Monument record MHG12802 - Broch, Doir A' Chata

(NC 5017 0239) Broch (NR) (remains of) OS 6"map, (1969)

The remains of a broch is now a structureless heap of ruins from which most of the large stones have been removed. No wall-faces are visible but Pococke (1887) who calls it 'Dun Cor', says internal diameter was 30ft with walls 6ft thick.
The mound now measures about 16m N-S by 14m and is 2-4m high on N side (visited by OS {EGC} 21 June 1963).
R Pococke 1887; RCAHMS 1911; A Graham 1949; J R C Hamilton 1968.

As described above; there is no local knowledge of name "Dun Cor".
Resurveyed at 1:10,560 Visited by OS (N K B) 30 August 1976.

Archaeological excavation May 1992, in advance of construction of extension to existing farmhouse. The presence of broch named Dun Cor was referred to 18th century, and substantial mound behind modern farmhouse was thought to relate to this structure. Surface traces indicate that mound is multi-phase and capped by substantial post-medieval occupation. A structure provisionally identified as a galleried roundhouse occupied E side of mound, outwith area of excavation. This was restricted to a relatively small area of 9m by 6m on southern edge of mound, together with a slit trench of 3.5m by 0.8m projectng westwards from NW corner of main trench. There were no clear surface traces of structures within excavated area. Excavation revealed a denuded length of drystone walling which projected westwards into trench for c2m. To W of this walling had been removed. Two distinct construction phases appeared to be present. The primary external face was of boulders backed by disturbed and partially voided core of cobbles, from which sherd of undecorated pottery recovered. No internal face survived within trench. These deposits were overlain by an irregular mound of midden material, possibly of comparatively recent origin. This walling had been refaced with carefully constructed boulder and coursed drystone face backed by earth and rubble. The primary wall-face was respected by compact cobbled surface up to 3.5m wide. Its underlying ground surface had been slightly scarped to create level surface. A secondary, roughly paved pathway lay above cobbled surface, and ran beside secondary wall-face. This pathway curved gently across trench from E- WNW and may indicate former course of removed walling in western half of trench. A continuous buried ground surface was sealed beneath these archaeological features, and lay directly above coarse sandy till subsoil. Other excavated deposits are more recent in origin, and relate to post-medieval activity. A near-level yard was created by stone-dumping directly above denuded walling. The excavated drystone walling is not inconsistent with that recorded in atlantic roundhouses of first millennium BC. Any such interpretation is, however, determined more by historically-recorded presence of broch than by diagnostic characteristics of excavated remains. The putative galleried roundhouse to E of excavated area may be structure to which historical records refer, and certain of excavated remains may represent features associated with it. <1> <2>

Further to work conducted in 1992 excavations were conducted in October 1996 in order to determine nature and extent of principal structures present within large grassed mound believed to contain remains of broch (NMRS NC 50 SW 2). Excavations 1992 had revealed remains of a substantial prehistoric wall, although it could not be conclusively interpreted as part of broch, and surface morphology of mound had indicated that wall was probably part of an ancillary structure. Four trenches opened 1996, with aim of removing modern material and collapsed remains to expose preserved prehistoric levels, but without excavating them. Trench 1 was designed to establish whether E end of mound contained a broch or related structure, as had been suspected from surface traces. Majority of deposits identified in this trench were post-medieval in date and related to dumping of angular stone, possibly to form a yard surface. Beneath this material was a bed of substantial rounded boulders, which may relate to prehistoric activity. These boulders did not appear to form part of a wall, and may represent terracing to E of roundhouse. The remaining three trenches revealed position of broch-related structure. It was found that structure lay immediately N of Durcha dwelling house, and that postulated post-medieval structure atop mound did not exist. Trench 2 was of sufficient size to provide a complete plan view through surviving remains of broch wall on its NE circuit. Trenches 3 and 4 were opened to detect position of inner wall face, and to determine diameter and extent of structure. It proved to have an internal diameter of c9m and walls c5m thick. The prehistoric walling excavated 1992 appears to have formed heavily denuded remains of S circuit of outer wall face. There is little doubt that identified structure is of type traditionally classified as broch. Where exposed in Trench 2, wall possessed inner and outer faces of large boulders, well fitted and with small chocking stones filling gaps between them. Here wall was of solid construction, containing discontinuous rough boulder internal faces, presumably devices used to ensure stability of construction, between which lay dumps of cobbles and boulders. Secondary face abutted outside of exterior face, against which midden deposit had been subsequently banked. It is estimated that inner wall face is preserved up to at least 2m high on N circuit, whereas outer face is nowhere likely to survive greater than 1m high. <3> <4>

All the investigations from 1992-96 have subsequently been published. <5>