Palau

Current archaeological evidence from Palau in western Micronesia indicates that the archipelago was settled around 3000– 3300 BP by normal sized populations; contrary to recent claims, they did not succumb to insular dwarfism.

Background: Previous and ongoing archaeological research of both human burial and occupation sites throughout the Palauan archipelago during the last 50 years has produced a robust data set to test hypotheses regarding initial colonization and subsequent adaptations over the past three millennia.

Principal Findings: Close examination of human burials at the early (ca. 3000 BP) and stratified site of Chelechol ra Orrak indicates that these were normal sized individuals. This is contrary to the recent claim of contemporaneous ‘‘small-bodied’’ individuals found at two cave sites by Berger et al. (2008). As we argue, their analyses are flawed on a number of different analytical levels. First, their sample size is too small and fragmentary to adequately address the variation inherent in modern humans within and outside of Palau. Second, the size and stature of all other prehistoric (both older and contemporaneous) skeletal assemblages found in Palau fall within the normal parameters of modern human variation in the region, indicating this was not a case of insular dwarfism or a separate migratory group. Third, measurements taken on several skeletal elements by Berger et al. may appear to be from smaller-bodied individuals, but the sizes of these people compares well with samples from Chelechol ra Orrak. Last, archaeological, linguistic, and historical evidence demonstrates a great deal of cultural continuity in Palau through time as expected if the same population was inhabiting the archipelago.

Conclusions: Prehistoric Palauan populations were normal sized and exhibit traits that fall within the normal variation for Homo sapiens—they do not support the claims by Berger et al. (2008) that there were smaller-bodied populations living in Palau or that insular dwarfism took place such as may be the case for Homo floresiensis.

In this research we examined the isotopic signatures of sequential growth increments from 10 modern and 11 archaeological Atactodea striata shells (21 shells with a total number of 112 samples). Modern shells and record- ed sea surface temperature measurements (SSTs) were used to evaluate the geochemistry of A. striata as a suit- able candidate for recording ambient SST. Pairing oxygen isotopes and recorded SST of modern samples with x- ray diffraction (XRD) allows the identification of the biomineralogical composition of A. striata and provides the necessary information to select the most appropriate carbonate temperature equation to convert oxygen isotope values to estimated SST. This SST conversion was then applied to isotopic data from 11 shells recovered from a ~1700 year old component at the Chelechol ra Orrak site in Palau, Micronesia. We discuss the biomineralogical composition of A. striata, modern and prehistoric sea-surface temperature variation, and the importance of using modern shellfish analogues in conjunction with archaeological samples for paleoenvironmental recon- structions. Isotope values from modern A. striata, when converted using Grossman and Kus' (1986) temperature conversion equation, was found to be similar to measured SST. Estimated SST from archaeological shells showed that the mean SST at 1700 cal. BP were similar to modern values, but SST ranges may have been slightly greater.

Archaeologists routinely reconstruct the types of marine environments fished by past human societies in order to understand economic systems, foraging behaviour, maritime technology and seafaring abilities. These reconstructions are based on ecological data provided by archaeofish identifications, but can be problematic where coarse-grained designations, such as inshore or pelagic, are used, or the influence of fish behaviour and life history traits on movement between habitats is overlooked. In tropical waters, intra-family diversity complicates habitat reconstruction by precluding lower-level taxonomic identifications that provide precise habitat information (e.g. surge channels, dropoffs). Consequently, a single generalised habitat may be imposed on fishes that could be caught in multiple environments, thereby eroding the reliability of fishing zone reconstructions. This study employs the archaeofish assemblage from Chelechol ra Orrak (c. 3000–0 BP), Palau to examine the analytical impact of these variables on fishing habitat reconstruction based on a blind assessment of taxon-derived environmental data by two analysts. We assesses how analysts' variable decision protocols for the handling of imprecise environmental data impact resulting habitat exploitation profiles. Our results address the issue of spatial resolution in habitat information gleaned from fish assemblages like Orrak's, with important implications for the interpretation of foraging practices and maritime adaptations. R ´ ESUMÉ Très souvent les archéologues reconstituent les types de milieux marins pêchés par les sociétés humaines du passé pour comprendre les systèmeséconomiques, le comportement alimentaire, la technologie maritime et les capacités maritimes. Ces reconstructions sont basées sur des donnéesécologiques fournis par les identifications des poissons archéologiques, mais cela peutêtre problématique lors des désignations générales, tels que les eauxcôtì eres ou pélagiques sont utilisés, ou l'influence du comportement des poissons et les caractéristiques de l'histoire de la vie en mouvement entre les habitats sont ignorés. Dans les eaux tropicales, la diversité intrafamiliale complique la reconstruction de l'habitat en excluant le niveau inférieur des identifications taxonomiques qui fournissent des informations plus précises de l'habitat (i.e. les canaux de surtension, des tombants). Par conséquent, un seul habitat généralisé peutêtre imposée sur des poissons qui pourraientêtre pris dans multiples milieux, affaiblissant ainsi la fiabilité des reconstructions de la zone de pêche. Cetté etude utilise l'agglomérat des poissons archéologiques de Chelechol ra Orrak (environ 3000–0 AP), Palau pour examiner l'impact analytique de ces variables de la reconstruction de l'habitat de la pêche, basée sur uné evaluation aveugle des données environnementales taxon-dérivées par deux analystes. Nousévaluons comment les variables, décision-protocoles des analystes dans le traitement des données environnementales imprécises impactent les résultats des profils de l'exploitation de l'habitat. Nos résultats portent sur la question de la résolution spatiale de l'information de l'habitat, extraitè a partir de l'amas de poissons comme l'Orrak, avec des implications importantes pour l'interprétation des pratiques de recherche de nourriture et des adaptations maritimes. Mots Clés: ˆ Iles Carolines, reconstruction paléoenvironnementale, archéo-ichtyologiques, archéozoologie, pêche des poissons pélagiques, pêche sur le récif

Past research has suggested that the humped conch (Strombus gibberulus), a species common in many prehistoric archaeological sites in the Pacific, declines in size and/or abundance over time. Explanations for this phenomenon largely revolve around the possibility that they were overharvested by human populations. In this study, we measured the length and width of over 1400 individual specimens of S. gibberulus shells recovered from the site of Chelechol ra Orrak in Palau, western Micronesia, in deposits dating from ca. 3000 BP to the present. Statistical analysis indicates that in contrast to previous reports, there is a significant size increase for this taxon through time which may be the result of a combination of anthropogenic and environmental factors. We discuss variables influencing mollusc size and suggest that, given the complexities of their interactions and the data limitations of archaeomalacological assemblages, unambiguous determination of the cause(s) of molluscan size change may not always be possible.