Font Size

100% 150% 200%

Background Colour

Default Contrast
Close Reset

Castle O'er

Enclosure(S) (Period Unassigned), Fort (Early Medieval), Fort (Iron Age), Bead(S) (Glass), Spindle Whorl(S)

Site Name Castle O'er

Classification Enclosure(S) (Period Unassigned), Fort (Early Medieval), Fort (Iron Age), Bead(S) (Glass), Spindle Whorl(S)

Canmore ID 67376

Site Number NY29SW 10

NGR NY 24190 92850

Datum OSGB36 - NGR

C14 Radiocarbon Dating


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 Dumfries And Galloway
  • Parish Eskdalemuir
  • Former Region Dumfries And Galloway
  • Former District Annandale And Eskdale
  • Former County Dumfries-shire

Archaeology Notes

NY29SW 10 24190 92850

(NY 2419 9285) Castle O'er (NAT)

Hill Fort (NR)

OS 6" map (1957)

See also NY29SW 32 for linear earthworks radiating from the fort.

For other linear earthworks on Castle O'er estate, see NY29SE 2, NY29SE 68, NY29SE 69.

The remains of a rampart or wall encloses the summit of Castle O'er Hill, an area c. 300' by 150'. The base of the knoll is in turn defended by a rampart thrown up by an internal quarry ditch. This combined work, measuring c. 500' by 350', lies within outer double ramparts with median ditch which enclose an area 800' in length which includes a well. (R W Feachem 1963). It could be significant that at Castle O'er, the one fort in E Dumfries other than Burnswark (NY17NE 2) that has been excavated - by D Bell, the proprietor (D Christison 1898; D Bell 1895) - a carbonised post was found beneath the inner rampart. Here, no timber-lacing is mentioned, and no vitrifaction of the rampart is observable, so that an early rampart is always possible.

Whatever may be the precise context of the innermost enclosure, the number of visible hut positions does not exceed 9, and if timber-built dwellings were confined within this area, there could never have been more than 12 at any time. There are problems of interpretation which may only be solved by excavation. it could be a defended settlement with an outer annexe, or an earlier site upon which later walled settlements with timber-built huts have been superimposed. (c/f NY29SE 8).

A Charles II coin, dated 1672, was also found during Bell's excavations.

R Bell 1906; J and R Hyslop 1912; RCAHMS 1920, visited July 1912; R C Reid 1930; G Jobey 1971.

This locally large fort is now completely swallowed up by a forest of conifers, and, although not planted itself, is to a certain extent blurred and defaced by the tussocks into which the grass which covers it has coagulated. A visit is nevertheless worthwhile, and the lay-out can be best appreciated from the summit of the rocky knoll which is included in the defended area.

The fort recalls that on Bonchester Hill (NT51SE 10) in that the inner part, occupying the rocky summit, is surrounded at a considerable distance by an outer line of ramparts. The summit of the knoll is girt by the remains of a rampart or wall which encloses an area measuring about 300 ft (91.4m) by 150 ft (45.7m), and the base of the knoll is in turn defended by a rampart thrown up from an internal quarry ditch. The combined work, measuring over all about 500 ft (152.4m) by 300 ft (91.4m) lies within the outer pair of ramparts with a median ditch which enclose an area over 800 ft (244m) in length, which includes a well.

R W Feachem 1963.

Artifacts from the Castle O'er area, which are in Dumfries Museum, include seven melon beads, several paste and jet beads and discs, part of a glass finger-ring, a flint blade, a bead of Antrim bauxite, and a number of Iron Age spindle whorls.

A E Truckell 1964

This complex multivallate fort can be divided into four component parts.

A - The innermost enclosure, measures 97.0m by 56.0m within the much reduced remains of an earth-and-stone bank. Two entrances are visible, on the NE and SW. Four timber house sites were located in the interior, the best preserved being a double ring- groove type 10.6m in overall diameter with slots 1.3m apart.

B - Surrounding 'A' are the main defences, consisting of double ramparts and medial ditch on the N and W, and on the S and E a single rampart with an internal quarry ditch. The two entrances are nearly aligned with those of 'A'. The remains of a wall, now only consisting of large outer facing stones, flanks the SW entrance.

C - Around the base of the knoll on the S and E, a bank with an internal ditch encloses an area of 210.0m, by 35.0m, including a spring on the E. D - A similar stretch of unattached bank and internal ditch encloses an open area to the SW of the fort.

It is probable that 'A' pre-dated 'B', due to the former's poor condition. 'C' is not attached to 'B', and its open entrances on the NE and SW, and internal ditch, indicate a primarily pastoral use. 'D' also seems to be a pastoral enclosure and is probably contemporary with 'C'. A hollow-way runs from the fort on the NE down to the Brunt Sike. To the SW a length of bank continues as a hollow-way for about 200.00m before being lost in forestry. Its ditch cuts through the bank of enclosure 'C' and it possibly represents a later boundary work.

Revised at 25".

Visited by OS (DWR) 22 October 1973

Castle O'er fort is as described; the finds are in Dumfries Museum. Surveyed at 1:10 000.

Visited by OS (BS) 17 October 1978

Castle O'er: This multi-period fort is situated on a knoll in a clearing in a forestry plantation. In the earliest phase it was defended by twin ramparts and a medial ditch which enclosed an area measuring 120 by 60m. Subsequently the inner rampart was replaced by a wall, reducing the enclosed area to 95 by 53m. There were entrances on the E and SW in both phases. In the interior, and probably contemporary with the second phase defences, there are at least three ring-groove houses (one of which has double grooves) and several possible house-platforms. A crescentic bank of bedrock NE of the SW entrance through the wall may be a result of the quarrying that has disturbed the W part of the interior.

On the S of the fort, and probably contemporary with the ramparts, there are two annexes, each formed by a bank accompanied by an internal ditch and slight inner bank.

RCAHMS 1980, visited August 1980.

Resurvey of this complex site has added significant detail to the existing record: a phase of refortification can be identified and at least some of the timber round houses on the site can be shown to postdate the defences.

Two major phases of defensive earthworks enclose the summit of Castle O'er, the earlier being a large oval fort defended by twin ramparts and a medial ditch; this was superseded by a smaller, oval fort, whose single wall follows the line of the earlier inner rampart along the N and S sides of the summit. At the W end, however, the rampart of the earlier fort appears to have been refurbished to create a forework around the entrance of the later fort. A similar feature may exist at the E end, but here the defences are even more dilapidated and have been obscured by the construction of timber houses. The outworks are an addition to the earliest circuit of defences, but their relationship to the later fort cannot be established from surface evidence alone.

The stances of at least twenty-nine timber round houses representing several phases of construction, can be identified within the interior of the fort. The character of the stances range from simple platforms to structures of single or double ring-groove type, and some of these have also been built on, or enclose platforms. At the eastern end of the fort a further four houses have been built on a terrace between the earlier and later lines of defence; these houses not only overlie the rampart of the earliest fort, but two of them, the northernmost and the southernmost, have also been cut back into the outer face of the wall of the later fort, and the former underlies the short length of bank which links the earthworks of the two forts on the NE. Ground survey has clearly demonstrated that multiphase occupation of Castle Oe'r continued after the hill ceased to to have any defensive significance. It cannot show, however, whether any of the surviving structures within the interior of the fort are contemporary with any of the phases of fortification.

Visited by RCAHMS (JRS), 16 April 1993.

Listed as forts, palisade and whorls ('Castle O'er and area').

RCAHMS 1997.

Scheduled (with NY29SW 32) as 'Castle O'er, fort and linear earthworks...'

Information from Histopric Scotland, scheduling document dated 13 March 2010.

Castle O'er

Fort [NAT]

OS (GIS) MasterMap, April 2010.


Publication Account (1986)

Keeping at bay a besieging force comprising massed ranks of conifers, the most remarkable of the iron-age hillforts in Eskdale still displays an impressive show of strength. It also continues to pose problems of interpretation as to the nature and sequence of its defences, the numerous linear earthworks in its vicinity, and of the domestic settlement contained within its ramparts.

The earliest phase is probably represented by a half oval summit area which was defended by a pair of ramparts and intermediate ditch. This enclosure measures 120m by 60m, and has always been entered from the east and from the south-west. It was later reduced to corresponding dimensions of 95m and 35m upon the construction of a drystone wall in place of the inner rampart; the nine circular house stances visible in the interior are probably contemporary with this phase. Two outer annexes to the south of the fort are enclosed by a system of banks and medial ditches, the outer banks being more pronounced. No credence is now given to the belief that this outer perimeter may have been a Roman siegework.

Hollow-ways run south and east from these outworks, and part of a third linear earthwork survives just to the west of the fort. These form part of an extensive system of/trenches' recorded in 1896; many appear to have been old tracks and are now absorbed within forestry plantation.

Information from ‘Exploring Scotland’s Heritage: Dumfries and Galloway’, (1986).

Ground Survey (21 October 2013 - 24 October 2013)

NY 24190 92850 A topographical survey was carried out, 21–24 October 2013, of the surviving earthworks of Castle O’er fort. The earthworks consist of a multi-phase, late prehistoric, multi-vallate fort and associated annexes and linear features interpreted as an enclosed settlement with boundaries forming part of a more extensive land management system. Information provided by the survey was used to create a digital terrain model, contour-based plans and oblique terrain models of the site in its landscape setting, which will be used to inform future conservation and management of the monument.

Archive: RCAHMS

Funder: Forestry Commission Scotland

Louise Baker and Enda O'Flaherty - Rubicon Heritage Services Ltd

(Source: DES)

OASIS ID: rubiconh1-188154

Note (1 July 2014 - 25 November 2016)

This fort, which crowns the summit of Castle O'er Hill, displays evidence of at least two major phases of fortification and has a large annexe on its SE side. In its first phase it was an irregular oval on plan, measuring a maximum of 125m from ENE to WSW by 60m transversely (0.6ha) within twin ramparts with a medial ditch, which have been deployed to enhance the natural topography, particularly on the precipitous SE flank, where the inner rampart follows the crest of the slope, and the ditch and outer rampart the foot, and even the apparently natural sectors of the escarpment between them have probably been modified and steepened. In its second phase, the core of the interior was reduced to an area measuring about 105m from ENE to WSW by 52m transversely (0.4ha) within a single rampart, though the earlier defences at the WSW end were also adapted to form a hornwork around the entrance; there is no evidence of a similar embellishment of the second entrance at the ENE end, but this has been mutilated by the construction of stances for later round-houses, and indeed more recent stone-robbing. No fewer than 29 round-house stances can be seen within the interior, some represented by levelled platforms, others by ring-grooves, in one of the latter comprising no less than three concentric lines. Some of these stances intercut and represent a sequence of occupation, while the platforms apparently overlying the ramparts at the ENE end suggest a final phase of open settlement. The annexe is probably an addition to the fort, enclosing an area of 0.5ha within twin banks with a medial ditch up to 6m in breadth; while an entrance has been left between the annexe and the fort defences at the WSW end, on the ENE its inner bank appears to ride up onto the outer rampart of the phase 1 fort. In plan, it is clearly conceived in conjunction with the linear earthworks that run up to the fort from the N and E, while an evaluation trench dug about 1985 by Roger Mercer demonstrated that a third linear earthwork on the S was constructed after the annexe. Five radiocarbon dates from this and other trenches relate to occupation and use of the fort and annexe during the Roman Iron Age.

Information from An Atlas of Hillforts of Great Britain and Ireland – 25 November 2016. Atlas of Hillforts SC1103


MyCanmore Image Contributions

Contribute an Image

MyCanmore Text Contributions