C.J Tabraham's failure to find any trace of the Deil's Dyke in the parish of New Cumnock in 1981 suggests that further destruction of the Dyke may have taken place since Graham and Feachem's assessment of the Dyke in 1956. However, armed with the ordnance survey map co-ordinates published by Graham and Feachem it is relatively straightforward to pick up the trail of the Deil's Dyke in New Cumnock as it heads up the west-facing slope of Dalhanna Hill.

The Dyke shoots up the hillside as straight as an arrow, like a swollen grassy scar, leading directly to the summit of the hill. Along the way, parts of the Dyke have been severely damaged, providing in some instances Gaussian-like cross-sections, which expose several large stones embedded across the core. At stretches where the Dyke has been obliterated, stones and boulders are often found breaking through the surface. There is no evidence of the course of any burns or sikes running adjacent to the Dyke, nor of any significant ditch either side of the dyke. Close to the summit of Dalhanna Hill the Dyke heads due east, towards Redree Burn in the distance, but soon all trace is lost, as ably illustrated in the RAF Aerial photograph (1946).

As to the function of the stretch of Dyke on Dalhanna Hill, it is from first impressions less problematic to suggest what it wasn't designed to be. It certainly evokes no sense of an ancient defensive earthen rampart, indeed the nearby dry-stane dykes twinned with barbed-wire fences appear Hadrianic-like in comparison. Parallels can be drawn with Sir Herbert Maxwell's assessment of the defensive qualities of the Catrail, the Deil's Dyke equivalent in Roxburghshire, 'If that was the origin and purpose of the Catrail, it is to be hoped that the Britons were better Christians than they were military engineers' (in G.V. Irving's notes on the Devil's Dyke).

With regards to the local tradition, preserved by Thomas Kirkland, that the Dyke was used to hide cattle from Border raiders, then this too can be discounted. Cows and their calves (and a big white bull!) freely roam the slopes of Dalhanna Hill, and the Dyke offers no barrier to their wanderings. Kirkland's tale may or may not have had its roots in those observations made by George Irving about the Deil's Dyke in Dumfriesshire. He suggested that the Dyke was erected to frustrate foragers returning with their booty of cattle, sheep and horses, thereby giving the plundered owners time to raise a posse and recover the stolen beasts.  






.Graham and R.W.Feachem
The work of A. Graham and R.W.Feachem [G&F] is now revisited and assessed for its relevance to the Deil's Dyke in the parish of New Cumnock. In summary they suggested that the Deil's Dyke proper from the parish of Durisdeer in Dumfriesshire to the parish of New Cumnock could be considered as a unitary earthwork (as opposed a discrete series of earthworks). They discounted both the medieval dynasties of Douglas and Crichton as the possible 'single authority' for erecting the dyke, electing instead Strathclyde Britons of the Dark Ages.

Douglas 14th century:
The Douglas hold on the lands of the Deil's Dyke proper was not as far-reaching as that suggested by [G&F]. The Earls of Dunbar or a cadet branch of that family, as Barons of Cumnock, held the lands of New Cumnock (which once formed part of the Barony of Cumnock) from the 13th century (and probably before then) until the early 17th century. Although the source of the River Nith is in the parish of New Cumnock, Ayrshire, these lands have never formed part of that political unit known as Nithsdale, the latter referring only to lands in Dumfriesshire. So neither the 3rd Earl of Douglas or his son William, Lord of Nithsdale were overlords of New Cumnock. Indeed when in 1369, the 3rd Earl of Douglas acquired 'all the lands in Galloway between the River Cree and the River Nith' (i.e. those lands in that part of Galloway that had reverted to the crown), George Dunbar, 10th Earl of Dunbar had a royal charter (25th July 1368) for the lands of Cumnock (Ayrshire), Blantyre (Lanarkshire) and in Galloway at Glenken (Kirkcudbrightshire) and Mochrum (Wigtonshire).  

Strathclyde Britons
In any case, [G&F] dismissed both families as the single authority responsible for erecting the Deil's Dyke proper on the grounds of lack of motive for building a boundary on land that they owned . Instead they offered a Dark Age solution, and visualised the dyke as a political statement by Strathclyde Britons (from Douglasdale or upper Clyde) ' asserting rights in this region against Northumbrian invaders from southern Dumfriesshire or Galloway'. Why the Britons of Douglasdale and upper Clyde were considered in favour of the Britons of Kyle is unclear. Certainly by the 13th century these lands were in that part of the Sheriffdom of Ayr, known as Kyle.

Under Eadberht, the Northumbrians successfully annexed the plain of Kyle in 752 AD, and evidence of Anglian settlement in New Cumnock is found in the form of a 9th century Anglian cross discovered near Mansfield farm, on the north bank of the River Nith. Nearby Corsencon hill* (Corswintoun, 1488) appears to contain the Anglian place-name element Old English swin-tun 'pig-farm' (pre-fixed by the later Gaelic cor 'rounded- hill') indicative of the advance made by the Northumbrian invaders up through Nithsdale. An advance witnessed by the discovery of several other Anglian crosses in Dumfriesshire and that county's Corsencon place-name equivalent of Dalswinton.

Consequently, a case could be made for considering the Deil's Dyke proper as the political equivalent of the Catrail earthwork, as opposed to the converse as suggested by [G&F]. That is to say the Deil's Dyke proper is an assertion of Anglian advance into Nithsdale and not a British statement of territorial claim.

* Corsencon hill - It also worth recalling Thomas Kirkland's reference to the sod fences on Corsincon, which can be seen to this day and will be discussed later.


J. Barber and C. J. Tabraham

Medieval Forest Management

Turning now to the work of J.Barber, I.D.Mate and in particular the observarions of C.J. Tabraham. Central to C.J.Tabraham's suggestion that the Deil's Dyke ' may be a tangible sign of medieval forest management' was his understanding that in the early 13th century Walter Steward held three forests, i.e. at Renfrew, Ayr and Sanquhar. Renfrew was indeed a Steward stronghold as was that part of Ayrshire known as Kyle Stewart, but not so the lands of Sanquhar (or Kirkconnel) in Dumfriesshire. The Stewart's foresta de Senecastre is in fact a reference to Sanquhar in the parish of St. Quivox, Ayrshire, which is found in the aforementioned lands of Kyle Stewart. James Paterson in 'Histories of Counties of Ayr and Wigton' writes 'In a charter of Walter the Steward to the Monks of Paisley, about 1208, liberty is granted to them to take wood for burning out of his forest of Seanccathre (Sanquhar?) in Kyle at sight of his forester.'

  (N.B. this is not to say that there was not at one time a deer-park at Sanquhar, Dumfriesshire for indeed there once was in the vicinity of the Sanquhar Castle.)  

Nevertheless Tabraham's observations are still worthy of consideration, for in the words of G.W.S. Barrow 'Deer hunting has left physical traces on the landscape in the form of banks and dykes used either as park boundaries or for trapping the deer'. John M.Gilbert in 'A Historical Atlas of Scotland' provides an excellent overview of the development of hunting reserves in Scotland which were introduced into Scotland by David I (1124-1153) in the 12th century. As well as royal reserves the king initiated the concept of baronial forests, whereby through a royal grant barons were allowed to have and rule over their own forests, where the word 'forest' meant hunting reserve regardless of whether it was a wooded area or not. Hunting rights were reserved for greater game like deer and boar and warren rights for lesser game such as fox, rabbit, wild cat and hare. Hawking and fishing were also subject to control. So too were grazing rights on the forest lands, such as the right of pannage, the grazing of pigs during the acorn drop . Gilbert goes on to explain that barons could create parks without a royal grant.

Earls of Dunbar
The Earls of Dunbar were the Barons of Cumnock possibly from as early as the 12th century and certainly throughout the 13th and 14th centuries. The title passed on to a cadet branch, the Dunbars of Cumnock, who continued to hold the barony until the early 17th century. The baronial seat was Cumnock Castle situtated strategically at the confluence of the Afton Water and the River Nith, at the heart of the modern day village of New Cumnock . Although no historical records survive of a royal grant pertaining to a baronial forest at Cumnock (foresta de Comenagh ?) there is a tenous link with the Dunbars and hunting activities in this part of the Southern Uplands.

Patrick, 8th Earl of Dunbar (b.1242 - d. 1308), a contemporary of James Steward, played host to Edward II of England at Cumnock Castle for eight days in August, 1307. During his stay Edward II held court, at which time, ' To a poor woman of Crathgork near Stirling, keeping 2 greyhound whelps of the king for 1 year and bringing them to him at Cumnock, for the expenses of the woman and the dogs, by the king's gifts at Cumnock, on 27 Aug;, 40s.' These greyhound whelps are not to be mistaken for today's breed of greyhound that hare round the dog-track at nearby Auchinleck. They were bred for chasing bigger game and in particular deer, as witnessed by their alternative name of the Highland deer-hound. Edward II was extremely passionate about hunting and was a frequent visitor to the numerous royal forests, reserved for the king's hunting, throughout England. Perhaps Edward II took to Dunbar's baronial forest at Cumnock to feed his passion during his eight-day stay, although Dunbar may not have had the pleasure of his company, since 'The king often chose not his courtiers but low born people to share the excitements of the chase' . The bringing of the whelps to Cumnock Castle may also form the basis of John Barbour's account of John of Lorne hunting Robert the Bruce with dogs in the Cumnock hills. It is to be expected that Edward II was more successful in a-hunting the deer than catching the Bruce!

Crawford of Cumnock
Another tenuous yet stronger association with medieval forest management in the lands of New Cumnock is found in the records of the Crawford family. The Craufurds held the lands of Dalleagles and Beoch in the west of the original parish of Cumnock. One of their number Roger de Craufurde was a contemporary of the aforementioned Patrick Dunbar , 8th Earl of Dunbar and in turn his descendant Sir John Crawford of Cumnock was a contemporary of George Dunbar, 10th Earl of Dunbar. As discussed above, George Dunbar also held lands in Galloway including Glenken in Kirkcudbrightshire, which he resigned to David II, following the king's release from English captivity in 1357. Andrew McCulloch in Galloway, 'A Land Apart' explains these lands included Robert the Bruce's (David's father) hunting reserve which had fallen into a state of disrepair. In 1358, David II appointed  John Crawford of Cumnock its forester or keeper. It was not uncommon that such an office was hereditary and perhaps the Crawfords of Cumnock had held a similar position for the Earls of Dunbar and their baronial forest of Cumnock.






Left : Scotch greyhound . Right The date-stone of Hugh Douglas and Margaret Crawford on Knockshinnoch Farm , New Cumnock showing the Crawford stag

The Forest of Cumnock ' foresta de Comenagh'

Tabraham's suggestion that the Deils's Dyke represents a tangible sign of medieval forest management presents a possible fit with the purpose of the section of the dyke in the parish of New Cumnock. Did the dyke represent one of the boundaries of the Dunbar's baronial forest  or park? A forest managed by Crawford of Dalleagles (Cumnock), and a forest in which Edward II engaged in his favourite sport of deer-hunting during a week long stay at the Cumnock Castle in the summer of 1307. 







If so what were the other boundaries of the forest or park of Cumnock? The Deil's Dyke enters the parish of New Cumnock  at the March Burn ('boundary burn'). This burn takes its name from the boundary of the sheriffdoms of Ayr and Dumfries (later Ayrshire - Dumfriesshire), and not from a forest or park boundary. However the Park Burn runs parallel to the March Burn and this may have marked the eastern boundary of the park. In 1897 fragments of a 10th century cross of Scandinavian influence, were found by the Park Burn, at Overcairn on the estate of the Marquis of Bute (and Baron of Cumnock). Could this cross once have by the Dyke? The Deil's Dyke terminates at the Afton Water which would have provided a natural boundary, i.e. the western boundary, whereas the River Nith to the north , may too have formed a natural boundary.


In the aforementioned Steward heartlands of Renfrew, near Barrhead are the Ferenze hills , a name which G.W.S. Barrow in 'The Uses of the Place-Names' explains as 'a misunderstanding of the Old French word fermeson 'close season', a time at which the monks of Paisley abbey were 'allowed to take hinds and not stags from the Forest of Raiss ' (possibly Gaelic ros 'woods'). As such a survey of the place-names in and around the suggested boundaries of the forest of Cumnock may prove useful.

Accepting the Park Burn as the eastern boundary , it is perhaps no coincidence that at the western boundary, where the Deil's Dyke meets the Afton Water, once stood the lands of Parck, now represented by the farms of West Park, formerly the ferme-tounes of Laigh Park or Nether Parck; and High Park formerly Over Parck. Here,another Park Burn drains off Dalhanna hill and flows gently into the Afton.

The modern day farm of Blackwood and its predecessors the ferme-tounes of Nether and Over Blackwood (now Burnton) doubtless owe their names to a black or dark woodland that once covered the lower northern facing slopes of The Knipes. The baron could charge a toll to permit others to cut timber from the wood for fuel, or to charge 'pannange' for the grazing of pigs in his woods during the acorn crop in autumn, or grant the right of warren (lesser game) in the woods or his forest in general. By way of example G.W.S Barrow in 'The Acts of William I' uncovers a grant to the monks of Coldingham of ' warren anent the possession of 'woods and waste', where waste is a reference to waste-land. Interestingly the ferme-tounes of Westland and Nether Westland ( now Meikle Westland and Waistland ), names all of which may represent a corruption of wasteland, lie adjacent to the lands of Blackwood. 







Burnton,  i.e. the burn-toune (formerly the merkland of Over Blackwood) has given its name to the burn it stands beside, i.e. Burnton Burn. On modern Ordnance Survey maps a tributary of this burn goes by the name of Elder burn. If this burn takes its name from elder bushes that adorned its banks , the Scots bourtree or Gaelic craobh fhearna would have been expected to feature in the name. Could elder  be a local corruption of Gaelic eilid 'hind', eildeag 'a young hind, roe' or eildeach 'abounding in hinds'? Probably not, since in any case the name Elder Burn is of questionable antiquity and the common place-name element in this context is elrig (Gaelic eileirig 'deer-trap').

To the west of Burnton is the triumvirate of Polquhirter farms, i.e. East , West and High  (formerly the ferme-tounes of Nether and Over Pohwhyrtyr) . These farms take their name from Polquhirter* Burn, as witnessed by the first element Old British  pol 'burn, stream'. The meaning of the second element is less clear and is probably associated with the nature of the stream e.g. Welsh carthu , Gaelic cairt 'cleanse', or Gaelic cartair 'dance'. Other suggestions include Gaelic  cairtear ' carter waggoner' (although this is pronounced carster), perhaps a carter of peats or timber, both of which suggest economic activity wtihin the Baron's park. 

To the west of Polquhirter, on the outskirts of the modern-day village of New Cumnock is the hillside known locally as The Coupla with Couplegate nearby. i.e. the road or way to the Coupla. The Scots coup 'tip, a place for emptying cartloads' comes to mind, cartloads from Polquhiter perhaps? However Coupla or Couplaw is possibly 'a bargaining hill' from the Scots coup 'to exchange, barter' and Scots law 'roundish hill'. (Cf. Couplaw near Kirkmuirhill, Lanarkshire, which coincidentally sits off Dykehead Road; Copin Knowe near Minigaff which Sir Herbert Maxwell gives as 'bargaining hillock'.)  Further evidence perhaps of economic activity within the confines of the park.  Adjacent to The Coupla is Stellhead Rig which appears to contain the Scots element stell ' enclosure for cattle or sheep'. Alternatively, stellhead could be a local corruption of the Scots stellage 'the ground on which a market is held', situated conventiently on the bargaining hill. 

Cairn Hill sits between Stellhead Rig and the Afton Water and since there are no records of ancient burial cairns associated with this site, it is perhaps a reference to a clearance cairn, i.e. a pile of stones formed by clearing the hillside in preparation for agricultural purposes. Other names indicative of land-clearing are found in the vicinity of the Deil's Dyke. Burnt Hill across which the dyke traverses, is doubtless land that was at one time cleared by burning. The adjacent Redree Hill may be a form of place-name tautology comprising two elements with the same meaning i.e. Scots rid, redd and Gaelic reidh 'clearance'. 

Unfortunately this brief survey of the place-names in and around the Deil's' Dyke has failed to unearth a place-name of the same significance (with respect to supporting the presence of a medieval baronial forest) as that of the aforementioned Fereneze in the Stewart forest of Renfrew. Neither has it been possible to offer a chronological development of these names. Of course,  those place-names assessed above have been scrutinised in an attempt to support the presence of a park or forest and it is to be expected that other more or less obvious explanations for these names may exist. Before leaving the place-name study, A. Graham's assessment of the the name Deil's Dyke is worthy or recording. 









It would be splendid to conclude that the Deil's Dyke in the parish of New Cumnock once marked the boundary of the medieval baronial forest or park of Cumnock. A place where the Earls of Dunbars, when based at Cumnock Castle enjoyed the thrill of the chase, hunting deer and boar with hounds across the lower slopes of The Knipes. Perhaps Edward II of England during his week long stay at Cumnock Castle indulged in his favourite pastime (no not that one!) , putting his two newly acquired greyhound whelps through their paces for the first time. The management of the forest was under the control of the Crawfords of Cumnock who held land at Dalleagles at the other end of the parish. Anyone found hunting within the forest, effectively atwixt the Deil's Dyke and the River Nith, without the baron's permission could face heavy fines, administered by the Crawfords at the baron's court. As time passed the barons of Cumnock came under greater pressure to accommodate economic activity within the confines of their reserve and much of the land was cleared for agricultural purposes, The trees of the Blackwood felled for fuel and timber and peats were cut on the hillsides and carted to the market on the Coupla, where lesser game such as rabbit and hare were also bargained for. The Deil's Dyke itself would eventually serve as a head-dyke , a form of economic boundary which divided outfield from common grazing.
Indeed it would be splendid to reach such conclusions. In doing so it is important to recall C.J. Tabraham's note of caution that his observations on the dyke as ' a tangible sign on medieval forest management' were based on surmise. Needless to say my own research, built chiefly on Tabraham's assessment, adds to these surmises!

If we recall J. Barber's study of the stretch of Deil's Dyke in the neighbouring parish of Kirkconnel, then it may well overlay an earlier Iron Age earthwork, the work of the Novantae tribe perhaps, a tribe that took their name from the River Nith. A man-made territorial earthwork raised by a single authority, i.e. the Novantae, to mirror the great natural boundary of the River Nith. The later linear earthworks appear to be the equivalents of modern-day drystane dykes. They look the same from field to field, from parish to parish and from county to county. Erected to have similar functions, .ie. to demarcate one piece of land from another and raised at the behest of the landowners of the the time and not by a single authority. Today drystane dykes follow much of the course of the Deil's Dyke, helping to emphasise the contrast of the green-yellow coloured lower slopes of the Knipes and the reddish-brown upper slopes (see below).









Other Earthworks in the Parish of New Cumnock

The study thus far has focussed on the earthwork known as Deil's Dyke part of which traverses Dalhanna Hill and therefore only partially addresses Thomas Kirkland's question ' What is the story of the of the sod fences or embankments to be seen on Wee Dalhanna and Corsincon, said to be for the hiding of cattle from the Border raiders?'  To learn of the Corsencon 'sod fences' and other earthworks then click on the lick to the left.












































' Confrims to Paisley Priory the whole land between the Old Patrick Water and the burn of Espedair (Paisley); the disafforested land between the Maich Water and the River Calder (Lochwinnoch);
a part of the land on the east of Paisley Mill; the land in Renfrew granted free in burgage (all Renfrewshire); timber and fuel in the forest of Sanquhar (St. Quivox, Ayrshire); and
the quitclaim of  Clohhary'  (Lochwinnoch?); all granted by
Walter son of Alan as his charter bears witness.
Edinburgh (10 August, 1211 x 1214).'
Acts of William I,
Ed. GWS Barrow








The Greyhound
'This animal has been known by various apellations, as the Irish wolf-dog, the Irish greyhound, the Highland deer-hound and the Scotch greyhound, for there appears no doubt that all the dogs thus denominated were essentially of the same breed. Its original home is supposed to have been Ireland'

'King David I commonly called the Saint, being a-hunting on Holyrood-day near to Edinburgh, there appeared a hart or a stag with a cross betwixt his horns, which run at the king so furiously, and dismounted him from his horse, that he was in hazard of being killed, if one of his attendants, Sir Gregan Craufurd, had not interposed. The armorial figures of Sir Gregan Craufurd, and all his descendants , who carry argent , a stag's head  erased with a cross crosslet to perpetuate the happy event '.

David II gave William Gordon, a major landowner in the Glenkens 'the whole lordship, administration and safekeeping of the New Forest of the Glenkenne'.  William Gordon, a descendant of this William Gordon was created 1st Baronet of Afton in 1706 at which time he acquired lands in the parish of New Cumnock, in the Afton valley and the Dalleagles estate.







































*Scots quh is typically pronounced as wh,  as is the case with the New Cumnock place-names Polquhirter, Polquheys, Monqhuhill but can also be pronounced as k e.g Sanquhar which if applied to Polquhirter gives Pol- kirter.

























































Click for full size image
Click for full size image
Deil's Dyke

Click on MAP
to see Full Image

Deils' Dyke
Dalhanna hill

Deils' Dyke
at Redree Burn

Deil's Dyke
Afton Water
River Nith

Park Burn

Cumnock Castle

Click for full size image
Click for full size image
Click for full size image
Click for full size image
Click for full size image
Click for full size image
Click for full size image

'The question remains as to how the idea of this dyke ever came to be conceived, and an answer to this can be put forward as the result of a fortunate accident. A shepherd's wife in Kirkcudbrightshire, when asked about a neighbouring portion of the Deil's Dyke, alluded to it in her reply as "the Deil-Dyke"—omitting the possessive "s", and placing the stress on "Deil" as if the words were hyphenated. The expression, thus sounded, immediately called to mind the old Scots word "deil" (spelt variously, and derived from O.E. dael), meaning "piece," "part" or "portion," and used in particular of land,  with a corresponding verb meaning "to divide into portions" and an allied word "deilisman" for one who divides or apportions.  Another word, "dale" or "daill" (O.E. dal), has the same meaning as "deil" when applied to land, and is understood to be pronounced "deil" in Galloway. I have further been informed by Mr and Mrs Smith of Pittodrie that "deil-dyke" is used to-day in the Garioch district of Aberdeenshire in the sense of a dyke marking a land-division or boundary; and by Mr C. B. Glen, Irvine, that "deil-ground" was in use a few years ago to describe a strip of ground dividing two properties near Portincross, Ayrshire'. A. Graham



Railway Line
A76 Road
A76 Road
The approximate course of the Deil's Dyke
Click for full size image


History of the parish of New Cumnock, Ayrshire, Scotland

© Robert Guthrie




Dumfriesshire & Ayrshire


New Cumnock


Click for full size image
Aerial Photographs
by permission of RAF
Z-I Wood

Other Earthworks

Click for full size image
Click for full size image
Click for full size image
Click for full size image

Map Co-ordinates
of Deil's Dyke

New Cumnock


Dumfriesshire & Ayrshire
New Cumnock

Selection of images of
Deil's Dyke
Dalhanna Hill

Johan Blaeu
'Atlus Novus' 1654

West Park

'The Park' from
Dalhanna Hill


Blackwood (centre) with Burnton to the right  and Z-I wood on Dalhanna Hill in the distance Meikle Garclaugh (front right)

West Polquhirter
with Corsencon Hill beyond

Stellhead Rig with
High Park in the distance, from
The Coupla

Redree Burn