Other Names: lynx du Canada (French); Kanadaluchs (German); lince del Canada (Spanish)

Description and Behavior Ecology Biology Habitat and Distribution Population Status Protection Status Principal Threats References

Description and Behavior: The Canada lynx has a flared facial ruff, black ear tufts, and long hind legs which lend a slightly stooped posture. The pelage is reddish-brown to grey; the hairs are tipped with white which gives the fur a frosted appearance. There is a rare pallid color phase which suggests partial albinism, known as the blue lynx in the fur trade (Quinn and Parker 1987). The Canada lynx's large spreading feet act like snowshoes, and are twice as effective at supporting its weight on snow as those of the bobcat (Parker et al. 1983). The lynxes show remarkable similarity of appearance compared to other related groups of cats, and the Canada lynx is often treated as conspecific with the Eurasian lynx (Kurtén and Rausch 1959, Tumlison 1987). However, the Canada lynx is only half the size of the Eurasian lynx: average adult weight of Canada lynx males is 10.7 kg (n=201) and females 8.9 kg (n=183) (U. Breitenmoser and C. Breitenmoser-Würsten in prep.). While the Canada lynx is probably a descendant of a Eurasian lynx ancestor which migrated into North America during one of the last two major glacial periods (Werdelin 1981, 1983b), the Breitenmosers (in prep.) argue convincingly that the two should be considered separate species, as they now show marked adaptive differences for prey capture. Whereas the larger Eurasian lynx preys mainly on ungulates, the Canada lynx relies almost exclusively on snowshoe hares, and is uniquely adapted, both behaviorally and physiologically, to exploit a cyclic prey base. Ecology: Among felid predator-prey relationships, there are none as closely tied as that between the hare and the Canada lynx (Van Zyll de Jong 1966, Nellis et al. 1972, Brand and Keith 1979, Parker et al. 1983, Ward and Krebs 1985). The lynx-hare cycle was first discovered from harvest records of the Hudson's Bay Company dating back to the 1800s (Elton and Nicholson 1942: Figure 6). Numbers of snowshoe hares peak approximately every ten years, and lynx numbers follow the same pattern with a short lag, typically one to two years (Keith 1963, Bulmer 1974). While the populations of many prey and predator species are cyclic and roughly synchronous in the northern latitudes, the snowshoe hare and lynx correlation is particularly close (Keith 1963, Mallory 1987: Fig. 7). The amplitude of the lynx population cycle is greater than that of any other predator (Bulmer 1974), and lynx density during cyclic highs and lows can differ by up to 15-fold (Breitenmoser et al. 1993b). As hares decline, fewer lynx breed, producing smaller litters with few, if any, surviving kits. As hares increase, so do lynx reproduction and recruitment rates (Nellis et al. 1972, Brand and Keith 1979, Parker et al. 1983, O'Connor 1984, Slough and Ward 1990, Breitenmoser et al. 1993b, Mowat 1993). In captivity, female lynx do not show such an early onset of sexual maturity or such high litter sizes as during hare peaks in the wild (Breitenmoser et al. 1993b). While lynx will switch prey during periods when hares are scarce (Brand et al. 1976), turning to small rodents, ground birds and, exceptionally, ungulates such as white-tailed deer, caribou, and Dall sheep (Saunders 1963, Bergerud 1983, Stephenson et al. 1991), lynx populations only reach high densities when supported by snowshoe hares (Brand and Keith 1979, Mech 1980, Ward and Krebs 1985). There are several competing hypotheses to explain the hare cycle. The most widely accepted explanation is that winter food shortage (Keith 1974) depresses hare reproduction (Carey and Keith 1979) at the population peak and starts the cyclic downturn, and hare numbers are subsequently further reduced due to predation (Keith et al. 1984, Boutin et al. 1986). Gilpin (1973) and Schaffer (1984) modelled harvest data mathematically, and concluded that the cycle is more complex than a simple predator-prey interaction, involving at least a third additional factor. Another suggested influence involves changes in the nutritional quality of vegetation in response to hare browsing (Bryant 1981, Sinclair and Smith 1984, Bryant et al. 1985). Nevertheless, in some areas, hares have declined even when food resources appear sufficient (Keith et al. 1984, Krebs et al. 1986). Preliminary results achieved from long- term field experiments (Krebs et al. 1992) now favor the hypothesis that predation alone, by a variety of specialist and generalist carnivores, is the driving force behind the cycle, as has been suggested for microtine rodents (Hanski et al. 1991). Biology: Birth season (W): May-Jun, exceptionally Jul (Saunders 1961, Nava 1970, Nellis et al. 1972, Mowat 1993). Estrus: no published information on duration or length of cycle (G. Mowat et al. in litt. 1993). It is possible that ovulation in the Canada lynx may be at least partly spontaneous, although this is controversial (Van Zyll de Jong 1963, Quinn and Parker 1987, G. Mowat et al. in litt. 1993). Lynx may be induced ovulators when prey density is low and there is less chance of meeting a mate, and spontaneous ovulators when prey density is high, improving prospects for breeding and raising young (Kitchener 1991). Gestation (W): 63 to 70 days (Saunders 1961). Litter size (W): higher (average 3.8-5.3) when prey is abundant, and reduced (2.3-3.5) when prey is scarce (Brand and Keith 1979, Slough and Ward 1990, Mowat 1993); range 1-8 (Tumlison 1987, Breitenmoser et al. 1993b). Yearling lynxes give birth to smaller litters (0-4.2: Mowat 1993) Age at independence (W): generally around 10 months: kits typically leave their mother's range in March-April (Saunders 1961, Bailey et al. 1986, Slough and Ward 1990, Mowat 1993). Age at first reproduction (W): 10 months (first winter) when prey is abundant, more generally 22-23 months (second winter) (females); second year (males) (Saunders 1961, Van Zyll de Jong 1963, Stewart 1973, Nava 1970, Brand and Keith 1979, O'Connor 1984). Reproductive rates (W): up to 100% during hare peaks, and as low as zero during cyclic lows (Mowat 1993). Interbirth interval (W): generally one year, rarely two (Tumlison 1987). Recruitment rates (W): 60-80% when hares are abundant and increasing, and approaching zero during lows (Brand et al. 1976, Brand and Keith 1979, Parker et al. 1983, Mowat 1993, Poole 1994). Koehler (1990a) found low recruitment rates of around 12% from 1980-1987 in mature forest in north-central Washington (sub-optimal habitat at the southern edge of lynx range). Mortality rates (W): adult mortality rates average 55% for exploited populations in Canada (R. Ward in prep. cited in Slough and Ward 1990). Rates vary dramatically with the hare cycle. Poole (1994) estimated survival rates in an unharvested population to be 90% before and during the decline in hare densities; 25% during the first year of low hare densities; and 37% during the second year of the low. All radio-collared lynx resident prior to or during the hare decline dispersed and/or died by the end of the first winter of low hare densities. Longevity (W): up to 15 years (Nava 1970; K. Poole, B. Slough unpubl. data). Habitat and Distribution: Lynx are distributed throughout the broad boreal forest belt of North America (Banfield 1974) and south into the American Rocky Mountains (Koehler 1990b), with a total range of some 7.7 million km2 (Parker and Quinn 1987: Figure 8). The historic range is largely intact, although it has shrunk in the south due to human settlement and forest clearance (Banfield 1974, Quinn and Parker 1987). Lynx will inhabit farming country, but only if it is interrupted by sufficient areas of woodland (Todd 1985). Bobcats appear to be expanding northwards, and have displaced lynx in some areas (Parker and Smith 1983, Rolley 1987). Lynx will travel long distances during both phases of the hare cycle seeking out patches of hare abundance (Ward and Krebs 1985), with movements of up to 1,200 km recorded (K. Poole, B. Slough & G. Mowat, unpubl. data). Extralimital records have documented lynx in the northern tundra and Arctic islands (Banfield 1974, Govt. of Canada 1988), and in Iowa, South Dakota, Nebraska and West Virgina (Koehler 1990b). Snowshoe hares prefer new growth vegetation, such as after forest fire or logging, and lynx may cluster in these areas (Quinn and Parker 1987). Population Status: Global: Category 4 Regional: Category 3 IUCN: Not Listed The status of the lynx is generally satisfactory (Quinn and Parker 1987, Govt. of Canada 1988). In Canada, it is considered endangered only in New Brunswick, and has been extirpated from Prince Edward Island and mainland Nova Scotia. The largest populations are found in southern Quebec, northern Alberta, northern British Columbia, Yukon, the Northwest Territories and Alaska (Govt. of Canada 1988; K. Poole, B. Slough in litt. 1993), although there is some concern that trapping pressure during the 1970s-1980s has reduced population levels (see Part II Chapter 4). The main US lynx population is found in Alaska. Elsewhere, they are more sparsely distributed, occurring in low numbers in the states of Washington, Montana, Idaho, Wyoming, Colorado, Minnesota, Wisconsin, Michigan, New York (reintroduced), Vermont, New Hampshire, and Maine, with the largest populations in the Rocky Mountains. Washington state recently listed the lynx as Threatened, and will take more active measures to aid population recovery (Anon. 1994b). Much of the lynx's American range consists of National Forest lands (Koehler 1990b). Lynx density fluctuates dramatically with the hare cycle (Breitenmoser et al. 1993b). An ongoing long-term study of an unexploited population in good quality habitat in the Yukon found densities of 2.8 individuals (including kittens) per 100 km2 during the hare low, and 37.2 per 100 km2 during the peak (G. Mowat and B. Slough, unpubl. data). Poole (1994) obtained very similar figures for his study area in the North-West Terrritories: 30 lynx per 100 km2 at the peak, and around 3/100 km2 the winter following the hare crash. In the south of their range, where snowshoe hare populations appear to be non-cyclic and stable at low densities, Koehler (1990a) reported lynx density at 2.6 individuals per 100 km2 (north-central Washington). The study was conducted in mature coniferous forest where fires had been suppressed, and the early successional growth preferred by snowshoe hares was limited to isolated pockets. Home range sizes for lynx range from 4-25 km2 for females, and 4-70 km2 for males (G. Mowat and B. Slough, unpubl. data). On the Kenai peninsula, Alaska, Kesterson (1988) found larger home ranges - 107 km2 for females and 225 km2 for males - but seasonal ranges were smaller, with females only 9.4 km2 in summer. Male ranges usually encompass those of females (Saunders 1963, Berrie 1973, Parker et al. 1983, Ward and Krebs 1985, Kesterson 1988, Slough and Ward 1990), but same-sex overlap has also been found (Berrie 1973, Mech 1980, Carbyn and Patriquin 1983, Noiseux and Doucet 1987; G. Mowat, B. Slough and K. Poole unpubl. data). Breitenmoser et al. (1993b) suggest that same-sex overlap reflects a high degree of tolerance of independent offspring by resident lynx, another unusual adaptation of the Canada lynx to a predictably cyclic prey base. Protection Status: CITES Appendix II. National legislation: managed for exploitation over most of its range. In Canada, trapping is regulated through closed seasons, quotas, limited entry and long-term trapping concessions. (See Part II Chapter 4 for a more detailed discussion of harvest management.) In the United States, trapping is permitted only in Alaska, Idaho, and Montana (Koehler 1990b). Principal Threats: In general, the future of the lynx looks more promising than for many other felids. Quinn and Parker (1987) do not believe that habitat alteration has had significant impact on lynx populations, although in the southern portions of its range optimal habitat for snowshoe hares is more patchily distributed (Wolff 1980, Sievert and Keith 1985). Modified logging, leaving interspersing areas of good tree cover, can actually benefit both lynx and their prey (Koehler and Brittell 1990). However, suppression of forest fires limits early successional growth favored by hares (Fox 1978), and may ultimately reduce hare abundance (B. Slough in litt. 1993). Lynx are easily trapped in comparison to other furbearers (Quinn and Parker 1987). At the low point of the hare cycle, lynx become more vulnerable to exploitation as they disperse in search of food - travelling greater distances increases the chances of being caught in a trap. Recruitment is also falling during this phase of the cycle, and it is possible that trapping pressure could severely disrupt lynx population dynamics by reducing numbers to the extent that recovery to previous levels is not attained when hares again increase (Brand and Keith 1979, Parker et al. 1983, Bailey et al. 1986). Several management options have been recommended to prevent over-trapping, including prohibiting exploitation in hare refugia (small patches of optimal habitat) throughout the cycle (Slough and Ward 1990, Poole 1992). In the past when lynx pelt prices were high (US$ 685 in 1981), trappers would seek out these refugia and concentrate their trapping efforts there (Carbyn and Patriquin 1983). Brand and Keith (1979) recommended that harvests be completely suspended for the 3-4 year low of the hare cycle, so that potentially more lynx are available for harvesting in peak years. Bailey et al. (1986) recommended a combination of harvest suspensions in the more accessible trapping areas during low hare years, and a quota system as lynx numbers increase. Government authorities have either implemented these recommendations or initiated harvest impact research programmes, as discussed in Part II Chapter 4. In addition, demand for lynx pelts, particularly from key European importing nations, appears to be declining, and pelt prices fell to less than US$ 80 during 1992-93 (K. Poole, B. Slough in litt. 1993). The threat of unsustainable trapping pressure is likely to diminish in the future.

Available from IUCN Publications Service, 119c Huntingdon Road, Cambridge CB3 0DL, UK. T: +44 1223 277 894; F: +44 1223 277 175; iucn-psu@wcmc.org.uk

Island Press Box 7, Covelo, CA 95428 T: 800 828 1302 (toll free in US) 707 983 6432 ipress@igc.apc.org

Peter Jackson 24-04-1997