The red squirrel ( Sciurus vulgaris ) is an arboreal mammal with an extensive Palaearctic distribution. The species can be found within pine ( Pinus spp .) and spruce ( Picea spp .) dominated forests across Europe , as well as in mixed deciduous broadleaved stands. Adult densities typically range between 0.8 and 1.2 animals per hectare, whilst greater numbers (2 - 4 ha -1 ) have been reported from some pine woodlands. These levels contrast sharply with the relatively low densities (0.1 - 0.3 ha -1 ) typically documented for commercial conifer plantations that contain a significant proportion of Sitka spruce ( Picea sitchensis ). However, despite commercial plantations being a resource-limited habitat (Wauters et al ., 2000), reproductive rates within the range of habitats occupied by red squirrels are broadly similar. A breeding adult female is likely to produce either one or two litters annually, each containing an average of three young.

Tree seed forms the bulk of the red squirrel's annual diet, but animals also consume such ‘secondary foods' as buds, shoots, flowers and the cambial tissue of trees, as well as fungi and the occasional bird's egg. Fluctuation in the availability of tree seed is reflected in seasonal variation in dietary composition, and it also directly affects the size of local populations. For example, poor seed years typically result in low reproductive rates and a decline in red squirrel numbers (Gurnell, 1987).

Red squirrel in Britain

The red squirrel has historically shown marked population fluctuations in Britain. These were probably caused by habitat modification, annual variation in weather conditions and the abundance of tree seed, and through outbreaks of epidemic disease. Widespread control of the red squirrel was carried out during the period 1890 to 1910 because populations were so large that bark-stripping had become a major problem in many woodlands (Shorten, 1954). In sharp contrast, during the eighteenth century, the species had declined to such a low level that in many regions, particularly Scotland , animals were re-introduced in order to restock woodlands.

In the last seventy years, the distribution and abundance of the red squirrel has again declined sharply throughout much of England and Wales . Habitat loss and woodland fragmentation have inevitably played their part, however, ecological replacement by the introduced grey squirrel ( Sciurus carolinensis ) is now known to be the most important factor. The substantial fluctuations in red squirrel populations long before the arrival of grey squirrels explain why the expansion of greys was initially not identified as the main cause of local decline and extinction. Indeed, there are a few well documented, if rather unrepresentative, examples of mixed red/grey populations persisting over several decades. However, the failure of red squirrel populations to recover any of their lost geographical range indicates that the progressive pattern of local extinctions has not simply been part of a natural fluctuation. The result of intensive scientific research, particularly from the 1980s onwards, has enabled us to firmly point the finger of blame at the grey squirrel.

Grey days

The first authenticated record of a grey squirrel introduction is for 1876 within the grounds of Henbury Park , Cheshire . Records of introductions from America then continued until the late 1920s, whilst translocations within the UK are documented until 1937. The increasing numbers of grey squirrels soon led to conflict with the timber industry, because of the species' tendency to strip the bark from softwoods. The grey squirrel's status as a major forest pest was recognised by Parliament through the Grey Squirrel (Prohibition of Importation and Keeping) Order 1937 whereby it became an offence to release or transport grey squirrels without a licence.

The grey squirrel is an attractive and charismatic creature, and many people enjoy watching the acrobatic antics of these animals. Indeed, grey squirrels are often deliberately attracted into gardens with supplemental foods. Having evolved in the oak/hickory forests of the eastern United States , the behaviour of the grey squirrel differs markedly from that of the Eurasian red squirrel. Adult greys are typically twice the mass of red squirrels and the species spends more time active on the ground (86% cf. 33% for red squirrel) (Kenward and Tonkin, 1986). The fact that they forage predominantly upon the woodland floor may explain why grey squirrels can put on twice as much body fat (a 20% body weight increase) in the autumn as red squirrels. Such a dramatic increase in body weight would incur a greater penalty upon the foraging efficiency of red squirrels, because their activity is concentrated within the forest canopy. However, despite the larger size of the grey squirrel, there is no evidence that reproductive rates are intrinsically better than those of red squirrels (Kenward and Holm, 1993). Adult female grey squirrels generally breed only once or twice each year and litters are the same conservative size of three young.

Mechanism of ecological replacement

Rapid replacement of red by grey squirrels was observed in southern England in the 1940 –1960s. This landscape is dominated by scattered broadleaved woodlands that often contain a large proportion of oak ( Quercus spp .). Wauters et al . (1992) observed that red squirrels consumed relatively few acorns in the autumn and winter months despite the fruits' abundance. It was known that acorns contained significant amounts of tannin, but field and laboratory studies later revealed that grey squirrels were better at neutralising acorn polyphenols and hence are more efficient at digesting the protein within acorns than their congeners (Kenward and Holm, 1993). This finding probably reflects the fact that grey squirrels evolved in broadleaved forest, and it would also, to some extent, explain why their densities (2 - 8 ha -1 ) are considerably greater than those reported for the red squirrel in broadleaf habitats.

The competitive advantage of grey squirrels over reds is clearly high in broadleaved stands, and this may be exacerbated by their piracy of red squirrels' food stores. In one study, grey squirrels were found to have exploited up to 66% of food hoards cached by red squirrels (Wauters et al ., 2002), and this was associated with a decline in red squirrel reproductive rates. The grey squirrel doesn't quite have everything its own way though. Coniferous stands, especially commercial spruce plantations, appear to be relatively inhospitable habitat for them (Lurz et al. , 1995).

In areas of Britain where coniferous stands were widespread, it was noticed that red squirrel populations persisted for longer despite the presence of greys. Studies of grey squirrel populations within such habitats indicated that densities were often equivalent to those of red squirrels recorded in similar habitat elsewhere. Kenward et al. (1998) observed that grey squirrel populations in conifers had higher levels of mortality than red squirrels, and they suggested that grey squirrel densities were being maintained by immigration from adjacent broadleaved habitats. The proximity of large-seeded broadleaves to coniferous stands was clearly of significance, and their removal is now generally advocated as part of local red squirrel conservation measures (JNCC, 1996).

Despite the fact that coniferous stands appear to favour the red squirrel, these populations are still frequently displaced by grey squirrels. Recent research in coniferous plantations indicates that the presence of an established grey squirrel population can greatly reduce the recruitment rates of juvenile red squirrels (Wauters et al. , 2000) but appears to have little effect upon adult survival rates. Consistently poor recruitment rates would begin to explain why grey squirrels have replaced red squirrels even within extensive conifer plantations, e.g ., Cannock Chase in Staffordshire. However, if direct competition and interference with population dynamics were not enough, the grey squirrel has another trick up its sleeve; germ warfare.

There is growing evidence that the grey squirrel can act as a vector for the transmission of the para-pox virus into red squirrel populations (Sainsbury et al. , 2000). Grey squirrel populations typically show high sero-prevalence (c. 60%) for antibodies to the para-pox virus (a pustular dermatitis). Infected animals show no clinical symptoms, indicating that the virus is present as a latent or sub-clinical infection. Anti-bodies are not found within red squirrel populations because exposure to the virus results in the development of extensive skin lesions and mortality. Death from secondary infection of the lesions occurs within a few days of the symptoms, and outbreaks have played a significant role in the extinction of several local red squirrel populations. As yet the mechanism by which the virus is transferred from grey to red has not been identified, but the potential for parapox epidemics illustrates the risk involved in allowing mixed red/grey populations to persist.

Reds on Anglesey

Figure 1 The location of Welsh forests containing red squirrels. Wales is 20,768 sq km in area and contains 286,000 hectares of woodland (productive and all other types). [ Red ] Key populations, [ Blue ] sites where the species is most likely to become extinct in the next few years.

The red squirrel population in Wales is highly fragmented (Figure 1) and may contain fewer than 1000 adults. Despite being located within extensive commercial conifer plantations, all the mainland red squirrel populations are sympatric with grey squirrels. It is likely that several of the smaller remnant populations may disappear within the next few years, whilst the long-term future of all of the populations remains uncertain. In 1997, it had appeared that red squirrels would be lost from the only Welsh island that contained them, the 720 km 2 island of Anglesey , but, through a systematic programme of grey squirrel control, the population has been rescued from the brink of extinction.

Historically, red squirrels had been distributed throughout the island's woodlands (c. 2000 hectares in area), in particular the mixed deciduous stands on the southern coast. Needless to say, in the mid 1960s, the grey squirrel crossed the narrow Menai Straits by way of the road bridges. Red squirrel numbers soon began to go into decline. During the 1970s -90s, the pattern of local extinction reflected that typically observed elsewhere in Britain. In 1995, the red squirrel was restricted to two commercial conifer plantations, and by the winter of 1997/98, animals could only to be found within one of these sites, the 244 hectare woodland called Mynydd Llwydiarth (Figure 1). However, it was recognised that, providing grey squirrel control was systematic, the island of Anglesey offered a unique opportunity to reverse locally the national trend of red squirrel extinctions. Rates of grey squirrel immigration onto the island would be relatively small and therefore control operations would deal with a discrete and isolated population.

It may be surprising to learn that the results of previous field studies aimed at quantifying the effect of grey squirrel control upon red squirrel populations had all been equivocal, e.g., research in Thetford Forest . Nevertheless, control operations are still advocated as the major tactic within red squirrel conservation strategies (JNCC, 1996). The fact that previous grey squirrel control programmes had failed to prevent extinction of local red squirrel populations meant that many potential funding bodies were reluctant to invest financial resources in the control of greys on Anglesey .

Figure 2 Eradication of the grey squirrel from the Mynydd Llwydiarth plantation on Anglesey .

Locally, it was recognised that, because of its rarity, the red squirrel had considerable socio-economic value as a tourist attraction. As a result, a rural regeneration company, Menter Môn (Anglesey Enterprise), decided to invest funds in the project. Work began in 1998 and initially targeted the Mynydd Llwydiarth plantation and surrounding broadleaved stands. In total, 68 grey squirrels were removed from the plantation in the first year. This represented a considerable reduction in grey numbers as only six, two and a single animal were trapped within the forest in the three following years (Figure 2). With grey squirrels removed, the remnant red squirrel population responded quickly and positively. Adult numbers increased sharply as a result of favourable survival (65-74%), high reproductive rates (58-73%), and through the extremely high recruitment of young produced in 1999. At least 75% of these young were known to have been present in 2000; in other published studies, recruitment rates are typically 20-50%. However, as the adult population expanded within the plantation the proportion of young recruited fell to 40% (2000/01) and 16% (2001/02) and the population began to level out (Figure 3). A Habitat-Suitability Model suggested that, under average conditions, the site could support just over 100 adults. The population approached this saturation level (Figure 3) during 2000 and this coincided with red squirrels being sighted within broadleaved stands adjacent to the forest during that autumn.

Figure 3 Estimates of red squirrel population density within the Mynydd Llwydiarth plantation.

During 2001, the first litter was discovered within oak/hazel woodland where grey squirrels had previously been controlled. It was almost certain that these were the only young weaned within broadleaved woodland within the whole of Wales . Red squirrels were then regularly sighted foraging within the hazel stands during the autumn and a total of ten were caught within the five-hectare woodland. To their delight, several local householders had red squirrels visiting their garden bird-tables.

The presence of grey squirrels within the Mynydd Llwydiarth plantation had clearly limited the abundance and distribution of the red squirrel. The pattern of adult and yearling red squirrel captures showed that animals had either avoided stands containing large numbers of greys, or that they had been prevented from using them (Figure 4). Data on stand tree species composition revealed that red squirrels had originally been associated with stands containing a significant proportion of Sitka spruce and lodgepole pine ( Pinus contorta ). In contrast, the largest concentration of grey squirrels was centred on stands dominated by Corsican ( Pinus nigra ) or Scots pine. Both Sitka spruce and lodgepole pine produce relatively smaller seeds in comparison with the two other pine species. Several studies have shown that the highest densities of either red or grey squirrels in conifers are often found within Scots or Corsican pine woodland (Lurz et al ., 1995). Following the removal of the grey squirrels within Mynydd Llwydiarth, adult red squirrels were more frequently trapped within the pine dominated stands (Figure 4). Data collected in the spring of 2002 revealed that the red squirrel distribution largely mirrored that originally observed for the grey squirrel in 1998.

Figure 4. The spatial distribution of adult & yearling red and grey squirrels within areas (A-H) of the Mynydd Llwydiarth plantation The presence of grey squirrels in 1998 greatly reduced the capture of red squirrels and hence data are presented for subsequent years only. Where a red squirrel was caught within two adjacent areas of the plantation a value of 0.5 was applied to each area in the calculation of a density estimate. (Data 1998 – August 2002).

The habitat utilisation by red and grey squirrels in the mixed population on Anglesey supports the findings reported in a detailed red/grey study conducted within Hamsterley forest in the north of England . Wauters et al . (2000) observed that pure Scots pine stands offered a relatively high quality seed source and that they contained the highest grey squirrel densities within the whole forest. Radio-tracked red squirrels avoided these stands and were associated with areas of dense Sitka spruce. The researchers felt that it was possible that inter-specific competition had caused red squirrels to avoid the pine areas.

Mynydd Llwydiarth is now the only site in Wales where the red squirrel population is not sympatric with grey squirrels. Continued geographical expansion of red squirrels is, however, limited in the mid to long-term because of the highly fragmented nature of the surrounding broadleaved woodlands. This has made the establishment of red squirrels in the large (c.700 hectare) Newborough forest a high priority. Red squirrels were present within this coastal Corsican pine plantation until the mid 1990s but its isolation precludes natural re-colonisation and hence a re-introduction programme is planned. Previous UK reintroductions have been unsuccessful, primarily because of a failure to exclude grey squirrels, but also as a result of disease and heavy predation (Kenward and Hodder, 1998). It is anticipated that grey squirrels will first be eradicated from Newborough, but the re-introduction will still obviously be a challenge.

Grey squirrel control on Anglesey removed c.5000 individuals during the first five years. Initially, trapping focused upon Mynydd Llwydiarth and a buffer zone of broadleaved woodland. However, by 2002, approximately 95% of woodland sites were trapped and the labour available had increased from a single trapper to a team of four. Those broadleaved stands repeatedly trapped have shown a sharp decline in the number of animals caught annually and it is hoped that grey squirrels will eventually be completely removed from Anglesey . Studies have shown that 65-75% of the island's grey squirrels are sero-positive for para-pox antibodies and, therefore, even a small number of grey squirrels within a plantation represent an unacceptable disease risk to red squirrels.

The current small size of the Anglesey red squirrel population means that it remains vulnerable to stochastic events and adverse environmental conditions. The short-term future is bright though, as there is still modest scope for continued natural expansion. With a red squirrel release programme planned for 2003, it is anticipated that, within the next five years, the island could contain the largest single population in Wales . A large red squirrel population has considerable potential as an attraction for eco-tourists, and hence could offer considerable socio-economic benefits to the island that is within one of the most economically depressed areas of the UK . The tourist potential is illustrated by the National Trust red squirrel reserve at Formby in England which receives several hundred thousand visitors each year.

Maintaining the red squirrel on mainland Wales will be more challenging. Habitat manipulation will be a valuable tool in the long-term strategy, but there also needs to be a policy change towards more aggressive grey squirrel management. In Wales, conifer plantation makes up 64% of the 286000 hectares of woodland cover and, yet, most of the mainland red squirrel populations have been lost. Those that remain are sympatric with grey squirrels. This demonstrates that, although conifer stands may offer a refuge, it is often only temporary in nature, and, consequently, none of the remaining Welsh red squirrel populations can be considered invulnerable to replacement by the grey squirrel. Whether any type of coniferous habitat in the UK can indeed offer a long-term refuge for the red squirrel is currently being investigated, with the focus upon Kielder forest in Northumberland, England.

To change the fortunes of Welsh mainland red squirrel populations will require significant financial resources. Grey squirrel control is routinely carried out to reduced damage to broadleaved timber crops, and is also advocated as a tactic in the conservation of red squirrels (JNCC, 1996). However, it is unfortunate that in red squirrel conservation terms, control is often seen as a pointless exercise, the view being, clear a large area and it will simply act as a vacuum that will attract greys from surrounding habitats. The large scale (2000 - 15000 hectares) of the commercial plantations that still contain red squirrels is also seen as prohibitive, because of the amount of labour required to trap the entire area systematically. So what is the solution?

Everyone recognises that the long-term conservation of red squirrels in Wales is at best an ambitious target, but the alternative is to accept the loss of this popular and endearing creature. Consequently, although mainland grey squirrel control programmes are unlikely to be sustainable, the cost of intervention is increasingly being set against the socio-economic value of red squirrels. Scientists trying to conserve equally threatened, but less charismatic species will cringe, but the high value placed upon the red squirrel by society has as much to do with its fluffy and cute characteristics (Figure 5) as it has with rarity. If ‘cute' gains public support, then this may ultimately generate the funds necessary to at least maintain the key populations in Wales (Figure 1).

It is of course true that a large section of the general public also consider grey squirrels to be ‘cute' despite the fact that the animal is the villain of the piece. In regions where it is the only squirrel species present, it is very unlikely that householders would accept widespread control. However, the work on Anglesey has illustrated that people are willing to accept systematic grey squirrel removal if it will help save a red squirrel population nearby.

Maybe, just maybe, Welsh red squirrels will still persist in decades to come.