Shirley Moor wooden structure (an interim report)1
Shirley Moor is an embayment of marshland on the north-west side of Romney Marsh. The ground rises to a height of more than 10m around its periphery. The south-east side of Shirley Moor is devoid of surrounding high ground and connects Walland Marsh, which is part of Romney Marsh. There is a hill towards the northern side of Shirley Moor, which rises to a height of 18 m OD and has upon it Shirley Farm and Glover Farm. Shirley Moor is now a well-drained area primarily assigned lo arable farming.
Wood was first seen in the autumn of 1996, protruding from the side of a drainage ditch on Shirley Moor at National Grid Reference TQ 9295 3255 (approximately 250 m west of Shirley Farm). It was visible for a length of 22 m in the south-east bank. Its extent was subsequently traced by auger. The known area of exposure was found to exceed 140 square metres, excluding the area that is eroded by the ditch. The drainage ditch runs down the Oank of the hill (described above) from the north-east to the south-west where it joins a sewer at a T-junction, the sewer widening to approximately 5m this point. This sewer follows the lowest part of the present land surface and therefore takes drainage water from the fields on either side.
An application was made to the Romney Marsh Research Trust to fund a trial excavation lo determine whether the feature was man-made and to recover and date some samples of the wood.' After digging, it transpired that the wood, originally seen as a single layer in the side of the ditch, was a natural assemblage some 40 cm thick consisting of three distinct elements.
(I) Water rounded wood with the bark eroded away
(2) Peal balls
(3) Wood with the bark still in situ (consisting of alder [alnus], holly [ilex] and ash fraximus]). The predominant species being alder.
This assemblage was resting on a bed of grey marine sand from which one valve of a scrohicularia was retrieved. This bivalve burrows in mud and sand between the tide marks. The overlying horizons consisted of silty/fine sand with the present day surface being 63 cm above the wood layer. The wood with the bark in place, because it was not rotten or abraded, was considered to be a primary deposit. Accordingly, two samples were sent to Beta Analytic Inc. for Carbon 14 analysis. The dates were found to be BP 4098 and cal BP 4170 (Table 1).
Previous work had identified a laterally persistent peat bed of between -4.71 m OD and 0.27 m OD about 2 km south of the wooden feature.[fn2] This peal was sealed by up to 4.5 m of marine brackish sediments, largely sand. The peat formed between cal BP 6200 and cal BP 3600. Provided that the sand at the two locations is the same deposit (and there is no reason to suspect otheiwise), the 14C date of the wood is clearly too old for it to be a primary deposit. It is likely that the dated wood samples had been preserved in the peat from cal BP 4098 until some time after the end of peat formation. During the subsequent marine inundation it had been released from the peat bed by erosion and redeposited at its final resting place. It is evident that large-scale peat bed erosion took place in Shirley Moor area because of the large quantity of peat balls that were found during the excavation.
Soil samples were taken at intervals in a vertical column on the north-east facing section of the excavation. The results of a simple qualitative test for free calcium carbonate (high, medium, low, nil) indicated a high level in all of the samples except for 2 showing a medium level. Both of these samples were from Horizon 5, containing the wood and peat balls, which being of an acidic nature have reacted with the calcium carbonate and reduced its quantity. Bearing these remarks in mind, both the PH and the free calcium carbonate levels were found to be similar in all soil horizons with no sudden changes, thus showing that no relict soil surfaces have been encapsulated above the marine sand.
Table 1
Radio Carbon Dates
sample RB-SM4B RB-S M9B Beta Analytic Lab. no. Beta-12'.'.824 Beta-12'.'.825 uncalibratcd con\'cntional 1-JC age AD/BC calibrated result {2sigma 95% probability] BP calibrated rcsultJ2sigma 95% probability) mean of BP calibrated result 3750 +/-70 BP Cal BC 2350-1945 Cal BP 4300-3895 Cal BP 4098 3800+/-80 BP Cal BC 2465-1975 Cal BP4415-3925 Cal BP4170
Greg Priestly-Bell & Robert Beck Lydd
1 The authors of this report would like to thank The Romney Marsh Research Trust for the grant. Also in consideration of the fieldwork: Mr Robert Orpin, Mr Stan Reeves, Mr and Mrs Chris Lacki, Dr Chris Salisbury, Mrs Liz Owen and Mr Rene Rcgandanzc; and fnr lhcir advice and support Dr Antony Long, Dr Martyn W.illcr and Dr Mark Gardiner.
2 This earlier investigation has found evidencc of peal redeposition which was also consistent with the authors findings; 'Long, D., Wallcr, M. & McCarthy, P., 'The Vcgctational History, Stratigraphy and Pollen Data for the Shirley Moor Region', in J. Eddison, M. Gardiner & A. Long (eds.), Romney Marsh: Environmental Change and Human Occupation in a Coastal Lowland, OUCA Monograph 46 (Oxford, 1998), 31-44.