Reviewed by: The Prehistory of Buka: A Stepping Stone Island in the Northern Solomons Simon Holdaway The Prehistory of Buka: A Stepping Stone Island in the Northern Solomons. Stephen Wickler. Terra Australis 16. Canberra: Department of Archaeology and Natural History and Centre for Archaeological Research, The Australian National University, 2001. xix + 360 pp., 68 figs., 40 plates, 175 tables, bibliography. Paperback. ISBN 0-7315-5500-7. This work reports on archaeological fieldwork conducted on the small offshore islands of Sohano and Pororan over eight months in 1987, which formed the basis of Wickler's doctoral dissertation. As is spelled out in Chapter 1, "Introduction and Research Design," Wickler sets out to test the ceramic sequence proposed by Jim Specht in his 1969 doctoral dissertation, examine the temporal relationships between the pottery styles that Specht proposed, and document more fully the rather insubstantial evidence for Lapita occupation that Specht discovered. In addition, Wickler seeks to provide both a more detailed understanding of occupation and settlement before Lapita through the analysis of nonceramic artifacts and faunal remains. Wickler's discovery on Buka of pre-ceramic occupations dating well back into the Pleistocene at Kilu Cave is by now well known, but in the book full details of the excavation are provided together with an analysis of the artifacts and fauna. Details are also provided for excavations at Palandraku Cave, where a brief pre-Lapita occupation, dated to approximately 5000 B.P., is inferred. Both caves also provided faunal evidence. However, to document the ceramic sequence, Wickler turned away from the caves, concentrating instead on a series of coastal sites with limited occupation spans. Significantly, some of the sites used are reef deposits, not a site type investigated by Specht, but targeted because of discoveries in similar situations on Nissan and in the Bismarcks. Ceramics from these sites are analyzed using attributes that relate to temper and paste, as well as vessel form and decoration. Analyses aimed at determining the nature of exchange and interaction during and after Lapita are partly based on these pottery studies, but also are based on an analysis of obsidian artifacts. In studying these materials, Wickler is concerned with isolating long-term trends in the movement of pottery and obsidian, and similar long-term patterns are sought in the faunal evidence, particularly information related to the shift from hunting and gathering to agriculture. The Prehistory of Buka is organized in a way similar to other regionally based Pacific archaeology studies. Chapter 2 backgrounds the natural and cultural environments, while Chapter 3 provides details of the site survey and excavations at a series of sites. Rather than attempt a wide-ranging survey, Wickler targeted the coastal portion of the raised Sohano formation within the Lonaha and Soraken land systems as areas in which suitable places to excavate might be [End Page 153] found. The sites studied are broken into two groups: the reef sites that provide most of the evidence for Lapita occupation, and excavated sites on land. The reef sites consist of scatters of pottery and stone artifacts distributed across the intertidal reef flats at three locations. Four of the excavated sites are beach deposits while the other three occur in caves or rockshelters. Full descriptions are provided for each of the excavations to a generally high standard, although the reproduction of photographs is poor. A section at the end of the chapter summarizes chronological information based on a sample of 16 radiocarbon determinations, nine of these obtained from the pre-ceramic levels at Kilu Cave and a further two from the preceramic deposits at Palandraku Cave. This leaves only six determinations for the excavated ceramic sites. Chapter 4 deals with Lapita pottery design and analysis, comparing assemblages from three reef sites, one each on Sohano Island, Buka Island, and Nissan Island. The chapter reports on the analysis of more than 42,000 sherds (the vast majority from one site), although only a much smaller sample (around 15 percent) of diagnostic sherds: those that are either shaped, decorated, or both, are subjected to the full range of analyses. Analysis involves a consideration of vessel morphology, manufacturing and finishing techniques, paste, composition, and decoration and is aimed at answering...