Woodlands

On the way from the Hermai the entire place is full of [deciduous] oaks. The name of the place, Skotítas [the Dark], did not come from the density of the trees, but from Zeus surnamed Skotítas, and there is a sanctuary of Zeus Skotítas on the left of the road if you turn aside approximately ten stadia. Going back from here, going on a little and turning left again, there is an image of Herakles and a trophy … The third turning off the direct road, to the right, leads to Káryai and the sanctuary of Artemis … and the Nymphs.

Reader, go to the remote byways of Mount Párnon in the Peloponnese, above the village whose modern Greek name is Káryai, previously known by the Slavonic name Arachova – both meaning ‘nuts’. The sanctuary of Artemis and the Nymphs is now a chapel, with wonderful plane-trees, among the biggest in Europe, beside a spring.¹ As far as I know, this is the earliest account of a named wood in Europe. Skotítas Wood was ‘rediscovered’ by the explorer Jochmus and a local priest in 1834. It is now probably bigger than in Pausanias’s time. The dominant tree, as in other ancient woods in the Peloponnese, is the deciduous oak Quercus frainetto. It is a rather open coppice (Fig. 44), with some ancient stools, and a rich flora of shrubs and herbs in glades and newly felled areas. Some of the oaks have been shredded.

Skotítas, as far as I know, is the only surviving wood in Europe to have its own god. Pausanias, unusually among ancient writers, was writing to tell posterity where things were in the landscape. Far more often we have to make do with incidental mentions in records written for some other purpose. This chapter explains what archives can and cannot achieve.

The earliest writings in Britain to describe specific pieces of landscape are perambulations, descriptions of boundaries appended to conveyances of landed estates. For example, the bounds of Long Itchington, bearing date 1001, run:

… from Ycenan [the Warwickshire river Itchen] to the cress well; to the alder-stub; to a high oak in the middle of Wulluht grove; to a withy-bed; to a barrow; to a little barrow …²

A thousand years later the alder-stubs and the barrows have gone, and so has the high oak, but its site is on a massive bank that now divides what was Wulluht Grove into Ufton Wood and Long Itchington Wood. (The orientation of the bank shows that it was made by the lord of Itchington.)^fn1

About 840 perambulations are attached to charters bearing dates from about 600 to 1080 AD. For many the authenticity of the title deed is questionable, which in effect means that the date is uncertain.³ They are usually written in Old English, sometimes Latin.

Perambulations exist in later centuries, especially for defining the legal boundaries of Forests. Many other countries have them: for medieval Hungary they are a large part of all the documentation.⁴ In Europe a few are earlier: in Crete (where they are set in stone) they begin in Ancient Greek times. In the United States, survey perambulations and their ‘witness trees’ – a few of which survive – are an important source for the landscape just before settlement.

Anglo-Saxon perambulations are numerous enough to provide material for statistical analysis, though this is not straightforward because the distribution is very uneven: there are dozens for Berkshire, but none for Norfolk.⁵ They are the earliest evidence for names, locations and boundaries of woods, and for the distinction between woodland and wood-pasture.

Many perambulations can still be followed on the ground (p.387f). Something like one-quarter of the woods are still there. Charters also mention hedges and hedgerow trees and freestanding trees in downland. The trees most often mentioned, in descending order, were thorn, oak, apple (or crab), willow (including sallow and withy), ash and elder. Oak, lime and birch are among the trees associated with well-wooded areas later to be Ancient Countryside (Table 7); thorn, blackthorn, apple and elder are mentioned especially in areas without woodland, later to be Planned Countryside.

Place-names, including names of woods, are derived from Old English, Old Norse or (rarely) Norman-French or Old Welsh. They offer a second window into the age of the charters. They once meant something, but over the years they have diverged in spelling and pronunciation from the everyday language: their meaning has been forgotten and sometimes misinterpreted.

To interpret place-names one needs to look at the spelling in early documents. The county volumes of the English Place-Name Society (not yet complete) give early spellings and offer interpretations.

Anglo-Saxon names of villages and hamlets sometimes indicate substantial amounts of local woodland. Those ending in -ley, -hurst, -field, and Old Norse -thwaite give an insight into the general distribution of woodland.

Place-names are only vaguely datable, and their study is not an exact science. The earliest document containing a name is only the latest possible date for either the name or the feature itself. One should beware of over-interpreting names. A Brentwood ‘burnt wood’ could arise at any time from the sixth to the fifteenth century; it should not be interpreted as ‘wood destroyed by fire’, for there is no knowing whether the fire was a bracken fire or merely charcoal-burning.

A peculiarity of England is that almost any old wood, however small, has a name, which is often the name of the wood itself and not of some nearby farm or village. The further one goes back in history the more such names emerge. Hayley Wood is still called by the name that it had in the Ely Coucher Book; but in 1251 Hardwick Wood, now called after the parish, had its own name, Bradelegh; Hockering Wood in Norfolk was called Swinnow. Gamlingay Wood (see Fig. 179) was called Short Wood, and the two parts into which it was subdivided before 1601 were called Mertonage Wood and Avenells Wood. Eversden Wood was divided among three parishes, and the part in Great Eversden was subdivided into Granditch Wood, Stockings Wood and Snapdean Wood. Some wood-names are forever lost. Epping Forest used to be Waltham Forest, both being names of nearby towns: presumably it had a name of its own before it became a Forest in c.1100, of which no memory survives.

Names of recent woods are often distinctive. ‘Plantation’ is normally just that, although the name may stick long after the planted trees have died out. ‘Hundred Acre Wood’ is probably recent: an ancient wood of that size would have had a more distinctive name. ‘Spinney’ means a wood of thorns – Latin spinetum – and remembers an abandoned field in the process of turning into a wood. ‘Fir Wood’ implies a pine plantation, but can be confused with ‘Furze Wood’ referring to gorse. ‘Conifer Wood’ usually alludes to a conyfare or organised rabbit warren.

Wood-names sometimes allude to uses or management. ‘Coppice’ or its spelling variant ‘Copse’ or its dialect equivalent ‘Spring’ often mean coppice compartments in a park or wooded Forest. There was a wood called Colyers Tayles in Castle Camps (Cambridgeshire) in 1411:⁶ collier being a charcoal-man, and tailz a coppice (Modern French taillis).

Besides wudu, ‘wood’, the Anglo-Saxons had words for what were presumably different kinds of wood-lot: bearu, fyrh∂e, græfe, graf, hangra, holt, hyrst, sceaga, strod, wald.⁷ Although most of these still survive in the names of ancient woods – Frith, Grove, Holt, Shaw – their meanings are doubtful, except for hangra, ‘Hanger’, a wood on a slope. ‘Shaw’ is now used of a narrow wood, but Old English scaga seems not to denote any particular shape. The common wood-name ending ‘-hay’, usually reduced to ‘-y’ (e.g. Easty and Northy Woods in Cavendish, Suffolk), apparently means ‘enclosed wood’. There were many unenclosed common-woods, whose names might include gemæn, ‘common’, as in Man Wood or The Mens (Sussex). ‘Grove’ (graf) usually means a small wood, but the word is once used for a type of woodland produce; and the present area of the wood Wulluht graf is 194 acres (78 ha), which even by the standards of the eleventh century was not small. Hris (now Rice or Royce) meant ‘underwood’. The Norse word for grove, lúndr, survives as ‘Lound’, ‘Lownde’ or the ending ‘-land’.

Tree names usually indicate the presence of the tree, but not how much there was of it. One knows how many oaks to expect in ‘Oaken Grove’ and perhaps in ‘Notoaks Wood’ but not in mere ‘Oak Wood’. Does ‘Birch Wood’ mean a wood composed of birch or a wood distinguished by having one conspicuous birch tree? A wood composed of birches would be ‘Birchen Wood’ or ‘Birchet’ (like Biolet in French or Birchat in German); Birch Wood in Dedham (Suffolk) appears as Birchetum in medieval Latin.

Many woods are named after owners or tenants, but were not kept up to date. Alsa Wood (Stansted Mountfitchet, Essex) is the medieval Asishey and Anglo-Saxon ‘Ælfsige’s hay’. One of the Hindolveston woods, Norfolk, was still called ‘Robin’s Wood’ in the eighteenth century, after John Robynes, an unfree tenant of Norwich Cathedral Priory, who had leased it five centuries before.

Wood- and tree-names are liable to misinterpretation. The parish of Lindsey (Suffolk) is corrupted from Lelesey and cannot be named after the lime (linde) trees in the local woods – although Groton Wood in the next parish may have given rise to the surname de Lyndewode. The wood next to Hayley Wood was called Littlehound Wood when it was grubbed in the seventeenth century, and gave rise to the later place-name Houndwood Field. The Ely Coucher Book of 1251 (Fig. 46) spells it Litlelund; it has nothing to do with hounds, but is the little lúndr. The 1251 scene was not new, but remembered Viking times. If there was a little lúndr there must (must there not?) have been also a great lúndr, which can hardly have been other than Hayley Wood itself.

Woods called ‘Long Wood’ are long, but a ‘Round Wood’ can be angular and a ‘Short Wood’ is not necessarily short.

Wood-names in Wales are disappointing, usually being no more than the name of an adjacent farm. Cornwall has a few wood-names in the extinct Cornish language: the great Merthen Wood on the Helford River was thought of as three woods, Coesenys (‘Island Wood’), Coose-Carmynowe and Cosabnack. In Ireland a few ancient names of big woods are known, but even if the wood survives it seldom has a Gaelic name today. Gaelic wood-names are sadly lacking in the Highlands of Scotland.

In France only woods of more than 100 hectares or so have names, which are often those of some nearby village or town. Very big woods (more than 400–1,000 hectares) are called forêts, but there is no apparent difference between bois and forêt except size. This seems to be a nineteenth-century corruption: in the eighteenth century there were fewer forêts, and even a huge wood (or group of contiguous woods) could be a bois. Bois de la Reyne, ‘Queen’s Wood’ (near Toul) of Cassini’s map (c.1755), about 50 square kilometres (12,000 acres) in extent – as big as the Forest of Dean – had become Forêt de la Reine by 1834. (It had been through a period when queens and Forests were both politically incorrect.)

A stage in investigating an ancient wood is to see whether it is ‘mentioned in Domesday Book’. Numerous modern editions make it possible to trace the record, if any, of almost any settlements in England. If, as often, there was more than one manor in a parish, or two parishes with the same name, the names of the landowners will help in working out which was which.

Domesday rarely names woods, and never records trees. In counties where it gives the dimensions of woods – x leagues long by y furlongs wide, at 8 furlongs to the mile and 1½ miles (2.4 kilometres) to the league – these sometimes make sense when compared to the present size of the wood. Where an estate had more than one wood, Domesday somehow combined them into one pair of dimensions. The woodland belonging to a place was sometimes located at a distance.⁸

My own interest in woodland history was inspired by Hugo de Northwold, Bishop of Ely 1229–54. Forty years ago I was involved with Hayley Wood, recently acquired by the then Cambridgeshire & Isle of Ely Naturalists’ Trust. Through Reaney’s Place-names of Cambridgeshire I found a reference to the Ely Coucher Book, compiled by the Bishop’s orders in 1251. Having been taught medieval Latin at school by a friendly master, Samuel Bate, I ordered the manuscript in Cambridge University Library and found what is now a well-known description of the wood:

There is there one wood which is called heyle, which contains fourscore acres.

This was a startling discovery for a young don who had been taught (as one was taught 40 years ago) that the countryside had always been changing and that the present woods, hedges, etc. were the product of farming practices no older than the eighteenth century.

It took some years of research to document boscus de heyle through the intervening centuries and to show that it was indeed the present Hayley Wood. The next step was to show that it was not a unique survival. To commemorate the three-quarters of a millennium of the Coucher Book I published an analysis of what had happened to each of the 49 woods in it (p.122ff).⁹

Hugo de Northwold was apparently a pioneer. Since Domesday Book there had been sporadic surveys of institutional estates, but the Coucher Book seems to be the first to descend to the detail of giving the names, areas, boundaries and uses of woods (Fig. 46) – besides fields, meadows, fens, heaths, and naming thousands of people who held lands from the Bishop, with their feudal duties, such as carting his underwood or gathering his nuts. This wonderful volume (called a coucher book because it lies down, being too big to stand on a shelf) is still unpublished.¹⁰

The Hundred Rolls of 1279 were a greater and more detailed Domesday Book, setting out to list the assets of all estates in England, including the area of woodland (though not usually the name) belonging to each manor. Alas, the rats in the Tower ate most of it (or was it ever finished?); but the Cambridgeshire part survives and was a link in demonstrating that the survival of Hayley Wood was not especially unusual. The surviving parts were transcribed and published in 1818.

Estates found their way temporarily into the hands of the king: those of a bishopric, during vacancies of the see; those of private lords, whenever the owner committed high treason or died without an heir; those of dependencies of French monasteries, when England was at war with France. The Crown would hold an inquisition into the property, making extents of the manors, including their woods and parks. The king’s agents would administer the estates; they would file accounts, occasionally (as with Staverton Park) for useful runs of years. Thus in 1356 Thomas de Lisle, Bishop of Ely, was on the run for the murder^fn2 of William de Holme, and Edward III’s valuation of his estates makes a useful comparison with that of a hundred years before: for ‘Heylewode’ it gives the coppice cycle (seven years), which the Coucher Book omits.¹¹

Lands belonging to monasteries were seized by Henry VIII in the 1530s. Most of them he sold off and frittered away the proceeds on obscure French wars. A department of state was set up, the Court of Augmentations, for receiving the king’s stolen property. Anyone making an offer for monkish land had to get a valuation, which is filed in the Public Record Office, for example:

Wherof 10 acres of 3 yeres growth, 10 acres of 7 yeres growith, 15 acres of 10 yeres growthe and 6 acres resydue of 14 yeres grouth, all whych woods be dyuysed [leased]… except great trees growing in the same by the late abbott and convent [of Bury St Edmunds Abbey] to Wyllyam Gooche [in 1533]…

And in the seyd woodes be growinge 200 sapling okes of 30 and 40 yeres growthe whereof 80 reseruyd for tymber for the ffermor [lessee] to repayre the howses … and 120 resydewe valluid at 4d. the tree wch is in the holle

[He was the ancestor of the Winthrops of Massachusetts and Connecticut. Half the wood is incorporated in the present Groton Wood.]¹²

Among documents containing incidental references to woods are cartularies, registers of land transactions kept by monasteries and other institutions. The transactions themselves seldom involve woodland, but woods appear among the abuttals that define other pieces of land. These may identify the location of the wood with respect to fixtures such as manor houses, roads or churchyards.

Many cartularies have been published. It is worth searching the index for incidental mentions of woodland. However, mentions tend to be scarce, and only those fluent in reading medieval manuscripts should try searching unindexed cartularies.

Although any large estate, from Roman times onwards, presumably had some sort of book-keeping, estate accounts survive from the thirteenth century onwards. An account roll typically covers one financial year (New Year’s Day often being Michaelmas, 30 September). It lists income and expenditure, grouped under heads such as ‘Carts’ or ‘Buildings’, to arrive at a net profit. On the back of the parchment roll, worn by many grubby fingers, is a list of stock, live and dead, from swans to tree trunks. There may be a heading covering the wood, with sales of faggots and the cost of making and carting them, and sales of timber. Other references, including to non-woodland trees, can be scattered through the roll, for instance purchases of hurdles or underwood for making them, or trees felled or bought for repairs or new buildings.

Account rolls can give a detailed picture of woodland management and the place of woodland in the economy. They record functions that involved money. Trees felled in the wood for repairs to the lord’s barn may appear under the cost of felling, or from sales of the resulting branches or bark. Sometimes there is a separate little account of the ‘servile works’ owed by tenants by way of rent – someone might work off x works by acting as woodward.

In the late Middle Ages, unfortunately, estates were often ‘farmed’, leased as a whole to a tenant whose records rarely survive. Estate accounts flourished again in the eighteenth century. The account books prepared by agents of the Ashburnhams, woodland owners in Sussex and Suffolk, record, year by year, income from so many acres of underwood which produced so many hundred hurdles (at six score to the ‘hundred’), ‘pease sticks’, ‘hurdle brush’, ‘short wood’, ‘cow bins’ etc., etc. – a mass of minute detail whose exact meaning was then familiar to everyone and is now forgotten.

Scottish native pinewoods are documented in estate records from the seventeenth century onwards. Some records are preserved because of lawsuits, especially over lands confiscated from earls who committed high treason in the Jacobite Rebellions. We learn of the traditional local uses of pine, birch and other trees, of commercial logging, and of the arrival of sawmill technology in the seventeenth century. They are the basis of Steven & Carlisle’s study in the 1950s – a work ahead of its time, not least because it went through the evidence wood by wood instead of working downwards from generalities – and Smout’s more recent work.

As far as I know the earliest surviving map to show a British wood is the famous Boarstall Map of c.1445. Boarstall in Buckinghamshire was a curious place, embedded in the wooded Forest of Bernwood. The map shows a very small village with its church, moated manor house and wayside cross, surrounded by open-field arable and also hedged fields. The periphery is almost all woods and wood-pastures, some of then named (‘Ffrith’, ‘Costowod’, ‘Pannsale’ etc.) and subdivided by hedges. A fallow buck and doe are drawn – to show that this is a Forest – and there are pollards and coppice stools of two unidentifiable kinds of tree. In the foreground is an anachronistic scene showing an early Forester [!] presenting the gory head of a wild swine to King Edward the Confessor, and in return getting a coat of arms [!] and a grant of land.

The purpose of the map was not topographical accuracy; much ink has flowed in a vain effort to reconcile it with what is known from other sources.¹³ It seems to represent a compartmental wooded Forest, with areas of grassland and pollard trees and other areas, hedged off, with coppice stools.

It might be thought that the medievals could not produce detailed maps of woods, but for two mysterious surveys, of Sibton (Suffolk) in 1325 and Leaden Roding (Essex) in 1439, each of which gives the areas of woods to within a quarter of a square perch, that is to within one part in 10,000 for a 15-acre (6 ha) wood.¹⁴ Such precision would be hard to attain even now (supposing the boundary of the wood could be precisely defined). However, at Leaden Roding three of the woods still survive, and have areas within 1 to 4 per cent of those cited in 1439: as close an agreement as could be expected, given that woodbanks have a finite width. People who could measure the area of an irregular wood with such accuracy would necessarily have been able to construct a detailed map. (Measuring woodland is more difficult than for other land because the trees get in the way.)

The craft of mapmaking ought to have been stimulated by the dissolution of the monasteries and privatisation of their lands. Thousands of new owners of monkish land and woodland would have wanted to know what they had acquired. However, it was 40 years later, in the 1570s, that detailed maps began to appear, showing all fields, hedges, woods, roads and buildings.

Early maps are usually of landed estates, at a scale of around 1:5,000, showing the shapes of fields and woods. They were often made when an estate had changed hands. Historians may philosophise eloquently about maps as expressions of ‘power’, but mere information was often needed: complexities of land tenure required research before a purchaser knew exactly what he had bought.

Mapmaking did not evolve out of crude beginnings, and accuracy does not go with late date. Elizabethan maps range from the sketchy to the extremely accurate. The best surveyors, such as Thomas Langdon who mapped Gamlingay (Cambridgeshire) in 1601, and John Norden who mapped the complex Stan-hope estates, including Staverton Park, in 1602, bear comparison with any cartography since.

Maps were made down the centuries, with changing fashions in style but (as regards woodland) no great changes in content. The existence of a map, its quality and survival are matters of chance. Maps show the outlines of woods, sometimes indicating the nature of the boundary. Individual trees are not shown, apart from a very few acting as boundary marks, or occasional mentions of service and other rare trees. An early convention appears to indicate timber trees scattered among underwood, and in wood-pasture cartographers sometimes indicate where the trees were dense or sparse. Many cartographers indicate trees in hedges, but usually at an even spacing, which must be conventional.

Areas of woods are often shown (in acres (often, but not always, the modern acre of 0.405 hectare), roods at four to the acre, and square perches at 40 to the rood). Very rarely we are told of how many years’ growth the underwood was.

Small-scale county maps of the sixteenth and seventeenth centuries, such as the Norden series, show conventionalised parks, but are of little use for woods. Patches of trees fill inelegant blank spaces rather than indicating reality. Only Rutland and Middlesex, tiny counties, are mapped at a scale big enough to show individual woods. Woods near roads may be shown on road maps, beginning in 1675.¹⁵

Knowledge of the Scottish Highlands has been transformed since Steven & Carlisle’s time by the rediscovery of the sketch maps made by Timothy Pont, indefatigable explorer of the wild Highlands between 1585 and 1596. These maps, recording every shieling, burn and wood, were never published. Later they were worked on by the learned Robert Gordon of Straloch, and sent to the Dutch cartographer John Blaeu, who published some in a heavily revised form in his great Atlas in 1654, and lost others. Blaeu’s version is simplified from Pont’s original, and woodland is usually exaggerated; whence (in part) the belief that as recently as the seventeenth century the Highlands were much more wooded than they are now.¹⁶ The next maps to show individual woods in Scotland were General Roy’s Military Survey, made in 1747–55 after the Jacobite Rebellion.¹⁷

Mapmakers later published county maps, at scales of 1 inch to the mile (1:63,360) or larger, showing the sizes and rough shapes of woods. These include Rocque of the London area in the 1750s, Chapman of Nottinghamshire in 1774, Hodskinson & Donald of Cumberland in 1774, Chapman & André of Essex in 1783, and Faden of Norfolk in 1795. Woods, however, are often shown inaccurately (as compared to contemporary large-scale maps). Any wood that was not visible from a public road, or had recently been felled, might be omitted. County maps are often better for Forests, wooded commons or parks than for woodland; some have a special symbol for differentiating wood-pasture (see Fig. 183a).

Each Enclosure Act required hiring a surveyor to make a large-scale map showing the proposed state after reorganisation. Such maps, dating between c.1720 and c.1850, exist for nearly half the parishes of England. Although woodland was seldom directly involved, many enclosure maps cover the entire parish and thus show woods: the earliest map to portray Hayley Wood is the enclosure map of 1816 which shows the wood, already surrounded by ‘Ancient Inclosures’. Wood-pastures, especially wooded Forests, often had separate enclosure proceedings and maps (which usually led to destruction of the site).¹⁸

Between 1838 and 1845 there was a campaign of mapmaking on account of the Tithe Commutation Act. Tithes, originally every tenth sheaf, lamb or faggot, paid to the provider of the parish priest – or, by then, often to some third party whose ancestors had purchased the entitlement – were to be replaced by a ‘rent-charge’, a money tax on almost every field, pasture, meadow and wood in England and Wales, which therefore had to be measured. If there was not already an enclosure-award map a new map was made, together with a schedule or list of lands.¹⁹

The student will probably find either an enclosure or a tithe map covering any wood. The acreage, owner and occupier will usually be given, but the land use seldom goes beyond differentiating ‘wood’ from ‘plantation’. Usually these maps are reliable, but they do not always prove the non-existence of a wood. Small areas of woodland among moorland, if they do not have definite boundaries, may be overlooked; some tithe maps, particularly in Dorset, are rather perfunctory and may leave woods out.

The national survey, at 1 inch to the mile, began in 1795, and worked northwards, reaching the north of England by the 1860s. Scotland was covered from 1847 to 1878. At first the os was similar to county maps; some of the same people worked on both. The early sheets, of the southern one-third of England, suffer from the same weaknesses. Often there survive ‘surveyors’ drawings’ (not field notes but an intermediate stage in publishing the map); these may give details of woods that disappeared from the published version, but they can also contain fictitious details such as field boundaries that bear no relation to the real fields as known from larger-scale maps.

After 1830, fieldwork was done more conscientiously and in greater detail; as far as I know woods from this point on are reliably shown. Some of the previous surveys were left unpublished; for a few sheets there were two published versions, before and after 1830. The metal plates from which the maps were printed remained in use (though sometimes copied electrolytically) for most of the century, railways and other new details being added in later impressions.

The large-scale os, at 6 inches to the mile (1:10,560), began in Ireland in the 1830s. This was followed by the tithe-award maps; then the surveyors were employed on a 6-inch and 25-inch (1:2,500) survey of Britain, beginning in Scotland and northern England where the 1-inch coverage was still unfinished. The 6- and 25-inch scales show almost the same information, except that the 25-inch adds acreages of fields and woods.

The 25-inch os of the mid- to late nineteenth century is the most detailed and accurate map of Britain ever made, especially regarding trees and woods. Wood-names and acreages are given. Perhaps the most detailed of all the sheets are those covering Hatfield Forest in 1874, distinguishing woods (with rides), scrub, grassland, small ditches, hundreds of wood-pasture trees (each one, as far as one can verify from air photographs, measured in by men dragging a chain), and even woodbanks. This is exceptional, but in general the first edition 25-inch tried to show every non-woodland tree in Britain, and as far as it can be checked got them right.

The os convention for ‘normal’ woodland depicts standard trees among underwood, but there are many other signs, which vary with date and are not fully explained. They include bracken, rough grassland, bushes without trees and trees without underwood. Conifer symbols must indicate planting, except in parts of the country where native yew or pine are possible. Even if the conifers have now disappeared, their former presence can explain other trees planted along with them, such as plantation-type oaks (Fig. 203), or trees such as birch that are successors to the defunct conifers.

Later editions of the os decline in content and accuracy: the first details to go were individual trees. Nevertheless, they are the chief documentary source for woodland in the twentieth century. The primary material is the 6-inch or 25-inch surveys, from which other scales are derived. A 1-inch map published in the 1950s can thus show woodland as it was in the 1920s. A wood may be shown as ‘rough grassland’, the result of a 1920s felling, and yet long ago have reverted to being a normal wood with surviving ancient stools. The date of survey, not of publication, should always be quoted.

In most countries governments, in their vanity, have tried to regulate the behaviour of trees with masses of statutes. At international conferences on forest history about two-thirds of the papers are on the history of forest legislation and one-third on the history of forests.

Although this began in the Middle Ages, legislation vastly expanded after 1750, the time of Enlightenment, when writers on modern forestry sought to bend the ear of government and invest their theories with the majesty of law. Typically they sought to replace the earlier, often communal, links between people and trees with something reflecting the Age of Reason (whether it was good or bad reasoning would emerge later). Woods were to be separated from pasture and farming, and to be managed henceforth on ‘scientific’ principles of modern forestry (Chapter 18).²⁰

Britain was a disappointingly un-Enlightened country with no tradition of top-down regulation until the Forestry Act of 1919. Charles Mynors has written a wonderful book, The Law of Trees, Forests and Hedgerows. Exhaustive and very readable, it goes back in some detail to the nineteenth century, but before 1919 deals almost exclusively in case law.

One can almost count the relevant English statutes on one’s ears. The two that concern us are medieval Forest Law and the Statute of Woods of 1543. Forest Law, though mainly concerned with deer, in theory forbade people to cut down their own trees within the legal bounds of a Forest. The Statute of Woods required woods to have a minimum of 12 timber trees per acre, to be fenced after felling to prevent them from turning into wood-pasture, and not to be grubbed out.

It is an anachronism to read early statutes as if they were intended to stop people from doing things; their purpose was often to raise revenue from fines. They are of value only in conjunction with records of prosecutions: what cases were brought? were the fines deterrent? People did cut down trees and woods within the legal bounds of Forests, and used or sold the timber and underwood. ‘Offences’ were prosecuted only if there was an aggravating circumstance, such as the perpetrator being the Prioress of Felsted (representative of a French religious house), or there being a large commercial interest, or the penurious and greedy Charles I being on the throne. Prosecutions under the Statute of Woods seem to have been very rare, although it could be cited in the terms of leases even 200 years on.

The legal documents relevant to the woodland historian are not the top-down ones. The minutes of local manorial courts record things people were not supposed to do, how often they did them, and how much they were fined. Non-woodland trees appeared before the courts more often than woodland: disputes over who owned a hedge, trees allowed to grow out and obstruct the highway, turning timber trees (which belonged to the landowner) into pollards (whose regrowth belonged to the tenant), and so on. Manorial courts could and did amend the rules from time to time.

Read with patience, court rolls can yield a vivid picture of how a wood or wood-pasture actually functioned. Species of tree are often named. One must note whether a type of ‘offence’ was a regular and repeated occurrence. Court proceedings seem often to have been used to collect a reasonable grazing or woodcutting rent under the guise of a fine.

Administrative records of English Forests, especially wooded royal Forests, attracted much attention a hundred years ago. Selections can be read in many published works, such as J. Charles Cox’s book The Royal Forests of England in 1905, and chapters by Cox and others on ‘Forestry’ in the earlier volumes of the Victoria County History. Many a wood was saved from oblivion because it happened to lie within a legal Forest and its owner committed a prosecutable breach of Forest Law.

For Epping Forest there are numerous fragments of the proceedings of lower Forest courts, which in this particular Forest handled the sort of business that manorial courts would have done were it not a Forest – people cutting wood who were not entitled to it, grazing illegal goats, removing wood under woodcutting rights on an ‘unreasonable great Cart’ instead of the prescribed sledge, etc.²¹

However, most of these are selections, made on an unknown basis, rather than a systematic analysis of all the Forest files in the Public Record Office. An exception is the compendium, compiled c.1400, of the archives of Sherwood Forest and edited by Helen Boulton, with its extensive records of timber, deer and Forest offences.²²

The earliest botanical record of a plant in a definite place is of nearly the rarest British plant, Sorbus domestica, the Whitty pear or ‘true service’, which has ash-like leaves and fruits like tiny apples. The author of the seventh-century Historia Brittonum, ‘Nennius’, records ‘apples … found on an ashtree’ on cliffs near the mouth of the Wye.²³ This wondrous tree was not noted again until found far away in Wyre Forest (Worcestershire) a thousand years later. After three centuries of further oblivion it has lately been found again on various cliffs, including the Wye, where it doubtless lurked unseen all the time.

Systematic records begin in a small way in the sixteenth century. In the seventeenth century Thomas Johnson recorded the plants in Ken Wood next to Hampstead Heath, and John Ray in Madingley and Kingston Woods near Cambridge. These can be compared with later records for those woods and with what is there now. They establish the important fact that woods lose species with time: thus in Madingley Wood, out of 31 species recorded in c.1660, 14 are no longer there. Extinctions are not at random, but are especially of non-shade-bearing plants of woodland grassland and coppiced areas, such as saw-wort and the grass Calamagrostis canescens.²⁴

Other historic records lurk in herbaria. Besides being biased towards rare plants, herbaria are organised by species rather than locality: it is easier to find out localities where Serratula used to occur than to find out what plants there were in Hayley Wood in the nineteenth century. Computer catalogues may help to solve this problem.

Each English and Welsh county has one or more record offices, run by the county council and open to the public. Their map collections are a good place to begin. They should have sets of nineteenth-century Ordnance Surveys, and maybe copies of the survey drawings. They will have most of the Enclosure-Act and Tithe maps for the county, and many earlier estate maps. They usually have a map index, parish by parish, as well as indexes to other types of document.

Eighteenth- and nineteenth-century county histories often reveal whether a wood, or the estate of which it formed part, belonged to the Crown, or a private family, or a monastery, or a diocese, or a Cambridge or Oxford college. This indicates the type of archive in which its earlier history is likely to be found. The Victoria County History – still incomplete – contains detailed accounts parish by parish, often identifying the archive and giving a reference to each document. English Place-Name Society volumes give references to documents mentioning particular woods.

The Public Record Office (National Archive) at Kew contains the records of Crown (and Duchy of Lancaster and Duchy of Cornwall) estates. For the Middle Ages many items have been published as abstracts in the hundreds of volumes of the king’s business correspondence known as the Rolls Series, including (among others) the Close and Liberate Rolls. These are to be found in big libraries; the volumes have excellent indexes of places. For many purposes it is unnecessary to see the original documents. The pro has many other types of record, including early maps and surveys. Consulting the catalogues used to be a research task in itself, but many of them are now on computer files. The British Library contains some important manuscript maps, such as Roy’s Military Survey. Pont’s maps are in the National Library of Scotland.

Records of monastic estates were scattered to the four winds at the Dissolution. By now they may have got into almost any archive: a useful source is Davis’s Medieval Cartularies.²⁵ Most monastic estates make an appearance in the pro, either in the Crown records (if Henry VIII retained them) or in those of the Court of Augmentations (if he frittered them away).

Diocesan archives have sometimes been transferred to public repositories: thus those of Ely diocese are in Cambridge University Library, but Canterbury has an excellent record office in the Cathedral. Every Cambridge (and Oxford?) college, and the university, has its own archive, containing records of each estate they have ever owned, some inherited from before the college was founded.

Private estates sometimes still have their own archives; they vary from those with muniment rooms staffed by professional archivists to those that are so secretive as to cast doubt on their confidence in the title to their lands.^fn3 Deposited private archives form the bulk of the collections of county record offices (not always in their own county). Private estates appear sporadically in the Public Record Office, for example the administrators’ accounts of high traitors or vacant bishoprics. Entries in the Calendar of Inquisitions Post-Mortem (Rolls Series) may mention an extent, but to find out its contents one has to consult the original document.

Useful pieces of information often turn up in unexpected contexts: woods may be named in the context of highwaymen, murders, heretics, accounts of unrelated estates, inquests after accidents, and woodsale advertisements in newspapers.

Archivists are responsible for producing documents, but not normally for reading or translating them, though they give friendly help with the occasional difficult word. Some record offices offer a research service for a fee.

The subject is plagued by factoids or canards: statements that look like facts, are treated with the respect due to facts, and have all the properties of facts except that they are not true. I do not mean conflicts of evidence or differences of scholarly opinion, such as the Vera controversy, but beliefs that persist even though they are easily disproved. Examples include:

The landscape has always been changing (instead of periods of stability alternating with more or less widely spaced phases of change).
Forests were ‘cleared’ by setting fire to them.
Hedges are no older than the eighteenth century.
The roofs of Westminster Hall and King’s College Chapel, Cambridge are of chestnut, not oak.
Erosion is necessarily the ‘fault’ of people destroying forests (rather than of ploughing land, or being part of the normal order of nature, the process that gives sedimentary geologists a job).

Not all factoids are relics of past ignorance: in 2001 it became accepted (despite public denial by the Director of the Animal Health Laboratory) that casual passers-by can transmit foot-and-mouth disease from animal to animal.

The ecologist needs to assess what weight to put on each item. Oak is often mentioned in documents: it was a common and expensive tree; it was easily recognised; it had many specific uses; it belonged to landowners, who kept records, rather than tenants who did not. Hornbeam is rarely mentioned: it was difficult to identify; there was little reason to recognise it, because it did nothing that other trees would not do; it sank into the general anonymity of underwood. Hence any mention of hornbeam is worth many mentions of oak.

Misunderstanding landscape history. The history of the countryside is not the same as the history of country people nor of what people have said about the countryside. There are other actors in the theatre besides Man. Landscape history is the history of human default as much as of human action.
Tradition of plagiarism: writers have copied one another. If John Evelyn said something in the seventeenth century, that was good enough to be repeated in the twentieth, without checking whether Evelyn was rightly informed or drew the right conclusions.
Expecting the history of the landscape to be simple. If some chestnut-woods can be shown to be planted in the eighteenth or nineteenth century, that does not make all of them post-1700 or all artificially planted.
Using only documentary evidence, which shortens perspectives and precludes knowing what was happening at times when people were not writing. In Britain most of the really big steps in making the landscape had already happened before the earliest documents. Even in the best-documented sites, such as Gamlingay Wood or Hatfield Forest, the fieldworker discovers things that are not in the written record.
Using documents of a generalised nature. The history of woodland is the sum of the histories of hundreds of individual woods. The history of people’s attitudes to woodland, however interesting, is not the same thing: it can hardly be understood without first knowing what it was they were attitudinising about.
Obsession with the history of forest laws. Writers accept legal documents at face value and interpret them in terms of modern legal practices.
Not knowing what to make of tradition. Almost any popular book on trees has masses of ‘legends’, often treated as evidence. These are colourful stories, which it seems a shame to ignore. The question is not so much ‘are legends reliable?’, but ‘is a particular story a legend or not, and where did it come from?’ For centuries, in a cosmopolitan country like England, country folk have been reading books and acquiring facts and factoids from other countries or from works of fiction.
Using generalised ‘human activity’ as a let-out. Biologists and historical geographers too often invoke human activity as an explanation of any pattern or phenomenon that they think is anomalous, as if this dispensed them from further investigation. But details are needed. If tree-planting is to explain an apparently anomalous tree distribution, one has to demonstrate that the pattern of planting activity fits that distribution.
Translation. Translators tend to know little of woodland history and fail to get the technical terms right. They can use ‘clearing a wood’ without saying whether they translate amputare boscum, felling a wood (expecting it to grow again), or evellere boscum, digging up a wood (to make a field). ‘Brushwood’ properly means birch twigs or other material for making brushes, but translators sometimes use it for the Latin subboscus, underwood in general. The student should make every effort to test the translation, if not to go back to the original record.

fn1 I am indebted to Dr David Morfitt for showing me the spot. For a discussion of woods in Warwickshire charters and later documents see Wager (1998).

fn2 This was, it seems, a genuine murder: the gangster-Bishop’s hand did not strike the blow, but did arrange the contract.

fn3 Another reason for being refused access to an archive is that the documents are lying in sacks on the floor.