What Images Return: A Study of the Stratigraphy and Seasonality of a Small Shell Midden in the West Isles of New Brunswick

Abstract

<p>This thesis addresses several aspects, both methodological and substantial, of the archaeology of coastal shell middens in the Passamaquoddy Bay and West Isles areas of southern New Brunswick. The inquiry takes the form of a detailed analysis of the stratigraphy and faunal assemblages of two shell midden sites, BgDr 48, a prehistoric site, and BgDr 49, an historic site, located on Partridge Island, one of the West Isles. Throughout the study aspects of these sites are compared to previously excavated sites in the area.</p> <p>With respect to the analysis of shell midden sites, it is concluded that: 1) there is no evidence for serious stratigraphic disturbance or mixing in midden sites not affected by recent agricultural activity; 2) the use of extensive column sampling methods is useful and desirable in excavating such sites, as it results in the recovery of significant faunal and botanic remains missed by the traditional excavation methods employed in the area; 3) at least some shell middens can be analysed in terms of culturally meaningful analytical units smaller than site size, defined on the basis of stratigraphic and faunal information.</p> <p>With respect to aboriginal subsistence, this study suggests that the importance of soft-shelled clams relative to other invertebrate prey species may have been overemphasized in previous midden studies in the area; evidence is presented which indicates that shell fishing was carried on during both summer and winter seasons.</p> <p>With respect to the settlement patterns of prehistoric peoples in the area, the study suggests that: 1) the preservation of Early Maritime Woodland components may be strongly selected against by rising sea levels, but that the measurement of foreshore slopes may be useful in predicting where such components may be preserved; 2) Early and Middle Maritime Woodland components are distinguishable on the basis of their respective faunal assemblages, and that discrete occupations within Middle Maritime Woodland components may be distinguishable on the basis of faunal variability; 3) these patterns in faunal variability are similar to patterns noted by archaeologists conducting research on the coasts of Maine; 4) the extant models of prehistoric coastal settlement for the area are probably simplistic, but more research is necessary before these can be replaced by more adequate and better substantiated models.</p> <p>As regards the historic archaeology of the area, the study suggests that some of the shell midden sites recorded on this coast may have resulted from historic occupation, and that these may be distinguishable from prehistoric middens by virtue of the low species variability in historic shell deposits.</p>