Consideration of the behavioural ecology of the shellfish and crustacean species identified may indicate harvesting strategies. Two genera, limpet (Patella spp.) and periwinkle (Littorina spp.), dominate the west coast midden assemblages. The relative abundance of the major species identified at each of the sites generally reflects that of modern shore populations in the region (Little and Kitching 1996). However, the virtual absence of dogwhelk (Nucella lapillus), a common carnivorous gastropod on modern shores, at An Corran may indicate food avoidance practices (Pickard and Bonsall in press). Overall, the evidence points to the adoption of least effort harvesting strategies at most sites (Connock et al. 1993; Pickard and Bonsall in press). The majority of the species identified are epifaunal littoral species, i.e. species that live on the substrate surface and occupy tidal regions of the shore. They are most readily collected at low tide with little or no equipment.
Infaunal and sublittoral shellfish and crustacean species that live below low water and/or buried in the substrate are rare at midden sites on the west coast of Scotland, although a recent marine survey attests to their abundance in coastal waters (McKay 1992). Some species such as the king scallop (Pecten maximus) were collected not as food but as empty shells for use as raw material (Russell et al. 1995). Infaunal bivalves are more characteristic of Mesolithic shell middens on the east coast of Scotland, where soft shorelines are more prevalent. In traditional and commercial shellfisheries bivalves are the most highly valued shellfish in terms of palatability and quantity of flesh (e.g. Gosling 2003). In the Late Mesolithic midden at Morton in Fife one of the most abundant shellfish was Baltic tellin (Macoma balthica). Its presence implies foraging in low salinity coastal environments (Coles 1971) consistent with the evidence of extensive estuarine areas near the site around the time of midden accumulation (Chisholm 1971).
Few unequivocal fishing-related artefacts have been recovered from Scottish Mesolithic sites. However, fishing strategies may be inferred from the size of specimens and range of species identified. The small size of many fish recovered and the abundance of gadidae in middens is suggestive of line fishing from the shore at these sites. One of the few items of dedicated fishing gear attributed to the Scottish Mesolithic, a fish hook from Risga (see Morrison 1980: plate XIV), would, if this identification as a fish hook is correct, lend support to this suggestion. Stationary traps or weirs, which are relatively commonplace finds at Mesolithic sites elsewhere in North-West Europe, are not recorded archaeologically. Such installations represent a considerable investment of labour, both in terms of construction and maintenance, and would likely only have been set up near long-term residential camps. It is not clear if any of the known Mesolithic shell midden sites from Scotland fall into this category.
The role of sea mammals in Mesolithic subsistence is equivocal. Their remains are never numerous, but at some sites, e.g. Cnoc Coig on Oronsay, they outnumber those of land mammals. Though likely used for food, seals and other sea mammals could also have been exploited for fur, skin, and intestines, which have traditionally been used in the manufacture of waterproof clothing and vessels (e.g. Lantis 1938). Evidence from Cnoc Coig suggests that skin and blubber were extracted from very young seals, whereupon the carcasses were discarded (Grigson and Mellars 1987). Both harbour seal (Phoca vitulina) and grey seal (Halichoerus grypus), which are commonly found in near-shore waters, could have been sought on haul-outs and skerries (Grigson and Mellars 1987). By contrast, rorqual (Balaenoptera spp.), which was recorded at Risga (Lacaille 1954) is a generally open sea species. Fin whale (Balaenoptera physalus), which was tentatively identified at Caisteal nan Gillean I and Priory Midden (Grigson and Mellars 1987), is seldom encountered in inshore waters. As there is no clear evidence for subsistence fishing in open-sea waters (Pickard and Bonsall 2004) and given the size of the fin whale and other rorquals, it is likely that Mesolithic groups exploited occasional stranded animals on the shore - a practice widely documented historically (e.g. Olsen 1999). Although marine mammal remains do occur in some middens, the frequency and pattern of occurrence suggest that their exploitation was often opportunistic.
Seabirds, prized by some historically documented forager societies (e.g. Meldgaard 1988; Oakes 1991) are relatively common in the middens. While they may have been taken primarily for their feathers or skins, their meat may have been consumed and the eggs collected where and when nesting sites were accessible. An interesting but rare occurrence in Mesolithic shell middens is the now extinct great auk (Pinguinus impennis). Its bones have been reported only from sites on offshore islands (Oronsay, Risga; Lacaille 1954) these birds spent their lives at sea and came ashore only to breed, and rocky islands were their preferred breeding grounds. Greenland Inuit are known to have exploited great auks intensively, consuming their meat, intestines, fat, and eggs, and using their fat as oil for lamps, and their skins for clothing and bags (Meldgaard 1988). Like penguins, the great auk is reputed to have been a powerful swimmer, but awkward on land, and their exploitation by Mesolithic people was likely a seasonal pursuit with the birds being captured at their spring breeding sites.