The end of days for the Cistercian grange of Beaubec was due to the devastation caused by the Great Plague of 1348–49. The discovery of the fuming pot made from mid fourteenth-century Drogheda ware and late-medieval ridge tiles was, up to now, the strongest evidence we had for this catastrophe. Now, thanks to the science of dendrochronology, a date for the abandonment of the service tower has been pinpointed to ‘After AD1356‘! The timber was also discovered to be from Scotland, perhaps it arrived in Ireland with ‘Richard de Preston de Beaubek’ who tried to confiscate this property in the decades following the Famine.
Another timber, lintel of the doorway into the service tower, was dated to the 1630s. This date shows that the service tower was restored or repurposed in the early-seventeenth century by the Draycotts of Mornington.
David Brown of the QUB dendrochronology lab takes a core sample from the lintel of the door into the service tower. This timber date to the 1630s.
David Brown inspects the service tower timbers.
The full report can be read here:
Queens University Belfast, School of Natural and Built Environment, Dendrochronology Laboratory Report 16/2021
Dendrochronological report on the wood samples from, Beaubec Tower House, Co. Meath by David M. Brown (School of Natural and Built Environment, Queens University Belfast, Belfast, BT7 1NN)
Samples from two timbers from Beaubec service tower were taken for dendrochronological dating. Both wood samples had enough annual growth rings for dendrochronological analysis. The wood samples provided dendrochronological dates from the middle of the seventeenth century and from the later part of the fourteenth century.
On the 11 November 2021, a number of tree-rings cores from a re-used oak door lintel and a cut slice of a plank from the latrine system were taken by a member of the Dendrochronology laboratory, Queens University Belfast. The dendrochronology laboratory reference numbers for these samples are Q12769 and Q12770.
Methods at Queens University Belfast dendrochronology laboratory in general follow those described by Baillie (1982) and English Heritage (1998). Details of the exact methods used are described below.
In the laboratory, a surgical scalpel was used to remove wood from the surface of the core and slice to expose the tree-ring pattern. Where the wood sample was soft or needed to be made clearer a razor blade was used. Finely ground chalk was spread and rubbed onto the prepared surface. This was to define the annual tree-ring boundary more clearly for measurement.
The tree-ring patterns on the samples were measured to an accuracy of 0.01mm using a microcomputer based travelling stage. The tree-ring series obtained for each sample was plotted on a computer screen to facilitate visual comparison. In addition, cross-correlation algorithm CROS84 (Munro 1984) and Cros73 (Baillie and Pilcher 1973) was employed to search for positions where the tree-ring series were highly correlated. These positions were then checked visually using the computer. All the measured sequences were compared with each other and any found to match would be combined to form a site master chronology. These and any remaining unmatched tree-ring series were tested against a range of regional and local chronologies using the matching criteria: high t – values, replicated values against a range of chronologies at the same position, and satisfactory visual matching. Where such positions are found these provide calendar dates for the tree-ring sequence.
The tree-rings dates produced by this process initially only date the measured series. The interpretation of these dates relies on the condition of the final rings in the sequence. In oak wood, if the sample ends in the heartwood of the tree, then the date of the last ring plus an addition of the minimum expected number of sapwood rings, indicate a terminus post quem date. Where some sapwood or the heartwood-sapwood boundary is present, then a death date range can be calculated using the maximum and minimum number of sapwood rings likely to have been present. The sapwood estimates are a minimum of 10 and a maximum of 46 annual rings, where these figures indicated the 95% confidence limits of the range. These figures are applicable to oaks from Britain and Ireland. In the Belfast laboratory, we use an empirical estimated sapwood range of 32 ± 9 years. If the bark edge survives then a death date can be directly obtained from the date of the last ring.
This sample is a mean sequence of five oak cores taken from different parts of the timber. The centre or pith of the tree is not present on the sample. There is no sapwood but the heartwood-sapwood boundary is present on the sample. The five measurements were averaged together to produce a 183 ring tree-ring series. This averaged tree-ring series was compared with a suite of regional and local tree-ring chronologies from Ireland. Extremely significant and consistent correlation values (t = 8.38*** cf. Early Irish AD Chronology (EARLYAD); t = 7.28*** cf. Belfast Index Master (BELIM) and t = 5.52 cf. Tintern Abbey, Co. Wexford) were found. These and other results indicate that the averaged tree-ring series dates from AD1421 to AD1603. The best-estimated felling date range for the tree from which this timber was taken will be AD1635 ± 9 years.
When this sample was measured, it yielded 167 annual growth rings. The centre or pith of the tree is not present on the sample. There is no sapwood or heartwood-sapwood boundary on the sample. The measured tree-ring series was compared with a suite of regional and local tree-ring chronologies from Ireland. No significant and consistent correlation values were found. The measured series was then compared with tree-ring chronologies from Britain. Extremely significant and consistent correlation values (t = 5.94*** cf. Scottish Index Master (SCOTIM); t = 4.82* cf. British Isles Index Master (BRITIM) and t = 4.81** cf. Carlisle Medieval Chronology)) were found. These and other results indicate that the measured tree-ring series dates from AD1158 to AD1324. The best-estimated felling date range for the tree from which this timber was taken will be AD1356± 9 years or later or After AD1356. This is using the Belfast sapwood estimate range. It could be slightly shorter using the English sapwood estimate of 15 to 40 annual growth rings.
The tree-ring series from sample Q12769M gives extremely significant and consistent correlation values with many of the regional and local site chronologies from Ireland. This re-used timber’s tree-ring series does indicate a felling date in the 1630s.
Sample Q12770 does not give any significant or consistent correlation values with the regional and site chronologies from Ireland. The measured tree-ring sequences however does give extremely significant and consistent correlation values with tree-ring chronologies from Scotland. This would indicate that this timber in the latrine is from Scotland.
It is rare to find Scottish timbers in an Ireland context. Scottish timbers were however found in Holm Cultram Monastery, Abbeytown, Cumbria. These timbers were from a latrine sluice gate from the Cistercian monastery and dated to the twelfth century.
Baillie, M. G. L. and Pilcher, J. R. 1973. A simple crossdating program for tree-ring research. Tree-Ring Bulletin, 33 7-14.
Baillie, M. G. L. 1982. Tree-Ring Dating and Archaeology. Croom Helm. London.
English Heritage. 1998. Guidelines on producing and interpreting dendrochronological dates. London.
Munro, M. A. R. 1984. An improved algorithm for crossdating tree-ring series. Tree-Ring Bulletin, 44 17-27.