From the charters examined so far there are some interesting findings. At Ednam near Kelso there was a road running to Kelso and to Sprouston ford. Tenants at Reddon were required to take corn to Berwick and return with salt and coal which shows there was a route to Berwick. At nearby Hadden there was a road to Carham where the Tweed could be crossed. Near Hadden there was a Prestre-bridge though the associated route is uncertain.

At Kirk Yetholm, a road ran into England and there was a bridge at the border. The road no doubt would have gone to Kirknewton and Wooler and beyond. In the direction of Kelso it is interesting to see that a parish boundary runs along a short stretch of road which may relate to the likelihood of the road from England continuing to Roxburgh. From Kirk Yetholm there was a road leading into England along the line of today's Pennine Way. It is thought that Kirk Yetholm was part of a lost Anglian "shire" dating from the 7th century so it is quite possible that these roads existed at that time.

South of here, in the present day Morebattle and Hounam parishes, Kelso and Melrose were gifted extensive lands around Mow and Hounam. These are in two parallel valleys, the Bowmont and the Kale, on a north-south alignment. Both valleys had roads, with one running to Roxburgh and one running to Melrose through Eckford, and it is likely that the roads ran high up in the valleys given the number of settlements and cultivation terraces to be found there. In fact, the road in the Bowmont valley is known as Clennell Street and runs over the border to Clennell and Alswinton in Northumberland. As it partly forms the parish boundary there, it is of mediaeval date at least but its other name of Emspeth indicates that it is much older.

There was a road between Mow and Hounam, and in the case of The Street, an example of a road forming the parish boundary. It continues into Northumberland as does Dere Street.

"near the limits of their land of Kelso, on the north side of the petary of Ednam, reaching thence along the boundary of the parishes to the southern bounds of Newton, and thence along the said bounds to the river Eden, and along the Eden to the bridge on the west side of Ednam, and thence to the road leading to the hospital (NMRS record), at the forking of the road which comes from the north side of the petary, and thence along the road to the place first mentioned; with the pasturage of a piece of ground between the petary and the bounds of Kelso." They were also granted a half ploughgate: "on the east side of the quarry belonging to the abbey, between the 14 acres of Paganus de Bosseville (granted to the abbey and lying in Ednam), the hospital land, the petary, and the road leading to Sprouston ford." (from Monastic Annals)

Ednam lies a couple of miles NNE of Kelso and the hospital about one mile south of Ednam on the Kelso road.

Unlike today, the parish boundary used to run a short distance north of the hospital then directly north-westwards to the Eden Water, passing south of Sydenham House. It is possible that the petary was in this area just west of Sydenham House as this could fit the charter description of following the parish boundary to the Eden.

The bridge to the west of Ednam is clear enough as is the road running to the hospital, and it is likely enough that it ran between Kelso and Ednam. The rest of the description is a little ambiguous but it seems that a road ran from the north side of the petary to join the road to the hospital, presumably very close to Ednam itself.

Reddon is to be identified with the present day farm of Reddon about 4 miles NE of Kelso and just south of the River Tweed (see Derek Hall, Scottish Monastic Landscapes, Tempus Publishing, 2006, page 157). The Hauden mentioned in the Monastic Annals is the hamlet of Hadden, a mile or so SE of Reddon rather than Hounam (sometimes Howden), 11 miles to the south.

Origines Parochiales Scotiae, Vol. I, page 440; Monastic Annals, page 114/5; Liber Kelso, charter 217, page 178

Hadden is the present day hamlet of Hadden, 4 miles ENE of Kelso. Although the Blindewelle and Croc cannot now be identified, Carram must be Carham one and a half miles to the north.

One would imagine a relatively straight road to Carham where the Tweed could be crossed and very probably give access to a road to Berwick on the north side of the Tweed.

There is a Pressen Burn that runs in a south-westerly direction from Pressen down to the English border at Pressenhill and partly forms the parish boundary of Sprouston. While it is likely that the Pressen Burn is the Prestre, there is no certainty that the bridge was at either of these places.

Shotton and Colpinhopes
A charter from the time of Alexander II gave Kelso Abbey:

The present day farm of Shotton lies just over the border about a mile north of Kirk Yetholm, and there would seem no reason to doubt that a road ran along this valley much as today.

Colpenhopes area (Coldharbour Hill)

"namely, from Hulaueshou (Elliesheugh) as far as its rivulet, and from there by Hulaueshou as far as the ford of Bolebent, opposite the church, and from that ford upwards as far as Hulaueshou, and from there by the road as far as Hunedune, and from there to the head of the Hulaueshou." Scilicet, ab Hulaueshou usque ad rivum ejus, et a rivo per Hulaueshou usque ad vadum Bolebent, contra ecclesiam, et a vado illo, sursum versus, usque ad Hulaueshou, et inde per via usque ad Hunedune, et inde usque ad capud rivi Hulaueshou. Image:near Molle (present-day Mowhaugh)

"that part of her land which lay next the rivulet descending from Brademedue as far as the Bolbent, and half an acre called Crokecroft next the road that led to Persouth..."       Image: Clennell Street at Cocklawfoot

JSM Macdonald (Placenames of Roxburghshire, page 30) says that Mollestele means "the cattle field at Molle." It is likely enough that it was in the vicinity of the settlement. The 1:25000 map shows numerous cultivation terraces in the area and these together with the mention of a rivulet running into Bowmont Water suggest that Brademedue could have been by the Hall Burn, Belford Syke or Calroust Burn. Persouth does not appear on any of the old maps but it may have been in the Sourhope area where there are cultivation terraces. Another possibility is that Crokecroft relates to the placename Croke Law near where there are remains of settlements and cultivation terraces in which case the road could have been up the valley of the Calroust Burn and Persouth would have been somewhere in the valley or over the watershed. However, it is more likely that Persouth (also written as Persouh) is in the vicinity of Percy Law, one kilometre north of Attonburn, and the road was therefore a local track (cf. Barrow, op.cit., page 261).
Wherever Persouth was, it is clear enough from the 1:25000 map that there are numerous remains of settlements and cultivation terraces in the vicinity of Calroust, Sourhope and Cocklawfoot and it highly likely that they were served by tracks. In fact, Clennell Street passes through Cocklawfoot coming from Alswinton in Northumberland and is more than likely to be the road that follows the Bowmont valley. It had an older name of Ernespeth (see also) noted in the chartulary of Newminster Abbey in Northumberland which takes it into Anglian times but could easily have its origins in prehistoric times. In later times it was well used as a drove road (see the RCAHMS Inventory for Roxburghshire, No.376, page 183 for details of another very old route, The Street, and some reference to Clennell Street - both tracks started near Alwinton in Northumberland but had quite distinct courses a few miles apart).

"his whole petary which was between Molope (Mowhaugh) and Berope and Herdstrete, which separated the land of Molle from the land of Hunun." About the same time he granted "the land with the meadows which he and Glai the nephew of Robert Avenel with the cellerar and brethren of Melrose perambulated, namely, as the furrow on the north side goes from the road which leads from Hunedune towards Molle, and goes as far as the rock...." (OPS)

If Molope is correctly identified as Mowhaugh, Berope is more likely to be a lost name nearby rather than the Beirhope four miles west of Mowhaugh. Herdstrete is difficult: it might be the path running up beside the Hall Burn directly between Mowhaugh and Hownam although the parish boundary running at right angles across it at the Howden Burn tells against this somewhat. It could also be The Street along which the parish boundary runs though it seems a little too much to the south (the RCAHMS Inventory suggests this identification). In any case the fact that the parish boundary does run along The Street indicates that it existed at that time and the other reference to the road between Hunedune and Molle is very likely to be on the line of the present day track running up the Hall Burn (see the RCAHMS Inventory for Roxburghshire, No.376, page 183 for details of The Street).

"From where the Stelnburn falls into the Blakburne, and by the Blakburne upwards to the two stones lying by the bank of the Blakburne, and opposite the house of Ulfi the steward, on the west part, and so upwards to a ditch, and to two standing stones in that ditch, and from these stones to another ditch filled with stones, and from that ditch to Heselensahe (Elliesheugh), and from that by the footpath under Heselensahe to the shallow at the waterfall of Alernbarhe, and from thence to the ford of the Steinburn, and so by the Stelnburn till it descends to the Blakburne." (Lees, page 46) Image:Blakedean

The Abbey of Paisley from its Foundation to Its Dissolution, J Cameron Lees, Paisley 1878, page 46
Origines Parochiales Scotiae, Vol.1, page 426; Registrum de Passelet, page 75

Fortunately Blachedane has survived as Blakedean, a mile or so north of Mow. It is likely enough that the Blakeburne is Atton Burn. Although it is not clear which of the streams shown on maps running into the Atton Burn is the Steinburn, the wording of the charter suggests the ford was on a local route only, perhaps to Molle itself.

The next charter below refers to a road between Molle and Roxburgh which must have passed through the lands of Clifton and Primside. It is not certain, however, that it is the same as that mentioned here, as Clifton and Primside are a little way off the direct route to Roxburgh. The mediaeval settlement lay a little way upstream from present day Clifton. There was a ford at Primsidemill which may have been used.

Crukehou is close to Crookedshaws, Prenwensete is Primrose and Grubbheud is Grubbit but generally the charter is difficult to follow because the names of streams have changed. The map shows some of the named places, the parish boundaries and some of the roads. However, the mention of the road is clear enough and presumably ran from Moll up the Bowmont valley to near Clifton and Primside then across to Roxburgh. It is not clear if it went directly to Roxburgh from here, i.e. through Linton, or went over to Eckford. Interestingly Jeffreys (page 321) says that there was a route between Melrose and the Cheviot granges that went by Eckford (see also Jeffrey, Volume IV, page 188 and Hardie, Roads of Mediaeval Lauderdale, page 33&ff).

As the course of Dere Street (noted here as Watling Street) is known, we need not be concerned with the details of the charter.

Grubbeshead is marked on early maps as Grubet a mile or so SSE of Morebattle, and less than a mile north of the grange.
The charter indicates that there was a track from the grange, and no doubt, Hownam, to the Morebattle area. As said, a route continued to Melrose Abbey through Eckford.

Maxton
Several roads are mentioned in these charters, viz: Derestreet, the road from Eckford to Melrose (coming from Hownam in the Cheviots), a road running to the east, the King's Highway between Roxburgh and Annandale, a road between Newtun and Roxburgh and a road forming the parish boundary between Maxton and Faringdon.

Interpretation
As the course of Dere Street is known, the charter need not concern us too much but Morrig survives in the placename Morridgehall, and Lilisyhates in Lilliardsedge.

Interpretation
At first sight it seems a very promising charter but it proves difficult to follow the boundaries given in the charter. Ultimately, however, we can gain a good idea of the course taken by the three roads referred to.

As Hardie suggests, Kill Law (see 6" OS map - Roxburghshire, sheet XIV) and Steelmoor, both north-east of Muirhouselaw may echo Kelfsete and Celfesetestele - they are in the right area as the charter has the heading "of Muirhouse" and "seat" and "law" are terms used of hills.

If this is correct we can note that the Eckford to Melrose road must have ran past the eastern side of Steelmoor Plantation on a direct route to Maxton. In view of this, Hardie's routing of it by the Duke's Strip and the reported remnants of a road heading from the end of the strip towards Rutherford Burnside may be too far to the east.

The charter's wording for the "road that goes to the east" is ambiguous. Hardie suggests the road is identical to the Newtun to Roxburgh road below, a road that ran south of the Tweed. This accords quite well with the mention of Rutherford and with the possibility that the use of "via" for the "road that goes to the east" and "stratam" for the Annandale road was intended to differentiate between them; that is, that the road to the east could not have been the Annandale road.

Against this, the position in the charter to the reference to the "road that goes to the east" implies that it was south of Kelfsetestele and so could have been the Annandale road.

That aside, it doesn't matter too much as there seem to be only the three roads referred to in the charter; the Newtun to Roxburgh road, the Annandale to Roxburgh road which forms the parish boundary between Mackustun (Maxton parish) and Faringdun (Fairnington in Roxburgh parish), and the Eckford to Melrose road.

It is not absolutely clear where Newtun was. Hardie (page 52) suggests it was Newton St Boswells, where Melrose had a grange. If so, this road must have merged with the Eckford road near Maxton.

Liber Melros Charters 282, 283, pp 249,250; Roads of Mediaeval Lauderdale, Hardie pps 18,19 ; The History and Antiquities of Roxburghshire and Adjacent Districts, Alexander Jeffrey, Volume 4, chapter VI, page 207, also 167; Monastic Annals, page 271; OPS, volume I, page 309; Lilliesleaf and Its Old Roads, Rev. James McKenzie, Trans. Hawick Archaeological Society, 1947


Jeffrey (vol.iv, page 167) refers to a highway running by Midlem, Milnrig, Chapel, Hillhead and Greenhouse and on to the Teviot, but Hardie has the road mentioned in the charter more to the west of this. His route is from Midlem practically due south to cross the Curlewude (now the Chesterknowes Burn in its upper part and the Shaw Burn in its lower part) and the Alne, into Lilliesleaf. The further course of the road may be shown by a relatively straight section of the boundary between Minto and Ancrum parishes that runs down to the Teviot, and it may have joined Jeffrey's road at the north end of this parish boundary.

McKenzie gives a more detailed and different account of the route. It can be followed easily on the 6" map (Roxburghshire, sheet XIII) running from Whitmuir past Whitmuir Hall where a path leads down to Kersknowe. From there it went by a still existing track to Curling Farm then crossed the Curlewude (Chesterknowes Burn) just east of Greenside Moss, and west of Calla Plantation (which he suggests is the Caldelawe of the charter) to make its way to Friarshaw Farm. A path then leads down to stepping stones on the Ale from where it is a short distance to Lilliesleaf itself. He makes the interesting point that the path leads precisely to Bishop's Close in the village where the Bishops of Glasgow (who had lands in Lilliesleaf) may have had a tithe barn and around which the village grew. In general he sees the route as part of a longer route from Glasgow to Lilliesleaf and Jedburgh, where Glasgow also had holdings.

South of Lilliesleaf (see sheet XIV) he continues the road by a short path that meets a wider road dating from the 18th century and which he suggests incorporates the mediaeval route. It skirts Lilliesleaf Moss to pass through East Middles then turns south-east to Craggs. From there it crosses the Mire Burn by a substantial bridge and runs down to Netherraw. Beyond Netherraw he routes it by the present day road past Greenhouse and Standhill then down through the Minto Policies (Lambslair Plantation) (sheet XX) to a ford at Linton Mill. Once over the river it would have taken a direct line to Jedburgh through Lanton. It is not clear from his text why he routes it along the river to Linton Mill when there was a nearby ford at Spittal and two near Barnhills. The ford at Spittal would access a direct route to Jedburgh by Ruecastle and Lanton Hill shown on the Military Survey map. As already said, there is a possibility that the course of the road beyond Standhill was along the parish boundary down to the Teviot although this needs confirmation.

McKenzie also notes the existence of another old road (the same as that mentioned by Jeffrey) that went from Whitmuir to Midlem then by Milrighall, Linthill and Chapel to Craggs where it joined the road above. He suggests that although still mediaeval it may have been an easier route for wheeled traffic and that Midlem itself being a larger village could have generated some traffic of its own.

In summary then, a road came from Selkirk through Whitmuir to run directly to Lilliesleaf, and had its origins in the connection with the Bishops of Glasgow. It continued SSE to the Teviot and then to Jedburgh although there is some doubt about the details of its route on this stretch. Another route went from Whitmuir through Midlem and joined the other road just beyond Lilliesleaf and probably developed after the first route.

The monks had their centre at the Monk's Tower, some 200 metres SE of Hassendean village. Despite the clear reference to pilgrims and wayfaring poor, and the implication of a route between (presumably) Hawick and Melrose, it is not clear what course this route would have taken. Possibilities would be the Roxburgh to Annandale road and then to Dere Street although this appears unnecessarily long, or up to the Lilliesleaf to Jedburgh road and from there to Melrose.

Another road is mentioned in a Dryburgh charter, where John, son of Ylif of Ylifston, gave them "ten acres of land of his demesne in the town of Yliston, viz., two acres in toft and croft nearest to and east of the rivulet which ran below his garden, four acres in Rokflat, next to and west of the road leading to Boulden, and three acres in Greenrig."(Jeffrey, iv, 174)

As said, neither of the Dere Street references allow us to determine the exact course of the road. The other road was in the vicinity of Elliston, two miles south of Newtown St Boswells and suggests a road ran in a north westerly direction up to Bowden, one and a half miles distant. Hardie suggests it would have led to Roxburgh, though the route seems awkward; a route through Maxton would be shorter.

One important question is the route taken by Dere Street and where it crossed the Tweed. Chalmers (Caledonia, Vol 3, page 91) had it cross at a ford near Gattonside then up to Clackmae, one and a half miles NW of Earlston. Roy (Military Antiquities, page 103) routed it from Gattonside up the valley of the Allan Water but appears to have been following the Girthgate.

Another proposed location was at Newstead and there was a tradition of there having been a fine stone bridge there quoted by Milne in his history of Melrose (1743).

Dryburgh Abbey

Looking towards Sorrowlessfield, Kedslie, Blainslie, Milkside etc. Malcolmesrode generally keeps close to the ridge line - the lands run from this to the valley of the Leader in the middle distance.

The various estates can be seen from the map. Apart from Gattonside and Fauhope immediately north of the Tweed, there is a cluster bounded by the Leader on the east and, on the west by a road called Malcolmesrode, that may have been Dere Street. These are Sorulesfield (Dryburgh), Kedsley (Dryburgh), Elwinseley and Herdesley (Dryburgh) forming an outlier of Earlston parish on the west side of the Leader, and Blainslie and Milkeside (Melrose) just south of Lauder parish.

Course of Malcolmesrode, and of Dere Street after Chalmers

Chalmers, referring to Kinghorn, routes Dere Street from a ford opposite Melrose up the west of the Leader, close to the turnpike as far as a camp called Chesterlee (Kedslie-NT54SE 20) - the road in this area could easily be seen. It crossed the turnpike and a small stream that joined the Leader below Chapel which would place this about 400 metres above today's Kedslie.

It then ran up to a Roman site called Wass or Walls at New Blainslie (site NT54SE 18 is 200 metres from New Blainslie). Kinghorn says that the road was very evident for a mile and a half north of here when it crossed the turnpike again, and a nearby stream, half a mile ENE from Chieldshiels chapel. It then ran towards Lauder.

Details of the Roman camps (all temporary camps) can be found on the Canmore database (Drygrange-NT53NE 33; Kedslie-NT54SE 20; Blainslie-NT54SE 18; South Blainslie-NT54SW10; St Leonards Hill-NT54NW 14) - Canmore entries for Dere Street on this stretch refer mostly to Chalmers. It is interesting to see that the camps are all on or close to the presumed line of the road. While this is suggestive evidence it is not absolutely conclusive. By their nature, temporary camps indicate the passage of a military unit rather than a settled base served by roads, although of course they could use an already existing road. The St Leonards camp (currently the largest known camp in the Roman empire) has been dated to the Severan campaign between 208 and 211.

There is another suggestive alignment, which complicates the issue somewhat. This is that the Blainslie and St Leonards Hill camps are exactly on the line of the Birkenside road which forms the county boundary and is mentioned in charters as a magnam stratam which implies that it was a made road. The question is whether it is Dere Street as Kinghorn suggests, or if it is a road to Birkinside and perhaps Roxburgh as Hardie suggests, though whether Roman or mediaeval is uncertain.

Clearly this is problematic, but to explore it further is too speculative given the available evidence.

With regard to Malcolmesrode, the fact that it is clearly named leads us to assume that it was either constructed or remade by one of the Malcolms before the first mention of it in the charters. Malcolm I (943-954) seems too early and Malcolm IV (1156-1165) too late. Both Malcolm II (1005-1034) and Malcolm III (Malcolm Canmore, 1058-1093) are more likely - Lothian passed into Scottish hands after the Battle of Carham in 1018 and the road no doubt helped control the new territory and allow a rapid response in the event of incursions from the south. As implied above, it has not been determined whether it had originally been a Roman road, or dated from early mediaeval times. As said, its ridgeway characteristics could indicate a very early origin.

The Roads (click map for larger image)

Several roads are mentioned although one or two may be the same road under different names.
• Malcolmesrode

Malcolmesrode

Looking towards Colmslie

Hardie takes the road from the Leader up the Kedslie Burn over to Colmslie and Whitelee, to cross the Gala Water near Crosslee and reach Windydoors some two miles further on.

This is mentioned in two Dryburgh charters relating to Elwinsley which lay adjacent to Kedsley. Willelmi de Lyndesay gave Dryburgh:

Two views of the Girthgate, just south of Threepwood Farm

Large dyke on the county boundary 50 metres east of Threepwood bridge on the south side of the burn

"Along the east side of the river Galhe upwards in the direction of their own property as far as the boundaries of Wedale, and also along the right boundaries of the land of Richard de Moreuille, viz., as the Mereburne falls into the Leder up to the source of the same Mereburne, and thence along the sike which issues from the Mereburne to the spot where that sike falls into the rivulet of Standene, and thence as far as Pot, and from Pot to Standande Stan, and thence as far as the King’s way where it enters the wood and divides the wood of Standene and of Threpwude, and thence by the same King’s way to Fairforde, and afterwards along that way which goes to the right as far as the foresaid bounds of Wedale, and thence by the right bounds of Wedale to the Galhe." The wood of Threpwude was excepted from this although Melrose was allowed the pasture rights - it was bounded as follows: "From Fairforde down by the Aloent to the moss which is between Threpwude and Cumbesleie Cnol, and thence by the same moss as far as the foresaid Pot." (OPS)

Liber Melros, charter 111, page 100; OPS, volume 1, page 282; Hardie, page 72

Note: Faireforde must have been on the Allan upstream from Threepwood Moss (Cumbesleie Cnol must be Colmsiehill). Given that the Standene was the Lauder Burn (Hardie), a good fit with the wording would be a road running west towards Fairford (the King's way), and meeting a road going to the right, which is presumably the Girthgate.As the presumption must be that the wording fits the parish/county boundary, one would follow the Standene down to near Muircleugh (Morclow) where the Pot would have been, so that the King's way may have been the straight stretch of boundary leading to the Girthgate - Fairford would then have been on the continuation of this line. It is, however, not clear where this road originated or where it ended. Gilbert links it to the "road to the wood" of Dryburgh charter 112 and has it run over to join a road running up to Stow along the west side of the Allan. However, it seems too far north to be the same road, especially as it is said to have divided the wood of Standen (which seems to be to the north of Threepwood) and Threepwood. Perhaps it came from Lauder.

Road to the side of the church at Stow
This is referred to in a charter giving the results of an arbitration following a dispute between Melrose and the men of Stow, viz:

Top