Early to Mid-Holocene marine fauna Species representation in Mesolithic shell middens

Aquatic faunal remains form a significant proportion of many Early to Mid-Holocene shell midden deposits. The range of species represented in each midden is variable, reflecting factors such as local microenvironments, resource selection by humans, and archaeological preservation. For example, at Ulva Cave 36 distinct taxa of shellfish have been identified (Pickard and Bonsall 2009); at Morton the shellfish assemblage was similarly diverse with 37 taxa recorded (Coles 1971). By contrast at An Corran, Skye, 14 genera were recovered from Early to Middle Holocene deposits (Pickard and Bonsall 2009) and at Sand, Applecross, only eight taxa were recorded (Milner 2009).

Fish are reported to be abundant at all sites. Unfortunately, detailed reports of the fish bone assemblages remain scarce. With the exception of Sand, Applecross (Parks and Barrett 2009) species diversity is generally limited, but this may reflect publication standards as much as prey specialization. Gadidae are recorded at all sites for which information is available, and saithe (Pollachius virens) dominates the ichthyofauna recovered from the Oronsay sites (Anderson 1895, 1898; Connock et al. 1993; Mellars and Payne 1971; Mellars and Wilkinson 1980).

Sea mammals were recorded at the Oronsay sites, Risga in Loch Sunart, and Sand, Applecross (Anderson 1898; Grigson and Mellars 1987; Lacaille 1954; Parks and Barrett 2009). Finds include the very large rorqual (Balaenoptera spp.), and several smaller species — seals (Phocidae) and dolphin or porpoise (Delphinus delphis/Phocaena phocaena).

Brachyurans, or true crabs, are present in most middens. Generally, where data are available, the presence of edible, swimming and green shore crab is reported (e.g. Anderson 1898; Coles 1971; Lacaille 1954). At Ulva Cave at least six species of crab were identified (Pickard and Bonsall 2008). Lobster (Homarus gammarus) was recorded at the Oronsay sites (Mellars and Payne 1971). Fine sieving of midden samples from Ulva Cave also led to the recovery of fragments of sea urchin (Echinoidea) tests (Pickard and Bonsall 2008).

Retouched carving from Michael Cave, Wemyss, Fife (Edwards 1933, 173). The carving has been interpreted as a terrestrial mammal as well as a seal (Kitchener et al. 2004, 77).

Other resources foraged on the shore include aquatic algae, inferred from the presence of small shellfish that live on seaweed but would have little food or decorative value. The seaweeds may have been harvested as food, fuel, or for medicinal properties (e.g. Banga 2002; Turner and Clifton 2006). Such shellfish species were particularly abundant at Ulva Cave (Pickard and Bonsall 2009). This may indicate a particular emphasis on seaweed collection at this site, but more likely reflects the sampling strategy adopted (Pickard and Bonsall 2009).

A wide range of aquatic bird species were identified at Morton with sea birds such as razorbill (Alca torda), and cormorant (Phalacrocorax carbo), comprising a significant proportion of the assemblage; however the relative abundance of the remains was not quantified (Coles 1971). Similar diversity of bird species is recorded at Risga although a distinct range of seabirds, including the now extinct great auk (Pinguinus impennis), were documented (Lacaille 1954). The majority of the species represented are seabirds that nest on cliffs or inhabit inshore waters.

3.3 Fauna

There is no comprehensive published account of Scottish fauna through the Lateglacial and early Holocene, and it has to be admitted that the database for reconstructing any such account is inadequate on account of the relatively poor survival of skeletal remains from these periods (Kitchener and Bonsall 1997; Kitchener 1998).

Lateglacial faunal remains in particular are scanty. It is considered, however, that some of the species known as, or suspected to have been, present in Scotland during the Devensian prior to the Late Glacial Maximum did not recolonize in the Lateglacial. The so-called Pin Hole Mammal Assemblage-Zone (‘Mammoth steppe’) of pre-LGM OIS3 stage Britain has been defined by Currant and Jacobi (2001, 2011) and it included among the larger fauna the woolly rhinoceros (Coelodonta antiquitatis), the woolly mammoth (Mammuthus primigenius), bison (Bison priscus), giant deer (Megaloceros giganteus), lion (Panthera leo),  reindeer (Rangifer tarandus),  and spotted hyaena (Crocuta crocuta). Of these species the woolly rhinoceros importantly is documented in Scotland from finds made in glaciofluvial sand and gravel deposits in the Bishopbriggs area just north of Glasgow, and the species has a most recent direct radiocarbon date from there of 31,140±170 BP (Jacobi et al. 2009a).  OIS3 dates for bear bones have also been obtained from Uamh nan Claonite, Assynt, Highland (Birch and Young 2009). Remains of woolly mammoth, giant deer, and the ringed seal (Phoca hispida — an Arctic species) have been found in Scotland (Kitchener 1998) and most probably relate to the OIS3 stage, but they have no associated radiocarbon dates. 

New research is suggesting some faunal presence during parts of OIS2 and elements of a tentative Dimlington Stadial Mammal Assemblage-Zone, to include a species resisent to extreme cold - the musk ox (Ovibos moschatus) - has been proposed (Currant and Jacobi 2011). Thus far there are no dated specimens from this period in Scotland.

The Lateglacial post-LGM mammal fauna (the so-called Gough’s Cave Mammal Assemblage-Zone) as known from Britain includes reindeer (Rangifer tarandus), wild horse (Equus ferus), aurochs (Bos primigenius), elk (Alces alces), brown bear (Ursus arctos), lynx (Lynx lynx), wolf (Canis lupus), arctic fox (Vulpes lagopus), saiga antelope (Saiga tatarica), and red deer (Cervus elephas) among the larger species (Currant and Jacobi 2001). The principal Scottish location to have produced radiocarbon-dated Lateglacial examples of any of these species is the limestone cave system of the Creag nan Uamh, Assynt, Highland, from where skeletal remains of brown bear, wild horse, and reindeer have been obtained (Murray et al. 1993; Birch and Young 2009). Other now extinct species known from the Creag nan Uamh caves are the arctic fox, the collared lemming (Dicrostonyx torquatus), the narrow-skulled vole (Microtus gregalis), and the northern vole (Microtus oeconomus), but their remains have not been directly dated (Kitchener 1998), The need for caution over assigning undated skeletal remains to early periods is shown by another example from the Creag nan Uamh caves, where a lynx bone has a radiocarbon age of 1770±80 cal BP (Kitchener and Bonsall 1997).

There is as yet insufficient evidence to speculate on possible Younger Dryas/Loch Lomond Stadial faunal presence (Currant and Jacobi 2001).

When it comes to the Postglacial period, Kitchener et al. (2004) have appraised the evidence for the mammal fauna in Scotland during the Mesolithic. As during the Pleistocene there is a paucity of preserved bone assemblages, although from comparative studies elsewhere in Britain the anticipated species can be more accurately predicted than for the Lateglacial. Definite evidence exists early on for three species which, apparently, then became extinct in Scotland during the Mesolithic (probably in the very early Mesolithic): giant deer, reindeer, and wild horse (Kitchener 1998; Gonzalez et al. 2000). This apparent time lag in vertebrate faunal changes across the Pleistocene/Holocene transition is thought possibly to reflect the complexity of the pattern of biome development in the early Holocene, which could have allowed for the persistence of Lateglacial-like refugia for flora and fauna (Coard and Chamberlain 1999). Resolution of the uncertainties of species survival across the transition will depend upon new finds and more accurate radiocarbon dating (Jacobi et al. 2009b, 21; cf. Street and Baales 1999).

Reconstruction of Mesolithic fauna in Holyrood Park, Edinburgh, © Crown Copyright Historic Scotland reproduced courtesy of Historic Scotland.

Survivors from the Lateglacial which also then survived the Mesolithic period before becoming extinct in Scotland at later periods were the brown bear, elk, aurochs, and lynx, while the red deer, a long-term survivor which was probably present in the Lateglacial is first positively documented during the Mesolithic (Kitchener et al. 2004). The beaver (Castor fiber) and stoat (Mustela erminea) were present in Mesolithic Scotland, but have not been recorded from any actual Mesolithic sites (Kitchener et al. 2004). Kitchener et al. (2004, table 5.3) have listed the species recorded from various Mesolithic sites, and further discussion of individual animal bone assemblages can be found in Coles (1971; 1983b), Mellars (1987), Hardy and Wickham-Jones (2009), and Bartosiewicz (in press).

There is in fact a relatively substantial amount of information available about the various small mammals, birds, fish, molluscs, amphibians, insects, and microfauna present, or suspected to have been present, in Scotland during the Mesolithic period (see for example Coles 2010; Kenward and Whitehouse 2010; Kitchener 2007; 2010; and see section 3.3.1 below). For the most part these data have not been collated, however, so there is no Scottish equivalent, for example, for Price's work on the small mammals of SW Britain (Price 2003; also listings in Schreve 2004). Also the ability to be certain that particular faunal remains are definitely Mesolithic will normally depend upon radiocarbon dating of individual specimens, since, in the absence of securely stratified Mesolithic horizons, the danger of contamination from more recent material is always present. Furthermore, for archaeologists the evidence which does exist can be fairly obscurely or only partially published, requiring determined research to track down, for example in the case of the Mesolithic amphibia from the Creag nan Uamh caves (Gleed-Owen 1999) or the early Mesolithic ducks from Puggieston, Angus (Smith and Jones 1976).

What needs to be emphasized is the archaeozoological importance of Quaternary faunal remains, even when found in contexts with no direct human connection, because of their potential for chronology and palaeoenvironmental reconstruction (e.g. Kolfschoten 2006). It is also worth noting the wider benefits which can accrue in terms of public interest and education when significant discoveries of Pleistocene mammals occur (Ashwin and Stuart 1996; Stuart 1997).