and extinctions, and it is noteworthy that lam prepared the figures. drought is not an important problem in the Australs, but hurricanes have been known to Literature Cited do considerable damage historically. Forest clearance by humans caused increased erosion Aitken, R. T. 1930. Ethnology of Tubuai. and greatly reduced the food supply for in- Bernice P. Bishop Mus. Bull. 70. digenous species and available roosts or nest- Anderson, A. 1989. Prodigious birds: Moas ing sites. The Pacific rat, also introduced by and moa-hunting in prehistoric New Zea- humans across (Matisoo-Smith and land. Cambridge University Press, Cam- Robins 2004), was an important predator of bridge. ground-nesting birds (Brooke 1995), and it Athens, J. S., and G. Ward. 1993. Environ- may have competed for similar vegetarian mental change and prehistoric Polynesian foods of flying foxes. Humans undoubtedly settlement in Hawai‘i. Asian Perspect. contributed both directly (through predation 32:205–224. and habitat destruction) and indirectly (by Banks, J. 1962. The Endeavor journal of Jo- the introduction of a potential competitor, seph Banks, 1768–1771, ed. J. C. Beagle- the Pacific rat) to the extirpation of the Pa- hole. Angus and Robertson, Sydney. cific Flying Fox on Rurutu. Barrau, J., ed. 1963. Plants and the migrations Because the Pacific Flying Fox is a gener- of Pacific peoples: A symposium. Bishop alist forger that feeds on fruit, it disperses Museum Press, Honolulu. seeds and acts as a pollinator. In this capacity, Bollt, R. 2005. Peva: The archaeology of a it may aid plants in the colonization of an is- valley on Rurutu, Austral Islands. Ph.D. land (Elmqvist et al. 1992) and disperse seeds diss., University of Hawai‘i at Ma¯noa, Ho- to regenerate areas affected by hurricanes. In nolulu. essence, the Pacific Flying Fox is a good ‘‘ba- Brooke, M. deL. 1995. The breeding biology rometer’’ of the overall health of an island. of the gadfly petrels Pterodroma spp. of the The extirpation of the species is one of many Pitcairn Islands: Characteristics, popula- signals that an island underwent stress from tion sizes and controls. Biol. J. Linn. Soc. human-induced and natural phenomena (e.g., 56:213–231. hurricanes and drought). Buck, P. H. (Te Rangi Hiroa). 1930. Samoan material culture. Bernice P. Bishop Mus. acknowledgments Bull. 75. Burney, D. A., H. F. James, L. P. Burney, R.B. thanks the National Science Foundation S. L. Olson, W. Kikuchi, W. L. Wagner, for supporting his research; Pierre Ve´rin for M. Burney, D. McCloskey, D. Kikuchi, introducing him to Rurutu; the community F. V. Grady, R. Gage, and R. Nishek. 410 PACIFIC SCIENCE . July 2006

2001. Fossil evidence for a diverse biota Koopman, K. F., and D. W. Steadman. 1995. from Kaua‘i and its transformation since Extinction and biogeography of bats on human arrival. Ecol. Monogr. 71:615–641. ‘Eua, Kingdom of Tonga. Am. Mus. Burrows, E. G. 1937. Ethnology of Uvea Novit. 3125. (Wallis Island). Bernice P. Bishop Mus. Loeb, E. M. 1926. History and traditions of Bull. 145. Niue. Bernice P. Bishop Mus. Bull. 32. Craig, P., P. Trail, and T. E. Morrell. 1994. Martin, H. B. 1968. The Polynesian journal The decline of fruit bats in American Sa- of Captain Henry Byam Martin. Austra- moa due to hurricanes and overhunting. lian National University Press, . Biol. Conserv. 69:261–266. Matisoo-Smith, E., and J. H. Robins. 2004. Dickinson, W. R. 2004. Impacts of eustasy Origins and dispersals of Pacific peoples: and hydro-isostasy on the evolution and Evidence from mtDNA phylogenies of the landforms of Pacific atolls. Palaeogeogr. Pacific rat. Proc. Natl. Acad. Sci. U.S.A. Palaeoclimatol. Palaeoecol. 213:251–269. 101:9167–9172. Elmqvist, T., P. A. Cox, W. E. Rainey, and Miller, C. A., and D. E. Wilson. 1997. Ptero- E. D. Pierson. 1992. Restricted pollination pus tonganus. Mamm. Species 552:1–6. on oceanic islands: Pollination of Ceiba Olson, S. L., and H. F. James. 1982. Fossil pentandra by flying foxes in Samoa. Bio- birds from the : Evidence tropica 24:15–23. for wholesale extinction by man before Flannery, T. 1995. Mammals of the South- Western contact. Science (Washington, west Pacific and Moluccan Islands. Cornell D.C.) 217:633–635. University Press, Ithaca, New York. ORSTOM. 1993. Atlas de la Polyne´sie fran- Flannery, T., P. V. Kirch, J. Specht, and M. c¸aise. . Spriggs. 1988. Holocene mammal faunas Pirazzoli, P. A., and B. Salvat. 1992. Ancient from archaeological sites in island Melane- shorelines and Quaternary vertical move- sia. Archaeol. Oceania 23:89–94. ments on Rurutu and Tubuai (Austral Gifford, E. W. 1929. Tongan society. Bernice Isles, French Polynesia). Z. Geomorphol. P. Bishop Mus. Bull. 61. N. F. 36 (4): 431–451. Green, R. C., and M. I. Weisler. 2004. Pre- Reimer, P. J., M. G. L. Baillie, E. Bard, A. historic introduction and extinction of an- Bayliss, J. W. Beck, C. J. H. Bertrand, imals in Mangareva, Southeast Polynesia. P. G. Blackwell, C. E. Buck, G. S. Burr, Archaeol. Oceania 39:34–41. K. B. Cutler, P. E. Damon, R. L. Edwards, Hedges, R. E. M., and G. J. Van Klinken. R. G. Fairbanks, M. Friedrich, T. P. 1992. A review of current approaches in Guilderson, A. G. Hogg, K. A. Hughen, the pretreatment of bone for radiocarbon B. Kromer, G. McCormac, S. Manning, dating by AMS. Radiocarbon 34:279–291. C. B. Ramsey, R. W. Reimer, S. Remmele, Henry, T. 1928. Ancient Tahiti. Bernice P. J. R. Southon, M. Stuiver, S. Talamo, Bishop Mus. Bull. 48. F. W. Taylor, J. Van Der Plicht, and C. E. Hill, J. E. 1979. The Flying Fox Pteropus ton- Weyhenmeyer. 2004. IntCal04 terrestrial ganus in the Cook Islands and on Niue Is- radiocarbon age calibration 0–26 cal kyr land, Pacific Ocean. Acta Theriol. 24:115– B.P. Radiocarbon 46:1029–1058. 122. Rolett, B., and J. Diamond. 2004. Environ- Kirch, P. V. 1982. The impact of the prehis- mental predictors of pre-European defor- toric Polynesians on the Hawaiian ecosys- estation on Pacific islands. Nature (Lond.) tem. Pac. Sci. 36:1–14. 431:443–446. ———. 1983. Man’s role in modifying tropi- Seabrook, F. A. 1938. Rurutuan culture. cal and subtropical Polynesian ecosystems. Manuscript. Bernice P. Bishop Museum, Archaeol. Oceania 18:26–31. Honolulu. Kirch, P. V., J. R. Flenley, D. W. Steadman, Steadman, D. W. 1989. Extinction of birds in F. Lamont, and S. Dawson. 1992. Ancient eastern Polynesia: A review of the record, environmental degradation. Natl. Geogr. and comparisons with other Pacific Island Res. Explor. 8:166–179. groups. J. Archaeol. Sci. 16:177–205. Eastern Limit of Pteropus tonganus in Prehistoric Polynesia . Weisler et al. 411

———. 1991. Extinct and extirpated birds Weisler, M. I., and R. Gargett. 1993. Pacific from Aitutaki and Atiu, southern Cook Is- island avian extinctions: The taphonomy lands. Pac. Sci. 45:325–347. of human predation. Archaeol. Oceania ———. 1995. Prehistoric extinctions of 28:85–93. Pacific Island birds: Biodiversity meets Worthy, T. 1998. A remarkable fossil and ar- zooarchaeology. Science (Washington, chaeological avifauna from Marfells Beach, D.C.) 267:1123–1131. Lake Grassmere, South Island, New Zea- Steadman, D. W., and P. V. Kirch. 1990. land. Rec. Canterbury Mus. 12:79–176. Prehistoric extinction of birds on Mangaia, Worthy, T., R. Walter, and A. Anderson. Cook Islands, Polynesia. Proc. Natl. Acad. 1998. Fossil and archaeological avifauna Sci. U.S.A. 87:9605–9609. of Niue Island, Pacific Ocean. Notornis Stuiver, M., P. J. Reimer, E. Bard, G. S. Burr, 45:177–190. K. A. Hughen, B. Kromer, F. G. McCor- Worthy, T., A. Anderson, and R. Walter. mac, T. V. D. Plicht, and M. Spurk. 1998. 2002. Appendix D. Fossil fauna from INTCAL98 radiocarbon age calibration, Niue Island. Pages 165–167 in R. Walter 24,000 cal BP. Radiocarbon 40:1041– and A. Anderson, The archaeology of 1083. Niue Island, West Polynesia. Bishop Mus. Walter, R. 1998. Anai‘o: The archaeology of Bull. Anthropol. 10. a fourteenth century Polynesian commu- Wragg, G. 1995. The fossils birds of Hen- nity in the Cook Islands. N. Z. Archaeol. derson Island, Pitcairn Group: Natural Assoc. Monogr. 22. turnover and human impact, a synopsis. Walter, R., and A. Anderson. 2002. The ar- Biol. J. Linn. Soc. 56:405–414. chaeology of Niue Island, West Polynesia. Wragg, G., and M. I. Weisler. 1994. Extinc- Bishop Mus. Bull. Anthropol. 10. tion and new records of birds from Hen- Weisler, M. I. 2001. On the margins of sus- derson Island, Pitcairn Group, South tainability: Prehistoric settlement of Utro¯k Pacific Ocean. Notornis 41:61–70. Atoll, northern . BAR In- Yen, D. E. 1973. The origins of Oceanic ag- ternational Series 967. Archaeopress, Ox- riculture. Archaeol. Phys. Anthropol. Oce- ford. ania 8:68–85.