Font Size

100% 150% 200%

Background Colour

Default Contrast
Close Reset

All our staffed properties, sites and offices, including the HES Archives and Library, are currently closed, but we’re working on plans to gradually reopen. In the meantime, you can access our services online. Find out more.

Scheduled Website Maintenance 14/07/20 00:00 – 04:00GMT – There will be periods of time during this window when this website will be unavailable.

Camp Tops

Cord Rig (Prehistoric), Fort (Period Unassigned)

Site Name Camp Tops

Classification Cord Rig (Prehistoric), Fort (Period Unassigned)

Alternative Name(s) Camp Tops Fort; Cock Law

Canmore ID 59026

Site Number NT81NE 8

NGR NT 86022 18019

NGR Description From NT 86046 18045 to NT 86011 17977

Datum OSGB36 - NGR


Ordnance Survey licence number 100057073. All rights reserved.
Canmore Disclaimer. © Copyright and database right 2020.

Toggle Aerial | View on large map

Digital Images

Administrative Areas

  • Council Scottish Borders, The
  • Parish Morebattle
  • Former Region Borders
  • Former District Roxburgh
  • Former County Roxburghshire

Archaeology Notes

NT81NE 8 from 86046 18045 to 86011 17977

For adjacent cultivation terraces, see NT81NE 30.

(NT 86022 18019) Fort (NR)

OS 6" map, (1962).

The small fort which occupies the summit of Camp Tops is D-shaped on RCAHMS 1956 plan, fig.441, measuring internally 200' E-W by 156', and is defended by double ramparts on the SW and elsewhere by triple ramparts with two medial ditches. The ramparts are of dump-construction, and show no stonework except on the S, where short lengths of outer facing, one course high, are visible in each of the scarps of the wasted double ramparts. Since the ground falls away fairly steeply on this side it is possible that these facings are simply the remains of kerbs designed to prevent the rubble ramparts from slipping.

On the N side, where the defences are best preserved, the inner rampart, which encloses the natural summit-area of the knoll, is only 6" higher than the surface of the interior but rises 9'6" above the bottom of the inner ditch. There are two entrances. Within the interior six huts, either circular or slightly oval, are defined by shallow trenches 2' - 3' wide. The trench of hut 3 is continuous, but those of the others are only visible for roughly half the perimeter, normally on the higher side of the hut. Nos 2 and 3 each contain an oval hollow, no doubt a hearth, while no 6, which has been cut into by a later quarry, is scarped internally on the upper side. An isolated crescentic scarp (7) on the S side of the enclosure may represent a seventh hut, while slight indications of another (8) were observed on the opposite side.

All the features noted are parallelled in the huts at the nearby settlement on Hayhope Knowe (NT81NE 18) which have been dated by excavation to the Early Iron Age. On the analogy of the sequence revealed at Hayhope Knowe it is possible that the huts were originally enclosed by a wooden palisade, and that the ramparts represent a later addition during the same phase of occupation. A trial pit (X on plan), revealed no trace of a palisade, but in view of the short length uncovered this negative evidence is not absolutely conclusive. Two pieces of local Iron Age pottery, now in the NMAS, were found in the trial pit.

The "earthworks" indicated by hachures on the OS plan lower down the slope on the N side of the fort are not part of the fortifications but simply two deep, hollow tracks of an old road, partly destroyed by cultivation.

RCAHMS 1956, visited 1949.

As described above. RCAHMS plan revised.

Visited by OS (JLD) 24 August 1960 and (RD) 27 June 1968.

No change to previous field report.

Visited by OS (DWR) 17 July 1973.

No change.

Visited by OS (TRG) 9 August 1976.

Camp Tops fort crowns a prominent knoll at the N end of a long ridge that reaches back to The Border. The ridge separates the valleys of the Kingseat Burn and the Kelsocleuch Burn and the fort overlooks their confluence at Cocklawfoot some 800m to the NW and controls ancient routes along the ridge, S to The Border. A second fort, Hayhope Knowe (NT81NE 18), is clearly visible 400m to the S. Eccentrically placed around the knoll, the fort occupies a gentle SSW-facing slope above a moderate slope on the SSE; elsewhere the slopes become moderately steep.

The fort, which is D-shaped on plan, measures 61m from E to W by 47m within multivallate ramparts. Triple ramparts, with two medial ditches, may be observed around the N part of the knoll with double ramparts visible around the remainder. Indeed the SE to S angle of the fort's ramparts has almost been levelled through subsequent ploughing thus hindering interpretation. Within the enceinte , there are traces of at least ten houses, of varying type. The houses appear to be laid out in two rows along a 'street' on the gentle SSW-facing slope.

The defensive sequence

Possible palisade

In the Inventory account, the RCAHMS (1956) drew attention to similarities between the sequence observed here with that at Hayhope Knowe, and suggested that the earliest element in the defensive sequence here could also be a wooden palisade within the line of the ramparts. Excavation of a trial pit on the supposed line revealed no traces, but the absence of a palisade-trench over such a short stretch cannot be taken as conclusive. Re-survey of Camp Tops, in 1986, also failed to locate any positive evidence for a palisade-trench at the core of the fort's defences. Nevertheless, the argument that the visible defences developed around a palisaded enclosure remains attractive on analogy with Hayhope Knowe. Additionally the orientation of the 'street', which is not aligned on the entrances through the upstanding ramparts, perhaps offers a tantalising hint to the presence of a palisaded enclosure. Without more substantive evidence, however, such thoughts can only be speculative.

Phase 1

The innermost rampart is seen as the earliest extant phase of the defences. It comprises a stone wall which is relatively intact along the N part of the circuit especially over the highest part of the knoll. Here the rampart is no more than 0.2m high over the interior but stands over 1m above an exterior shelf. Entrances on the E side of the enclosure and, possibly, on the W, may well be secondary as they are continued through the outer ramparts. Indeed the original entrance may well have been blocked during the construction of the phase 2 rampart. On the SW and S, the rampart itself is almost negligible, only surviving as a much reduced scarp returning across the knoll immediately below a shallow quarry scoop to the base of a later quarry which cuts into the houses of the interior. The rampart only picks up again some 15m before the entrance on the E side of the knoll.

Phase 2

Subsequent ploughing has much reduced the outer ramparts of the fort but oblique aerial photographs (D Harding, University of Edinburgh; RCAHMS) suggest that a further complete circuit of double ramparts was added outwith the innermost defence. The phase 2 ramparts are not concentric with the original circuit but diverge in the S sector. There are slight traces of a counterscarp outside the outermost bank in the N half of the fort. Entrances, corresponding with those in the phase 1 rampart were observed on both E and W sides of the phase 2 enclosure. That on the E measures 8m across, whereas that on the W is 3m wide. A further break on the SSW side of the enclosure leads into a sharply-defined quarried area which penetrates the interior for some 14m. Over the S side of the knoll traces of outer facing stones are visible on the inner of the two ramparts.

Subsequent activity and cultivation remains

Destruction of the ramparts through cultivation is most pronounced over the SE to S sector of the circuit where the upstanding banks have been levelled leaving only scars visible on aerial photographs. The exact nature of this cultivation is not clear from the oblique photographs but it does not resemble the cord rig visible on the N side of the knoll. Here there is a significant area of cord rig cultivation which extends up to the rampart and which appears to be covered by a counterscarp bank.

The interior features

At least ten houses were recognised within the multivallate enclosure during the 1986 survey of the site. They are all levelled into the gentle SSW-facing slope of the knoll. The relationship between activity in the interior and the rampart sequence is difficult to establish but it appears that two rows of houses were laid out on either side of an irregular terrace creating the appearance of a 'street'. Before discussing the sequence between the houses, it is worth observing that the 'street' does not align on the present E entrance but slightly to the S of it at a point where the innermost rampart is uncharacteristically thick. It is tempting to look for 'blocking' of an original entrance at this point, whilst a new (phase 2) entrance was opened up just to the N or perhaps in an even earlier opening in a, now invisible, palisade buried by the phase i ramparts.

The earliest houses lining the 'street' appear to be houses (2, 5, 6 and 7) on the N side and houses (4, 9 and 10) on the S side of the 'street'. House (2) is eventually replaced by houses (1 and 3) in succession; their construction effectively blocking the 'street' and possibly reflecting the uncertain status of the W entrance through both phases of rampart. A tenth house, house (8), lies downhill from the lower side of the 'street' and appears to be cut by the late quarry through the S side of the ramparts. Additionally, at the W side of the site there is a small platform-scoop 6m in diameter and a 'stance' of similar proportions. It proved difficult to distinguish house-types on the ground, particularly between ring-grooves and true ring-ditch houses. This was in part due to the complexity of the overlapping of the sequence between buildings and because of later quarrying and plough damage.

A possible ring-ditch house (1) measures about 12m overall. The entrance may lie to the NW. It overlies house (2) but is, in turn, overlain by house (3).

House 2 is a ring-groove house which measures 10m in diameter. Again the entrance is to the NW. It is overlain by house (1) (and house (3)).

House (3) appears to comprise a double ring-groove and measures 10m in diameter with an entrance to the E. It overlies (houses (2) and (1)).

House (4) is another double ring-groove house measuring 10m in diameter with an entrance possibly to the E.

House (5) is a ring-groove house which measures 10m in diameter with an entrance to the E.

House (6) is a ring-groove house which measures 10m in diameter with an entrance to the E.

House (7) is a possible ring-groove house which measures 10m in diameter with an entrance to the E.

House (8) comprises a platform with a groove along the rear (N). It measures about 9m in diameter and has an entrance on the W.

House (9) is a double ring-groove with a central platform measuring 10m in diameter. The position of the entrance could not be determined.

House (10) comprises a platform with a groove at the rear. It measures a maximum of 9m in diameter.

Information from RJ Mercer (University of Edinburgh) 15 March 1986

RCAHMS MS 2598. No. 46/596


Note (14 September 2015 - 16 August 2016)

This small fort occupies a hillock forming the summit of the long spur dropping down from White Knowe above Hayhope Knowe. The defences comprise three ramparts and ditches, which step down the steep N flank of the hillock and swing round onto the E before disappearing beneath an area of cultivation, but at an entrance on the WNW the innermost and medial rampart contract into a single line and there are only two on the SW; there are also traces of a counterscarp bank on the N. Whether this contraction of the defences on the W and SW was part of the original design of a multivallate scheme, or whether it indicates several periods of modification and rebuilding is unclear, but it is difficult to sustain Roger Mercer's suggestion in 1986 that the inner rampart was a primary defence to which the outer ramparts were subsequently added. The D-shaped interior of the fort, measuring 61m from E to W by 47m transversely (0.02ha), is packed with traces of timber round-houses representing several periods of construction. At least twelve can be seen, comprising shallow ring-ditches, platforms encircled by shallow grooves, and simple platforms, and several have a pit in the centre, which in 1949 RCAHMS investigators identified as possible hearths. The round-houses are broadly disposed in two lines, though whether this is an accident of the sloping topography or about access from an axial 'street' is unclear. The only other features visible within the interior is a faint hollow immediately to the rear of the rampart on the N, but a trial excavation in 1949 by RCAHMS investigators revealed that it was no more than a shallow quarry behind the rampart. Surrounding the fort there are extensive traces of cultivation, including cord rig, which extends right up to the foot of the defences on the N, where Mercer suggested it might be overlain by the possible counterscarp bank; the relationship remains untested.

Information from An Atlas of Hillforts of Great Britain and Ireland – 16 August 2016. Atlas of Hillforts SC3438

Sbc Note

Visibility: This is an upstanding earthwork or monument.

Information from Scottish Borders Council


MyCanmore Image Contributions

Contribute an Image

MyCanmore Text Contributions