The subject of this study falls into two parts. Firstly an examination of the methodology of the analysis of archaeological fish remains. It considers the problems of recovering and quantifying material and then the aims and difficulties of analysis. Several lines of evidence can be used to study how the fish were caught and utilised but much of it is ambiguous and natural agencies can produce similar patterns. One major use-for fish remains lies in the study of seasonality; behavioural evidence is not as reliable as is often claimed but growth rings and sometimes fish size are clear indicators. Integrated with this is a case study of a large and well-recovered assemblage of fish remains from a series of five fourth millennium bc. shell middens on the small island of Oronsay(Inner Hebrides). The fauna is dominated by the young age stages of one species, the saithe(Pollachius virens) but at least fifteen other fishes are represented; they are mostly found along rocky shores or in inshore waters. The size of the assemblage and the lack of selectivity, both in species and sizes, suggests a technique of mass capture such as a weir or nets; however, a combination of methods including line fishing from boats seems likely. The behaviour of the species and evidence from the traditional fisheries demonstrates that it could be caught for most of the year, except the 'winter' quarter. Fish size and growth ring data reveal a consistent pattern of a single principal fishing season at each site and differences between them. Collectively, the period of fishing spans much of the year and, as the sites are broadly contemporary, there is a strong possibility that they functioned as part of a single economic system.