Dating the Gundulf Door, c.1080
/Tim Tatton-Brown and Daniel Miles investigate the remarkable ‘Gundulf Door’, dendrochronologically dated as one of the oldest doors in the country. Featured in The Friends of Rochester Cathedral Annual Report for 2001-2002.
Twelve years ago Dr Jane Geddes came to Rochester at the suggestion of the Cathedral Archaeologist, Tim Tatton-Brown, to look at a remarkable early door he had discovered on the back of the door into the stair-turret in the north-east transept. She looked carefully at the door, and wrote a brief article on it, for the Annual Report of the friends for 1989/90 (pages 19-22).
Her conclusion was that the door must have been made in the 11th or 12th century, based on her expert knowledge of carpentry and, in particular, of the decorative ironwork on it. (Dr. Geddes has recently published her definitive work, Medieval Decorative Ironwork in England (1999), and this sets the Rochester door in a wider context).
3D model of Gundulf’s Door in 2013. Laser scan produced by Dr Ben Edwards (Director, AS&C, and Manchester Metropolitan University) and Dr Andrew Wilson (Director, AS&C, and Bangor University), commissioned by Keevill Heritage.
In late July 2002, Dr Geddes is returning to Rochester (she lives in Scotland) to lecture on the door at the British Archaeological Association Conference. In preparation for this, it was suggested that it might be worth trying to date the door more precisely using dendrochronology. Consequently, with a grant for the cost of the work from the Society of Antiquaries of London, Daniel Miles of the Oxford Dendrochronology Laboratory, came to Rochester on the 25th of April to examine the door and to take miniature core samples for dating. With the help of Dave Baker and his men, the whole door, which is very heavy indeed, was lifted off its hinge pintles and taken outside the Cathedral and laid on two trestles. By using special drilling equipment, a 5mm core was drilled through the thickness of the board, passing through into the next board, and so on across the door. In this way the four original boards were sampled through a single small hole which was then plugged with an oak plug and stained.
Dendrochronology is the study of tree-rings through which it is possible to ascribe calendar years to each ring if the timber is successfully cross-matched. In order for this process to work, one must have a good selection of reference chronologies for the period and geographical area under study. Cross-matching is done by matching the annual climatic variations in the growth rings using a suite of specially-developed computer programmes. However, the tricky part is to determine when the trees were cut down, and more importantly, when the timber was used to construct the door. If the timber being sampled still retains bark, then the matter of interpretation is straight-forward, in that the outermost ring beneath the bark gives the year the tree was cut down.
However, in finer objects such as doors, the sapwood would normally be trimmed off in that it is much more susceptible to beetle attack. This leaves us with a major problem in interpretation in that we do not know how many rings were lost in the conversion of the tree into boards. Fortunately, studies of sapwood have shown that in 95% of trees from southern England, sapwood would have been between 9 and 41 years. By applying this estimate, one can be fairly confident that a board with no sapwood at all would not have been felled any less than 9 years beyond the last growth ring, and if there is some incomplete sapwood, or the waney edge denoting the heartwood/sapwood transition, then a felling date range can be derived by using the above estimate. The boards would not have been seasoned for any significant length of time, as evidenced from the shrinkage of the boards after being made into the door.
In the case of the Gundulf Door, four boards were sampled. Three of these boards matched so well together that they were likely to have originated from the same tree. The fourth board originated from a different free and source, and Was much slower grown. Taking into account any unmeasured rings to the edge of the board, the three boards from the same tree all had a last measured ring date of 1066, coincidentally the date of the Norman Invasion. This would suggest that a minimum number of outer-most rings were trimmed from the tree during conversion, and this was confirmed by the lucky presence of a heartwood/sapwood transition on one of these. By applying the 9-41 years sapwood estimate, a felling date range of 1075-1108 can be given for the group of three boards. The fourth board did not have any evidence of sapwood, but by taking account of any unmeasured rings, a terminus post quem or felled after date of after 1045 can be given for this. This particular board had an earliest measured ring date of 822, which means the tree would have been almost 300 years old when felled.
Thus, three of the boards from which the Gundulf door is constructed originated from a single tree which was felled sometime between 1075 and 1107. This is highly significant in that Gundulf (who also advised on the building of Rochester Castle as well as the White Tower at the Tower of London) was bishop of Rochester from 1077 to 1108. He founded a new Benedictine priory here in 1083, and shortly after which time the cathedral was completely rebuilt. Thus, the tree-ring evidence shows that this door is virtually without question the sole surviving piece of decorative iron and woodwork from Gundult's cathedral, and probably dates to just before the turn of the eleventh century. More importantly, out of only a handful of possible pre-1100 doors surviving anywhere in Britain, the Rochester door has the honour of being the earliest example to be scientifically dated.
Tim Tatton-Brown and Daniel Miles
Featured in The Friends of Rochester Cathedral Annual Report for 2001-2002.
Friends of Rochester Cathedral Annual Reports
The Friends of Rochester Cathedral were founded to help finance the maintenance of the fabric and grounds. The Friends’ annual reports have become a trove of articles on the fabric and history of the cathedral.
