Woodland Flowers

Woodland cover in Great Britain has expanded from about 4–5% of the land surface around 1900 to about 13–14% today. A century ago woodland cover was predominantly broadleaved and concentrated in southern England, whereas now conifers predominate in many areas and our largest forests are in northern England, Wales and Scotland. This expansion has been the result of Government policies to reduce our dependence on imports of wood and wood products, although we remain the second- or third-largest importer of such materials in the world. However, even our most ancient woods were new once. Any direct descendants of the wildwood (if they exist) are on land that was covered by ice or tundra some 15,000 years ago. Other ancient woods are underlain by Bronze or Iron Age field systems, Roman and medieval farms. So how long does it take for the ground flora to move into new woodland, and where do the plants come from?

As a teenager, I helped my maths master, Bill Dawkins, weed and dig up Christmas trees that he had planted in a field behind his house. Intermediate rows of mixed hardwoods and conifers were to be left to grow into a wood. Fifty years later I went back to see what had happened to this wood. I found myself looking up into the canopy of trees that I remembered as barely overtopping me. The ground flora was mostly of Bramble and Nettle, but there were also daffodils (cultivated), bluebells (native), Pendulous Sedge, Soft Shield-fern, Common Figwort, Lesser Celandine and Primrose. The daffodils, and perhaps the bluebells, were planted in; the primroses had spread from remnants of an orchard that had been in one corner of the field and the other species had come in from hedges and gardens around the new wood. While nowhere near as rich as nearby ancient woods, the formerly grassy inter-rows are turning into a proper woodland ground flora.

During the 20th century large-scale conifer plantations were created on lowland heaths, upland moors and bogs (Gambles 2019, Tsouvalis 2000). There was extensive spread of broadleaved woodland through natural regeneration on the downs of southern England following the sharp decline in rabbits from myxomatosis in the 1950s. Oak, Ash and Thorn sprang up on commons as livestock grazing became uneconomic or too risky because of traffic collisions on nearby roads. Replacing steam trains by diesel meant fewer line-side fires, and railway embankments became lines of trees, exacerbating the annual problem of ‘leaves on the line’. Miles of new woodland have been planted or grown up naturally along motorways and other major trunk roads. Abandoned land in or on the edges of towns and cities has turned into scrub. Thousands of small broadleaved woods have been planted under various farm woodland schemes. Some estates in Scotland have promoted large-scale natural regeneration of Birch and Scots Pine where they have fenced deer out of wide areas or increased the levels of deer control.

Conservationists have been ambivalent about this forest expansion. There were objections from the 1930s onwards to large-scale conifer afforestation because of its effects on the landscape and the restrictions on where people could walk (Symonds 1936). In the 1970s and 1980s major campaigns were organised against the planting up of lowland heaths, bogs and upland moors because of the losses to breeding bird populations and to moorland and bog vegetation (NCC 1986, Warren 2000). Conservation organisations regularly seek to remove or suppress developing native woodland on grassland and heath reserves through the judicious use of cutting and grazing. Small-scale farm planting on improved pasture has been dismissed as a waste of time because such woods may not support woodland specialist species in our lifetimes. Some previous colleagues even formed a group called Conservationists against Tree Planting or CAT-P (say it!).

At the same time, we applaud woodland restoration projects such as that at Carrifran Woods, which aims to ‘re-create an extensive tract of mainly forested wilderness’ in the Southern Uplands (Ashmole & Ashmole 2009). In the Highlands, the charity Trees for Life seeks to link up ancient native pinewoods through establishing islands of ‘healthy young forest scattered throughout the barren, deforested glens. As these new trees reach seed-bearing age, they are expected to form the nuclei for expanded natural regeneration in the surrounding area’. Another charity, the Woodland Trust, has been creating thousands of hectares of new woodland across Great Britain, both on its own land, and through helping others to plant trees and new woods for a wide range of benefits. The spread of scrub on the rewilded Knepp Estate in Sussex has been associated with increases in local populations of nightingales, Purple Emperor butterflies and turtle doves (Tree 2018).

New woodland can contribute to the targets for woodland expansion in the Biodiversity Action Plans (reflecting international commitments under the Convention for Biological Diversity) by helping to improve conditions for threatened woodland species and communities. Woodland expansion fits with the recommendations for more, bigger, and more joined up, conservation areas proposed in the Making space for nature report to Defra (Lawton 2010), and also with climate change adaptation strategies (chapter 16). The increase in overall woodland cover may be part of the reason that woodland plants have shown fewer declines and more increases compared to species associated with unimproved grassland or arable weeds (Preston et al. 2002). We need, therefore, to look at how the flora in new woods develops.

Many new woods in the south and east of Great Britain are small, and planted on former arable or pasture. Before the canopy closes, the nutrient-rich ex-farmland soils that often underlie such woods favour competitive grasses and weeds, rather than shade-tolerant species. As the trees grow, their shade reduces the cover of the open ground plants, but the soil fungi content may still not be suitable for the establishment of woodland species. There are seldom any woodland species already present on site and the seed bank is usually of non-woodland plants, although there may be woodland species in nearby hedges that can colonise the new woodland.

A major programme of such lowland woodland planting was stimulated in the early 1990s by the then Countryside Agency through the Community Forests programme. This sought to encourage environmental and social regeneration through creating woods around towns, and more than 10,000 hectares of new woodland were planted (Mell 2011). One of the younger examples is Thames Chase, on the floodplain close to the Dartford–Thurrock River Crossing in south Essex. The Chase was formerly farmland, but even in the early days, when it was just rows of trees with rough grass in between, people came to walk in it. Some of the plantings have now closed canopy and a woodland flora is just starting to appear: the odd bit of Bramble, Ivy and Lords-and-Ladies. Compared to the bleak fields there previously, it is a considerable improvement.

The National Forest was a parallel initiative to the Community Forest programme, but on a bigger scale. Spread across 500 square kilometres, it covers parts of Derbyshire, Leicestershire and Staffordshire. Since 1995, its woodland cover has increased from about 6% to just over 20% in 2016 (Cloke et al. 1996). The trees are in the ground, but again the ground flora has taken its time to arrive. At Battram Wood near Ibstock in Leicestershire 20-year-old Oak plantations have mainly Bramble in the ground flora, the odd clump of Male and Broad Buckler-fern and occasional plants of Wood Avens. Remnants of the early open vegetation phase hang on in places – Rosebay Willowherb, dandelions and grasses such as Cock’s-foot, Common Bent and Red Fescue. There are also large areas of bare leaf litter with no plants growing at all. The same sort of pattern has been reported from woods created in the Central Belt of lowland Scotland.

Nevertheless, over a longer time-scale more interesting species do appear. Thirty-three small farm woods (0.1–31ha) planted on fields in the first half of the 20th century in the Vale of York were surveyed for their ground flora in 1990 (Usher et al. 1992). The surveyors found 115 species in total. Bramble, Nettles, Yorkshire-fog, Rosebay, Male and Broad Buckler-fern were widespread, but amongst the other species recorded were Wood Avens, Enchanter’s-nightshade, and specialists such as Common Dog-violet, Dog’s Mercury, Sanicle and Primrose. Nearly half the species were found in fewer than four woods, but Wood Speedwell and Wood-sorrel were in nine woods each and Bluebell in 24 out of the 33.

In 1918, the Reverend Adrian Woodruffe-Peacock described the species found in Poolthorn Covert, North Lincolnshire (Woodruffe-Peacock 1918), which was planted in 1797 on a rough pasture of Tufted Hair-grass. The Reverend noted various species that came and went, including Male-fern, Lesser Burdock, Common Hemp-nettle, Water Figwort Scrophularia auriculata, Marsh Thistle, Rosebay and Great Willowherbs. He thought that the Wood Anemone came in on shooters’ clothes, while Spurge-laurel and Honeysuckle were probably brought in by blackbirds. Bluebell might have come in with starlings: it first appeared under ‘fetid blackthorn bushes’ used as their roost. Unfortunately, that wood is no longer there to allow us to follow up his work, but the fortunes of plants in other new woods can be observed.

At Rothamsted Experimental Station in Hertfordshire (Harmer et al. 2001) there are two small self-sown woods which developed from arable fields abandoned in the 1880s. There was an initial increase in the species richness of the ground flora but many of these early colonists were light-demanding species that disappeared as the trees and shrubs grew up. By 1915, one of the woods, Broadbalk Wilderness (0.13ha), had Ivy throughout with Lords-and-Ladies, Sanicle and Dog-violet. Dog’s Mercury with traces of False Brome were present, mainly towards the woodland margins, along with Hogweed, False Oat-grass, Hedge Woundwort and Nettle (Brenchley & Adam 1915).

By 1998 the flora included Bluebell, Black Bryony and Lesser Celandine, species that might be found in long-established mixed woodland in the Midlands. Other species were still only at the woodland edges – Ground-ivy, Wood Avens, White Bryony. The somewhat larger Geescroft Wilderness (1ha) had more woodland species. Bluebell and Lesser Celandine dominated the ground flora and formed dense patches throughout the wood, although again other woodland species were mainly confined to the margins.

The flora in these two small woods is limited by their size, closed canopy and lack of open space, hence the many species found mainly at the margins. Under the canopy, competition from ground-growing Ivy may also be a limiting factor, and in Geescroft the spread of Holly in the understorey will slow further expansion of the woodland flora. The various surveys include some species that established for a while but had died out again by 1998, including False Brome, Pignut, Sanicle and Bittersweet in Broadbalk, and Wood-sedge, Giant Fescue, Sanicle and Sweet Violet at Geescroft. Presumably these could colonise again if conditions within the wood became more suitable for their growth; their current absence is not because they cannot reach the woods.

Another study of woodland development is under way at Monks Wood National Nature Reserve in Cambridgeshire. A four-hectare field next to the wood has been left to colonise naturally since the 1960s with Oak, Ash, Blackthorn, Hawthorn and Dogwood. After 38 years Kevin Walker and Tim Sparks recorded 89 ground flora species in it, including both woodland generalist and specialist species. Some were largely confined to the edges (Bugle, Dog’s Mercury, Wood Avens, Three-veined Sandwort), but others had spread through most of the area (Ground-ivy, False Brome, Lords-and-Ladies) (Walker & Sparks 2000).

Oxlip, bluebells and Dog’s Mercury have spread into the self-sown triangular field (believed to have developed as woodland since the 1920s) next to Hayley Wood in Cambridgeshire (see also chapter 4). In 2017, I also saw False Brome, Wood-sedge, Spurge-laurel, Wild Strawberry, Wood Avens, Honeysuckle, Sanicle, Black Bryony and Common Dog-violet, which is not a bad plant list for a small piece of young closed-canopy woodland! Bird’s-nest Orchid has also been recorded there.

These studies show that a surprising number of species may make it even into small woods. They allow woodland plants such as Enchanter’s-nightshade to occur in landscapes where they would otherwise be rare. Moreover, compared to the improved fields and arable crops that they generally replace, small woods provide opportunities for common species such as Bramble and Nettle to spread, so increasing the potential food plants for a wide range of invertebrates.

Enchanter’s-nightshade

Despite its common name, Enchanter’s-nightshade Circaea lutetiana is not related to the other (poisonous) nightshades but is in fact part of the same family as Evening Primrose and Fuchsia. It is found throughout Great Britain, except in parts of northern Scotland, as well as occurring across much of the rest of Europe. It favours fertile, disturbed soils, including shaded gardens (Grime et al. 2007).

Enchanter’s-nightshade is one of the later plants to emerge in the spring, and its small white flowers provide some summer variety for the woodland flora. It has rather thin, weak leaves compared to Dog’s Mercury, with which it often grows, but is more able than the Mercury to colonise new woodland sites. The rounded fruits are covered by hooked bristles – miniature burdocks – that are easily picked up on the fur of passing animals or surveyors’ socks, giving the potential for long-distance spread. Once established, creeping above-ground stems allow the plant to fill its immediate surroundings. Its tolerance of shade means that it can form dense stands under conditions that exclude more vigorous, taller-growing species such as Rosebay Willowherb.

Enchanter’s nightshade, a fast-colonising woodland species.

After the First World War, the Forestry Commission was looking for areas to plant up and the Breckland, a large sandy area on the Suffolk/Norfolk border, was identified as being of little value for farming. Its heaths and acid grassland were known for their unusual flora, the closest perhaps we have in Great Britain to central European steppe-type grass-heath (Dolman et al. 2012). However, this did not stop large areas being planted with Scots and Corsican Pine to become what is now Thetford Forest.

Fortuitously, we have a detailed account of the vegetation development for one area. West Tofts on the eastern side of the Forest had been divided into small blocks, each of which was planted with a different tree species. In the 1950s, some 20 years after the trees had been established, John Ovington, a researcher with the fledgling Nature Conservancy, recorded what was growing under the trees (Ovington 1955). The unplanted plot was the richest, with over 43 vascular plant species recorded and 99% ground flora cover. Deciduous plots (Birch, Alder and European Larch) were generally similar to each other in richness and cover of the ground flora (17–21 species; 64–77% ground flora cover), with grasses (Common Bent, Sheep’s-fescue, Yorkshire-fog) being the most frequent species. Nettles were common in the Alder plot, probably reflecting higher soil nutrient levels as a result of the nitrogen-fixing nodules on Alder roots. The Corsican Pine and Douglas Fir plots had only six and two vascular plant species recorded respectively, and less than 3% cover. There was little Bracken.

In the early 1970s Mark Anderson of Forest Research resurveyed this area. The stands were still fairly grassy, but Bracken had increased under the Larch and Douglas Fir. The two stands that had been species-poor in 1952 (Corsican Pine and Douglas Fir) had been thinned and gained species, whilst the Larch and Alder plots had lost species, probably because they had become more shaded (Anderson 1979). Anderson concluded that the vegetation under different tree species tended to converge around a dozen common species, although this pattern was less clear at West Tofts than in his other two sites (Bedgebury in Kent, Abbotswood in the Forest of Dean).

When I visited West Tofts in 2017 some stands in the original trial had been felled and replanted and others had opened up through windthrow. Most of the plots consisted of tall Bracken with Raspberry, grasses and a little Nettle, almost exclusively woodland generalists. The unplanted plot in the original Ovington study had become a young pine stand over Bracken.

With relatively uniform site conditions (flat topography, free-draining soils) and a predominance of one type of trees (pines) in the crop, the woodland flora that developed in Thetford Forest was always likely to be uniform. Large areas are dominated by Bracken as in the West Tofts section, some areas by Bramble, with a scatter of Wood Sage, Heath Bedstraw, Bent Grasses and Sweet Vernal-grass. Where denser canopied conifers, such as Spruce, have been planted, or in the young pine stands, there is very little flora. Beech stands, often planted as strips along the forest roads, show a slightly different development with small patches of False Brome and Enchanter’s-nightshade – perhaps heralding the start of a richer woodland flora in future.

The creation of Thetford Forest certainly destroyed large areas of the Breckland flora. However, the Breckland flora has also been lost from much of the unplanted surrounding landscape because of agricultural improvements since the Second World War. At least in the forest some of the commoner open ground species remain along the rides and timber stacking areas or reappear after disturbances such as felling. Of the 43 species that Ovington recorded from the unplanted area in 1955 I saw about half, for example Lady’s Bedstraw and Wild Mignonette, along a track leading up to Ovington’s study area in 2017.

Rosebay Willowherb – a plant of young plantations and other disturbed ground

Rosebay Willowherb Chamerion angustifolium was once considered rather rare, although familiar enough to be mentioned in Gerard’s herbal. The last century then saw a major expansion of this species. It spread dramatically in London after the Blitz, on the mass of rough ground, broken walls and roofs. This affinity with burnt or disturbed sites such as along railway lines is reflected in its common American name of Fireweed.

It is generally common in the early stages of new woodland, although by 1955 in Thetford it survived mainly in the unplanted areas (Ovington 1955, Grime et al. 2007). In new woodland, or after felling in established woods, it takes over bare ground to create dense stands over a metre tall. Below ground Rosebay has long, horizontally-spreading, roots and in America these can help to stabilise areas of bare ground after fires or logging that would otherwise be subject to soil erosion. In British woods it may also help reduce loss of nutrients and soil on disturbed ground.

Rosebay’s reign may be brief, because it does not survive long under shade. In the Thetford study, it was hardly recorded where the planted stands had closed canopy. Similarly, in Wytham it was found in 38 (nearly a quarter) of the permanent plots in 1974, when many stands had only just closed over following a period of post-war planting. By 1991, 17 more years of shade and little disturbance reduced Rosebay to 13 plots, and it was completely absent from the 1999 records. It has since turned up occasionally in one or two plots, usually where there has been thinning or windthrow. These disturbances provide the right conditions of bare ground and lots of light that allow the Rosebay to bounce back again from wind-blown seeds, whose drifting white plumes are a common sight in autumn.

Rosebay’s abundance and association with derelict ground mean that it is sometimes dismissed as just another weedy species by botanists, but that is to overlook the brilliance of its massed flowers. Londoners voted it as their County Flower in Plantlife’s 2002 competition.

Though often thought of as a weedy species, Rosebay is attractive en masse.

There are fewer differences in the vascular plants found in recent versus ancient broadleaved woods in the uplands of Great Britain. Upland woods often have many gaps in the canopy, making them quite open, and may be heavily grazed like the adjacent hill-land. Woodland boundaries may be less fixed, with periods of woodland expansion and contraction in response to changes in grazing pressures. There are also more places in the uplands where woodland ground flora species can survive outside woods, such as along stream-sides and under Bracken.

Carrifran, in the Southern Uplands, was typical of many valleys and glens in having very little tree cover in the late 1990s, but, as mentioned earlier, a fantastic new woodland is now developing (Ashmole & Ashmole 2009). Bracken, Bilberry and Wavy Hair-grass were widespread on the open moor prior to planting, and a range of other woodland species, including Wood-sorrel, Common Dog-violet, Primrose, Sanicle and Opposite-leaved Golden-saxifrage, were scattered through the valley. On a visit in 2014, I also saw Wood Anemone, Bluebell, Early-purple Orchid, Globeflower and Water Avens. As the trees grow, there will be the potential for these woodland species to spread.

Birch, one of the main trees planted at Carrifran, is also common in naturally-developing woods in the uplands, such as in the strips along railway lines that were fenced against stock. Very dense young Birch stands may have just litter below, but usually there is more of a ground flora in new birchwoods than under many new lowland woods. This flora may be mainly Bilberry, Wavy Hair-grass and Bracken, with Purple Moor-grass where conditions are wetter, or Bramble if the soils are a bit richer.

The changes in the vegetation under a developing birchwood were studied in north-east Scotland in a series of woods that had been growing on moorland for 20 to over 60 years. Heather initially increased, because the grazing was reduced, then declined with increasing canopy closure because Heather is not very shade-tolerant (Hester et al. 1991, Mitchell et al. 2007). Common Bent and Wavy Hair-grass spread as the woodland matured and opened out. The changes in the vegetation were mainly driven by the changing light regime under the trees. However, some soil nutrients were also more available under old Birch than on moorland. Where the woods continued to be grazed Bilberry was reduced, giving the less palatable but slower-growing Cowberry more of a chance.

A pioneering exercise in woodland creation under extreme conditions was carried out in the 1960s to extend the native woodland on the National Nature Reserve of the Island of Rum in the Inner Hebrides. Some areas were ploughed, and Lodgepole Pine used as nurse crops that were eventually to be removed. Peter Wormell, the warden, was involved from the outset and some of the first trees were hand-planted by his wife in 1959. In 1994 he wrote enthusiastically of the thousands of trees established, their growth, and the variety of structures developing in this new woodland (Wormell 1994).

There were woodland ground flora species already on the island of Rum, in isolated fragments of woodland on crags and in gulleys, and in places these spread. In the Kilmory Fank plot Peter described bluebells bursting with seeds under the planted trees; in the Kilmory North Plot Wood Sage, Wood-sorrel and Bilberry were spreading below the trees, Golden-scaled Male-fern and Hard-fern more in evidence and natural regeneration of Honeysuckle abundant. Underneath young alders, Marsh-marigold, Primrose and Meadowsweet were extending along a streamside. A later report on the Rum plantings was more circumspect (Stiven & Smith 2005); while a significant area of woodland had been established, tree growth was slow with little sign that the trees were regenerating outside the original plantings. Nevertheless, a gradual development of woodland plant communities was taking place.

When I visited Rum in 2018, I could see how both sets of conclusions might be drawn. There are some mature 19th-century mixed plantings established in sheltered sites on relatively fertile soils. These have well-developed carpets of Bluebell, Primrose, Pignut, Wood-sorrel and Honeysuckle. On an undercliff on the east coast there is a strip of Birch and Rowan, with occasional Alder and willows, which has a flora of Bluebell, Creeping Soft-grass, Bracken and Wood-sorrel. It looks largely semi-natural but runs into a patch of young Oak that was clearly part of the conservation plantings and has a similar ground vegetation. Some of the patches of Alder and Birch, established in the 1960s along stream gullies, have mainly Purple Moor-grass, but with Bramble, Lady Fern, Broad Buckler-fern, Hard-fern, Tormentil, Self-heal and Soft Rush. Other 1960s pine stands have just Purple Moor-grass beneath them, with the old plough lines still visible. However, Purple Moor-grass is the major component of some western native pinewoods, such as at Sheildaig Forest in Wester Ross on the mainland, so perhaps even these species-poor stands on Rum are in one sense developing the ‘expected’ woodland ground flora.

There are, in Great Britain, about 126,000ha of Larch and 665,000ha of Sitka Spruce, making the latter now our commonest forest tree. Many were established, both by the Forestry Commission and private owners, in the 20th century, in regions and on soils where semi-natural woodland was scarce. In the first few years after planting there was often increased vegetation growth because the rabbits, sheep and deer were fenced out (Hill 1979, Coote et al. 2012). However, once the trees got away and closed canopy, vascular plants were more or less eliminated over large areas under crops of Hemlock Tsuga sp., firs and the ubiquitous Sitka Spruce. In a study at Kielder Forest, 17% of stands had no ground vegetation, while in a further 40% of stands the vegetation was too sparse to assign to any specific vegetation type (Wallace & Good 1995). The plantations were often managed on short rotations (30–50 years), generally under ‘no-thin’ regimes to reduce the risk of windthrow. This gave little opportunity for the flora to reappear in older stands before they were clear-felled.

Where a ground flora does develop under conifers, the commonest species tend to be Broad Buckler-fern, Wavy Hair-grass and Bilberry; less frequent species include Heather, Rosebay Willowherb, Heath Bedstraw and Purple Moor-grass. Mosses and liverworts can be abundant (though limited in species richness). Thicker patches of grasses and sedges often pick out small gaps in the canopy. More extensive vegetation cover is generally found under mature Larch and Pine, with Bramble and Bracken on fertile sites. In Ireland, a grass-dominated community has been described under mature Larch with Common Bent, Sweet Vernal-grass, Sheep’s-fescue, Wavy Hair-grass (French et al. 2008) that is similar to some of the grassy stands in lowland pinewoods in Great Britain.

After the trees are felled, the ground flora spreads back, mainly from species in the soil seed bank, for example Heather, various sedges and rushes, Heath Bedstraw and Foxglove. Incoming wind-dispersed species such as Rosebay Willowherb can be important. Woodland species that were already present in the stand benefit from the increased light after felling, but these are usually just a small range of grasses and Bramble. At Kielder, Purple Moor-grass tends

Broad Buckler-fern

If a species is to become common in upland conifer plantations it needs to be able to tolerate generally acid soils, heavy shade and have an efficient system for dispersing across the landscape. Broad Buckler-fern Dryopteris dilatata meets these criteria. It is one of our most common ferns, in all types of woodland, along with Male-fern and Bracken (Page 1988).

Its much-divided fronds emerge from the underground stem, often looking a little untidy or straggly compared to Male-fern. The clumps may be quite dispersed or locally abundant. On deep humus-rich soils the clumps can be massive, with fronds up to waist height. In wet alderwoods the fern may be clustered on the raised drier ground around the tree bases. Like some other ferns it contains toxins that reduce its palatability to herbivores, but it may be eaten none the less. In winter it dies back, the fronds being susceptible to hard frosts that leave them blackened.

In late summer, small round structures develop on the underside of the fronds which contain the spores. Millions of them may be produced from a single plant! These are wind-dispersed and can be found in the top few centimetres of many woodland soils. They germinate in response to light, but what is produced is not a new fern as we usually see it. Instead, a different type of plant grows that has only half the number of chromosomes. It is much more like a moss or liverwort and needs moist, humid conditions. Cross-fertilisation occurs during this stage and a sporeling with the more familiar fern structure then develops, often to be seen on old tree stumps.

Like many common woodland plants Broad Buckler-fern may be dismissed as not of much interest, but the tracery of its leaves is worth a second look.

Broad Buckler-fern with its finely dissected fronds.

to dominate the clear-fells at higher altitudes, Wavy Hair-grass on drier sites with shallower peat and Tufted Hair-grass at lower levels. Bracken may spread in from plants surviving along ride-sides. Where Heather re-establishes quickly and in abundance, the replanted area may for a short time show some similarities to the former moorland vegetation. As the second-rotation crop starts to grow, however, the species of open conditions are shaded out again.

The second and subsequent rotations of conifers in these new forests are likely to be more varied than the first. Foresters are moving away from very large clear-fells on many sites and are trying to use a wider range of tree species to reduce risks from climate change and outbreaks of pests and diseases. Where the trees can be thinned and allowed to grow on for longer, the vegetation underneath them starts to become more interesting, with carpets of mosses, Bilberry and Wood-sorrel. The occasional wind-thrown tree, with its associated pit-and-mound, adds to the diversity. There may be, over time, a gradual accumulation of woodland plant species (Quine 2015), such that stands start to resemble more natural Larch forests on the Continent or stands in the Pacific North-West of America where Sitka Spruce is native (Tittensor 2016). For the time being, though, their flora may remain limited, with mainly woodland generalists such as Broad Buckler-fern.

The National Vegetation Classification (NVC) has been used, and sometimes abused, as a model of how we might expect new woods of native trees to develop. Some of the Rum plantings, and pine plantations elsewhere, do show similarities to types described by Rodwell (1991) and may contain pinewood specialists such as Twinflower. Mature/over-mature stands of Sitka Spruce can show some overlaps with upland Oak type communities (W11, W17) (Ferris et al. 2000), partly because these Oak communities themselves are poor in vascular plants. We should not, however, expect the vegetation of new upland plantations to be a direct analogue of that in nearby broadleaved woodland (Wallace 2003). Conifer forests have a different structure and create a different type of microclimate for the ground flora. Conifer woods are generally darker, there are changes in the water regime and the rain of needle litter is different to the autumnal broadleaf leaf-fall.

People sometimes compare habitat conservation to the listing and conservation of buildings. If we pursue that analogy ancient woods can be likened to old houses and churches that should generally be protected. Some new woods clearly show potential to become treasured features of the future, while other commercial Sitka stands may be seen as equivalent to the 1960s ‘brutalist’ period of concrete architecture. However, just as old churches and thatched cottages do not make good office space compared to modern designed buildings, upland conifer plantations can fulfil different and still valued functions for society. When there were proposals to sell off the state forests in England around 2010, the big plantation forests of Kielder and Thetford had as passionate supporters as did the New Forest or the Forest of Dean.

New woods and plantations are part of the 20th century’s additions to our landscapes. Our successors will have their views on what parts of this legacy they want to keep and which to discard. However, we should at least try to improve the quality of the new woodland we are leaving the next generation.

If the only objective in creating a new wood is to produce timber, what matters is getting the trees into the ground and growing well. However, most woods also fulfil other functions, and many woodland owners are interested in wildlife: they may therefore want their new woods to have a rich flora. New woods created where there are already a lot of woodland plant species on or near the sites are likely to end up richer than those formed on species-poor and isolated locations.

Some of the species moving in to Geescroft and Broadbalk Wildernesses, described above, were probably already present in the adjacent hedge. In Lincolnshire, woods created next to old hedges or ancient woodland were more likely to have been colonised by woodland species (Peterken & Game 1984). In Sweden, oak plantations showed decreasing species richness with increasing distance from the new site to the nearest ancient woods (Brunet 2007). Similar results are also emerging from a major study of the new woods created in the Central Belt of Scotland and in the National Forest in central England (Watts et al. 2016).

New woods are also likely to end up richer in plant species if they grow where there are already some woodland/woodland edge species present. John Rodwell, primary author of the National Vegetation Classification volumes, and Gordon Patterson from the Forestry Commission, identified the sorts of semi-natural open vegetation that would be best for developing different broad groupings of woodland communities (Rodwell & Patterson 1994). For example, to create (by planting or natural regeneration) a lowland Oak–Bramble–Bracken woodland the best sort of site might be one with rough grassland of False Oat-grass, Yorkshire-fog, Cock’s-foot, Hogweed, Rosebay, Nettle and Thistle. For an upland Oak–Birch–Bilberry woodland we should look for moorland with Heather, Bilberry, Bell-heather, Wavy Hair-grass, Sheep’s-fescue, Mat-grass, Common Bent, Sweet Vernal-grass, Heath Bedstraw and Tormentil.

The Reverend Woodruffe-Peacock’s list of species that were probably in the rough pasture where Poolthorn Covert was planted is quite long, as is the list for what was on Broadbalk Wilderness before the trees started to grow. Species such as Tufted Hair-grass and Marsh Thistle can carry over from grassland to the woodland edge/ride component of a new woodland; Wavy Hair-grass and Bilberry do the same over the moorland/woodland transition.

In Allt Gleann a’Chlachain in the central Highlands of Scotland a new woodland was planted with the intention that this would provide shelter for sheep and improve the farming productivity. John Holland from Scotland’s Rural College looked at how the vegetation changed after the sheep were fenced out. There was extensive growth of Heather, Bilberry and Purple Moor-grass. Tall herbs emerged and flowered, including Meadowsweet, Wood Crane’s-bill, Lady’s Mantle, Angelica, Melancholy Thistle, Common Valerian, Northern Bedstraw Galium boreale and Goldenrod. Large patches of Wood Anemone developed across the well-drained slopes. There was also an increase in butterflies such as Scotch Argus and Small Pearl-bordered Fritillary.

Rodwell and Patterson, however, stress that ‘existing vegetation may have great conservation and amenity value in its own right’. Often the decision is made that we should not sacrifice the existing open vegetation character and value for the sake of any potential increase in value of the new woodland cover. Woodland development then goes on to other land where the vegetation is not so rich, but this means that conditions may not be so suitable for developing woodland plant communities. No wonder new woods are often initially rather dull botanically!

High soil fertility in ex-farmland soils often limits the spread of species into and through new woods. Woodland specialists may grow better under high nutrient conditions in greenhouse experiments (Hipps et al. 2005), but in the field they are likely to be outcompeted by tall species, such as Nettle, False Oat-grass and Cow Parsley, that are better able to thrive in enriched soils. While it is easy to add nutrients through fertilisers, it is not so easy to reduce fertility levels. An approach trialled by the charity Landlife was to use a machine like a deep plough that turns the soil over, effectively burying the topsoil. The new surface layer consists of the less fertile subsoil and does not have as many buried weed seeds as the original upper layers. The vegetation that then grows up is less competitive for the trees and more attractive for visitors to the site, because it contains more flowers. The technique has been used to produce rich woodland edge communities amongst new tree plantings.

The Reverend Woodruffe-Peacock concluded ‘that in time most, if not all, woodland species adapted to any particular soil or soils gradually arrive, by chance means, at a covert … as isolated as Poolthorn.’ However, ‘in time’ may mean a century or more. There is a case for giving some species a helping hand, because ‘It is better to light a candle than to curse the darkness’ (William Watkinson 1907).

The principles around introducing wildflowers to new sites as a nature conservation practice were developed in the 1980s for creating species-rich grassland (Wells et al. 1981). There are now a wide variety of grassland wildflower mixes available from commercial seed firms. It has taken longer for woodland ground flora introductions to be accepted as a conservation technique in new woodland (Blakesley & Buckley 2010). It may still feel too much like gardening for some, but in 2008 the charity Flora Locale noted in one of its leaflets ‘On farms, in forests, city parks, suburban gardens, quarries and on roadsides – British wildflowers, trees and shrubs are being sown and planted on an unprecedented scale’.

People had long been introducing species for aesthetic reasons: there are clumps of daffodils and snowdrops on the top of Wytham Hill for example. Other species, such as Pendulous Sedge, might be introduced as potential food for pheasants. Then in the 1980s Laurie Boorman of the Institute of Terrestrial Ecology, whose colleagues were working on creating flower-rich grassland, set up a small-scale trial of woodland ground flora introductions for conservation purposes in Milton Keynes. Milton Keynes was chosen because there was a great deal of planting of native trees and shrubs along roadsides and amongst the houses. The initial work involved both the sowing of seeds and planting of young plants into six young plantations. Establishment was good: all, but one, species were found a couple of years later.

The work was then followed up by Joanna Francis as part of her PhD from the University of London (Francis & Morton 2001). Over the next ten years, regular visits confirmed that the introduced species were surviving and in 2000 a more detailed re-survey was carried out. Foxglove was no longer found, but it needs open woodland and the plantations had probably just become too dark for it. The other species were still in most of the woods where they had been introduced, although the spread to new areas was slow. Wood Avens had done well, as had Red Campion and Hedge Garlic where conditions were suitable. Bluebell spread was perhaps enhanced where the plantations were on embankments: the seeds could just roll downslope.

Since the initial trials, more than 100 hectares at Milton Keynes have had woodland species introduced to them, including specialists such as Wild Garlic, Primrose, Pignut and Sweet Woodruff. The species used were, with a few exceptions, all locally native plants. As Francis and Morton (2001) put it ‘A visit to these plantations in spring may surprise the majority of woodland managers. Those expecting rank "weed" species to dominate … will find instead a wealth of woodland field layer species growing vigorously beneath the trees’.

In April 2019 I was shown examples of the Milton Keynes roadside belts and their developing flora. Unthinned areas were typically densely shaded with little ground flora, or, where there was a bit more light, a ground-covering carpet of Ivy. Where the ground flora had been introduced, primroses were blooming in abundance; bluebells were less advanced, but other woodland species seen included Pendulous Sedge, Wood Avens, Greater Stitchwort, Wild Garlic and Red Campion.

The Highways Agency have taken an interest in further development of this work, because extensive tree planting and natural regeneration happens regularly alongside major roads. The Highways Agency (2005) guidance is that the plants should be put into relatively open ground conditions with extensive leaf litter, and only a low cover of competitive species such as Bramble, Bracken, Ivy and tussocky grasses. The sites must not be too shaded, or the plants will not grow, nor too open as competitive species will dominate the introductions. Moderately fertile soils are best – if conditions are too rich there will be lots of nettles. Various mixtures of species have been designed, comparable to the different grassland mixtures, and experience gained on which species may be best introduced as seed, and which as small plants. Generally, the aim is to establish ‘islands’ of woodland flora that can subsequently spread on their own.

About the same time as Boorman was starting his experiments in Milton Keynes, woodland flora introductions were being trialled on Rum (Stiven & Smith 2005). The 1984 proposals were for species to be introduced to the oldest planted plots, in isolated situations where possible confusion with any natural immigration or spread from the pre-existing flora would be limited. Red Campion, Wood Anemone, Dog’s Mercury and Hairy Wood-rush were used. The plants apparently established well, but unfortunately there is little information on how these trials did subsequently.

Recently there has been renewed interest in Scotland in introducing woodland ground flora to the many plantings that have taken place through the Central Belt and which are often overrun by just nettles, Hogweed or Rosebay Willowherb. Trials are being established to demonstrate cost-effective ways of integrating the introduction of the ground flora with other aspects of the woodland creation process. The idea is that the introduced plants should act as sources for further natural colonisation of the woodland (Worrell et al. 2016).

As happened with grassland wildflower introduction, the initial research sowings have progressed to the point where commercial woodland seed mixes are now available. A wide range of woodland plants may be included, ranging from woodland specialists such as Wild Garlic, Pignut and Nettle-leaved Bellflower, through generalists such as Foxglove, Hedge Garlic and Wood Sage, to non-woodland species such as Yarrow and Agrimony (Blakesley & Buckley 2010). If a species is to be included in the mixes it is helpful if it produces lots of seed on a regular basis. This is better for the nurseries who need it to bulk it up, and also for the subsequent spread of the plant in the sites to which it is introduced. One obvious species in this respect is Red Campion, which may have several thousand seeds per individual.

If woodland plants struggle to establish well in new woods because the soils are not suitable, could this problem be partly overcome by bringing in ancient woodland topsoil with its array of micro-organisms as well as buried seeds, bulbs and rhizomes? There are now instances where this has been done, usually associated with the granting of planning permission to destroy a patch of ancient woodland. Soil collected from the threatened site has been taken to a new location; there the existing topsoil is removed and the material from the ancient woodland spread in its place.

An early case involved Biggins Wood in Kent, a small ancient wood that had the misfortune to lie where the Folkestone terminal for the Channel Tunnel rail-link is now. In 1988, some topsoil from this coppice was moved to a nearby hillslope; the area was then planted up. In the new wood there was an initial increase in the abundance of open-ground species, mostly from the soil seed bank, but many shade-tolerant species survived and spread, including Wood-sedge, Primrose, Stinking Iris and Wood Speedwell (Buckley et al. 2017). Some woodland specialists such as Remote Sedge and Narrow-spiked Wood-sedge did not transfer well and were limited to small seepage areas on the new site, which was generally drier than the original wood. The trees planted on the new site created a rather uniform canopy, and a more variable light climate might have allowed a wider range of ground flora to develop.

Red Campion

Red Campion Silene dioica is a tall, opposite-leaved herb, commonly found in woods and hedgerows where it can provide a summer nectar source for bees and butterflies (Comba et al. 1999). It generally grows on damp fertile soils in lightly shaded situations. There is little risk of it being planted outside its native range because it is present through much of Great Britain, except the central and northern Highlands and south-west of the Wash. In Cambridgeshire and Huntingdonshire it has been described as rare in the past, possibly as these are low rainfall areas and the plant is not very tolerant of drought, but also because these counties have few woods!

Campion can survive in deep shade as non-flowering rosettes of leaves, but then flower profusely after coppicing, or along woodland edges (Grime et al. 2007), further characteristics that make it a good species for woodland introductions. There are separate male and female plants which have rather different behaviours. Male plants tend to be more common and to produce more flowers and leaves than female plants; female plants make up more of those flowering in June. Later in the summer males and females flower equally. However, it is unlikely that any sowing would be made up of just male or just female plants, so the separation of the sexes is not likely to be a problem. A final point in its favour for planting mixes is that it has attractive flowers.

Red Campion, a favoured species for woodland introductions.

A second example involved a five-hectare outlier of the Blean Woods north of Canterbury that was to be quarried. The soil was stripped off the old site and loose-tipped on to the prepared new site. Weedy species, such as Sowthistles Sonchus spp. and Redshanks Polygonum persicaria, quickly colonised the open, disturbed soil surface, although they soon declined as competitive species such as False Oat-grass, Soft Rush and Creeping Soft-grass increased. There were still differences from the pre-translocation state after ten years, but the development of the woodland vegetation was faster than if the new woodland had been left to be colonised naturally from surrounding areas. Some of the woodland specialists, such as Wood Anemone and Yellow Archangel, were less common than in the donor area, although no species was lost completely (Craig et al. 2015). Others, such as Bluebell, were more resilient and became dominant with Bramble over large areas. Autumn translocation of the soil was better for the woodland ground flora than moving the soil in spring, particularly in the early years, possibly because in autumn more of the plant material was dormant when the soil was moved and so less likely to be damaged.

A third example from Kent is where the A21 was widened past Tunbridge Wells in 2015–2017. Nine hectares of ancient woodland, mainly Sweet Chestnut, were destroyed in the process. The mitigation involved work to improve 27 hectares of existing woodland and the creation of 18 hectares of new woodland. Live stools of Sweet Chestnut and Hazel stools were transplanted into the new sites along with other stumps and logs so that the new woodland would start with a substantial legacy of dead wood. The top 20 centimetres of soil from the ancient woodland were spread on some of the new woodland areas.

Not all the Sweet Chestnut and Hazel survived the move, but enough did in the short term to give the new areas a very different structure from a standard plantation. The initial response from the soil translocation was also promising. Bramble and grasses, such as Sweet Vernal, dominated large areas, as they would in a clear fell or coppice, but when I was shown the site, I saw other woodland species such as Bluebell, Primrose, Pendulous Sedge and Wood Anemone amongst them. There was a clear contrast to the weedy flora found where no topsoil had been spread. None of these woodland species would have been on the site, a rather nondescript grass field, beforehand. I also saw more butterflies flitting amongst the soil translocation areas than elsewhere.

Soil translocation is a controversial, expensive and uncertain procedure, so has only been considered with major infrastructure projects such as new roads or airports. It may, for example, also become part of a package of mitigation measures applied where ancient woodland is destroyed along the route of the new HS2 London–Birmingham rail link.

Planting woodland ground flora and moving woodland soils may lead to claims that ancient woodland communities can be recreated in this way, weakening the need to protect such sites from development. This is not true. At best, only some of the relevant species, even among the vascular plants, are transferred and the translocated soil and its micro-organisms are much disturbed in the process. It should only be considered when all other options for avoiding the destruction of ancient woodland have been exhausted.

We are not ‘recreating ancient woodland’ but are (hopefully) speeding up the development of more interesting and richer woodland ground flora communities in these new woods. The experience gained may be relevant to situations where we might want to assist the spread of species beyond their current native range as part of adaptation to climate change, considered in the next chapter.