At Staosnaig, Isle of Colonsay, there is evidence for intensive processing of hazelnuts and other plant types, suggesting the harvesting of the woodland (Mithen et al. 2001), Figure 38. Colonsay is unlikely to have supported large game and hence visits may have been specifically directed towards the exploitation of the hazel-rich woodland. Likewise, Fiskary (CANMORE ID 299865) on the Isle of Coll can be interpreted as a fishing camp, visited on at least three occasions providing discretely dated activity events, centred on c. 9070, 8520 and 8280 cal BP (Mithen et al. 2007a; Wicks and Mithen, forthcoming), Figure 27 and Figure 28. The fish bones from Fiskary indicate inshore fishing, possibly using fish traps not unlike the cairidh that survives at Fiskary Bay today. The quantities of charred hazelnut shell and wood charcoal at Fiskary suggest that activities other than fishing were also undertaken. Unfortunately, the debris at Fiskary is an unstratified palimpsest and cannot be divided into the three (or more) activity events. As such, although one might assume the same range of activities were undertaken at Fiskary during each visit, we cannot exclude the possibility that one visit involved fishing, another involved hazelnut collecting, and a third might have involved neither or both.
The Mesolithic deposits at Rubha Port an t-Seilich (CANMORE ID 98306) on the Isle of Islay provides evidence for another form of activity: big game hunting (Mithen and Wicks 2010a, 2011b; Mithen et al. 2015). Although heavily fragmented, the faunal remains include those from red deer, roe deer and wild boar, as well as from fish and birds. Rubha Port an t-Seilich (CANMORE ID 98306) is one of only two non-shell midden sites where faunal remains have survived, the other being the much later dated Storakaig (CANMORE ID 304908), also on Islay. It seems likely, however, that other microlithic-chipped stone scatters on the larger islands, such as Bolsay (CANMORE ID 37342 AND CANMORE ID 37347) (Isle of Islay, Mithen, Lake and Finlay 2000), Kinloch (CANMORE ID 22202) (Isle of Rum, Wickham Jones 1990) and Creit Dubh (Isle of Mull, Wicks and Mithen 2010c , forthcoming) were also related to large game hunting.
One further specialised activity location appears to be Coulererach (CANMORE ID 74459) on the west coast of Islay (Mithen and Finlay 2000). Excavation revealed evidence for a predominance of the early stages of core preparation, with large numbers of split pebbles and primary flakes. The location is close to beaches that today have relatively large numbers of flint pebbles. Assuming this was also the case in the Mesolithic, Coulererach appears to have been a raw material acquisition site, from which partially prepared cores were taken to other destinations, especially on islands lacking in flint raw material. Comparative analysis of chipped stone assemblages from the region has yet to be undertaken in an attempt to trace such raw material movements, although evidence for this is already apparent from the presence of bloodstone from Rum and pitchstone from Arran at sites such as Staosnaig on Colonsay (Mithen, Finlay, Carruthers et al. 2000).
The rarity of evidence for structural remains implies a high level of Mesolithic mobility with limited investment in ‘place’. This might, of course, reflect no more than the limited area of excavation at any one site, the predominance of small trenches being unable to expose structural evidence other than isolated pits and postholes. But when relatively large areas have been exposed at Kinloch (CANMORE ID 22202 (Wickham-Jones 1990) and at Bolsay (CANMORE ID 37342 and Staosnaig) (Mithen et al. 2000) the number of substantial features in patterned arrangements has also been limited. Although these sites lack evidence for major structures, they have arrays of postholes, stake-holes and pits that suggest windbreaks and shelters, most likely erected for short periods of activity. More substantial structures are currently known from just two sites, Staosnaig and Creit Dubh. The large shallow pit containing charred hazelnut shells at Staosnaig has been interpreted as a secondary use of a hut floor (Mithen et al. 2001), while a dense concentration of postholes and stake-holes at Creit Dubh, some providing well-formed arcs and associated with large fire pits, suggests they may have been one or more structures erected (Mithen, Wicks, Pirie and Mariecevic, forthcoming ), Figure 39.
In summary, the residential period, 9400-7800 cal BP, appears to have been one of mobile foraging, with visits to specific locations to exploit specific resources most likely at specific times of the year. The archaeological sites provide an insight into some of the specific activities undertaken in the residential phase of Mesolithic settlement. These must be seen in the context of a constantly changing environment, the evidence for which might provide further signs of Mesolithic activity.