Jump to ContentJump to Main Navigation
A History of the Native Woodlands of Scotland, 1500-1920$

T.C. Smout and Alan R. MacDonald

Print publication date: 2004

Print ISBN-13: 9780748612413

Published to Edinburgh Scholarship Online: March 2012

DOI: 10.3366/edinburgh/9780748612413.001.0001

Show Summary Details

Woodland as pasture and shelter

Woodland as pasture and shelter

(p.102) Chapter 5 Woodland as pasture and shelter
A History of the Native Woodlands of Scotland, 1500-1920

T. C. Smout

Alan R. MacDonald

Fiona Watson

Edinburgh University Press

Abstract and Keywords

This chapter discusses the use of woodland for pasture and shelter. For many tenant farmers and cottars, especially in the Highlands and the Uplands, the main use of wooded areas was to provide pasture and shelter for stock. It is quite impossible to find a Scottish wood, either in the Highlands or the Lowlands, at least before the nineteenth century, from which domestic animals were excluded, except sometimes on a temporary basis. The woods evolved with grazing stock, and the distinction between a wood and a wood pasture is functionally meaningless.

Keywords:   woodland, wood pasture, tenant farmers, cottars, grazing stock

As some of the examples of wood use in the last chapter remind us, there was more to woodland than the production of wood. For many tenant farmers and cottars, especially in the Highlands and the Uplands, the main use of wooded areas was to provide pasture and shelter for stock. It is quite impossible to find a Scottish wood, either in the Highlands or the Lowlands, at least before the nineteenth century, from which domestic animals were excluded, except sometimes on a temporary basis. The woods evolved with grazing stock, and the distinction between a wood and a wood pasture is functionally meaningless. The latter term (not used in our period) has come to mean a very open wooded area with 20 per cent or less canopy cover where animals are grazed, but almost all woods were essentially pastures, and certainly all woods provided shelter. Of course, if any wood was too thick it might provide valuable shelter in a Scottish winter but not very satisfactory pasture. As Oliver Rackham reminds us, ‘most shade-bearing grasses and herbs have little nutritional value, and some are inedible or poisonous. Domestic livestock love tree leaves, but cannot climb for them.’1 On the other hand the presence of trees on the ground, especially of birch, undoubtedly enriched the soil, and where trees were relatively thin on the ground and there was plenty of light, there was often a flush of excellent grass which contrasted with heathy land beyond the woodland edge. Woods which were under little pressure from animals in summer were often valued as a source of hay.2

The presence of domestic stock in a wood modified its character by their grazing and browsing, and by their trampling and pushing, but in the original wildwood there had always been animals, deer, boar and auroch, who kept grasslands open and perhaps, as Vera suggests, restricted the woods to groves and thickets so that they never became an enveloping cover (see above pp. 11–12, 33–4). Over the centuries, the auroch and boar (p.103) were extinguished, though the deer remained: cows, horses, sheep and goats replaced them. The character of the woods continued to be modified by their grazing, as the animals ate up or trampled away a greater or smaller proportion of the young saplings. Most woods would thus have come to have rather an open character, despite occasional descriptions of very dense woods (see above, p. 73). This process could of course be checked by enclosure or regulation if the landowner so decided, but very often the maximum advantage to everyone must have appeared to be a regime where there were just enough trees in a wood to provide shelter, local timber and leaf drop to replenish the ground, but not too many trees to shade out the best grasses on the pasture.

Such open woods would have had spreading trees. The tendency of Scots pine in Scotland to form picturesque, open-crowned shapes (‘Granny pines’), compared to the tall upright growth more characteristic of Scots pine in Scandinavia and elsewhere in its range, was a consequence of the trees growing unobstructed by neighbours. It has sometimes been assumed that this was due to foresters leaving, after a clear-fell, series of isolated and perhaps already twisted trees to shed seed for the next generation. It is at least as likely to be a consequence of the pines growing well apart in heavily grazed woods.

Under upland conditions there was likely in any case to be a natural continuum from a relatively thick wood, albeit with rocky or boggy openings, on the valley floor, to a few trees thinning out on the open moor of the upper slopes, a tendency that would have been considerably reinforced by the use of the wood by domestic animals. Changes since 1800 have almost everywhere eliminated the gradation to a natural tree line, as can be seen for instance from an estate map of Eliock Wood in Dumfriesshire in 1767 which showed how the wood spread up the deep cut burns to 600 feet in altitude, with an outlier at 650–700 feet, spreading to 900 feet up one burn, after which the draughtsman showed first small trees then scrub to 1,000 feet. By the time of the First Ordnance Survey Map of 1856, the main wood had been fully enclosed, with the upper margin made straight, and those patches of the old wood left outside explicitly described as pasture.3 As noted in Chapter 4, it often created a problem for cartographers and surveyors as to the point at which they defined such ground as wood rather than as grazing with a few trees. Similar conundrums elsewhere have led to large inconsistencies in classification – Rackham cites such a case in Greece, where ‘official statistics of the extent of dhásos [open grazed woodland] in (p.104) Crete vary from four and a half per cent of the island in 1981 to thirty-three per cent in 1992’.4

To the landowner and the tenant, however, the more material question was whether the ground had greater value as pasture or as wood. The answer initially would depend on the interplay of three variables, the density of the trees and the rates of return on the timber and on the grazing, which in turn, would depend on other variables, the extent of local scarcities of wood and pasture, and the strength of external demands for timber products and animal products.

Sometimes it was perhaps possible for land users to enjoy the best of all worlds, if resources were not scarce. Relatively unrestricted grazing might both leave enough grass for animals and timber for people, if the balance was right. There are times in the seventeenth and early eighteenth centuries when local descriptions give a sense of a world delightfully filled with plenitude. Thus at Conaglen in Ardgour (perhaps around 1630) there was:

Great number of firr trees in this glen. And it is verie profitable to the Superior and Master of the countrie for it is good to feed guids [cattle] therin being of twall mylls of length or therby, and there is a water in the glen which doeth transport great trees of firr and masts to the seasyde.5

Similarly Glen Moriston was ‘a very profitable and fertill little glen … plenteous of corne and abundance of butter, cheeses and milk and great and long woods of firr trees’.6 At Cabrach in Aberdeenshire ‘there is a great wood in Old Doveran and a Forrest where there is frequent resort of Deer, Roes, Heathfowl and other Game, and which with Cattle, Sheep, Goats, Butter, Cheese and Wool are the Commodities of the Place’.7 In the 1680s, the Earl of Airlie had ‘a good interest’ in Glenisla ‘with two great woods called Crandirth and Craigiefrisch he has a large Glen for grassing with abundance of hay meadows with a free forrestrie which in those places they reckone much worth’.8 These are examples of co-existence between extensive woods and herds of animals, though how it was managed and whether it was sustainable in the long run remain obscure. Perhaps the (p.105) situation was not always as idyllic as it seemed, as at Ardgour the woods had been very much reduced in extent by the end of the eighteenth century.

The viability of any managed system without enclosure would depend on the relative abundance of trees and animals and the quality of herding. The number of animals could vary quite sharply if the human population rose, so that proportionately more subsistence animals were kept, or if, because of a growing external market for meat, hides or wool, more animals were kept with the same or smaller human population. Sutherland was another district where in the mid-seventeenth century trees and animals seemed to thrive together, the straths being ‘replenished with woods, grass, corns, cattell and deer, both pleasant and profitable’.9 But it was here that, around 1800, and therefore before the main clearances or any intensive commercial farming, but under circumstances of a rapidly increasing population of subsistence farmers, there occurred the clearest evidence of woodland collapse with apparently serious ecological and economic consequences (see below, pp. 119–21).

Very frequently, as at Eliock, there was worked out, in the long run, a compromise between the need for pasture and the need for wood, in which areas where the wood was thin were defined as pasture and eventually lost most of their tree cover, and areas of thicker wood nearer to transport or to habitation came to be enclosed and protected from stock, at least for part of the time. The outcome sometimes was a characteristic mosaic of small woods close to houses or farmtouns and scattered trees on the lower part of the hill and other woods or ribbons of tree cover surviving on screes or rocky slopes at a distance. Such a pattern is well exemplified in Assynt.10 Grazing lowered the natural tree-line as the upper part of the hill was totally abandoned to animals.

A further complicating factor was the existence of hunting reserves, ‘foresta’, in which game was preserved for the Crown or the nobility and from which domestic stock were often partially excluded, perhaps sometimes to the benefit of tree cover. In practice most ‘foresta’ would have been regularly grazed, even in the Middle Ages.11 By the seventeenth century the royal forests had been made over to private ownership. The nobility (p.106) nevertheless often enjoyed defined hunting rights in the hills, for example the Earls of Mar in Deeside, the Earls of Atholl in northern Perthshire, the Earls of Moray in Glen Finglas, and the Earls of Breadalbane in the Forest of Mamlorn, where their interests conflicted with those of local graziers.

The early form of deer hunt involved driving the animals with beaters and dogs through woods and open moor, into an ‘elrig’, a natural trap or man-made enclosure where they were shot with bows and arrows: James IV on his jaunt in Atholl in 1528 was said by Pitscottie to have killed thirtyscore red deer, and ‘uther small beistis, as ra and rebuck [roedeer], wolf and fox’. The total may have been inflated for effect on this occasion, but the slaughter (and cruelty) was always great.12

Increasingly more fashionable than the chase and ambush of deer on the open hill, however, especially in the Lowlands, was the park close to the great house. At Cadzow and Dalkeith this has produced famous late medieval and seventeenth-century survivals of Lowland wood pasture that were originally hunting parks (Fig. 5.1).13 There must have been many more. Falkland Wood was a royal deer park until the English let out the animals and cut down most of it in 1653 to build Perth Citadel. It lingered on to be shown in attenuated form on Adair's map of Fife in 1685, but had gone by start of the following century.14 Its site (on good brown forest soil) is now a substantial farm. There was perhaps a narrow line between a grand hunting park and a modest ornamental park where the laird surrounded his tower house, or later on his mansion, with scattered specimen trees and grass for fallow deer or cattle. When Abercumbrie wrote of Carrick in Ayrshire that ‘every Gentleman has by his house both wood and water orchards and parkes’ he was probably referring to the ornamental rather than the utilitarian.15 The maps of Pont, Gordon and Blaeu all show fenced wooded enclosures round the big houses. All parks, however, were wood pastures in which animals and trees existed together in a controlled environment, and the balance between economic use and pleasure could be varied according to circumstance.

From the early seventeenth century there occur other references to parks (p.107)

Woodland as pasture and shelter

Fig. 5.1 Cadzow's medieval hunting park, Lanarkshire, with trees dating at least to the fifteenth century. They appear to have been planted, or allowed to grow, on still older cultivation rigs. Anne-Marie Smout.

that were not obviously ornamental or merely for amusement, but probably implied a calculated combination of economic grazing and timber production. The records of the Campbells of Glenorchy provide the best evidence of this, with a whole series of parks referred to in the records of Baron Courts held at Killin, Finlanrig and ‘Crannich’ on Lochtayside as early as 1617 and 1618. We have little idea of how they were managed, apart from a determination to keep them free of the tenants' goats and pigs.16 It is possible that, like other parks constructed in Galloway late in the seventeenth century, they were intended partly as holding and fattening enclosures for black cattle destined for the drove south but we have no direct evidence of that. The most famous of the Galloway parks was that of Sir David Dunbar of Baldoon, Wigtownshire, from which in the 1680s he sold hundreds of cattle into England. It was said to be an enclosure two and a half miles long and a mile and a half broad.17 No doubt the trees within (p.108) this park were intended primarily to shelter the animals rather than to produce wood, but with pollarding the two uses were not mutually exclusive.

By the end of the sixteenth century, and probably in many places centuries earlier, the majority of woods in the Lowlands had become enclosed in order to protect timber supplies, and were the objects of carefully considered woodland management (see Chapter 7). Even these, however, were regularly open to grazing by animals, with the ground subject for two-thirds or three-quarters of the rotational cycle at least to some seasonal grazing. As late as the 1790s the haggs of enclosed coppice oakwoods in Argyll and Perthshire, managed on a twenty to twenty-five year rotation, were normally given from five to eight years' exclusion from cattle. In the woods of the Duke of Montrose, the annual value of the grass was normally less than 10 per cent of the annual value of the wood and bark, but at Cruixton and Mugdock it was worth over 30 per cent.18 Sometimes the modification of a long-enclosed wood in the interests of the stock could be considerable. Alexander Wight, ca. 1780, told of how Mr Steele of Gadgirth in Ayrshire cleared out the undergrowth of a natural wood that had become so overgrown that it ‘left no possibility of pasturing cattle in it’, and by clearing, burning and thinning, turned it into very good pasture: ‘the remaining trees are in a more thriving state than formerly and the field is very beautiful’.19 There was in fact no wood in Scotland unaffected by the grazing of domestic stock to a greater or lesser degree.

Grazing pressure was of three descriptions – summer only, winter only, and throughout the year. Summer grazing was mainly on the high ground, associated in the Highlands with the shieling system whereby animals were taken in May to summer pastures where herders tended them for several months, eventually returning to the lower fields from which crops had been harvested.20 Allocation of grazing was by ‘souming’, whereby each tenant was allowed a certain number of animals on the hill. Edward Burt in the early eighteenth century said of an area near Inverness:

A soume is as much grass as will maintain four sheep: eight sheep are equal to a cow and a half, or forty goats … the reason of this disproportion between the goats and the sheep is that after the sheep have (p.109) eat the pasture bare the herbs, as thyme, etc. that are left are of little or no value except for the brouzing of goats.21

There was, though, great variation. In other instances, ten sheep were reckoned to a cow, or a horse was considered equivalent either to two cows or eight sheep.22 It does not follow that there were necessarily about twenty-five goats, or four to ten sheep, for every cow on the hill: cattle were essential to pay the rent, and were often likely to be preferred alone. In late eighteenth-century Assynt, for example, the sheep-to-cattle ratio varied on ten farms from 4:5 to 1:5, with a mean of 1:3 in favour of cattle.23 Sheep, goats and horses were certainly important, but the eighteenth-century Highlands were seen essentially as a cattle-raising economy. Yet it was widely agreed that the hills were grazed far below what their carrying capacity would have been if the stock could have been supplied with ample feed in winter.24 As James Robertson expressed it in Glen Avon:

A man who pays £3 sterling of annual rent will perhaps have 20 black cattle, 3 or 4 horses, 20 sheep and 10 goats. During summer and autumn the pastures could maintain thrice the number, but they would perish during the winter or spring.25

This problem was solved when the turnip arrived in the Highland straths in the nineteenth century, and ever since there have been very much heavier grazing levels than previously on the uplands. The switch from a cattle economy to a sheep economy, that began in the southern edge of the Highlands around 1760 and swept the whole area in the first half of the nineteenth century in the period of the Clearances, exacerbated the problem by making dominant on the hills an animal that grazed very closely and destructively.

Shielings had at one time been common in the Southern Uplands as well as in the Highlands, until they were replaced by permanent shepherds managing sheep flocks all year. By 1500 shielings were rare in the Lowlands, (p.110) though they may have continued in some areas south of the Forth into the sixteenth century, as on the moorlands between Clydesdale and West Lothian or in the Ettrick Forest.26 They remained common throughout the Highlands in the seventeenth and eighteenth centuries, and only died out on the Isle of Lewis in the twentieth century.27 As population pressure increased, especially after 1700, some Highland shielings began to be cultivated, and even to become permanent, if marginal, settlements: this was happening before 1750 in the upper parts of the Dee, in Glen Shee, Glen Tilt, in Atholl and generally in the Tay catchment area.28 Wherever this took place, environmental change was accelerated by the reinforcement of grazing pressure by cultivation pressure.

Descriptions of the impact of summer grazing on trees are hard to find, perhaps in some cases because it had been going on for so long that in most localities few trees were left on the shielings themselves. They are not, however, completely lacking. In 1596, Sir Duncan Campbell of Glenorchy was given liberty from Privy Council to destroy unauthorised summer shielings in the Forest of Mamlorn in Glen Lochay. Graziers, apparently from Glen Lyon:

yeirlie in the sommer seasoun cumis and rapairis to the said forrest, biggis sheillis within and about the same and remains the maist parte of the sommer seasoun at the said forrest, cutting and destroying in the meantyme the growand treis within the said forrest and schuting and slaying in grite nommeris the deir and wyld beistis within the same.29

This was far from the end of the matter. In 1738 there were nearly 1,200 beasts grazing in and around the forest. The Earl of Breadalbane was determined to keep them out and wrote to his forester: ‘I as the king's heritable keeper, am in duty bound to preserve the forrest from all insults and you are to drive back all beasts out of the forrest as often as they come in there.’30 It seems unlikely that he had any lasting success: in June 1751 there was a ‘riot committed by the Glenlyon people in the forest of Mamlorn by pulling down the forester's house’.31 There are isolated trees in high Glen (p.111) Lochay now that show the same sort of evidence for low pollarding or ‘cutting high’ that is common in Glen Finglas.

Similar usage occurred in the Forest of Mar, though generally without the same levels of confrontation. In the interesting dispute over the forest between the Earl of Fife and Farquharson of Invercauld in 1758–60, which involved the tenant of the latter cultivating the ground next to the woods of the former, it was argued that the pinewoods gradually ‘shift their stances’ because they propagated by blowing their seeds onto the ground immediately next to the old woods, or into openings ‘where they have Freedom of Air’. It was further said that pines growing up on the moors beyond the woods ran little risk from cattle, because:

There is hardly any pasture on the moor adjacent to the woods, and the cattle are all grazed in the forrests, where there is the best outpasture for all kinds of cattle to be found any where in Scotland.32

The phrase ‘out pasture’ seems to imply that at least some of the animals in Mar were out among the pines in winter. There were many summer shielings in the Forest of Mar, some indicated in Farquharson's map of 1703, a number at the edge of the pinewoods but most along the edge of watercourses in open ground. More than 1,000 cattle were in the souming for the Forest in 1729.33 (See Map 5.1.)

The suggestion that cattle would not eat out the seedlings on the moor is undermined by a later eighteenth-century observation from Argyll. Here John Smith noted that the remnants of old pinewoods stood in the higher glens, and shed seed in winter which was driven to a distance by the storms: from these ‘a beautiful plantation rises up in the spring, but when the cattle are driven up to the mountains in summer, this precious crop, the hope for future forests, is for ever destroyed’.34 Yet he himself also mentions that, in the parish of Little Dunkeld in Perthshire, a ‘fir wood of three hundred acres has thus arisen from seed driven by the wind from old trees within these thirty years’.35 No doubt, as always, the truth varied according to the density and character of the animals on the ground.

Woodland on lower ground was potentially even more vulnerable, depending also on its age and degree of enclosure in addition to the other (p.112)

Woodland as pasture and shelter

Map 5.1 Distribution of shielings in the Forest of Mar. Royal Commission on the Ancient and Historical Monuments of Scotland, based on OS Map, Crown Copyright, in Mar Lodge Estate, Grampian: an Archaeological Survey (RCAHMS, 1995), p. 5, with permission.

factors. Young wood unenclosed was a meal for the taking. ‘To plant a wood without enclosing it’, wrote William Boutcher in one of the earliest Scottish forestry treatises (dedicated to the Duke of Buccleuch), ‘is not only amongst the idlest ways of throwing away money, but is laying up a fund of remorse and discontent that must necessarily occur from the devastation of cattle and sheep.’36 Bitter experience taught many a landowner that truth. Thus Sir Ludovic Grant of Grant was told by his factor in 1748 that of 100,000 (p.113) pines planted on Dulnain Brae, not 4,000 were left undestroyed by cattle and sheep. Sandy Fraser, the miller at Castle Grant had also been letting in his horses at night, but had not been stopped as, if he had been arrested, there would be ‘noe body to grind there corn’.37

Most of the damage, however, would occur to woods deliberately left beyond any fencing system because it was judged that their value to the tenant (and the rent-roll) as pasture outweighed their value as timber. It was this that stirred the anger of John Williams in 1784, when he spoke of the Gordon lands in Lochaber and the former Jacobite estates in Wester Ross, Inverness and Argyll, as containing ‘a great many thousand acres’ which:

form a rich stool of oak in a deep soil, where the most luxuriant shoots are produced in summer, while the goats are in the hills, but they are soon browsed down in autumn, and kept level with the heath, by the goats and other cattle, and if any plant chances to raise its head beyond the reach of the goats, it is soon destroyed by the axe of the Highlandman, who strips off about four feet of the bark quite round, a little above the root, and leaves the young tree standing, to die a lingering death, as a monument of his barbarous greed.

He includes among such lands Coigach, the north side of Loch Broom, Kinlochmoidart, ‘several thousand acres’ of the north shore of Loch Eil and ‘in the glens at the head’ of Loch Eil, in Glen Lochy and on the shores of Loch Lochy, around Loch Arkaig, ‘beautifully and richly covered with oak, birch and fir, where a good deal of oak is grown out of the reach of cattle’. He adds also the north shore of Loch Leven, with ‘a good stoop of oak’, Ardshiel (‘and a good deal peeps out of the heath on the braes of that estate in summer’), as well as parts of Speyside and ‘that extensive moor’ between Fort William and the Spey where ‘a thick stool of oak appears among the heath’.38

This highly important observation, by a man who had obviously travelled over a wide area with critical eyes, identifies the extent of grazing pressure on large expanses of shrubby woodland that was prevented from coming to maturity. Nor does it stand alone. A similar point was made by the Rev. John Walker in 1808 when he observed that the Highlands (p.114) contained plenty of vestiges of former woods: ‘the forest trees are still to be seen vegetating from large old roots; but what they shoot forth in summer, is eaten down and destroyed by the cattle in winter’. He lists the species: oak, ash, elm, birch, alder, holly, yew, rowan, hazel, whitebeam, goat willow, apple-leaved willow, grey willow, crack willow, bay-leaved willow, hagberry [bird cherry], water elder, blackthorn and whitethorn – all indicative of rich biodiversity. These, he says, have evidently formed in the past extensive woods and could do so again, ‘though now nearly obliterated’.39

A third commentator, specifically confining his remarks to Argyll, was John Smith, who in 1798 suggested that landowners wishing to cash in on the profitable contemporary market for oak products should turn their attention to ‘another kind of ground which ought to be planted, and of which we have large tracts’. This was at present covered with brushwood, such as hazel and birch ‘seldom allowed by the cattle to rise above two or three feet high’. Nature, he said, was never mistaken in the suitability of soil for wood, and sometimes such brush also already had sprouts of oak and ash. Besides, ‘patches of dwarfish oak, which the cattle never allow to rise one foot from the ground, are also common in many parts of the country’. All that land like this needed was a good dyke and a few extra acorns.40 It is important to realise that what is being described was not the outcome of extensive sheep farming such as overtook such land in the following century, but of a farming system still based on black cattle. Yet the coming of sheep naturally continued and exacerbated a process already under way. The words of Robert Monteath in 1827 are strikingly similar to those of John Williams forty years before:

It is as notorious a fact as the sun at noon day, that throughout most of the counties comprising the Highlands of Scotland, particularly the whole of Argyleshire, that there are millions of stools or roots of oak and some other kinds of trees detached throughout almost the whole of the extensive fields now appropriated to sheep pasture; nor is this to be wondered at, when we consider that many thousands of acres of land that was formerly carrying Natural Woods have of late years been left unenclosed and set aside for pasture lands; it is no less wonderful than true, that, the growths of these stools or roots, though devoured and eaten up in winter by cattle and sheep, are never wholly extirpated; as soon as the grass gets up so as to afford a supply of meat for the (p.115) sheep etc., the growth gets up, and so soon as the grass fails, the growths or saplings are eaten up.41

In the end, no doubt, even this scrub would vanish, certainly if muirburn was associated with the management of sheep pasture as it so often was. Yet, providing the pressure was not too intense, it is possible that such a system was sustainable, and a deliberate way of providing some sustenance to stock at a time when the grass had withered. In the absence of rabbits (rare in the Highlands before the nineteenth century) the stools could spring with new shoots year after year, and the soft growth of the season disappear once more in winter, as the writers described. Possibly it fulfilled a similar function to leaf hay in Norway or Sweden, where the winter was too cold for animals to be left out of doors in the Scottish manner. It must, though, always have been vulnerable to misjudgement and change: too many animals, and the stools were sure to die out.

Then there was the impact of grazing on mature woodland that was certainly also valued for timber production, but remained open to cattle in winter and sometimes to a limited degree in summer as well. The Forest of Mar we have already referred to. The arrangements at Comer Wood at the foot of Glen Cannich are explicitly laid down in an agreement of 1686, where the tenants were allowed to pasture the whole woods from 1 November until 5 May, with certain barren ewes allowed between 1 May and 1 September.42 This was probably fairly typical: Lindsay found that in Argyll and Perthshire most animals traditionally went to the hill grazings between 1 May and 1 November, but milk cows went later and returned sooner.43 Whether such arrangements prevented regeneration would surely depend primarily on the numbers of animals wintering – those left in summer would not normally be very numerous on the low ground. On the other hand, there were occasions when the summer grazing seemed to be important (Fig. 5.2). In 1758 the tenants of Craigroyston on Loch Lomond petitioned their landlord, stating that for eleven months in the year their only pasture was in the woods that were due to be enclosed, and that without them their tenancy would become impossible. Similarly in 1780 the tenants of Carnghouran in Rannoch claimed that if the Black Wood was enclosed they would lose their best pasture and only wintering ground. In both cases there was reference to summer as well as to winter pasture.44 (p.116)

Woodland as pasture and shelter

Fig. 5.2 Summer grazing of cattle among the woods of Lochlomondside, as illustrated in C. Cordiner in 1786. St Andrews University Library.

Agrarian experts often described open woods as pasture. Thus William Marshall explained that the townships on Lochtayside contained on average, within the head dyke, about twenty acres of infield, fifteen acres of outfield, ten acres of meadow, thirty-five acres of green pasture and ten acres of ‘woody waste’: he further explained that patches that are ‘too wet, too woody or too stoney to be plowed, are termed meadow, and are kept perpetually under the sithe and sickle for a scanty supply of hay’.45 On the estate maps of the area, ca. 1760, small woods are described sometimes as pasture, sometimes as woods, with little consistency or obvious reason.46 Similarly when George Langlands surveyed Kintyre (1770–7) he detailed all the small patches of woodland; for example, at Barrmolach in Carradale, a township of about 560 acres, he listed five little woods: (p.117)

5 acres – mostly birch and some hazel wood, midling grazing.

4 acres, 2 roods – birch, hazel with some alder, a midling good grazing.

11 acres, 1 rood, 10 perch – mostly birch, some hazel and trangent ash and oak – midling grazing.

9 acres – mostly birch with some hazel, tho' a midling grazing.

13 acres, 1 rood, 9 perch – mostly birch with some hazel, not very good grazing.47

Land like this in Argyll, if it contained oak, was quite likely to become enclosed and better managed, when, towards the end of the eighteenth century, the returns for tanbark and charcoal rose steeply. Tenants, however, were unlikely to see a benefit to their own pockets. Arrangements varied, but possibly few landlords were as extortionate as James Riddell, the new proprietor of Ardnamurchan, who in 1774 agreed with Donald Campbell, ‘tacksman of Ardnamurchan’, that if the proprietor enclosed any woods the tenant should receive a rebate equal to the value of lost grazing for five years, but then pay an increased rent at seven and a half per cent of the cost of enclosure when the grazing was restored to him. Nevertheless, Campbell did not own the trees so protected, and was obliged to provide grass for the horses needed to transport the timber produce.48

Apparently more consensual was the appointment of two independent assessors on the Breadalbane estate in 1759 to judge compensation to a wadsetter with grazing rights in the wood of Letters near Dalmally when the Earl wished to enclose it. The wadsetter's opinion of the operation nevertheless comes through in his receipt for the money:

Received by me Donald Mcintyre … two years rents of the big inclosur of Leathers which is laid wast for the preservation of the woods.49

Thus might a modern farmer receive his set-aside money.

The Duke of Argyll's estate was particularly careful in exploring which woods would repay enclosure and which should be allowed to remain as open pasture, depending on their compactness, species composition and access. Thus a survey of parts of Morvern in 1786 described all the local woods in detail and estimated appropriate rent reductions if enclosure was to be carried out. Some woods, like Ardtornish, were judged to be ‘in such rough ground and so difficult to enclose’ as not to be worth the expense, or (p.118) like those at Mechanach, ‘mostly good birch, some alder and very few oaks and ashes’, which ‘should be left open, as barren wood at any distance from the sea will not bear the expence of inclosing and deduction of rent’.50 Others were more suitable, as at Camusallach, where the woods were mostly ash and oak, with some birch, extensive and thriving, and ‘close on the shore’: it could form one great enclosure with the neighbouring wood, and rent deductions could be smaller if the tenant were allowed to cut hay in the wood and to labour with spades the patches of arable ground along the shore. Nevertheless, ‘it takes in a considerable part of Ardslignish's wintering’.51 There was some sensitivity to the needs of a working farm: at Gortanbeg it was necessary to leave a walled lane in the enclosure (‘a lonie’) so that the tacksman's cattle could ‘have access to the shore in time of snow and bad weather’ – though probably the farmer would rather have had the beasts in the wood than on the shore.52

Rebates of rent could be quite substantial. In 1699, when some ironmasters from Cumberland bought certain woods in the parish of Canonbie in Dumfriesshire, their contract specified that the landowner, the Duchess of Buccleuch, would pay damages to the tenants for grazings lost due to the enclosure of newly cut coppice. In the 1720s, such compensations could be as high as £18. 15s. sterling to a single tenant in one year.53 In the Highlands, the Duke of Argyll's chamberlain's accounts of 1775, list several payments to tenants for loss of pasture due to enclosure for wood, ranging up to £8 a year.54 Nevertheless, by the exclusions the farms were inevitably made less valuable: even if stock might be allowed back after five to seven years, there was a substantial problem of wintering them in the interim. Scottish practice compares badly with that in Denmark, where by a statute of 1805 all woods were to be enclosed and awarded to one owner, but landowners were also obliged to make available to farmers pasture equivalent to that lost in the forest.55

All, however, was not necessarily well for the husbandman even where the tenants got free run of the woods for their stock all winter. There are (p.119) some most interesting and significant instances in the northern Highlands, where, in the absence of any attempt to protect the woods, the quality of the land appeared to be deteriorating so notably that it imperilled the viability of the holdings.

The first of these was in Wester Ross, from Inverinate on Loch Duich in 1753, when Torquil Mackay, a wadsetter, wrote to the factor of the Earl of Seaforth in these terms:

As we in this and the neighbouring countries depend more upon the profite of cattle than that of tillage, we must studdy to have our grasings as good pasturre and as convenient as possible … I hear from sensible honest men that other places in the country besides my tack have now less wood and more fern and heath than formerly, so that cattle want shelter in time of storm (as we never house any) and their pasture is growing more course and scarce. I know of severall burns that in time of a sudden thaw or heavy rain are so very rapid that they carry down from the mountains heaps of stone and rubbish, which by overflowing their banks, they leave upon the ground next them for a great way and by this means my tack and others are damnaged, and some others more now than formerly.56

It was an explicit statement that, in a situation where animals were always outwintered, the woods were diminishing, pasture deteriorating and stream erosion and deposition becoming a bigger problem.

By the end of the eighteenth century there was a wide recognition that, in the north of Scotland, the decline of woodland was an agrarian problem. Sir John Sinclair in 1795, for example, spoke of the ‘natural birch woods’ as ‘much in decline’ and likely to give out within a generation or two, which would be a ‘melancholy situation’ for the people.57 John Henderson, writing of Sutherland in 1812, gave the fullest account of the phenomenon and its consequences. He spoke of the ‘remains of a shrubbery of birch, hazel, aller, willow and some oak bushes, in the straths of the several rivers and burns in the country’ that was ‘not of so great extent as formerly, and is rapidly decaying in some places’. He continued: ‘the natural woods on the several straths in this country to the southern, western and northern coasts are decaying fast’.58 (p.120)

He appeared uncertain as to the cause. ‘Naturalists aver’, he said, that it was owing to severe winter and spring frosts for many seasons past. At first sight the tree species that he names appear well adapted to far more extreme climates than Sutherland, and a run of freak cold seasons seems an unlikely explanation. On the other hand, if an increase in precipitation had coincided with an increase in grazing pressure, such is the unfavourable oceanic nature of the Scottish environment compared to Scandinavia that regeneration could fail completely. There can be little doubt about the grazing pressure. Henderson himself observed that ‘from the constant browsing of black cattle, it is not surprising that the oak is nearly gone’. He also described how until recent years, every farmer had had a flock of twenty to eighty goats.59

Whatever the cause, he was clear about the consequences:

It is a well known fact, that in the straths where these woods have already decayed, the ground does not yield a quarter of the grass it did when the wood covered and sheltered it. Of course the inhabitants cannot rear the usual number of cattle, as they must now house them early in winter, and feed, or rather keep them just alive, on straw; whereas in former times their cattle remained in the woods all winter, in good condition, and were ready for the market early in summer. This accounts for the number of cattle which die from starvation on these straths, whenever the spring continues more severe than usual: and this is one argument in favour of sheep farming in this country.60

He cites as support a letter from Alexander Sage, minister of Kildonan (but no advocate of sheep farming), who corroborated the decay of the woods as a ‘remarkable alteration on the face of this part of the county in the course of the last 20 years’. Sage adds a few different details: formerly the inhabitants ‘out-wintered their cattle till the beginning of January, whereas they must now house them in the beginning of November’, and when the wood was wasted away the ground became covered with coarse heather ‘in place of the fine strong grass with which the woods abounded’. This led, he said, to ‘a degeneracy of black cattle in the parts that were formerly covered with wood’.61 (See also Fig. 5.3.)

It was, of course, the north of Scotland that was least touched by the external market for wood products in the eighteenth century, partly due to (p.121)

Woodland as pasture and shelter

Fig. 5.3 A well-wooded glen at Strathmore in Sutherland, with a creel dwelling to the left of the ruined broch of Dun Dornaigil. The decay of the Sutherland birchwoods was a complaint of many at the end of the eighteenth century. From Charles Cordiner's Views, 1780. St Andrews University Library.

its geographical remoteness, but also because the woods there were predominantly composed of birch, hazel and other ‘barren timber’. Even in Inverness-shire where there was more oak, it was seldom enclosed, but left to the depredations of cattle, ‘all shamefully abandoned, after every cutting, in the same neglected condition’, in contrast to the situation of enclosed woods further south in Perthshire or Argyll.62 It was in the north, therefore, that the trees were most vulnerable to an unsustainable regime in using woodland as grazing: interestingly, it appeared as a priority to the new sheep farmers in Sutherland after the clearances that the woods should be rehabilitated, and initially they had some success (see above, p. 69). In the (p.122) south, forms of woodland protection were, at least for a time, more emphasised, but here the farmers had different problems.

It is not easy to get a long-term perspective on the impact of grazing on the history of Scottish woods. The first point to emphasise is that grazing was a natural component of the woodland ecosystem, ensuring much open land and much space between mature trees even in early prehistory. Auroch, wild white cattle and wild boar ultimately became extinct, but red deer and roe deer persisted and domestic stock replaced the vanished megafauna. In the Middle Ages, however, the numbers of grazing animals, particularly domestic stock, must have been checked in the uplands by the wolf, still reckoned a numerous pest in the fifteenth and sixteenth century.63 This was a situation in which montane scrub and upland woods, perhaps in the form of very open wood pastures, could have continued to reproduce themselves.

Between 1600 and 1900 a series of changes occurred to this situation. Firstly, climate was less favourable, especially in the seventeenth century, than it had been at the height of the Middle Ages: if it became more difficult for trees to regenerate, a large proportion of seedlings was at risk from being grazed off. Secondly, the wolf, now rare except in the northern Highlands after 1600, was eventually exterminated, possibly before 1700, certainly before 1750. The absence of any predator lifted the ceiling on the number of stock that could be kept, particularly for sheep and goats. The latter became extremely numerous in the seventeenth and eighteenth centuries, until eventually checked by improving landlords concerned at the damage they could do to woods. Thirdly, the population of people and therefore of their subsistence livestock grew steadily in the Highlands in the eighteenth century, and after 1740 the rapid increase in the price of black cattle must have tempted farmers to put more and more stock on the hill.64 This was the great age of the drove roads, with 60,000 beasts sold at the Falkirk Tryst by 1794.65 Consequently by the end of the eighteenth century, upland scrub and wood pastures show every sign of being under pressure, eventually becoming vestigial remains almost invisible in the heath. Conversely, woods at lower altitudes, long enclosed in the Lowlands, were now enclosed in Highlands, where access to markets for tanbark, charcoal (p.123) and timber justified it: domestic stock were not excluded, but were controlled and limited.

In the nineteenth century, despite the initial experiences in the Sutherland straths, the dominance of sheep and the possibility of keeping many more mouths on the hill once the problem of over-wintering stock out of doors had been solved by turnips, put paid to the chance of any long-term revival of montane scrub and wood pasture, especially when fire was used as the main tool to control and improve the hill grazings. As the price of charcoal and tanbark plummeted, and cheap timber came in from abroad, there was little point in keeping any regulations over stock in enclosed woods either, unless they were to be reserved as game shoots. There was a large drop in the amount of land with scrub and scattered trees on it, and the main cause for that decline must be overuse by cattle and goats in the first instance, and sheep in the second.

By way of postscript, it is intriguing that many of the areas where ancient upland wood pasture survives are in former royal forests, of which Glen Finglas is the prime example. Perhaps this is because in these areas trees were given precedence over stock for longer, since in the minds of their owners the deer that they hunted needed the woodland to survive: ‘no woods, no deir’ said the factor to the Earl of Moray, impressing upon his employer in 1707 the need to prevent poaching and uphold the need to keep the trees against stock.66 In districts where there was no such attempt at control, the trees are likely to have gone earlier. By the nineteenth century, the association of red deer with a wooded landscape was much less close, and while deer forest owners might then see some need for low-lying plantations or woods to shelter their animals in winter, stalking itself took place on the open moors, which afforded a clearer line of sight for the rifle if the land was totally unwooded.


(1) O. Rackham, ‘Forest history of countries without much forest: questions of conservation and savanna’, in S. Cavaciocchi (ed.), L'uomo e la foresta secc. XIII–XVIII (Prato, 1996), pp. 297–326: quote on p. 299.

(2) I. F. Grant, Highland Folk Ways (edn Edinburgh, 1995), p. 97.

(3) R. C. Blackwood, ‘An estate's forest history – Eliock, Dumfriesshire’, Scottish Forestry, 9 (1955), pp. 13–21.

(4) Rackham, ‘Forest history’, p. 299.

(5) A. Mitchell (ed.), Geographical Collections relating to Scotland made by Walter Macfarlane (Scottish History Society, Edinburgh, 1906), 2, p. 165. Conaglen appears to have been mistranscribed as ‘Cowglen’.

(6) Ibid.

(7) Ibid.

(8) Ibid.

(9) Ibid.

(10) Scottish Woodland HistoryR. Noble, Woods of Assynt (project report for the Assynt Crofters Trust, [2000]).

(11) J. M. Gilbert, Hunting and Hunting Reserves in Medieval Scotland (Edinburgh, 1979). The Bishop of Aberdeen's Forest of Birse contained seventeen shielings, reflecting the seventeen townships of the parish in the twelfth century.

(12) P. Dixon, Puir Labourers and Busy Husbandmen (Edinburgh, 2003), pp. 43–8; Anon. (ed.), ‘Ane fair pallice of greine tymber’, Reforesting Scotland, 16 (1997), pp. 33–4.

(13) M. Dougall and J. Dickson, ‘Old managed oaks in the Glasgow area’, in T. C. Smout (ed.), Scottish Woodland History (Edinburgh, 1997), pp. 76–85; O. Rackham, Trees and Woodland in the British Landscape (revised edn, London, 1995), p. 142.

(14) R. Sibbald, The History Ancient and Modern of the Sheriffdoms of Fife and Kinross (edn London, 1803), p. 387; G. Whittington and C. Smout, ‘Landscape and history’, in G. B. Corbet (ed.), The Nature of Fife (Edinburgh, 1998), p. 33.

(15) Mitchell (ed.), Geographical Collections, 2, p. 3.

(16) NAS: GD 112/17/4.

(17) J. A. Symon, Scottish Farming Past and Present (Edinburgh, 1959), p. 323.

(18) J. Robson, General View of the Agriculture of the County of Argyll (London, 1794), p. 58; J. Robertson, General View of the Agriculture in the Southern Districts of the County of Perth (London, 1794), p. 97; J. Smith, General View of the Agriculture of the County of Argyll (Edinburgh, 1798), p. 130; J. M. Lindsay, ‘The use of woodland in Argyllshire and Perthshire between 1650 and 1850’, unpublished University of Edinburgh Ph.D. thesis (1976), p. 482.

(19) A. Wight, Present State of Husbandry in Scotland (Edinburgh, 1778–84), 3, p. 199.

(20) A. Bil, The Shieling, 1600–1840 (Edinburgh, 1990); A. Fenton, Scottish Country Life (Edinburgh, 1976), pp. 124–36.

(21) E. Burt, Letters from a Gentleman in the North of Scotland (London, 1754), 2, p. 155.

(22) Lindsay, ‘Use of woodland’, p. 139.

(23) R. J. Adam (ed.), Home's Survey of Assynt (Scottish History Society, Edinburgh, 1960).

(24) Lindsay, ‘Use of woodland’, p. 141.

(25) D. M. Henderson and J. H. Dickson (eds), A Naturalist in the Highlands: James Robertson, His Life and Travels in Scotland 1767–1771 (Edinburgh, 1994), p. 161. When Rob Roy died in 1734 he left thirteen cows, twenty-three goats, twenty-one sheep and five horses, one blind. NAS: Dunblane Commissary Court CC/6/5/24, pp. 145–6. We are obliged to David Stevenson for this reference.

(26) I. D. Whyte, Scotland before the Industrial Revolution: an Economic and Social History, c.1050 – c.1750 (London, 1995), p. 142.

(27) Fenton, Scottish Country Life, p. 133.

(28) Bil, Shieling, pp. 255–77; Whyte, Scotland before the Industrial Revolution, p. 134.

(29) NAS: GD 112/1/342.

(30) NAS: GD 112/15/261/6–8.

(31) NAS: GD 112/15/320/1.

(32) J. G. Michie (ed.), Records of Invercauld (New Spalding Club, Aberdeen, 1901), pp. 143, 145, 153.

(33) Royal Commission on the Ancient and Historical Monuments of Scotland, Mar Lodge Estate, Grampian: an Archaeological Survey (Edinburgh, 1995), pp. 6, 9.

(34) Smith, Agriculture of Argyll, p. 146.

(35) Ibid.

(36) W. Boutcher, A Treatise on Forest Trees (Edinburgh, 1775), p. 243.

(37) I. D. Grant, ‘Landlords and land management in north-eastern Scotland, 1750–1850’, unpublished University of Edinburgh Ph.D. thesis (1978), 1, p. 173.

(38) J. Williams, ‘Plans for a Royal forest of oak in the Highlands of Scotland’, Archaeologica Scotica, 1 (1784), p. 29.

(39) J. Walker, An Economical History of the Hebrides and Highlands of Scotland (Edinburgh, 1808), 2, p. 277.

(40) Smith, Agriculture of Argyll, p. 145.

(41) R. Monteath, Miscellaneous Reports on Woods and Plantations (Dundee, 1827), pp. 53–4.

(42) J. Munro (ed.), Inventory of Chisholm Writs 1456–1810 (Scottish Record Society, Edinburgh, 1992), no. 560.

(43) Lindsay, ‘Use of woodland’, p. 137.

(44) Ibid.

(45) W. Marshall, General View of the Agriculture of the Central Highlands of Scotland (London, 1794), pp. 30, 32.

(46) M. M. McArthur (ed.), Survey of Lochtayside, 1796 (Scottish History Society, Edinburgh, 1936).

(47) Inveraray Castle Papers: Survey of Kintyre, George Langlands, 1770–1777, no. 200.

(48) NAS: GD 112/10/1/4/68, fo. 6.

(49) NAS: GD 112/15/363/6, fo. 3.

(50) E. R. Cregeen (ed.), Argyll Estate Instructions: Mull, Morvern, Tiree, 1771–1805 (Scottish History Society, Edinburgh, 1964), p. 132.

(51) Ibid.

(52) Ibid.

(53) NAS: RD 4/94, pp. 374–9; NAS: Buccleuch Muniments, Accounts charge and discharge Liddesdale 1722, GD 224/239/24.

(54) Inveraray Castle Papers: Chamberlain's accounts, 1774–5, fo. 242.

(55) T. Kjærgaard, The Danish Revolution, 1500–1800: an Ecohistorical Interpretation (Cambridge, 1994), p. 111; B. Fritzbøger, Kulturskoven: Dansk Skovbrug fra Oldtid til Nytid (Copenhagen, 1994), p. 341.

(56) NLS: 1359/100, Delvine Papers.

(57) J. Sinclair, General View of the Agriculture of the Northern Counties and Islands of Scotland (London, 1795), p. 161.

(58) J. Henderson, General View of the Agriculture of the County of Sutherland (London, 1812), pp. 83, 86.

(59) Ibid.

(60) Ibid.

(61) Ibid.

(62) J. Robertson, General View of the Agriculture of the County of Inverness (London, 1808), pp. 209–10.

(63) M. L. Anderson, A History of Scottish Forestry (Edinburgh, 1967) 1, pp. 269, 275.

(64) R. A. Dodgshon, Land and Society in Early Scotland (Oxford, 1981), pp. 277–320; A. J. S. Gibson and T. C. Smout, Prices, Food and Wages in Scotland, 1550–1780 (Cambridge, 1995), pp. 196–7.

(65) A. R. B. Haldane, The Drove Roads of Scotland (Edinburgh, 1952); H. Hamilton, An Economic History of Scotland in the Eighteenth Century (Oxford, 1963), pp. 88–96; J. Mitchell, The Shielings and Drove Ways of Loch Lomondside (Stirling, 2000).

(66) NAS: GD 50/149. Moray box Lord Doune no. 1/447. Glen Finglas, like many royal and noble forests, had a history of tension between foresters and graziers. In 1580, James VI issued a letter of protection against people who ‘compelled the forester to allow them to pasture a great number of their stock within the forest to the destruction of the king's deer and the fouling of the grass’ (Anderson, History of Scottish Forestry, pp. 194–5) and the factor in 1707 also referred to a previous occasion when the deer had been ‘destroyed’ and ‘your lordship's predecessors were forced to send Callum McGregor to the Muires of Atholl to bring deer out of it to replenish it.’ These references may refer to a single event.