Herefordshire Through Time
Herefordshire Through Time
Jay Lane Roman fort was discovered by aerial photography and has been tested by excavation. In 1978 the field was ploughed within hours of being bought. The field is very stony with Aymestrey limestone, which is not present in any quantity in other neighbouring fields. The site was excavated by S.C. Stanford in 1962, and a two hectare auxiliary fort of Ostorius Scapula's mid-1st century campaigns was discovered. Excavations revealed earth ramparts revetted with turf and with external ditches. There were timber gate-towers at the entrances, as well as corner and interval towers also constructed of timber. No internal buildings were excavated.
It is thought that the fort was developed in conjunction with the Leintwardine and Buckton forts.
The fort appears to have been dismantled around AD 70-80, based on Samian ware finds and the historical context.
The fort faces south-west, looking towards Coxall Knoll and the Upper Teme valley, and on this side the slope to the Clun is steep. The approach from the north-west is also fairly steep, but on the other two sides the ground falls away gently.
Excavation has confirmed the presence of a double system of ditches. The ditches are V-shaped, 7-8ft (2m) wide and 3ft (1m) deep where they were well preserved on the north-east side. A berm(level space) 5ft (1.5m) wide separates them, and upon this a small bank may have been raised to increase the counterscarp of the inner ditch. On the steeply sloping north-western and south-western sides the ditches are shallower, and the outer ditch is especially weak.
No trace of a rampart bank was found, nor were there any postholes for a timber-framed rampart or palisade. The rampart must have been provided with a turf revetment, a suggestion that is backed up by the character of the ditch filling.
A further clue to the form of the rampart comes from the unexpectedly small gate-towers, of two ranks of posts rather than the more common three. If these towers are a guide to the width of the rampart, then it will have only have been about 9ft 6ins (2.9m). A rampart of this width would have stood no more than 10ft (3m) high to the rampart walk, although a breastwork and merlons could have raised this height to 15 ½ ft (4.7m).
The gatehouses were identified by large post-pits in the natural clay and shale. The pits were square or oblong with sides 2-3ft long, and were 2.5-3ft deep from the modern ground level. Some of the pits showed as pink or brown clay with occasional flecks of charcoal against the bright yellow packing of clay and shale. This indicates that the post stumps rotted in situ, having perhaps been cut off.
These postholes show that three or four of the gatehouses were built upon two rows of five posts. At each end a tower 9.5ft square overall would have provided a guard-room at ground level. The space between the towers was divided centrally by two posts, leaving a 9.5ft passage either side. At the north-east gate two drains passed through these two middle areas.
Only the south corner tower was excavated in 1962. On the basis of the four posts located this is assumed to have been a six-post structure, measuring 15ft by 8ft. The surviving posts-pits were very shallow on this steeply sloping corner and the post positions ill defined, probably through weathering and root penetration rather than robbing.
Between the gates and corner towers were set interval towers with four posts each. Three were excavated, two on the north-west side and a third on the south-west side. The north-east tower on the former side was best preserved, showing that the posts were of the same scantling (cross-section) and set in pits of similar size and depth as those for the gateway timbers. The excavated towers show a size range of 8ft-10ft square.
The double ditches were interrupted by causeways for the four gates. The dimensions within the lip of the inner ditch are 537ft by 448ft, creating an enclosed area of 5.6 acres. Within the ramparts the dimensions are 507ft x 413ft, i.e. 4.8 acres. The shape of the fort is very nearly rectangular.
The width of the major roads may be assumed to be no less than about the 20ft of the gateways. In the retentura the areas either side of the via decumana would measure 122ft by 164ft, with dimensions of 200ft by 164ft in the praetentura. Planned thus, Jay Lane could accommodate six buildings of barrack size in the retentura and ten barracks plus two narrower buildings in thepraetentura. This is more accommodation than was required by the military infantry regiment at Fendoch in Perthshire, Scotland, and presumably indicates the presence of stables and extra stores. The site is too small for either a cavalry or part-mounted military regiment. Cavalry alae of 500 men are known to have occupied sites at Chesters (5.75 acres) and Benwell (5.64 acres) (both in Northumberland), so Jay Lane is only just below the known limits for such units. It is most probable that Jay Lane held an ala quingenaria of 500 cavalrymen.
The tactical advantages of this emplacement on the hill are appropriate for a front-line fort established at the time of the conquest; it marks the earliest military use of the Leintwardine position. Such as it is, the Samian ware found is all of the 1st century, with two pre-Flavian pieces and four that could date to the reign of Nero. Since there is no reason to believe that most of these finds were deposited late in the fort's history, we may use the latest items to provide a terminal date of c. AD 70-80 for the site. Since some of the gateway towers were considered not worth digging out, the occupation was probably of 15 or more years duration. As Jay Lane is a key point on the Watling Street West, its foundation is certainly to be related to a major development of the frontier rather than any minor campaign. Historically the bracket may be defined as AD 47-61, i.e. the start of the governorship of Ostorius Scapula and the reorganisation following the revolt of Queen Boudicca.
[Original author: Miranda Greene, 2004]