Accessibility

Font Size

100% 150% 200%

Background Colour

Default Contrast
Close Reset

Harpercroft

Fort (Period Unassigned)

Site Name Harpercroft

Classification Fort (Period Unassigned)

Alternative Name(s) Wardlaw Hill

Canmore ID 41986

Site Number NS33SE 6

NGR NS 3600 3252

Datum OSGB36 - NGR

Permalink http://canmore.org.uk/site/41986

Ordnance Survey licence number 100057073. All rights reserved.
Canmore Disclaimer. © Copyright and database right 2018.

Digital Images

Administrative Areas

  • Council South Ayrshire
  • Parish Dundonald (Kyle And Carrick)
  • Former Region Strathclyde
  • Former District Kyle And Carrick
  • Former County Ayrshire

Archaeology Notes

NS33SE 6 3600 3252.

(NS 3600 3252) Fort (NR)

OS 6" map (1967)

This fort, crowning a subsidiary summit of the Dundonald range, consists of an inner ring, formed by a ruinous stone wall about 300ft in diameter, with a second wall some 300ft outside it. Both are now low, stony banks spread to a maximum width of about 20ft. The plan is unusual if one period only is, as seems probable, represented, but it is comparable with that at Carwinning (NS25SE 6). An unfinished oval quern stone from here is in the NMAS (Acc No: BB 125).

R W Feachem 1963; D Christison 1893; Proc Soc Antiq Scot 1958

A large circular earthwork measuring some 100m in diameter, and which is in the form of a flattish area on top of the hill bounded by the course of a stone wall, 0.7m wide and 0.3m high, on top of a scarped terrace. This 'terrace' averages 1.7m in height and is composed of earth and stone. The earthwork is mutilated by radar installations; it is similar, but larger, in character to that a few hundred metres N of it (NS33SE 5). It has not the characteristics of a hill fort, but more of a settlement.

Visited by OS (JLD) 28 May 1954

A 'citadel' fort or minor oppida, occupying an entire summit plateau, at about 145m OD, comprises an inner fort or settlement on the highest point of the plateau, and at an average distance of about 100m, an outer defensive circuit.

The inner work measures some 105m E-W by 90m transversely within a collapsed stone-built rampart mostly showing as a scarp up to 1.6m high. The rampart has been severely mutilated around the N side, where there may have been an entrance; and a roadway has been cut through the E side. A ruined wall of later date follows the top of the rampart around the W half of the fort. The interior is occupied by buildings of a transmitter station within an enclosure fence; what may be part of an original internal division on the E side (see Christison's plan) has been cut by this detail.

The outer work, of similar construction, follows the natural crest of the plateau and measures some 320m E-W by 260m transversely (about 6 hectares - 15 acres). It utilizes steep natural scarps around the N and W but around the S a stony rampart with an outer scarp up to 1.5m high survives. On all bu the W side the rampart scarps are topped with a continuous dry stone dyke. The undulating interior has been heavily cultivated and is featureless, as is a lower terrace area to the E of the fort (area NS 363 325) where RAF APs (F21/58/2712: 0070-1, flown 1959) possibly indicate the former vestiges of a rectangular field system. Also of possible significance is an old bank scarp which extends eastwards away from the fort rampart at about NS 3621 3247.

The form of this work differs in many aspects from its suggested parallel at Carwinningm, and a closer comparison may be Carman (NS37NE 2) and possibly Bowden Hill (NS97SE 1).

Resurveyed at 1:2500.

Visited by OS (JRL) 4 May 1982

A watching brief was undertaken during the laying of communications cables, within an existing trench, to the radio-mast station on the summit of the hill. The re-excavated trench ran directly W to E and crossed both the inner and outer rampart, although their was a large quantity of boulders downslope, possibly representing rampart tumble.

At that inner rampart that upper part of a possible stone faced wall with a rubble core(1.6m wide) was uncovered. The trench did not penetrate the lower levels of the rampart. No artefacts were recovered.

Sponsor: Mercury Communications Limited.

CFA 1992

Activities

Field Visit (16 March 1953)

Earthwork, Harpercroft (see the plan made by the Commission in 1953, and Proceedings, vol. Xxviii, 390).

Inner enclosure: This is approximately circular and is defended by a single rampart. Except on the north side, where it is drawn across flat ground, the rampart runs along a well-marked shoulder which takes the form of a scarp about 6ft in height. One or two of the outer facing stones are still in situ, and a great deal of tumbled stone on the face of the scarp shows that the rampart has, in fact, been a substantial stone wall, but its thickness is uncertain in the absence of any sign of inner facings. What Christison took to be inner facings are actually the ruins of a later dyke, built of materials from the older wall and no doubt contemporary with the croft shown on the plan. There are no definite indications of the original entrance but it probably lay in the wide gap on the north. The stone-revetted track on the south-east is unquestionably modern, while the gap on the east seems to be due to ploughing. The croft and field walls clearly belong to a later occupation, and the mounds notice by Christison are presumably field-clearance heaps. The L-shaped wall on the south east probably belongs to an outbuilding connected with the croft. The whole of the top of the hill is enclosed within the rampart, and consists of a more or less level area which has been ploughed at some time in the past. Any original structures have been obliterated.

Outer Enclosure: On the west side, north of the point where a later dyke diverges from it, the outer rampart is indicated only by a crest-line running along the lower shoulder of the hill and overlooking a marshy hollow to the north west. The scarp is only 6 ft in height at the point of junction with the later dyke, but increases to 20ft in height at the gully, and then diminishes again. The bottom of the scarp is about 15 ft in from the dyke at the north end of the gully and continues at this distance from it as long as they remain parallel. All the stones have been robbed from the rampart in this sector to build filed walls. There is a break in the scarp, 33ft in width, situated opposite the hedge running down from Wardlaw Hill, and on the saddle which offers easy access to the fort. Another track, 20ft in width, occurs just before a modern wall across the line of the rampart and where a transverse gully cuts it at right angles. On the north east side the modern wall lies 12ft in from the edge of the scarp (the hachures can be made of arbitrary length here on the final plan, for the slope is long and unbroken). The first signs of tumbled stone appear where the scarp has broken away back to the line of the modern wall at the first tree. The break is 60ft in width, and beyond it the scarp is again 12ft in from the dyke with, however, a distance of 8ft between the top and bottom and a height of 6ft. The scarp becomes increasingly stony, and where the first kink appears in the wall, the outer face of rampart is visible for the first time, 5ft beyond the wall; it is 7ft in length and consists of a core of small stones faced with large boulders. The crest here is only 6ft from the dyke but the bottom of the scarp is still 8ft out and 6ft in height. The rampart continues in this state to the eastern apex, where the ruined wall branches off to the ESE: beyond this point the scarp is very slight and the wall is superimposed on the crest. The scarp next becomes a mound with a very slight inner slope some 10ft inside the dyke, and an outer slope 6ft beyond the dyke as shown on air photographs. Here loose stone is visible on both sides but there are no facings. The inner slope of the mound gradually decreases in width to the point where the modern wall diverges from its line, and a ruined wall takes its place. The latter starts on the line of the rampart then, after a short distance, cuts 15ft inside it before resuming again on the same course. In this sector the scarp is about 8-12ft out, but on this side the rampart does not follow a contour but is drawn arbitrarily across a slight slope. Some 60yds north of the starting point in this description, a track 8ft wide, and probably secondary, cuts obliquely through the crest-line from WSW.

The conclusion is that the outer rampart has been substantially stone wall, like the inner one, but its width is uncertain. Its line is in some measure dictated by contours but on the south west it deliberately chooses a curvilinear path. Air photographs and the OS map should be employed to complete the plan.

Visited by RCAHMS (KAS) 16 March 1953.

References

MyCanmore Image Contributions


Contribute an Image

MyCanmore Text Contributions