This site and its content are under development.

Quercus petraea (Matt.) Liebl., Sessile Oak

Account Summary

Native and frequent. European temperate.

1934; Praeger, R.Ll.; Co Fermanagh.

Throughout the year.

Growth form and preferred habitats

A relatively long-lived deciduous tree, Q. petraea can form the dominant top canopy species in mixed woodland, especially in more upland woodland or in coppice. It performs best on well-drained, shallow, moderate to strongly acidic mineral soils.

Fermanagh occurrence

Although this species is nowadays regarded as the native Irish oak, in the 19th century, probably partly for reasons of identification difficulty, it was considered a rarity (eg Cybele Hibernica 1898). Currently the Fermanagh Flora Database indicates that it is less frequent in the VC than the closely related Q. robur (Pedunculate Oak), representing only 35% of the total combined records of these two common oak species. Q. petraea has been frequently recorded in Fermanagh, but it is widely scattered across just 82 tetrads, 15.5% of those in the VC. It is mainly distributed in woodlands on demesnes around the major lakes, with a major concentration of records around the Crom Castle estate.

In comparison, Q. robur is represented in 173 tetrads, 32.8% of the squares in the county. Clearly, Q. robur is much more generally scattered throughout our woodlands than Sessile Oak. Two published studies have involved oaks in Fermanagh: the first was a comparative study of oak leaf characters across 35 woods in N Ireland, nine of them in the VC, carried out by Rushton (1983); and secondly and more recently, a genetic study by Kelleher et al. (2003) of Irish oak material from across the island that included samples from the Crom Castle National Trust estate.

Regarding oaks in Fermanagh, one of the oddest things is that Q. petraea is so well represented around the shores of Lough Erne, and particularly in the low lying area around the Lower Lough where limestone rock outcrops very frequently. It may be that these particular soils are too shallow and dry to support Q. robur, but clearly Q. petraea can tolerate the peculiar set of environmental conditions they offer. Nevertheless the two common oak taxa overlap in Fermanagh as they do elsewhere, both in their ecology and distribution. A species coincidence map of the Fermanagh data displayed 25 sites and subsites in the VC where the species had been recorded together.

Oak variation, hybrids and identification difficulties

Species overlap of Q. petraea and Q. robur is also the case throughout their geographical range in Europe (the distribution overlap being partially due to past phases of timber plantation in which Q. robur was the preferred timber tree). On account of the amount of variation expressed, many oak trees are very difficult to assign in the field to either Q. petraea or Q. robur. Indeed, many specimens may be fertile hybrids or extreme forms of one or other of these two genetically introgressive species (Cousens 1963, 1965; B.S. Rushton, in: Rich & Jermy 1998). Q. petraea is generally regarded as a relatively good, homogenous species at least in S Ireland and possibly also in S England. On the other hand, Q. robur is more variable and has been 'tainted' with gene flow from Q. petraea, and probably also from other European oak species earlier in history, probably during one or more of the 'Ice Ages' when all oak species were forced to migrate south into refuge areas in S Europe (Cousens 1965).

While there are areas in Britain and Ireland where one or other of these two native oak species is the more prevalent, they obviously overlap to a considerable extent in both their ecological and geographical ranges, and undoubtedly their natural distribution is modified by past timber plantation. Oak hybrids between the two species are also very variable and they further complicate the question of identification. The hybrids are frequent, fertile and are known as Q. × rosacea Bechst. As a result of introgression and backcrossing of the hybrid with both its parents, the precise limits of the three taxa (ie the two parents and their hybrid) are controversial, and they have been the subject of investigation for many years (Stace et al. 2015).

Some taxonomists faced with the oak identification problem have even tried to redefine the whole species concept in plants (Burger 1975). Van Valen (1976) has emphasised the notion of 'multispecies', a form of ecological species concept embracing the ecocline and gene transfer, rather than the reproductive species concept based on genetic isolation. This is a deep and provoking topic, but it is interesting and worth mentioning since it arose out of the difficulties faced when identifying oaks. The arguments and the pattern of variation involved in oaks has led to a continuing academic debate as whether Q. petraea and Q. robur should be considered two separate species or as two ecotypes within a common species (Muir et al. 2000; Thomas et al. 2002).

A study of the two oaks by Bacilieri et al. (1996) for instance, showed that both species were almost completely out-crossing (ie there was a high degree of self-incompatibility). It found that there was a considerable rate of unidirectional gene flow from Q. petraea to Q. robur, varying between 17% to 48%. These workers regard Q. robur ecologically as a pioneer species, which is, or was in past forests, progressively replaced by Q. petraea. They believed that the measured gene flow reinforces the succession occurring between the two species (Bacilieri et al. 1996).

At the same time these authors and others query how it is, that despite the high level of interspecific gene flow, Q. petraea and Q. robur somehow remain differentiated to the degree that they do throughout their natural geographical distribution (Zanetto et al. 1994; Bacilieri et al. 1996; Main et al. 2000). Herein lies the real mystery! As Stace (1975) and Stace et al. (2015) have shown in a British and Irish context, hybridization is a very frequent and normal process in flowering plants, and it is especially expected and developed in long-lived and clonal species.

Timing of flowering (anthesis) and occurrence of hybrids: While as noted there are differences in the distribution and ecology of the two oak species, there is also a small separation in the flowering time between the species in the same locality. Flowering in Q. robur is earlier than in Q. petraea by between two and ten days, the difference diminishing at higher altitudes. Despite some degree of timing difference, there is a considerable overlap in anthesis, and together with the substantial degree of self-incompatibility shown, this helps increase hybridisation frequency in mixed populations. Experimental hybridisation studies have been carried out by several workers in Britain, Germany and France, but the progeny typically appear small, weak and more prone to fungal disease, so that few, if any hybrid acorns produce healthy seedlings, let alone saplings (Rushton 1977; Steinhoff 1998; Bacilieri et al. 1996; Stace et al. 2015).

Selection of identification material

When attempting identification of these two oaks and their intermediate hybrids, it is vital to select suitable representative material from individuals for examination. Unfortunately the ideal material is well-illuminated central crown leaves from mature individuals, which is by definition, generally unobtainable from well-grown trees of the size to which these trees grow. When collecting samples, it is also important to realise that oak trees produce two batches of leaves per season. The later, late-July and August opening 'Lammas leaves', are more variable and they differ in shape from the first leaves of the season that expand in May and early June. It is the early season leaves that are the 'correct' or better ones to choose for identification. It is also important to recognise that the leaves on seedlings and on young trees are completely impossible to identify accurately (Jones 1959).

Based on his own work and that of several others, Brian Rushton has assembled a very useful table of the significant identification leaf characters used to separate the two oaks and their hybrid, together with helpful illustrations published in Rich & Jermy (1998, pp. 74-6). Potter (1994) has also formulated a reasonably straightforward identification procedure for non-specialists using a leaf index based on several simple measurements to be carried out on ten leaves from three segments of the crown.

Studies of Irish oak variation

A study of Irish oaks by McEvoy (1944) concluded that Q. robur is native on the island, noting that it is the principal oak of the central limestone area of the island in the rather limited areas where the rock has not become overlain by peat bog. McEvoy also found that the only extensive remnants of semi-native oak woodland in Ireland are all peripheral to this central area of the country, growing mainly on soils derived from acidic, siliceous rocks. He also discovered that the trees in these semi-native woods are of undoubted Q. petraea affinity. This agrees with the work of Jones (1959) on the occurrence of oaks in Britain, where he found that in general, Q. robur predominates over Q. petraea on more base-rich, fertile, low-lying soils, and it is more tolerant of waterlogging than the latter. On the other hand, Q. petraea is the principal species on both more acidic sandy soils, and on more upland terrain.

A more recent, detailed study of Irish oaks, which includes trees from the Crom estate in Fermanagh, is that of Kelleher et al. (2003). These workers looked at the genetic diversity of 26 Irish populations using chloroplast DNA for an analysis of the 'genetic fingerprints' of the individual tree. Chloroplasts are inherited exclusively through the maternal line in oak acorns, and thus carry genetic markers identical to those of the mother tree. While European populations of Q. petraea and Q. robur have a total of 25 distinct chloroplast types (ie haplotypes) present in them, only two haplotypes (haplotypes 10 and 12), were found in the Irish oaks sampled. Each of the two haplotype occurs in both of our oak species, with haplotype 12 present in 81% of Q. petraea individuals and in 62% of the Q. robur specimens examined. Using this type of data, indices of genetic diversity can be calculated within and between oak populations, and in this respect the Irish oaks have proved less diverse than mainland European populations. Nevertheless, the Irish oaks are in agreement with the overall pattern of haplotype diversity and distribution that is shown over the larger continental geographical range of the trees (Kelleher et al. 2003).

Studies of this type confirm that Q. robur and Q. petraea are closely related at a molecular level, and that they hybridize both experimentally and in nature, yet they do so without swamping the differences between them (except perhaps in parts of Scotland) (Cousens 1963, 1965). The nearest we can come to a current explanation of this phenomenon is to point to a degree of ecological separation between the species. Of the two oaks in question, throughout Europe Q. robur is the species of lowland, heavy soils, tolerant of waterlogging. In this larger geographical range, the difference in terms of substrate base status appears less consistent than it does in Britain and Ireland, but Q. robur is generally associated with the more fertile soils (D. Kelly, pers. comm., June 2003).

Fossil history

Fossil pollen studies cannot tell us which oak species (ie Q. petraea or Q. robur, or both), was or were initially present in the British Isles (Godwin 1975). However, after the last cold glacial period, oak first appeared in the SW tip of England around 9500 BP. Fossil material is first recorded in SE Ireland very shortly afterwards, around 9400 BP (Birks 1989). The two oak species and their hybrid appear to have migrated up along the western coast of France from a glacial refuge (refugium) in N Spain (Cantabria) or around the Bay of Biscay.

This was one of three such southern refugia shelters these Quercus species retreated to during glacial conditions, the others being the Balkan peninsula to the Black Sea shore, and to a much lesser extent, the Italian peninsula south of the Alps (Palmer & Birks 1983, p. 354). Q. petraea has close taxonomic affinities in both the Iberia and Balkans, whereas Q. robur has only one close relative in the Balkans, so the location of the refugia of the two oak species most common in Britain and Ireland is still a matter of discussion and macrofossil research.

Chloroplast DNA research has also shed light on the distribution and migration routes of oak species in the early post-glacial. Kelleher et al. (2003) found two main chlorophyll DNA haplotypes in Ireland (haplotypes 10 and 12). These correspond to the oak haplotypes that migrated northwards from Spain after the end of the last ice age. Haplotype 12 was found in oaks in the north and on the periphery of Ireland, while haplotype 10 was more central and southern in its occurrence. Five of the populations studied, including oaks on the Crom estate in Fermanagh, contained mixtures of these two haplotypes. Only one population in Glencar, Co Kerry contained a single tree with a non-native haplotype (number 7). The pattern of chloroplast DNA genetic markers found in Ireland was found to be consistent with that expected from a natural distribution of oak, and it was concluded that in the main Irish oaks are derived from indigenous (native) material, rather than from foreign, planted, introduced stock (Kelleher et al. 2003).

While the all Ireland study of Kelleher et al. (2003) only sampled a few Fermanagh trees at the Crom estate, in them it found both of the characteristic Irish genetic marker haplotypes 10 and 12. This made the Crom tree sample one of a minority of just five Irish oak populations out of the total 26 populations sampled which contained more than one haplotype marker in its chloroplast DNA. This fact indicates a higher genetic diversity in these five populations compared to the other Irish populations, yet the general levels of genetic diversity indicated by this study overall are low in comparison with Britain, where oaks have three main haplotypes and three subsidiary ones (Cottrell et al. 2002), and in France, where the oak populations of these two species contain a total of twelve haplotypes (Dumolin-Lapègue et al. 1999).

The fossil record shows oaks initially spreading into de-glaciated ground that was not bare, but already occupied by a mixture of birch, hazel, willow and juniper. Even though the ground was occupied, oaks invaded at an extremely rapid rate, advancing 350 to 500 m per year until around 8000 BP, by which time the species had reached S Scotland and N Ireland (Birks 1989). The rate of further spread then slowed dramatically to around 50 m per year, possibly because of the cooler summer temperatures typical of these more northerly regions.

Climatic limits to oak growth

To this day, climate remains a factor limiting oak growth in Britain and Ireland. For instance, a study of tree-ring samples from oaks at 13 sites across both islands showed marked similarities in their response to climate (Pilcher & Gray 1982). The study proved that high rainfall, particularly in the growing season, and high temperature in early summer, favour oak growth. High temperatures in the previous winter, however, are detrimental to growth in the following season, a finding which Pilcher & Gray suggested could well be due to depletion of the tree's starch reserves.

Acorn dispersal

Since the acorns of our two oak species represent very large heavyweight seed, the rapid rate of tree advance measured in fossil studies suggests that the species must have had considerable assistance with their dispersal. Most probably this involved a feeding relationship amounting to a mutualistic partnership with birds. Bird species such as jays, rooks and wood pigeons are considered the most likely oak seed vectors (Birks 1989).

Toxins

Oak trees have tannins in every tissue of the plant and there are relatively high levels of hydrolysable tannins in young leaves and in green acorns. This latter form of tannin breaks down in the gut of herbivores to form gallic acid and pyrogalllol. Pyrogallol oxidises blood haemoglobin and it is said to also attack the liver and the kidneys of animals. As toxicity levels are high in young tissues, animal poisoning is seasonal, occurring mainly in the spring (Cooper & Johnson 1998).

Oak mast years

It is not clear to the current author whether or not the calculations of oak dispersal rates quoted above from fossil studies allow for the occurrence of oak mast years which certainly occur nowadays. One would imagine that seed predation very probably did occur in the early post-glacial period as the large seed represents an important source of food for numerous woodland animals. In the case of beech and other trees including oak species, masting behaviour is very important in setting limits to seed predation, and thus permits and enables tree regeneration (Matthews 1955) (see the Fagus sylvatica species account).

Individual oak tree life-span

One of the myths associated with our oaks is their supposed longevity. The often met notion of the tree "being 300 years in the growing, 300 years in the being, and 300 years in the dying"' (Thomas 2000, p. 266), stretches credibility very thin. In the real world the lifespan of the typical oak could generally be truncated into more like 200 to 300 years in total, and the evidence suggests, or rather proves, that only the very exceptional individual survives longer, unless we widen our definition to include the epicormic growth on pollarded trees and coppice stools. The biggest and oldest oaks are all hollow pollards, and the oldest apparent 'maiden tree' (ie not pollarded, - and we cannot even be certain of that status in this particular case), is the Q. robur 'Majesty' at Fredville Park, Kent, which is considered to be around 450 years old at most (Mitchell 1996, p. 312 & 313; Pakenham 1996, p. 18 & 19).

Pilcher (1979) examined the ages of living oaks in nature reserve woodland slopes above Rostrevor, Co Down (VC 38). This woodland stands on steep rocky slopes and it contains trees growing under stress that are similar in size and appearance to many found today in Fermanagh. Using tree-ring counts and allied dendrochronological techniques, Pilcher found that the oldest trees he sampled began life in the early 1740s (ie maximum age 260 years at this sampling date). Historical evidence suggests that the price of oak timber in Britain had risen to such an extent in the 1730s that felling became economically driven and was happening extensively throughout Ireland (McCracken 1971). Thus Pilcher felt that the Rostrevor trees he studied probably represented regeneration that occurred after this particular spell of economic timber extraction. His results did not suggest that fresh planting occurred, but rather that natural regeneration took place.

Tree-ring studies of modern oaks in Ireland associated with carbon dating and dendrochronology, indicate that the majority of trees of both native oak species and their hybrid growing in semi-native habitats are only 150 to 250 years old, and that they are very heavily constrained by human activity (Baillie & Brown 1995). The oldest oaks in N Ireland are (or were) at Shane's Castle, Antrim, where two large trees felled in the 1980s, proved to be 307 and 340 years old. The well-grown oaks at Inisherk on the Crom estate in SE Fermanagh are almost as large as those surviving at Antrim, but they are quite a lot younger. Tree ring boring samples of the trunks of the largest Crom oaks produced a planting date of around 1720, making them nearly 300 years old. Most of the older Crom trees are 19th century (Browne & Hartwell 2000).

As a result of their study, Baillie & Brown (1995) concluded, "Overall, with the exception of a few comparatively young oaks sampled in hedgerows, it was apparent that the nineteenth-century Irish landscape outside enclosed estates, must have been almost devoid of oak trees." (my italics). Furthermore, these authors also suggested that, "despite the wishful thinking of many people, there does not appear to be any evidence of existing relict ancient oak forest anywhere in Ireland, including even those small patches on lake islands and inaccessible slopes." (again, my italics). Other woodland history experts disagree with both these statements, especially with the latter (eg Rackham (1995), and D. Kelly, pers. comm., June 2003).

Irish Ordnance Survey maps of the 1830s-40s show where woodland survived through the period of greatest pressure on the land. However, for several reasons the current author (Ralph Forbes) will go further and state that I reckon almost all woodland oaks in Fermanagh, as opposed to parkland trees, have a maximum age of around 200 years. When growing under stressed conditions, oaks do not develop massive boles. Observation shows them beginning to rot and break up while still of modest dimensions when compared with trees in 18th or 19th century parkland, or when compared with estate woodland specimens in Glenarm, Co Antrim (VC H39) or Killarney, Co Kerry (VC H2). Trees in estate parkland generally benefit from open and almost unrestricted growing conditions, so that uncut 'maiden' trees grow much faster and very much larger than upland trees growing on rocky slopes. Alternatively, estate trees may have been coppiced or pollarded at some stage and their trunk growth has thus been invigorated. Again, observation shows that most very old trees of any species, apart possibly from Yew (Taxus baccata), have been lopped in this way at least once during their lifetime (Mitchell 1996; Pakenham 1996).

Undoubtedly part of the difficulty in understanding oak biology and ecology is the extent to which the tree is a relatively long-lived plant, and secondly - and probably the crux of the matter - ever since the arrival of Neolithic farmers in these islands, it has seldom been allowed to live out its biological life span of somewhere between 200 and 400 years (or maybe somewhat more in exceptional cases), depending upon growing conditions, in anything even remotely approaching a natural manner (Shaw 1974; Minihan & Rushton 1984; Ellenberg 1988, Table 9; Ingrouille 1995). Throughout history oak timber has been far too valuable to allow the tree to die and fall naturally. Alternatively, the land on which the tree is growing becomes too valuable, or it becomes required for other purposes. Either way, management steps in and fells the tree or the entire woodland. Even if the individual tree survives into old age, during its long life some form of major disturbance of its growth almost inevitably occurs. Often the disruption is perfectly natural, but increasingly it is due to human activities. In any event, disturbance curtails the natural performance of the tree, resulting in distortion of its growth or its premature destruction.

The effect of past human pressure on woodland

It is easy for us today to forget that in the late 18th and early 19th century, Ireland's human population peaked at a level somewhere between 8.0 and 8.5 million. In comparison, the 2016 total population of Ireland (north and south combined) was 6.6 million. The great majority of Irish people nowadays live in urban conurbations, whereas in the 18th and 19th centuries, rural populations were very much larger and widespread. The Irish population doubled in the 60 years between 1780 and 1840, creating an enormous demand for all kinds of timber. Demand was particularly high for hard, durable construction timber, and throughout that period in rural communities shortages must have reached levels of desperation that are inconceivable today (McCracken 1971; Mitchell 1986). Before the development of the railways in the mid-19th century (the Londonderry and Enniskillen Railway reached Enniskillen in 1854), coal was very expensive and industrial and domestic fuel was almost entirely confined to wood and turf (dried spade-cut sods of moss peat). As a result of the demand for wood, even the most natural-looking and extensively visited and studied oak or mixed deciduous woodlands in Ireland today, are in reality totally secondary and planted. In some cases, as for instance in Killarney, secondary oak woodland stands on ground that previously supported clear-felled woodland containing oaks (Kelly 1981; Rackham 1995; Pilcher & Hall 2001).

The best Fermanagh woodlands containing substantial oak populations are the nature reserves at Correl Glen, Cladagh River Glen and Hanging Rock, plus the numerous wooded islands and shoreline woods of Lough Erne and Lough Melvin. Most of these woods are associated with past or present landed demesnes that belong, or used to belong, to wealthy families. Even in the case of these secondary semi-natural mixed oakwoods, at present we cannot be certain regarding the native provenance of the seed sources used in their plantation. However, modern allozyme coding genetic techniques may eventually allow us to discover which tree samples represent native local Irish oak genome regeneration (eg Kelleher et al. 2003), and which are derived from seed imported from Britain, or less likely, from continental suppliers (Gordon & Fraser 1982; Nelson & Walsh 1993, p. 117).

General oak ecology and biology

An enormous amount has been written about the ecology and biology of the two oaks native in Britain and Ireland, Q. petraea and Q. robur, and for instance, a Web of Science Database search on 21 Mar 2003 for 'Quercus petraea' turned up no fewer than 343 scientific papers dating from 1993 onwards! On account of the huge volume of published research and the fact that there is reasonably easy access to the general information in summary form (eg in the Biological Flora account of Jones (1959); papers in the Botanical Society of Britih Isles conference volume edited by Morris and Perring (1974) 'The British Oak, its History and Natural History.'; Rackham (1980) 'Ancient Woodland, its history, vegetation and uses in England., Chapter 17.'; plus concise accounts in Grime et al. (1988) and Milner (1992), and the excellent summary and review of forest ecology by Ingrouille (1995) in his book 'Historical Ecology of the British Flora.', pp. 170-203), I will not rehearse here the basic details of oak ecology. Instead I wish to present a review of the recent (ie, post-1980) changes in perceptions regarding the biology, ecology and conservation situation of these two oak species, which will be for the most part applicable both to them and to the complex of their hybrids, ie, what might broadly be referred to as the British Isles oak species aggregate.

Lack of oak regeneration and oak population decline

The lack of regeneration in oaks in Britain and Ireland was quite widely appreciated from around 1910 onwards (Watt 1919), and the near absence of young saplings and trees under 50 years of age is now understood to be more general in its occurrence. This phenomenon is now recognised as a widespread European 'oak decline problem', akin to, but not identical with that at present being recorded in North America (Thomas et al. 2002).

In C Europe at least, present day oak decline involves two syndromes characterised by either (i) cyclic episodes of rapid mortality in local but widespread centres, followed by decreasing and slower mortality. Such episodes may last for up to 10 years and sometimes are preceded by a predisposing phase of reduced growth; or (ii), general oak woodland decline which is characterised by increasing crown thinning of the trees in entire stands over large areas, but which involves only low levels of mortality (Thomas et al. 2002).

In Britain and Ireland we may be witnessing the latter form of oak decline, although it is not all that apparent in some areas, including N Ireland. The sudden oak death scenario we have certainly not encountered here, although there are worries that a new virulent lethal form of the fungus Phytopthera introduced with ornamental plants such as Rhododendron and Viburnum is actively spreading in Britain and Ireland. While this fungal pathogen is capable of attacking Quercus species, it is not yet known to have done so.

Late onset of seed production

A factor adding to the reproductive limitation of the oak tree is that younger trees do not become sexually mature until they are at least 40 years old, and acorn production is variable from year to year, following the mast pattern which is also met in the more shade tolerant beech tree (Matthews 1955; Shaw 1974). Having said this, acorn production itself is not likely to be a significant limiting factor for oaks, since, except in the very worst years, far more of them are produced than is required to maintain existing oak populations in woodland. The perils undoubtedly lie with the fate of the acorn, the seedling and the young sapling, the most critically stage being the first year after germination, when seedling establishment is taking place (Rackham 1980).

Ecological advantage of large seeds

Having said this, the large seed, while susceptible to damage or destruction early on from drying out, frost, or hungry herbivores unless covered, hidden and protected either by the soil litter horizon or by the surrounding vegetation, does provide the biological and ecological advantage of a relatively large "starting capital" for the tree embryo, making the seedling independent of soil nutrients for up to two years after germination (Ovington & MacRae 1960; Jarvis 1963). The large seed size also permits the pattern of seedling growth which provided the shade is not too dense places the taproot before the shoot and allows the former to thicken and accumulate storage materials. In turn this often enables the young seedling and the sapling to survive dieback and to resprout after meeting adverse growing conditions, or even after the repeated attentions of browsing herbivores (Jones 1974). In deeper shade on the other hand, the shoot etiolates and the root becomes too poorly developed to mechanically support it, so that the tall, spindly seedling generally ends up lodging, i.e., falling over sideways and dying, being pushed aside by competing stems of such plants as e.g., bracken, or the grass, Deschampsia flexuosa. If this latter fate doesn't materialise, root competition from older trees and from other species becomes of greater importance in limiting the growth of the young oak seedling than the low light intensities, but again, poor root development of the seedling does makes it very susceptible to any occurrence of drought conditions (Jarvis 1964).

It has also been shown in cultivation experiments that the larger the acorn, the more rapidly it grows and therefore the more likely it will escape through the very vulnerable small seedling development stage. "In view of the large number of ways in which small seedlings may be killed or stultified, the ultimate advantage to the larger seedling may be out of all proportion to its initial size advantage. For example, there is only a small margin between successful establishment and smothering by bracken or Deschampsia." (Jarvis 1963).

Saplings and regeneration

Saplings and young trees are occasionally present in more open woodland areas, for instance along rides, roads and wood margins. However, these saplings are usually present in numbers far too small to achieve effective woodland regeneration (Rackham 1980, p. 296). This is especially the case when there is pressure on seed, seedlings and saplings from grazing by both farm-stock and wild or feral animals. Small scale herbivores such as wood-mice, voles and specialist feeding invertebrates are frequently involved, together with the more obvious seed predation and herbivory carried on by larger vertebrates, such as squirrels, rabbits and deer (Crawley & Long 1995; Kelly 2002).

The first year seedling growing in woodland shade needs all its leaf area intact in order to be able to manufacture and store sufficient reserves for its overwinter survival, so that any damage is potentially lethal during its early months of existence. Young leaves are particularly vulnerable since they are succulent and initially they have low levels of the bitter tannins which make older, harder, more leathery leaves highly unpalatable or indeed toxic to grazing animals of all descriptions (Cooper & Johnson 1998).

A 25-year experimental study monitoring Q. petraea sapling survival in Killarney, SW Ireland by Kelly (2002) concluded that successful regeneration only occurred there in unshaded or lightly-shaded sites where grazing levels were kept low. Even so, oak mortality was high and the median proportion of saplings that survived in unshaded plots protected from grazing was extremely low, measured at just 0.2 after 25 years study. The study found Oak saplings were soon over-topped by faster growing species, mainly Betula. At the end of the study the surviving oaks had a mean height of 3.1 m (a range from 0.9 -7.0 m), and the birch which was suppressing them had developed a canopy at around 12 m (Kelly 2002).

It has been suggested that the decline in oak regeneration noted in the early years of the twentieth century might stem from the arrival of oak mildew, Microsphaera alphitioides, which was unknown in Europe prior to 1907. Rackham (1980) hypothesised that this probably made oak seedlings and saplings "succumb to a degree of shade which they would formerly have survived". Some subsequent studies however have failed to show this effect (e.g., Kelly 2002), and it is probably an over simplification to expect a single factor rather than a specturm of interacting causes to be responsible for the syndrome of regeneration failure.

The problems that the two Quercus species face in regenerating under oak canopy (or the near impossibility of this happening!), suggests that perhaps oak woodland was not the climatic vegetation 'Climax' in Britain and Ireland as suggested and believed by Tansley (1949), and was taken up by early palynologists interpreting their fossil pollen diagrams (Shaw 1974). Later fossil pollen studies suggested that a more mixed deciduous woodland assemblage, including elm, and, at least in England where they are indigenous, other more shade tolerant tree species, such as lime, beech and hornbeam, might co-exist with oak, or act as stages in some form of succession in which oak played just a part (Godwin 1975, p. 277-281, and Mitchell 1986, p. 34).

On the other hand, historical studies of British woodland and its uses by Rackham (1980) have led him to suggest that oak regenerated more freely in the historic past than it does now, and that perhaps some detrimental change in the biology of our native oaks took place around 150 to 200 years ago which reduced oak regeneration capacity. Rackham's researches found that medieval carpenters had an abundant supply of "small oaks", with a rapid turnover in just that age-class, ie, trees of less than fifty years growth. At present, this size of oak is very deficient in woodlands throughout these islands. While most of this small bore timber would have come from oak coppice, Rackham still points out that it wasn't until the eighteenth century that possible hints begin to appear in the written record, suggesting there might be a deficiency in the supply of oak timber from managed woods (Rackham 1980, p. 295).

A possible more pioneer colonising role for oaks

Studies on the European continent suggest that in the past oaks might have undertaken a more pioneering role, similar to that of the light-demanding birches today, colonising woodland gaps, newly available disturbed ground, and drying out areas of fens and bogs (Björkman 2001). Here the more rapid growth of accompanying birch might allow it to nurse oak saplings which would have colonising the ground at the same time as the birch with the assistance of avian seed predators. Oaks are very much slower to develop and gain height than birch, but if they survive (even in a stunted form), eventually they should outlive, outgrow and overtop the comparatively short-lived birch (Kelly 2002).

In the more cloudy oceanic and upland areas of Britain and Ireland, however, oaks are unable to regenerate under their own shade and they are constantly accompanied by a very rich variety of the phytophagous insect, lichen, fungal, bird and mammal communities they support. Oak trees under this sort of pressure eventually lose vigour and they would then very gradually be replaced in woodland by infiltration of more shade-tolerant, high canopy species, ie elms, beech, hornbeam and limes (Rackham 1980; Ingrouille 1995).

However the arrival of Neolithic man in Britain around 5000 BP and the onset of his woodland clearances and the prolonged exploitation, management and especially the favouring over other species of oak for timber usage, together with the more recent mismanagement and neglect of woodlands since other materials have replaced timber, has modified the ecological picture of woodland out of all recognition. The more recent changes in our use of timber and our appreciation of woodland have happened quickly, over just a couple of human generations, to the extent that we only now appear to be gaining some foggy notion at to how woody species and the woodland community naturally behave during an interglacial stage. No doubt we have a great deal more to discover before we can effectively manage upland and lowland mixed deciduous woods in changed circumstances purely for the conservation of the trees and the communities they support. We just hope and pray that it is not already too late to achieve this goal without major loss of the associated biological diversity.

British and Irish occurrence

The Q. petraea map in the New Atlas clearly demonstrates the accuracy of these assessments. It shows that Sessile Oak predominantly occupies peripheral and upland siliceous terrain in Ireland, while in Britain it has a pronounced western distribution, correlated with the more ancient, acidic rocks and with a wetter oceanic climate. The New Atlas map of Q. robur indicates that at the hectad level of discrimination this species is almost omnipresent on all soils in Britain, avoiding only the most shallow, driest limestone soils, deep acid peat and high mountains. Since considerable areas of Co Fermanagh have an underlying Carboniferous limestone rock structure, even though this is frequently buried beneath boulder clay or peat and the majority of the county is low-lying and with a preponderance of seasonally wet, clay soils, overall the existing data agree with the views of both McEvoy (1944) and Jones (1959) on the predominance of Q. robur under such conditions.

The New Atlas maps of the two common oak species clearly illustrate that the one area of Ireland where neither oak can survive is Co Mayo (H26 & H27) with its vast empty stretches of treeless bogland and quartzite mountain cones. In Britain, the same degree of absence is clearly apparent in the Scottish Highlands and their vast north-western blanket bog areas (Preston et al. 2002).

Oak and Man

The Oak tree was regarded as a 'Noble of the wood' in the 8th century Irish Laws of Neighbourhood, on account of the trees' size and quality of timber. In addition, the acorn fruit was a useful and important seasonal food for pigs – ie autumn pannage (Edlin 1963). For these reasons, oak was protected, and rather severe penalties would have been due from anyone found cutting branches, boughs or felling a tree (Nelson & Walsh 1993). Despite this, oak plays a relatively minor role in Irish traditions when compared with ash or yew (see Nelson & Walsh 1993).

In a lengthy article on 'The Sacred Trees of Ireland', however, Lucas (1963) points out the connection of sacred oak groves (in Irish, 'doire', from 'dair', meaning 'an oak tree'), with ecclesiastical sites. It is also the case that 'doire' (anglicized as 'Derry') and its diminutive 'doiríin' (anglicized as 'Derreen') are two of the commonest elements in Irish place-names, reflecting the fact that the forests of ancient Ireland were chiefly of oak. Furthermore, McCracken (1971, pp. 24-5, Map 2), has mapped all of the Irish townland place-names containing the word 'doire', and then used this together with the written historical record to estimate the extent and distribution of major pre-1600 AD oakwoods across the island.

Oak timber had very many uses in Ireland as elsewhere, including furniture, barrel staves, building and boat construction. In addition, less well grown pieces were converted into charcoal for iron smelting and for gunpowder, or they were used for fencing or as planks for wetland and bog walkways. McCracken (1971) has provided a very full historical account of Irish timber use from Tudor times to the present, while Neeson (1991) has likewise provided a detailed history of Irish Forestry.

Names

Grigson (1987) provides derivation of the English name 'oak' from the Anglo-Saxon/Old English 'ãc', a word which is cognate with numerous other Germanic languages and refers to the fruit of the tree, the acorn, which was regarded as one of the most useful products of the tree, the oak forest being widely used to fatten swine. Anglo-Saxon laws were in force to protect the trees for this very reason (Prior 1879). Britten & Hollland (1886) list 21 variations from 'aac' to 'atchorn' for the more widespread and familiar 'acorn', plus a further 24 English common names for the tree. The main ancient folklore beliefs, worship of, and superstitions associated with oak, are summarised in Grigson (1987), Milner (1992) and Vickery (1995), while there are 35 references to oak in various folk modes recounted in Friend (1883).

Threats

There is no immediate conservation threat to native oaks in Fermanagh, but the stock of trees is old and many individuals in woods and parkland are approaching senescence. An active programme of replacement is urgently required and the best option is to use local acorns. The 2009 arrival in Ireland of Phytophthora ramosum, the fungal pathogen responsible for so-called 'Sudden Oak Death' in N America, is a worrying development since it is reputed to affect many different woody species. The list of susceptible trees and shrubs includes Fraxinus excelsior (Ash), Betula spp. (Birches), Acer pseudoplatanus (Sycamore), Vaccinium myrtillus (Bilberry) and Viburnum spp. (Viburnum) as well as Larix spp. (Larch), Oaks (especially Q. ilex), Fagus sylvatica (Beech), Aesculus spp. (Horse-chestnut) and Castanea sativa (Sweet Chestnut). Rhododendron is also seriously affected and it can act as an evergreen carrier and spreader of the fungus. As a result of this, rampant Rhododendron is being more actively removed from some National Trust estates than ever before – proving that every cloud has a silver lining!