A List of Tree-Ring Dates for Building Timber in Southern England and Wales

Abstract

Table I and Map 1 show the places of origin in southern England or Wales of oak samples that have been collected from standing buildings, vernacular or otherwise, for dating by dendrochronology at the laboratories in Munich and at Oxford. List 3 (above) tabulates results for the samples that have been dated up to April 1980. For the dating attempted by V. Giertz-Siebenlist at Munich, 126 samples, mainly cores, were collected in England in 1969, 1970 and 1972 with F. W. B. Charles and W. Horn. For these, listed as Group S(CH) in Table I, success was under 10% using Huber's reference curve for central and southern Germany together with Hollstein's for Western Germany, as mentioned in the paper by Berger, Giertz and Horn (1971). Table I also shows that a much higher fraction of the relatively few samples taken from Welsh crucks-Group S(S)-were dated. After the research at Munich ceased, V. Kerner-Siebenlist kindly sent me all the samples and the ringwidths measured on them, for us to try to match the undated ones using chronologies developed at Oxford for southern and eastern England. Progress is being made with them, and List 3 includes as F(CH)3 and F(CH)4 results for timber from two buildings. Most of my own samples, F(F) in Table I and List 3, were collected from 1968 and 1971. They are mainly cut from building timber available through restoration or demolition. Radiocarbon measurements were made on a few of them (Currie and Fletcher, 1972). Some were kindly polished by Ian Gourlay of the Department of Forestry, Oxford, while ring-width measurements of several were made in conjunction with Frank Walker. Almost all consisted of fast-grown timber and contained under 100 annual rings. No serious effort was made to match and date the samples until recently (1978), by which time we had formed I chronologies not only for oaks of slow growth but also for trees of relatively fast growth that were likely to be appropriate for matching building timber. At present we regard the chance of matching short sequences from building timber as remote unless the likely date of construction is known to within about half a century. This application of our research is therefore being confined in the first place to such samples.