Font Size

100% 150% 200%

Background Colour

Default Contrast
Close Reset

In recognition of the essential restrictions and measures imposed by the Scottish and UK Governments, we have closed all sites, depots and offices, including the HES Archives and Library, with immediate effect. Read our latest statement on Coronavirus (COVID-19).

Galashiels To Rink Hill

Linear Earthwork (Period Unassigned)

Site Name Galashiels To Rink Hill

Classification Linear Earthwork (Period Unassigned)

Canmore ID 110753

Site Number NT43SE 39

NGR NT 4798 3495

NGR Description From NT 4798 3495 to NT 4758 3220

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 Caddonfoot
  • Former Region Borders
  • Former District Ettrick And Lauderdale
  • Former County Selkirkshire

Archaeology Notes

NT43SE 39 from 4798 3495 to 4758 3220. LIN 11.

Formerly LIN 528.

(NT 4800 3509 - NT 4798 3500) Earthwork [NR]

(NT 4798 3495 - NT 4792 3468) Earthwork [NR]

(NT 4792 3468 - NT 4789 3390) Earthwork [NR] (course of) [NAT]

(NT 4789 3390 - NT 4798 3367) Earthwork [NR]

(NT 4798 3367 - NT 4824 3290) Earthwork [NR] (course of) [NAT]

(NT 4807 3254 - NT 4791 3249) Earthwork [NR]

OS 1:10000 map, 1993

South of Stannis Burn, however, for just over a quarter of a mile, the work is in better preservation, (It deteriorated noticeably, as a result of deep tractor ploughing, between the visits paid in 1945 and 1946.), and in fact it is only in this section that the ditch and bank can be properly identified and measured (see NT43NE 50 from NT 4800 3510 to NT 4792 3469)). Though wasted at its lower end almost to the condition of a terrace, some 26ft wide over all, as it mounts the field towards a wood (NT 480 349) the work clearly shows its character as a ditch 9ft wide and very shallow, with a bank on its W side which is spread to a breadth of 14ft. Its alignment is typically irregular. Within the wood any possible confusion of this part of the work with a road is finally disposed of by the manner in which it ascends a steep and rather rocky pitch; no road would have followed such a course, and moreover, just below the ascent, the ditch ends and then immediately reappears about its own breadth to one side. Above the steep pitch the earthwork is overridden first by the hollow tracks of the old road from Fairnilee to Galashiels (RCAHMS 1957, No.94), and then for a short distance by the E boundary-wall of the wood; S of its point of re-emergence, and while still within the wood, the ditch is from 10ft to 13ft in width by up to 3ft 6ins in depth, while the bank is too much wasted for measurement.

On leaving the wood the earthwork quickly fades out, and no trace of it can be seen until after it has passed the site of the vanished earthwork (NT43SE 6) on the crest of the shoulder, which it seems to have incorporated in its line. Thence, however, for some 500 yds to the SE corner of the field (NT 479 340) sufficient indications remain to prove the correctness of the line as shown on the OS map. A terrace up to 20ft wide over all can be followed for much of the distance. Along the E margin of the adjoining field the work is obscured by a modern farm-track which runs along its course, but it again becomes visible as a ditch and bank in the wood (at NT 4790 3391 to NT 4798 3367) above (i.e. E of) Hollybush (NT43SE 52). For 185 yds from the NE end of this wood (NT 479 338) it is generally in fair preservation, the ditch being up to 16ft wide by 4ft deep and the bank, which is here on the W, being 14ft thick by 2ft high. At NT 480 337 it is crossed obliquely and destroyed by an old road and from here onwards for some 400 yds the 1st edition of the OS 6-inch map (Selkirkshire, 1863) marks the "site of" a second work duplicating the existing one at a rather lower level; it is possible that the surveyors may have mistaken this road for a duplicate earthwork, perhaps under Kemp's influence (Kemp 1877). Southwards from NT 480 337 the earthwork runs along the E edge of the wood - at first as a terrace about 12ft wide flanked by a stone wall on the E, to which side, being downhill, the bank appears to have changed, then progressively narrowing with the gradual encroachment of the wall, and finally lost where the wall begins to curve towards the SW (NT 480 335). It is possible that this section may at some time have been used by traffic; but if so this use can only have been incidental, and there can be no doubt that we are here still dealing with a linear earthwork.

The whole of the rest of the work, to its end beyond Rink Hill (NT 4791 3250), has been severely damaged, and in places obliterated, by ploughing, but air-photographs (visible on RAF vertical air photographs, RAF 106G/Scot/UK 142, 2067-8 and 5067-8, flown 1947), fortunately supply what cannot be seen on the ground and correct the misleading records of past observers. Thus, reinforcing ground observation with the photographs, we find that the earthwork leaves the corner of the wood on a course which runs fairly straight towards the SSE; it does not run W of S and then make a turn, as shown on the OS map. About 100 yds from the wood it becomes faintly visible descending into the small steep-sided valley in which Kemp (1877) placed his "double line of defence", the ditch appearing clearly at the bottom and measuring 15ft in width. The bank is no more than a trace on the E side of the ditch, but it too appears clearly at the wall on the parish boundary some 40 yds farther on. The next 300 yds carry the work up the opposite side of the valley, where it appears as a hollow up to 20ft in width; but that this is still the earthwork, and not an old road, is proved by the appearance on the photograph of a clearly marked angle, and not a gradual curve, where the alignment changes slightly at the crest of the slope.

On passing the wall that joins NT43SE 9 and NT43SE 7 (RCAHMS 1957, Nos. 121 and 122), (NT 483 329), the work dies out on a course running due S; but the air-photograph again shows that it swings SSW round the side of Rink Hill, then turns nearly W - another angular turn - at the wall that joins NT43SE 7 (RCAHMS 1957, no.122) to Rink Farm (NT 4826 3241), and so runs into the small strip of trees and bushes due S of NT43SE 7 (NT 480 325). Within the strip, and running along its whole length of rather less than 200 yds, there now appears little beyond a linear hollow or terrace, excavated along the face of the steep hillside and partly filled up with land-stones and rushed soil. Where measured, the terrace was 8ft wide, with a steep scarp 8ft high on its upper side and a bank 9ft thick by 2ft high on its lower one. On the whole this fragment is more like a road than an earthwork, but its continuity with the rest is not in doubt and its appearance may well be due to secondary use by traffic. That a road did once pass along it is strongly suggested by the alternative line of 'Catrail or Picts' Work Ditch' that is marked on the OS 1st edition map as emerging from the E end of the strip; this line swings round the E side of Rink Hill at a lower level than the earthwork, to vanish after passing close to NT43SE 9, and it is most easily explicable as an old road to Galashiels.

West of the strip the earthwork has again been obliterated, but air-photographs (visible on vertical air photographs RAF 106G/Scot/UK 142, 5067-8, flown 1947), show it as running W for about 100 yds, turning at an angle - again as no road would have turned - and continuing SW to the point on the Galashiels-Peebles highway to which the OS 1st edition map erroneously takes it direct. However, the strip of trees through which this piece of the highway runs has again preserved the earthwork - on the upper side a fragment only 20ft long, but on the lower a terrace about 10ft wide which descends the steep slope obliquely to the point shown on the OS map (NT 476 324).

[But for its close association with the rest of the work, combined, perhaps, with its straightness, this section might well be taken for a terraced track, and a connection looked for with the ford by the islands at Howden Pot (NT 4747 3224) (cf Lynn 1898; 69). It is, in fact, very similar to another oblique terrace in an analogous situation due S of Rink farmhouse (NT 484 322); and this latter is almost certainly an older version of the stretch of modern farm-road below which it runs, although Kemp (loc.cit.) appears to regard it as part of a linear earthwork connected with 'the Catrail' and decending from Rink Fort (NT43SE 7). If this old road once led to the ford at Howden Pot, it would no doubt have helped to create the impression that the earthwork did the same, as stated by Lynn.]

At the lower edge of the plantation the work is finally obliterated, and nothing can be seen either on the ground or on the photographs to show whether or how it ran on to the Tweed. Some very faint traces of what may be a filled-up ditch suggest, however, that it may have taken a slightly more easterly line than the one shown on the map, to end just opposite the mouth of the Howdenpot Burn.

Visited by RCAHMS 17 November 1945; 14, 17, 18 December 1946; 17 October 1947

W Kemp 1877; F Lynn 1898; RCAHMS 1957

NT 4800 3509 - NT 4798 3489. From a fence junction some 14m S of Stannis Burn, this earthwork can be seen on the hill slope to the S. This section appears more in the form of a terrace c.4.0m broad. If it ever was a bank and ditch, the former would have been on the W (lower) side. A little more than halfway up this slope the feature becomes more pronounced as a ditch or hollow-way.

NT 4798 3489 - NT 4792 3468. In the wood, there is no trace at first for about 40m; then, on a fairly steep slope, the ditch with bank on the W side appears. The bank soon ends as the crest of the slope appears, but the ditch continues for a further 20m. From here to a footpath which cuts across the feature, there are no definite traces. Beyond the footpath the ditch is well defined (c.5.0m broad by 1.0 deep), with an accompanying bank on the E side. This section is strongly reminiscent of a hollow-way.

NT 4792 3468 - NT 4781 3444. No definite traces can be found in a ploughed field, except for a short stretch of terrace, c.15m long against the dyke at the S end of the field.

NT 4781 3444 - NT 4790 3405. The earthwork appears as a terrace across an arable field for c.140m, and this is shortly mutilated as it enters a newly-ploughed portion of the field, but for c.80m before the dyke bounding the S side of the field is reached, a faint hollow is discernable on the hillslope.

NT 4790 3405 - NT 4789 3398. No trace; cart-track occupies course.

NT 4790 3398 - NT 4789 3390. No trace of the earthwork, cart-track occupies course.

NT 4789 3390 - NT 4798 3367. A well-defined hollow (c.6.0m broad by 1.0m deep) through a wood. In one section a bank accompanies the ditch for c.35m, on the W side. Some 80m beyond a causwayed track cutting across the ditch, the feature disappears entirely merging with the plantation dyke.

NT 4798 3369 - NT 4814 3323. There are no vestiges of the earthwork on a very steep slope on the line of the plantation dyke as shown on the OS map, and no traces of the course indicated by Auth 2 and AP's can be found on the steep SE slopes of Mount Skep.

NT 4814 3323 - NT 4824 3291. On the S side of a dyke at the foot of the slope a slight hollow c6m broad by 0.3m deep is found running for c30m southwards. Beyond this and up the opposite hillside, only faint indications of the hollow (c8m broad) are visible---with a noted curve to the SW near the crest at NT 4824 3301.

NT 4824 3291 - NT 4802 3255. No traces are visible on the slope of the steep hill above Rink Farm.

NT 4802 3255 - NT 4791 3249. A short way in a small strip of wood, the hollow-way resumes in a well-defined state being c.6-8m broad and 1-2m deep, with a strong bank on the lower (SE) side. Just before the earthwork leaves the wood, it turns quite sharply towards the S.

NT 4791 3249 - NT 4771 3241. All traces of the earthwork are lost in a ploughed field.

NT 4771 3241 - NT 4767 3239. On the S side of the road, there is a steep descent to the haugh below, and in the felled plantation at this point, there is a slight terrace, but this has almost disappeared due to soil slip.

NT 4767 3239 - NT 4751 3224. No trace across a ploughed field. (This course and that of the two preceding paragraphs, is shown on the 1st edition of the OS 6-inch map (Selkirkshire, 1863), sheet 8, and described as [The Catrail or Pictsworkditch] (site of). Another similarly-described courses lie SW and S of Hollybush farm: NT 4798 3372 - NT 4806 3333 and NT 4837 3329 - NT 4806 3256. No vestiges of these remain in arable fields; but it is doubtful if these ever did exist.

Visited by OS (WDJ) 19 January 1961.

In 1480, the ranger of yarrow got 2000 'vangorum et tribulorum' i.e. spades and forks. These may have been used for making ditches and banks to form the bounds of the stedes in Ettrick Forest, bounds which still seem to be shown on the OS maps. That they were used for planting is doubtful.

M L Anderson 1967


Sbc Note

Visibility: This is an upstanding earthwork or monument.

Information from Scottish Borders Council


MyCanmore Image Contributions

Contribute an Image

MyCanmore Text Contributions