Home > Roads and Tracks of Ayrshire > Post-Mediaeval












Roads and Tracks of Ayrshire

Back (Mediaeval Roads)



Early Legislation

Bridges and Fords

Early maps Speed 1610, Green 1679, Moll 1718 & 1725, Bowles 1735, Bowen 1747, Kitchen 1749

Commissioners of Supply

Burghs and parishes

Estate Roads

Drove Roads




Text file only

In this chapter, dealing with the period 1600 - 1750, we will look first at early legislation which had an effect on road building, and then at bridges, a large number of which were built in this period. We will then attempt to identify what roads there were at this time. There is no problem in doing this at the end of the period, i.e., about 1750, as the excellent maps of the military survey show all the roads. Because of their comprehensiveness, there is a separate section on these maps. The problem is more in the early years of the period as the evidence is quite scanty. Such evidence as there is, is found on some early maps along with what one can infer from the location of bridges and the role of the burghs and other settlements in road building.


Early Legislation

We use the term post-mediaeval loosely for the 1600's and the early to mid 1700's just prior to the turnpikes. It was a time when those in power became more aware of the need for roads and began to pass the first laws on roads.


As already noted, the first legislation other than two early statutes of William the Lion and David II was in 1555 with an Act that required highways to be kept open; any one stopping them up could be punished as an oppressor. This was reinforced in 1592 where fosses and dykes were to be made to enclose the common ways. This is interesting as it implies known tracks, and the act of enclosing them leads to a defined highway, even if not surfaced, as distinct from a route between two places which because people took the easiest path could be hundreds of yards wide.


In 1617, an Act similar to that of 1555 was passed. (1) It made the repair and upkeep of highways the responsibility of Justices of the Peace, a system that had operated in England since 1555 (see Moir (1) for the effect James VI had on roads on Scotland after the Union of the Crowns). Justices of the Peace were introduced in 1609 where appointed individuals represented the King. They had the power to appoint overseers and call out tenants and cottars to work on roads. The Act said that roads to the kirk and the market were to be twenty foot in breadth. Where this fell down was that although JP's were responsible for the highways, they did not have control of finances and local landowners were very reluctant to pay for something of no immediate benefit to them.


To counteract this, the statute labour system was extended in the 1669 Act for Repairing Highways & Bridges. It stipulated that the Sheriff and a Depute Sheriff of the Shire and the Justices should meet on the first Tuesday of May yearly and draw up a list of Highways, Bridges and Ferries and divide responsibility for these between themselves or an overseer. They were then "required and authorised to call upon all tenants and cotters and the Servants to bring horses, carts, seeds, spades, shovels, picks and mattocks required for repairing the highways." Overseers were to be appointed. Individuals were to work for 6 days for the first three years and 4 days a year after that (changed in 1685 to 6 days yearly over the next five years and then in 1719 to 6 days annually in perpetuity). The work was to be done between the "Bear-seed" and the "Hay-time or harvest". Those who did not turn up could be fined 20/- Scots daily or "punished in their persons" if they could not pay this. The money was to be used to hire others to do the work. The roads were to be at least 20 feet wide and were to be suitable for winter use. This was changed the following year to any time, except seed time and harvest, and payment could be made instead of labour at 6/- a man and 12/- a horse daily (Scots). In 1719, this changed again to 3 days after seed time and 3 days after harvest. The commutation rate by then was 6d per day (Sterling).


This system although flawed did result in some work being done although we cannot say for definite which roads were built as the records of the Justices for Ayrshire no longer exist. The corresponding records for Lanarkshire mention inspections of roads in several parishes early in the 1700's. As there is no reason why Ayrshire should be any different, we are probably safe to assume that it too had a fairly well developed system of roads at this time. As well as the roads shown on maps of this period, the old road excavated by MacDonald (2) between Dalmellington and Ayr could date from this time or have had work done on it if it was an earlier road and Clarke and Wilson (3) found pre-turnpike roads near Mauchline that may also be of this period. We can assume some work was done on roads to churches and markets (1617 Act). There is also a reference to work being carried out in 1695 on a road between Kilmarnock and Irvine via Dreghorn and Annick.


The problem with Statute Labour was that most individuals greatly resented having to work for no wages and it was a real hardship for old and infirm people. Overseers had next to no idea of how to make or repair a road. A typical repair would be to smooth out ruts and fill in holes with rock. In making roads, a heavy bottoming was used with large stones as kerbs, onto which smaller stones would be placed, and a topping of gravel. While this was fine for limited traffic, it was unsatisfactory for heavier demands as lack of drainage led to uneven settlement and undermining. Robertson says that the small stones work underneath the larger and raise them above the road surface. (4) The jarring from passing traffic allows more small stones to work their way under resulting in a road with projecting stones and corresponding hollows. Although from a different period, his picture of a road at this time is pertinent. It was "very narrow and deep, not unsimilar to an old broad ditch, much water worn and thickly strewn with loose stones." He complains about the older roads which were often made up of unbroken field stones and then when the land was enclosed forced the traveller to stay on them whatever lay ahead. As these roads usually took the shortest route often on the higher ground there was no thought to gradient that made travel difficult for wheeled transport.


In 1686, further legislation required the Commissioners of Supply to meet with the Justices to organise the repair of highways and bridges by statute labour.


By the early 1700's, a growth in trade and the beginnings of the agricultural revolution pointed to the inadequacy of the roads for the increase in traffic. The response of the authorities was to use the existing legislation to effect road improvements and the statute labour system was implemented fully at this time. Another act limited the number of horses drawing a wagon or cart and stipulated the minimum wheel breadth. The theory was that wider wheels would help flatten the ground.


A good example of local implementation of statute labour is provided by Kilmaurs in a burgh record of July 1706. (5) "It was statute and ordained the whole burgesses, cottars and inhabitants to send ane man out of every house wt. ther horses wtin ye toune to repair the high wayes and venniles…and that they come out man and horse preceassly agd. at nine houres in the forenoon upon the sd. day Being ye twentie-fourth day of July next under ye penaltie of six shilling scotts money to be taken and payd out of every house yt sendeth not out a man and 12/- scotts money out of each house qr ther are horses if they send not out ane horse."


In 1745, Maybole Town Council minutes note that the inhabitants are "to turn out and collect stones and sand: those who have horses to lead the same, and those who refuse to comply to be fined in eighteen shillings Scots for each absent day."


The inbuilt inadequacies of statute labour were still there, however, and by 1767 a new system was developed whereby this responsibility could be met by making a monetary payment. Landowners were liable to pay 30/- for each £100 Scots of valued rent; innkeepers and carriers had to pay 1/-per day per horse and 6d per man; and householders along with manufacturers and tradesmen not in the first two categories were liable to pay 3/- per year. All aged between 16 and 60 were liable except apprentices and those in "indigent circumstances." Although the payments were strongly resented, at least it meant that the money could be used to employ people who had some knowledge of road making and who would treat the task as everyday work rather than as an imposition. By 1800, statute labour in Ayrshire was generally commuted and contractors were often employed to do the work. The system was eventually abolished in 1878 under the Roads and Bridges Act of that year.


Bridges and Fords

These were the responsibility of the Commissioners of Supply and fortunately records of their proceedings exist from 1715. (6) The Commissioners were a committee of major landowners set up in 1667 with the responsibility for setting taxes and for building and maintaining bridges.


A "bridge money" assessment was decided upon, and this could be drawn upon for repairs or new bridge building. Typically, a committee would be formed to examine a bridge and obtain estimates for its repair. Thus, at the meeting of 17 May 1745 (7) we have the heading, Water of Finnick to be visited to build two bridges over it and the text, "There was produced a petition from Craufordland and others anent the building of two bridges over the Water of Finnick which being considered by the Commissioners they recommend to Crawfurdland Dunlop Landlands Gardrum and Robert Paterson or any three of them to try what contributions can be gott from the neighbouring towns and countrry, and to call different tradesmen to visit what places are propper for building the said bridges at, and to make estimates of the expences and report." Over time, regulations for the administration of the bridge money were developed and were strictly enforced.


The need for repairs of bridges at Doon and Barskimming is mentioned several times from 1715 onwards (references are from the first Minute Book, CO3/1/1) - Barskimming bridge at the time was suffering from a "profound ruining and falling down." Fail is mentioned at the meeting of 15 May 1718 and Coyle on 2 June 1719. £723.5.0 was paid towards a bridge of poorstone at Fail.


On 16 May 1720, the "falling and utter decay of the bridge that formerly was at Barrbeth" is noted; this was on an old road between Dalmellington and Straiton, which crossed the River Doon.


Over the next few years, there is mention of repairs to bridges at Hudostone (perhaps Hoodstone bridge on the A71 east of Hurlford), Killin, Galchoin and Carriosbridge. It was proposed to build bridges at Riccartoun, Kirkmichell over the Watter of Girvan, Dalrymplekirk over the Doon, and near Colmonnell or Ballantrae over the Stencher. The site finally chosen for the last-mentioned bridge was near the kirk at Colmonell and the bridge itself was built about 1745. Another was built at Ballantrae itself about 1770.


Image from Detroit Publ.Co., c.1900
The old bridge at Ballantrae. Ardstinchar Castle may have been built here to guard an early fording point. Image from the Detroit Publishing Company's Views of Landscape and Architecture in Scotland - see thumbnails on Library of Congress site here.

McDowell says that stone from Ardstinchar Castle on the hill above Ballantrae was used for this bridge and that stones from Carleton Castle may have been used for Colmonell Bridge.


In 1724, there is mention of a bridge near Renfrewshire, and one over Halfwatter. The following year sees the Riccartoun bridge started, and mention of bridges over the watters of Gogo, Rye and Stobonstoun (it is not clear where this last place was).


In 1728, it was decided to build a bridge near Barr Kirk. In 1729, the bridgemoney was allotted to Dailly, Colmonell. Barr, Dalrymple, Barskimming and Holmstoun. The next year mention is made for the first time of a bridge over the Lugar and in 1734 a bridge is proposed over the watter of Glazert at Gallowayfoord (between Kilwinning and Stewarton).


A petition was made to the Commissioners in 1738 for a bridge over the Garrer Burn between Kilmarnock and Irvine, and there is an interesting entry about Gogo bridge which a landowner says was carried away by "ane impetuous shower after a great thunder", which he repaired immediately "lest that breach should have carried away the whole bridge." Expenses were £9.1.7. He represented that it was "absolutely necessary for the intire safety of that bridge to have a strong buttress built at each end of the land stool to keep off all accidents of that kind." Approval was given to inspect the bridge for this work. Mention is also made of Noddle bridge.


Other bridges mentioned are Hoodstones and Sares, Clark's and Corses (in parish of Beith - Corses is probably Kerse), Maxholm and Lugtoun which needed repairs. New bridges were proposed for the Nith below the old mains of Cumnock (1742) and Stair (1745), and as previously mentioned, over the Water of Finnick (1746).


Apart from these records, there is reference in the literature to other bridges. A bridge over the Irvine on the route to Ayr was rebuilt in 1748 and a new bridge was planned for Ayr in 1782 as the old one was in serious danger of falling down. In a study of the new bridge, Taylor gives full details of the lengthy and expensive process of obtaining an Act of Parliament to build a bridge. It is noteworthy how unco-operative the Commissioners of Supply were; they seem to have viewed the bridge as the responsibility of the burgh alone. (8)


By 1666, the bridge in the town of Irvine had decayed to such an extent that it was dangerous to those crossing. Repairs that year helped but by 1695, it had become dangerous again; so much so that carts carrying coal were banned. A flood in 1712 caused severe damage and destroyed a number of bridges in the area. Rebuilding started in 1748: parapets were kept low so that pack horses could pass and there was the usual embayment in the middle for passing traffic. Its width was 11 foot. (9,10)


Down in Cumnock there was a bad accident in 1753 when they were building a bridge over the Lugar at Stepends. The men were sheltering under the arch from the rain when it collapsed and killed 4 men and 4 boys. (11)



The Roads

In this section, we will look at different sources of evidence for roads. Early maps are useful although their roads can be difficult to identify and in one or two cases do not even appear on Roy who is more or less contemporary with them. The Commissioners of Supply records from 1715 onwards are also useful, as it is reasonable to assume that the bridges mentioned were linked to roads or at least tracks. A number of new burghs were established in this period and the trade they generated must have led to some work being done on roads. Finally, there are occasional references of roads that may date from this period, including some from folklore.


Early maps

Routes shown on early maps

Reproduced from the 1935 Ordnance Survey map. © Crown copyright

Speed 1610

He shows Galloway extending to Girvan and along the Water of Girvan, with Carrick up to the Doon and Kyle up to the Ayr. (12) This has to be squared with Pont who was surveying about 1600 who shows three stones "depynding Galloway from Carrick" three quarters of a mile above Glenapp, i.e. about a quarter of the way from Ballantrae to Stranraer but it may just reflect what the people said to the cartographers. The three stones are named Tam Slowan; Friar's Kirk and the Taxing Stane (Carrick Gallovidian, McDowall).


Green 1679

Green's map of Scotland (on Charting the Nation site - search for Green) dates from 1679. (13) He only shows one road in Ayrshire running from Lanark over to Irvine and then south through Ayr to the border with Galloway.


In places, it is very hard to pin the route down. On entering Ayrshire somewhere near Loudoun Hill, it definitely does not follow the Irvine Valley but runs north of this over to Cunmninghamhead and Irvine. The most sensible looking route, going by rivers and placenames e.g. Braidley and Broun and comparison with other maps, would be on the hills to the north of the Irvine valley then running close to Waterside and Fenwick or perhaps nearer to Moscow and then past Kilmaurs (none of these places are shown) to cross the Annick just north of Cunninghamhead and then to Irvine. This might be consistent with Pont's crossing over the Glen just above Darvel. It should be pointed out that Green's road runs across the border into Renfrewshire but when one looks at Moll who has a similar county boundary but more place names this seems unlikely.


From Irvine it ran near to Dundonald and Monkton to Ayr. Given the topography this gives a likely route within a half-mile or so. South Of Ayr, it runs directly to Newark just south of Alloway, at the foot of the Carrick Hills. The next places shown are Trochraig and Killochan on the River Girvan. Given that the road does not go near to Maybole or Crossraguel it may well have run on the eastern slopes of the Carrick Hills over to near Kirkoswald and then taken to the hills again to climb over to Trochraig and Killochan. There is a possibility that the route over to Trochraig was the same as or close to the road shown on Moll except where it neared Trochraig.


South of this point, it ran through Carleton (in Lendalfoot) to Ardstinchar near Ballantrae and thence through Glaick to Stranraer. There is a distinct possibility that south of Carleton it is the same as Roy as far as Ballantrae. In fact, there is enough ambiguity for the stretch to Glaick (on the coast just above Little Laicht - cf. Bowles) for his depiction to fit a Glenapp route or one similar to Bowles running to Glendrissaig and Glenapp.


Moll 1718

There is a north-south route running from Glasgow via Renfrew, Irvine, Ayr and Girvan to Stranraer along with a route from Riccarton, south of Kilmarnock, to Sanquhar. A road leads off this to Hamilton. Other roads shown are Irvine to Largs and Irvine to Kilmarnock.


It is impossible to tell from the map what course the road took between Renfrew and Irvine but it may have been through the Lochwinnoch gap. The depiction of the route to Largs is sketchy but topography suggests it was near the line of later routes. The road to Kilmarnock and Riccarton may well have been on the old A71 line through Crosshouse and Dreghorn.


Heading south to Ayr, it is not certain if it took the more inland route through Dundonald or one nearer the coast, such as that shown on Roy only 30 years later. South of Ayr, it ran to Dunure and then is shown running directly to Killochan, near Girvan. This could fit a route near the coast or one heading across the hills between the Kirkoswald area and Killochan. South of Girvan, the route looks very similar to those shown on other contemporary maps. From Ballantrae southwards, it probably ran directly to the mouth of Glenapp.


The route from Riccarton to Sanquhar is interesting as it is a "lost road" and one or two clues suggest it may be mediaeval. It went first to Dunley where a road branched north-eastwards towards Hamilton, then to a place called Oygang, near to two lochs and then to Sanquhar. Pont shows Dunley close to Rottenrow and Skeoch, two miles northwest of Mauchline. This is a strong indication that the road leading from here to Hamilton is along the Crosshands to Galston line. The road from Riccarton probably went via Bridgehouse (NS434347) and out towards Carnell and Rottenrow, perhaps on the same route as Bowles' (1735) Kilmarnock to Mauchline road, though as we saw that may have ran farther to the west.


South of Darley, Moll's road went to Oygang, which Pont shows to have been near Barlosh, about three miles southwest of Ochiltree. Oygang must have been situated at approximately NS488180. There is another Rottenrow just to the north, suggestive of a "King's Highway" (Route du Roi). The map is too indefinite to say whether the Darley to Oygang stretch ran through Mauchline or Ochiltree. However, it may have picked up the Mauchline to Ochiltree route, probably of mediaeval date, because of the need to cross the river and the "Gaitsid" shown on Pont, which was near Barturk and Cawhillan.


From Oygang to Sanquhar, the most likely route would have been that shown on Armstrong as running from Auchencloigh over to New Cumnock.


Moll 1725

Moll's map of Ayrshire is in two parts, Cunninghame north of the Irvine and Kyle and Carrick below it (1745 imprint). He shows only one road, from Glasgow to Stranraer. (14)


It is based on Pont and therefore relatively easy to relate to the more detailed Pont map and so to modern maps. When this is done, the course of the road does not correspond to anything shown on Roy or Armstrong, or later maps.


From Kilmarnock northwards, it ran between Dalmunsternock and Rowallan then just west of Robertland and of Blacklaw. On crossing the county border, it headed towards Neilston.


It is very difficult to tie down the course of this road from the map and there are no obvious traces of it on the ground. He shows it running up from Kilmarnock and passing between Dalmunsternock and Rowallan, slightly closer to the former although this may just be the drawing of the map. As there is only 3/4 of a mile between the two places and Thomson shows a marshy area near to Mosside it must have run on one side of the moss - the west side would allow a straighter course. Interestingly there is a right of way in the area.


Beyond this point it went up to Robertland Castle. On a direct route, it would have gone through Gainford and West Pokelly and the topography would allow this with deviations here and there to make an easier path. Gainford may be significant but as it doesn't appear on Pont or Armstrong, it may just refer to the later Fenwick to Stewarton road. Armstrong shows a road branching off the Glasgow road south of Dalmunsternock and going through Birkton to Rowallan. Unfortunately, Birkton doesn't appear on other maps so this clue is lost. The road ends near Rowallan downstream from the confluence of two streams just east of the old castle. This is well off the putative route of our road but given the inaccuracies of map making in those days cannot be completely ruled out. Curiously it lines up with a road Armstrong shows running from Robertland up to near Whitelaw - this suggests a missing stretch and a northward extension in roughly the right direction. This might be so but it is inconsistent with Moll who shows the road as having a more westerly course.


From Robertland northward, one sees from Moll that it crossed the Annick Water near Robertland and swung up towards, but never crossing, the Clerkland Burn.


It is in fact possible to relate these two streams on Moll's and Pont's maps and it is a good fit. One can then plot the road on to Pont and see which places it passes. Unfortunately, this doesn't help much as Pont doesn't always have his farms in the right place. We can say it passed close to Maryhill, Foulshaws, Whitelaw, Gabrochill and Auchentiber but one would have assumed that anyway from its northward course.


Near to Robertland, there are a number of clues to an old road but again they are inconclusive. For a start there is Kingsford - King would be a bit stronger than Kyne or cattle as a clue to a road - and nearby a Spital. Then on the outskirts of Stewarton there is a Causeyhead and just above it a Thornhill which Thomson shows as Thornygate. North of this is Gateside, which is shown on Pont. While all of these are significant in themselves, they are hard to relate to this road in a conclusive way.


Another indication of the route is given if we assume that later roads took a similar course, as dictated by the topography. Thus, although the present day back road running up from Fulshaw to Townend of Fulwood is later (it appears on Thomson) it links reasonably well with Pont's farms, the course east of the Clerkland Burn and the general direction towards Neilston. From Townend, it would have gone either up towards Craignaught or Gabrochill avoiding marshy ground in the area to hit the line of the present road which Armstrong shows. It would have made sense for it to have been close to this road as there were extensive mosses north of here and it was much shorter to Neilston than a more westerly line. Beyond this point, Moll shows it crossing the Levern Burn some way from the source. This suggests a routing up to Aboon the Brae or further on before making the crossing into Neilston, rather than taking the modern line.


At present, there is no particular indication on the ground of an old road from Fulshaw northward, either as a holloway or as a made road. Against this the ground is currently open pasture which even allowing for it being a bit wilder and overgrown would be quite passable in good weather, so there may have been no particular need for a made road. The marshy ground north of here is very evident, being a silted up loch and would definitely have forced a diversion. At best then we have it pinned down to a strip of no more than 200-300 metres in width, sometimes less but with no real evidence on the ground to be certain of its course.


From Kilmarnock south, it is shown running just east of Craigie, then past Craigie Castle and Barnweil to Kessock (St Quivox) to Ayr, running to the south of Ladykirk. The mileage is shown as eight. It is hard to say if it is represented at any stage by a modern road, there is certainly no direct through route on this line today.


It is worth noting that Roy has a road running to Craigie. However, it cuts down to Fail rather than Barnweil and only takes a similar line again south of Ladykirk. More evidence would be needed before one could say they are in parts, the same road, but the similarities are suggestive.


South of Ayr, the road goes to Maybole, passing through Alloway. A reasonable fit, given the position of Newark, Bla (possibly Blairstone) and Knockdon would be the B7024 (or a similar line) which used to be the turnpike between Ayr and Maybole. Mileage given is six miles. It is also shown on Roy and may well be the same road.


From Maybole, it ran past Crossraguel Abbey and on towards Kirkoswald. However, his mapping is quite misleading as it suggests a straight road to Girvan. In fact, it seems to have gone from Kirkoswald up by the west of Craigdaw Loch to Kirkhill and then south of the Ladyburn to a place called Snaid (NS209013). It then passed west of Trochrague, crossing Girvan Water near present Girvan Mains. There is no hint of this stretch (shown as 15 miles) on Roy or Armstrong.


Beyond Girvan, it no doubt went to the east of the present town and then took to the high ground rather than the coast. It is shown east of Ardmillan and crossing the Bynehill Burn. However, it is unclear from the map whether it kept to the east or the west of the high ground here (i.e. Byne Hill, Cairn Hill, Grey Hill) although his depiction of a couple of hills suggests the road was on the west side. Kitchin (1771) shows the same road, and passing Balumnock, which ties its course down a little more.


It is clearly shown running between Knockdaw and Carleton and then north of Knockdolian to cross the Stinchar a mile or so from present Ballantrae. From there, it took a direct route to present Smyrton and then ran down between Carlock Hill and Milljoan Hill. This suggests it was very close to the present road. It then went to Mark but it is not clear where this was. There is a Mark in Glenapp, but his Mark is far to the south of Glenapp so this is unlikely. The route was probably that from Altimeg over to the Dalnigap area (NX130710) where there is a present day road that meets the coast just above Stranraer.


Roy's road south of Girvan kept to the coast as far as Ardmillan and then ran inland. As it also runs between Knockdaw and Carleton and goes north of Knockdolian Hill, it may be the same road though this is not absolutely certain. Roy crosses the Stinchar nearer to Ballantrae.


Bowles 1735

He has a road from Kilmarnock to Glasgow but oddly enough nothing to Ayr. (15) Its course past Fenwick and the Black and Brother Lochs, as well as Mearns, suggests strongly that it is the same as that shown on Roy and Armstrong. This would give it the line of the present A77 to Eaglesham road end and then along the minor road at Kingswell.


There was also a road up the Irvine Valley. This kept south of the river through Riccarton. Somewhere before Dallars it forked, with one branch continuing to Mauchline and the other crossing the Irvine before the confluence with the Cessnock. It then ran towards Loudoun Castle. This seems a circuitous route as the direct distance between Riccarton and the crossing point is about three miles, whereas the route shown is at least four miles. This may just reflect the presumed marshy state of the flood plain here. There is no indication of this route on Roy, although Armstrong has a road north of the Irvine.


From Loudoun, the road ran wholly on the north side of the Irvine, presumably close to the present road (shown on Roy). Bowles shows a crossing over the Glen (as does Pont) but his road crosses nearer the Irvine. Interestingly he continues his road into Lanarkshire on the north of the river, whereas Roy has the southerly sweep past Gorsebraehead.


The Mauchline road would fit the present minor road to Carnell. It then ran down to the west of Loch Brown, which lay southeast of Crosshands and so to Mauchline. Although the present road from Carnell to Rottenrow seems to fit the line, there is no readily identifiable track beyond this. As Roy doesn't show a road in this area, the present road may be more recent. Rottenrow is indicative of an old road but not necessarily the one in question.


Although this is the likely line, one puzzling feature on the map suggests a different route, viz. a depiction of a hill that could be Craigie Hill - the road runs just east of this to Adamhill (on the A719, Galston to Ayr road). Roy has a road running past Craigie to near Adamhill where his Irvine to Dumfries road is picked up which would take it to Mauchline. Additional evidence would be needed to say definitely which is the correct line.


South of Mauchline, it clearly heads for Cumnock and New Cumnock and the county boundary. It is practically impossible to tell from the map what its exact course was but it seems consistent with Roy, i.e., a crossing at or near Howford then a similar line through Auchinleck and Cumnock (neither is shown on the map). It also seems to have the same line as Roy south of Cumnock through Borland to New Cumnock where the distinctive change of direction is shown. However, his road runs north of the Nith for a couple of miles unlike Roy whose road is south of the river.


A road from Ayr is shown joining the above road at Cumnock. It appears reasonably consistent with Roy (and Armstrong) although he has a stretch north of Sundrum which would take the road north of Roy's line (effectively the A70) for three or four miles. His mapping here looks inaccurate, so this may just be a mistake.


The road south of Ayr is interesting because it does not go through Maybole. Rather it crosses the Girvan near Crosshill, and then runs south of the river to Girvan. Having said that, it cannot have been far from the other routes to Maybole as it passed through Auchindrain keeping to the west of the Doon to go east of Lochlands and just west of Dalduff to cross the river somewhere just south of there. It then ran directly to Kilkerran, where there is a distinctive turn that took it through Bargany and Dailly to Girvan. For a mile or so north of the Kilkerran to Girvan stretch, the line seems very similar to Roy and Armstrong, and is likely to be the same road.


South of Girvan, it goes to Ardmillan and then keeps fairly close to the coast. It is hard to say if it is identical to Roy's coastal route but it cannot have been far from it.


Beyond Ballantrae, there is a significant difference from other routes including the Roy/Armstrong route, which went up past Auchencrosh to run down Glen App. Bowles shows a more direct route passing just east of Glendrishaig (NX056762), crossing the App somewhere near Haggstone and then running inland (perhaps two or three miles from the coast) towards Stranraer. It does not appear to correspond with Roy's three routes over Haggstone Moor south of Glenapp, though it must have crossed them. The stretch south of Ballantrae presumably kept to the ridge as far as Glendrissaig, and then to the west of Penderry Hill.


Bowen 1747

Bowen's map is similar to Kitchin as it shows the Ayr - Kilmarnock - Glanders route and a route south from Drongan to Newton Stewart.(16) Kirkbride and Blair are shown, which along with the generally straight line of the road, suggests strongly that it went down to Patna, followed the present hill track to Straiton and then made its way over to Balloch and then south.



Kitchin 1749

Two roads are shown on this map, one a main north-south route through Ayr, the other south of Stair through Drongan. (17) The stretch north of Kilmarnock is of interest as the road is shown going to Glanders in Renfrewshire as on Moll's earlier map of 1725. Unfortunately, there are no obvious clues that would let us say the routes are identical and in fact, the routing past Blacklaw is different on each map. However, it is clear that there is a "lost" road or roads here which took a fairly straight course between Kilmarnock and the Neilston area.


The route to Ayr looks very similar to Moll, i.e., going through Craigie, as does the stretch to Maybole.


The Drongan route is probably the same as the one shown on Bowen. If so, it is of interest that it is shown going west of the Water of Coyle to a place marked as Caris C(astle). As this is shown on the Doon, it is probably Keirs just south of Waterside rather than Kerse, a couple of miles west of Littlemill. This takes it off the line of the Patna to Straiton track but it would still be feasible to reach Straiton. If from there, one links it to Bowen's map which shows Kirkbride and Blair, there is the interesting possibility that it is the same as the "lost road" shown by Roy which ran from the Nick O' the Balloch road near Drumyork towards Knockgardner (and ultimately Straiton).


Ainslie's 1782 map of Wigtonshire has a short stretch of coast road from Stranraer up to Glenapp where it crosses the river at Finnart. (18) There is also a road running from Cairnryan up to Little Laicht and then up to Haggstone Moor. A main road is shown running north from New Luce that splits into a Girvan and a Maybole road. The Girvan road is the Beneraird track that ran from the New Luce to Barrhill road up to Lagafater Lodge and Beneraird - it is clearly shown on O.S. maps. The other road continued to Barrhill. From Glenwhilly (NX170710), it ran about 300 metres east of the present road as far as Miltonise. North of here, it ran a few hundred metres to the west of the modern road to take up the modern line north of Chirmorie. It is the same road as that shown on Roy.


Commissioners of Supply

Although dealing almost exclusively with bridges we must not overlook the clues that these records provide. The bridges mentioned above, which date from 1715 onwards and possible associated routes, are:

Doon 1715 onwards This is the bridge at Alloway on the old route south to Maybole

Barskimming 1715 onwards "profound ruining and falling down". This is on the route Roy shows between Ochiltree and Mauchline

Fail 1718 This was probably on the Monk's Road from Mauchline over to Fail

Coyle 1719 This was on the Ayr to Cumnock route

Barbeth 1720 "falling and utter decay". This was presumably on the hill road from Dalmellington over to Craig south of Straiton

Hudostone 1720 repairs This may be the present day Hoodston Bridge east of Hurlford

Killin repairs Not identified

Galchoin repairs Not identified

Carriosbridge repairs Not identified

Riccarton proposed 1723 This is the old Riccarton Bridge. It is more than likely that there was a made road up to Kilmarnock and some way south of it

Kirkmichael proposed The road shown on Roy runs through Kirkmichael from Maybole over to a route running north to Dalrymple

Dalrymplekirk proposed This is on the north-south route just mentioned, a short stretch of which still remains

Colmonell proposed, built in 1745 This is on what later became a turnpike route from Girvan to Stranraer

Bridge near Renfrewshire 1724 mention

Halfwater 1724 mention

Bridges over Gogo, Rye, Stobostoun mentioned 1725 The Gogo bridge was likely to have been on Roy's route near Largs; the Rye bridge is problematic as Roy has no road north of Dalry and the town itself is south of the river. A possible candidate is an old bridge mentioned in the minutes of Ayr County Council's Highway Committee in 1934 that was on an old route up the Rye Water. Stobostoun not identified.

Barr (Kirk) proposed 1728 This was on the route up to Old Dailly

Sares not identified

Clarks Bridge This is on the Beith to Johnstone route shown on Roy

Corses (in parish of Beith) This is very probably Kerse on the present Kilbirnie to Lochwinnoch route. At the time, Roy shows no road here. The crossing would have been of the Maich Water

Maxholm repairs not identified

Lugtoun repairs Roy doesn't show a road at Lugton, although if the Lugton Water is meant it could be the Dunlop - Barrmill - Beith road shown on Roy.


Burghs and parishes

The earlier references to the responsibilities of the two burghs of Kilmaurs and Maybole point to what must have been similar arrangements in each burgh and parish in applying the statute labour and other requirements, however half-heartedly this might have been carried out. Indeed, we can surmise that the towns were probably better organised and more determined to carry out the work than country parishes, although much of it must have been focused on the town streets and perhaps a short way out of town. There are in fact a few suggestive Gateheads near towns, which may indicate where a made road terminated.


A number of new burghs were formed at this time, often in places where there was already an unofficial market. These were:

Dalmellington 1617

Kilbirnie 1642

Girvan 1668

Tarbolton 1671

Fairlie, Dundonald, Riccarton and Montgomeriestoun were parchment burghs, existing only on paper.


The four working burghs may well, like the other burghs, have taken their responsibilities quite seriously. However, the deficiency in records means we cannot say anything about what they did except what can be inferred from other sources. One thing that is definite is that they must have fulfilled a need for trading and so helped to establish or reinforce a network of local roads or tracks. This also applied to places which, although not burghs, had been given the right to have a fair or market. These were:

Ochiltree 1669

Dalry 1681

Dalvennan near Straiton 1685

Straiton 1695

Cocklebee in Stewarton 1707

Doghillock in Fenwick 1707

Galston 1707

Riccarton 1707.

Other, "unofficial", markets and fairs were reported from Beith, Kilwinning, Stewarton, Barr, Colmonell, Dailly, Kilbride, Kirkmichael, Muirkirk, Sorn and Stewarton. (19)


There is some hard evidence for roads that may date from this time. The Dalmellington to Ayr road has already been mentioned and there is likelihood that some of the roads round Ayr and Irvine would have been improved. There is a reference of 1696 to work being carried out on a road between Kilmarnock and Irvine, and the roads found by Clark and Wilson near Carnell seem to date from this period.


Estate Roads

Another impetus for road building came from landowners whose interests in agriculture and mining, as well as amenity, could only benefit from improved transport.


The Earl of Loudoun began to build a series of roads on his estates near Galston as well as a bridge over the Irvine in the mid 1730's. The Marquis of Bute and Lord Auchinleck made similar improvements later. Smaller landowners also played a part in this and the Statistical Account mentions several, e.g., in Kirkmichael and Sorn.


Drove Roads

Unfortunately, J R Haldane in his book the Drove Roads of Scotland doesn't deal with Ayrshire. There were certainly drove roads that led outwith the county to join the main routes south into England which may date from this time. Roy in 1750 shows one leading from New Cumnock over to Crawfordjohn and the Old Statistical Account of c.1795 refers to cattle being sold for driving to the English markets in the following parishes: Barr, Colmonnel, Coylton, Dailly, Girvan and Kirkmichael. Other parishes sent cattle to local markets such as Ayr, sometimes further afield as with Galston sending calves to Edinburgh and Largs sending cattle to Greenock. Muirkirk had 14,000 sheep in the parish and sent flocks to Linton, Lanark, Carnwath, Kilbryd and sometimes Glasgow and Edinburgh. No doubt a number of hill tracks were used to join the main routes but the existence of tolls for cattle indicates that turnpikes must also have been used when this was unavoidable.




Another source of some help is folklore where routes are mentioned in the stories.


A good example is Sawney Bean who with his wife fathered a family, which because of incestuous relationships grew to a large size. They lived in caves near Bennane Head from whence they would emerge to waylay travellers, rob them and then eat them. They are thought to have been responsible for 1000 deaths. The whole family was hunted down by James VI himself, taken to Edinburgh and executed, including the children. It is clear from the story that this route was well used in the 1500's although no doubt travellers took wide detours because of the dangers.


Robertson, in Old Ayrshire Days (20) tells the story of Maggie Osborne, the Witch of Ayr who lived in the later part of the 17th century. Maggie used to fly all the way from Ayr to Galloway for meetings beyond the Nick O' the Balloch accompanied by little imps whose feet were so hot they burned up the grass on the top of Carrick Hill and marked a highway known as Maggie's Gait to Galloway. There is a bridge north of Balloch known as Maggie's Bridge.


Another tale is of Michael Scott the Wizard (The Deil O'Ardrossan) persuading the Devil to build a bridge from Cumbrae to Hunterston Point. However, the Devil couldn't stand being looked at and when someone saw him at work the bridge, nearly at completion, fell down into the sea. (21)


There is a folk tale about the Murder Hole near Rowantree in the vicinity of Loch Doon. The area gained a bad reputation when people started disappearing in a mysterious fashion. The locals themselves, left their dangerous isolation on the moors leaving only one old woman and her two sons who couldn't move. The story goes that a pedlar youth in danger of being benighted (i.e.stranded overnight in open countryside) remembered the warm welcome he had received at the old lady's house the year before - so warm that they'd implored him to stay a little longer after the other visitors had left. Again he was welcomed and put up in a room but on hearing strange noises he spied on the woman and her sons and witnessed another traveller being murdered. He fled and reached safety. Under questioning, they admitted to killing nearly fifty travellers and flinging their bodies down the "murder hole" nearby. The location seems to have been near the Nick O' the Balloch and the pedlar reference suggests a 17th century date. (22)


It is said that James V was going to a wedding at Sorn and stopped at an inn or cottage to water his horse; the place taking the name of Kingswell. This confirms the route over the Fenwick moors, or perhaps one through the Eaglesham area.


There is a reference from the Fenwick Parish Records for 1693 which confirms the Irvine Valley route. (23) A John Steill appearing for "driving some kine to a farm in Strathavon, did carie most insolentlie, and upbraided the Session, instead of giving anie suitable confessione and acknowledgement of his sin."



As we have seen, evidence for this period is poor so that although Roy shows what must have been nearly all made roads and significant tracks at the end of the period we cannot say with certainty when they developed. Many were clearly mediaeval and no doubt had been improved and so justified being called a road. Others may have been minor tracks at that time but through greater use and improvement were becoming important enough to be shown on Roy. Others again may have been conceived and constructed entirely after 1600. In their differing senses, these are all post mediaeval.


There is no doubt, however, that from the 1600's onwards people were improving tracks and building roads. The introduction of legislation, statute labour, bridge building and the other factors we looked at are evidence of this. Indeed, a large number of turnpikes follow the line of the roads shown on the Military Survey maps of the 1750's which suggests that these were sufficiently made up to be adopted as turnpike routes. The implication is that they may well have been built from 1600 onward, though of course some may have been earlier. Certainly by the end of the period, the process of road building was well established as shown by the changes occurring in the twenty-five years between Roy and Armstrong, finally culminating in the development of the turnpikes.



1. for further details of this and subsequent legislation see James Ferguson, The Law of Roads, Streets and Rights of Way, Bridges, and Ferries in Scotland, Edinburgh, 1904, pps 105-110.

See also Donald G Moir, The Roads of Scotland, II. Statute Labour Roads: The First Phase, Scottish Geographical Magazine, Volume 73, Issue 2, 1957, Pages 101 – 110. His account of the effect James VI had on Scottish roads after accession to the throne of England is interesting.

2. James MacDonald, Notes on the "Roman" Roads of the One-inch Ordnance Map of Scotland, PSAS, vol.xxvii, 1893, pp.417-43

3. Clarke and Wilson, A Possible Roman Road between the Rivers Cessnock and Ayr, AANHS Collections, 2nd Series, Vol.5, 1959

4. Robertson, Rural Recollections, Irvine, 1829

5. D McNaught, Kilmaurs Parish and Burgh, Alexander Gardner, Paisley, 1912, p.251

6. CO3/1/1 - CO3/1/5, Ayrshire Archives

7. CO3/1/1

8. A.L.Taylor, The Braw New Coat - The Building of Ayr New Brig, AANHS Collections, 2nd Series, 1961

9. W Walker, History of Irvine - Collection of articles which appeared in the Kilmarnock Standard, see copy in Carnegie Public Library, Ayr

10. A McJanet, Royal Burgh of Irvine, Civic Press, Glasgow, 1938

11. Rev. John Warrick, The History of Old Cumnock, Paisley, 1899, p.322

12. John Speed, The Kingdome of Scotland, 1610

13. Green R and Berry W, A New Map of Scotland, 1679 (search for Green)

14. Moll H, The South Part of the Shire of Air, containing Kyle and Carrick; The Shire of Renfrew with Cuningham. The North Part of Air, Bowles and Bowles, London, 1745. See also H Moll, A Pocket Companion of ye roads of ye north part of Great Britain, called Scotland, London 1718

15. Bowles C, A new and complete map of Scotland with roads forts and military ways, 1735

16. Bowen, A New and Accurate Map of Scotland or North Britain, 1747

17. T Kitchin, Map of Coningham being the north part of Ayrshire; Map of Kyle and Carrick being the south part of Ayrshire, Geographiae Scotiae, 1749

18. John Ainslie, A Map of the County of Wigton, Edinburgh, 1782

19. G S Pryde, The Burghs of Ayrshire, AANHS Collections, 2nd series, Vol. 4, 1958

20. Robertson, in Old Ayrshire Days, Stephen and Pollock, Ayr, 1905

21. George Robertson, Rural Recollections, Irvine, 1829

22. Rev. C H Dick, Highways and Byways in Galloway and Carrick, reprinted 1972, EP Publishers

23. Alfred C.Jonas, Extracts from Fenwick Parish Records 1644 - 1699, PSAS, volume 46 (1911-12) p.48

see also Ayrshire Place Names Index 1775 - 2000, Compiled by J & E.W.Steel, 2001 (microfiche)




Next (The Pre-turnpike Period)