MEM23810 - Palaeoenvironmental evidence at Comerslade (Find Spot)

Palaeoenvironmental samples of the site taken in March 2008 were subject to pollen analysis and radiocarbon dating. Up to 1.8 metres of peat was noted at the site. The exact site from which core samples were taken is unknown.

Palaeoenvironmental samples of the site taken in March 2008 were subject to pollen analysis and radiocarbon dating. Up to 1.8 metres of peat was noted at the site and at least three distinct tephra horizons were also noted in the sequence. Four pollen zones were recorded but few testate amoebae were recovered and at insufficient concentrations to make a meaningful, representative count. The Radiocarbon results indicate that peat developed at the site during the later Mesolithic, shortly before 6350-6080 cal BC, and it is thought that the top of the Comerslade sequence falls between cal AD 1570-1890. The site is not currently accumulating peat but there has been little loss of peat from the upper levels of the sequences. It is one of only four sites known on Exmoor which include significant wood within the stratigraphy. The exact site from which core samples were taken is not specified in the report. [1] This record was included in gazetteer of known sites that preserve wood macro fossils on Exmoor and the samples gathered in 2008 were reexamined in 2012. The key results are that peat accumulation began shortly after 6500 cal BC, in a landscape that included mixed woodland (Betula, Quercus, Corylus and Salix) and some open ground. By 5500 cal BC open ground indicators had all but disappeared from the pollen record. At around 3310 cal BC significant reductions in the amount of tree pollen occurred (particularly Betula and Corlyus), coincident with an increase in grassland. There was a significant reexpansion of woodland at ~2600 cal BC (largely Corylus, but also Quercus), before major clearance again at ~1570 cal BC. The new radiocarbon dates from the section are duplicates from a piece of branchwood (Betula sp.) and the fine humin fraction from the peat. Similarly to the results from Halscombe Allotment, the humin date fits very well with the existing dates from the site, and the branchwood date is older (in this instance by around 500 years). Twelve samples from Comerslade were assessed for insect preservation. All samples produced well-preserved insect remains. Few of these fragments were identifiable to species level, owing to the small size of samples. These fragments that were identifiable are indicative of still or slow-moving water. The sample from 90-100 cm depth (c. 5300-5700 cal BC) includes Bradycellus ruficollis, a species that occurs on acid moor and heathland. This contrasts somewhat with the pollen analytical evidence, which has no pollen of heathland taxa in these levels, and provides a good indication of the added value of the insect analysis. [2]