The tree ring dating of a group of Waringstown timbers

by M. G. L. BAILLIE, Archaeology Department, Q.U.B.



Review

Journal of Craigavon Historical Society

Vol. 3 No. 2

To be completed ...

In the second volume of this Journal I discussed briefly the possibilities of tree-ring dating in the north of Ireland. The method depends on the similarities between the patterns of wide and narrow rings in oak trees growing over the same period of years. Since that article a 590 year standard chronology for Irish oak has been published where each ring can be assigned to a calendar year (Baillie, 1974). With the aid of a computer programme it is now possible to date the ring patterns from north Irish oak timbers on a routine basis with a better than 75 per cent chance of success, providing that the individual specimen exhibits more than 100 growth rings.

One building, the Grange, in Waringstown, has produced a series of oak timbers which adequately demonstrate the successes and limitations of tree-ring dating. The results of the study of these timbers form the main substance of this article and hopefully show how this dating method can allow the establishment of a framework of key dates to supplement, or in some cases supersede existing historical information. To expand this further, any information on the Grange can assist in two separate spheres, the first in the general history of Waringstown and the second in dating the particular style represented in this important house.

Waringstown Grange

While Waringstown House and the Holy Trinity Church are comparatively well documented, little is known of the remaining 17th century buildings except for the inferred early date for the cruck built cottages and the Grange which is variously described as c. 1698 or early 18th century on the basis of a vaguely associated date stone and style respectively.

There seems to be no question that the Grange as it stands to-day is a building of circa 1700. This is adequately reflected in two references. The first attributes it to the early eighteenth century (HMSO 1966) and undoubtedly the house does compare with Berwick Hall, County Down, which is attributed a date c. 1700, on the basis of a normally datable architectural feature. The second reference obviously weighing heavily on the basis of style quotes a date stone of 1698 which is built into the farmyard wall some distance from the house itself (UAHS, 1969). One point which is clear is that the house was thatched up until the beginning of this century. The present slate roof is considerably less steep than the original thatch and the long walls have been raised to accommodate this decreased pitch. This can be clearly seen in Plate l.

Re-roofed

When the house was re-roofed in the present century the original roof timbers were removed. However when the roof space was visited during renovations between 1972 and 1974 it was discovered that the two ends of the blades of one of the roof trusses had been left in place when the truss was cut away. These blades had right angled notches at their lower ends which rested on the top inside edge of the original masonry walls, without wall plates.

On examination it was found that both of the truss blades had been cut as squared halves of the same tree, a common building practice to ensure matching blades in a truss. One of these blades retained bark on an unsquared edge and from a dating view point this meant that an accurate felling date could be established. A slice was removed from the truss blade and its transverse surface polished so that the width of each successive ring from the centre to the outside could be measured.

A plot of these ring widths against a scale in years forms the ring pattern for the specimen. No. 1 in Fig. 1 shows the ring pattern of the Grange truss blade plotted in its matching position against the Belfast master chronology. This cross match was found visually by moving a trace of the ring pattern over the master chronology, it was then verified by an independent statistical test which showed that a match of this significance could be expected only once in 10,000 random attempts. By reading off the date from the master chronology corresponding to the outer, felling year of the truss blade it is seen that the tree from which the truss blade was cut last grew in 1692. This is verified by a small stray timber from another part of the house which also retained its total sapwood, No. 2 in Fig. 1.

The Grange, in former times, with thatched roof shutters.

Earlier times

Since it is now generally accepted that in earlier times timber was not seasoned for building purposes, we can infer that the Grange was built in 1693 or 1694. Examination of the truss blades showed that they had warped considerably since they were cut which increased the likelihood of their use immediately after cutting. This date is interesting on several counts, firstly it fits with a general building phase connected with the return of confidence in Ulster after the Revolution of 1688, and secondly it gives another date for the use of the truss roof in Ulster. In addition it opens up interesting historical possibilities, for example, is this a house built by or for one of the weavers who are supposed to have been brought to Waringstown from the Low Countries following Samuel Waring's visit there in 1688.

Early in 1974 the ongoing renovation of the Grange uncovered a large heavily adzed and worked oak beam from the lower floor of the house. This timber again exhibited bark along one edge and its ring pattern was obtained for tree-ring dating. This beam, No. 3 in Fig. 1, had put on its final growth ring in 1658.

This date opens up several lines of speculation, the first of which is that the Grange was originally a one storey dwelling, in line with almost all the other houses along the and the northern half of the village. In addition, because of its central position at the road intersection, in the centre of the village, it must have been one of the first houses to be erected and would have formed the nucleus of the settlement. This is of course much more reasonable than having the house built for the first time in the 1690s. However, it throws new light on the early stages of the settlement at Waringstown. Atkinson infers that the village did not begin to form until after the building of Waringstown House in 1667. It would appear from the date for this Grange beam that houses existed in the centre of the village from within a year or so of William Waring's purchase of the land in 1658. The exact nature of this early dwelling will probably never be known, but it is possible that it was a cruck built house like the others which lined the village up until the 1950s.

If all building timbers came with traces of bark or total sapwood it would be possible to build up an almost complete picture of building phases in this area. However, some brief notes on three other timbers from the Grange will show the shortcomings of the tree-ring method where inadequate material is involved.

Re-used

Of these three timbers, two come from a patently re-used context in a barn at the rear of the Grange and one comes from an unspecified location within the house. In each case the sapwood, the outer living rings of the tree, was missing. This is not serious in cases were a definite heartwood/sapwood transition remains, because it is possible to make some allowance for the sapwood rings, usually 32 + 9. However, if there is no trace of sapwood it is impossible to gauge the amount of heartwood missing and the felling date can then only be estimated as the date of the existing outer rings plus 32 + 9 or later.

In Fig. 3, No. 4, is the stray from the Grange and Nos. 5 and 6 are the re-used timber from the barn, the sapwood allowance has been indicated in each case, but it must be stressed that the actual felling dates could be later in each instance. It can be seen that these timbers fall into an earlier rather than a later group and most probably belong to the initial 1658 building phase, those from the barn having been re-used possibly at the enlarging of the Grange in the 1690s. They do not carry the same weight as the evidence for dates supplied by the complete timbers 1 to 3. It is of interest that another timber from the site of the old Police Station in the centre of the village dating 1645 + 9 or later, seems to support this early building phase.

In conclusion we can see the importance of the accurate historical information supplied by tree-ring dating. It is unfortunate that to all intents and purposes the cruck houses of Waringstown have disappeared. It would have been possible with careful selection to trace the expansion of the village from its beginnings in 1658 to the first map record in 1709. However, it is possible, that with time, occasional timbers may turn up from secondary contexts which will allow at least a few more dates to be fitted to the history of the development of the Waringstown area.

  • Atkinson, E. D. (1898) - An Ulster Parish: Being a History of Donaghcloney, (Dublin).

  • Baillie, M. G. L. (1974). A Tree-Ring Chronology for the Dating of Irish Post- Medieval Timbers, Ulster Folklife, 20.
  • Baillie, M. G. L. (1974). Dendrochronological Dating of Oak Timbers from the Gloverstown House, Ulster Folklife, 20
  • McCourt, D. and Evans, D. (1972) - A seventeenth Century Farmhouse at Liffock, County Londonderry, Ulster Journal of Archaeology, 35
  • HMSO (1966) The Archaeological Survey of Northern Ireland; County Down (Belfast).
  • UAHS (1969) - List of Historic Buildings in the Area of Craigavon (Ulster Architectural Heritage Society, Belfast).