Northern Two-lined Salamander
© 2004 Henk Wallays (1 of 32)
Eurycea bislineata (Green, 1818)
David M. Sever1
1. Historical versus Current Distribution. Northern two-lined salamanders (Eurycea bislineata) are the well-known “yellow salamanders” of the northeastern United States and eastern Canada. This species was one of the first North American salamanders formally given a Linnaean name (Green, 1818), and the numerous nineteenth century synonyms (see Harlan, 1826a; Holbrook, 1840; Dumeril et al., 1854; Cope, 1889) indicate that the species was widespread and well known to naturalists of that era. No subspecies are currently recognized. However, until Jacobs (1987) raised southern two-lined salamanders (E. cirrigera) and Blue Ridge two-lined salamanders (E. wilderae) to full specific status, these forms were considered subspecies of northern two-lined salamanders, so the historical distribution included the range of those species as well. Some workers, such as Petranka (1998), still recognize southern two-lined salamanders and Blue Ridge two-lined salamanders as subspecies of northern two-lined salamanders. I support the elevation of southern two-lined salamanders and Blue Ridge two-lined salamanders and, indeed, I believe both of those forms are polytypic (Sever, 1999b,c). In this report, E. bislineata is considered to include group C of Jacobs (1987), all Canadian populations (none were analyzed by Jacobs), and additional populations assigned to the species by morphological or genetic analysis in Ohio, West Virginia, and Virginia (Sever, 1989, 1999a, Mitchell and Reay, 1999). The distribution of northern two-lined salamanders, therefore, is from Labrador and northern Québec and eastern Ontario, through New Brunswick and New England to northern Virginia, west through eastern Ohio and the Kanawha River valley of West Virginia (Sever, 1999a). In Canada, northern two-lined salamanders still have not been recorded from Nova Scotia or Prince Edward Island. Sever (1999a) stated that the distributional limits of the species in Canada are unclear, and that the range of the species may actually be expanding. The boundaries between northern two-lined salamanders and southern two-lined salamanders in Ohio (Guttman and Karlin, 1986) and Virginia (Mitchell and Reay, 1999) have been established by allozyme analysis, and some geographic overlap occurs in both states. The specific identity of Maryland populations has not been determined, but allozyme analysis indicates that two genetically distinct, geographically overlapping forms occur (Miller and Hallerman, 1994). The contact zone in West Virginia has not yet been investigated by genetic analysis, but overlap occurs in the Kanawha River Valley (Sever, 1999a,b). Ecological studies are necessary in areas of sympatry between northern two-lined salamanders and southern two-lined salamanders to determine how reproductive isolation is maintained. Southern two-lined salamanders replace northern two-lined salamanders in the southern Blue Ridge Mountains in southwestern Virginia (Dunn, 1926; Jacobs, 1987), and no areas of sympatry are known (Sever, 1999a,c). Some studies on two-lined salamanders published prior to Jacobs (1987) do not specify localities from which specimens were examined; thus, the specific identity of animals used cannot be determined. In the present account, an effort was made to restrict comments to studies on populations within the range of northern two-lined salamanders as defined above.
2. Historical versus Current Abundance. Green (1818), in the original description of the species, simply reported: “Inhabits shallow waters; is found in numbers early in the spring, and is very active.” DeKay (1842), however, stated: "Although this species is said to be very common, both by Green and Harlan, I have never had the good fortune to meet with it." In many late nineteenth century and early twentieth-century reports, northern two-lined salamanders were considered common to very common (e.g., Allen, 1868; Morse, 1904; Reed and Wright, 1909). Cope (1889) stated that the species was “very abundant in Pennsylvania, and extends its range, with decreasing numbers, to Maine.” I collected my first specimen in 1961; the animal was lying upside down under a log, along a rivulet at the edge of a dirt road that was bordered by a fence-row of Osage orange. In 40 yr of experience with the species, I find that northern two-lined salamanders remain common to seasonally abundant in appropriate habitats. Obviously, habitat destruction (deforestation, pollution, and siltation of streams) has eliminated many populations, but northern two-lined salamanders ikely are lpresent in most clean-water, small, rocky, woodland streams within its range. Weller and Green (1997) report no indication of decline in Canada where for most of their range, northern two-lined salamanders are the only stream-side salamanders. The post-glacial dispersal of the species into Labrador and nearly to Hudson Bay in Québec and Ontario indicates tremendous dispersal ability in such a small amphibian.
3. Life History Features. Much literature occurs on these topics, from the first description of nesting in shallow streams (the usual situation) by H.H. Wilder (1899) to the most recent report involving nests at depths of 9–13.5 m in a mountain lake (Lake Minnewaska, Shawangunk Mountains, New York) by Bahret (1996). However, Bishop's (1941b) account on the natural history of northern two-lined salamanders in New York is still the most comprehensive available for the species.
A. Breeding. Reproduction is aquatic.
i. Breeding migrations. As noted by Bishop (1941b), no marked seasonal movements, such as those associated with mole salamanders (Ambystoma sp.), occur. In summer, especially during wet periods and at higher altitudes, adults may wander far from breeding areas in springs, seeps, and streams. Dry conditions, however, will cause these animals to return to aquatic habitats, independently of any mating or breeding behavior (see MacCulloch and Bider  for an extensive study of summer movements of northern two-lined salamanders in Québec). In the population that Bahret (1996) discovered nesting at considerable depths in a lake, the adults stayed in the lake the entire year. In any event, because the eggs are laid in water and the larvae are aquatic, females will congregate at appropriate nesting sites. Oviposition follows within 1 mo or so of mating, so males congregate at such sites as well. In Ohio, congregations of males and females in appropriate mating/nesting areas sometimes commence in fall prior to hibernation. The mating period is at its peak in March and April after hibernation. Northern two-lined salamanders are relatively easy to find in these generally wet, cool periods, and adults seem abundant at locales where they are hard to find in mid summer. Mating occurs on land (Noble, 1929a), but it is not unusual to find pairs under rocks in the middle of streams during daylight hours. All adult females at mating/nesting sites are gravid with large, vitellogenic eggs; this does not necessarily mean an annual breeding cycle. Non-breeding females have no compelling reasons to reside at mating/nesting sites and compete with sexually active individuals for food and shelter. The classic description of courtship and mating in northern two-lined salamanders is by Noble (1929a), which includes the first observations on any plethodontid of the tail straddle walk and use of courtship glands.
ii. Breeding habitat. Although small, swift-flowing, rocky, woodland streams are the most frequent nesting areas, nests can be found in a wide variety of aquatic habitats. The occurrence of nests in deep lakes already has been noted (Bahret, 1996), and I have found nests in gravelly spring-heads, wells, and boggy areas.
i. Egg deposition sites. In eastern Ohio, I commonly find the first nests during the last week of April to the first week of May, and nesting is nearly synchronous among females in a population. The eggs most frequently are found attached to the undersides of rocks and logs in flowing water. In Lake Minnewaska, however, Bahret (1996) found clutches on the topmost leaves of water moss and not deposited cryptically.
ii. Clutch sizes. In Massachusetts, I.W. Wilder (1924b) found that the usual complement was 12–36 with an average of 18. In New York, Noble and Richards (1932) reported a maximum of 41 eggs, and Bishop (1941b) found a maximum of 43 eggs with a mean of about 30. Stewart (1968) examined the number of enlarged ovarian follicles in New York specimens and found 19–86, with a mean of 46 follicles. However, some ovarian follicles may not mature, so follicle counts probably overestimate the number of eggs ovulated and fertilized. Often times, > 1 nest can be found under a single large rock. One should consider strongly the possibility that any nest that contains > 50 eggs likely represents the complements of > 1 female. The female stays with the eggs; I have found females guarding eggs that were nearly ready to hatch. Incubation period is 4–10 wk (Petranka, 1998).
C. Larvae/Metamorphosis. Larvae are aquatic; excellent descriptions and illustrations are provided by Trapido and Clausen (1940) for specimens from Québec and by Bishop (1941b) for specimens from New York.
i. Length of larval stage. For New York stream populations, the larval period is generally 2 yr with a mean size of 30 mm SVL (45.7–80.0 mm TL) at metamorphosis (Stewart, 1968). In Pennsylvania, some individuals take 3 yr to metamorphose, but despite whether 2 or 3 yr are necessary, metamorphosis occurs at 27.1–34.1 mm SVL and 54.6–60.9 TL (Hudson, 1955). Trapido and Clausen (1940) reported that in Québec the larval period in streams may be 3 yr and the smallest metamorphosed individual was 77 mm TL. Larvae from still waters seem to attain a larger size than those from streams. For example, the largest larvae Bahret (1996) collected from Lake Minnewaska were 43–46 mm SVL and 84–92 mm TL. Whether the larger size of these larvae is due to a longer larval period is unknown.
ii. Larval requirements.
a. Food. Smallwood (1928) stated that the larvae of northern two-lined salamanders from the Onondaga Mountains of New York eat caddisfly larvae and beetle larvae. Burton (1976) reported the most important food items found in larvae from a New Hampshire stream are midges, stoneflies, cladocerans, and copepods.
b. Cover. At night I often have observed northern two-lined salamanders larvae at the bottoms of shallow, quiet pools that form below riffle areas in rocky streams. At these times, the larvae are either stationary on the substrate or slowly crawling along the bottom. Larvae can be found in such pools in the daytime as well, but usually then they are found under sticks, leaves, and other debris that accumulate in areas of such pools. Larvae also are collected frequently after scrapping through rocks and logs in riffles and holding a dip-net downstream from the disturbance. Hudson (1955) in southeastern Pennsylvania noted that larger larvae are less secretive than smaller larvae. Larvae are prone to drift downstream (Johnson and Goldberg, 1975), which might be an important factor in the extraordinary dispersal ability of this species.
iii. Larval polymorphisms. Unknown.
iv. Features of metamorphosis. Some observations suggest that a great deal of variation occurs in the time to metamorphosis within a given cohort, independent of environmental factors. In Massachusetts, I.W. Wilder (1924a) noted that the majority of larvae pass through their second winter in a pre-metamorphic state, whereas others lag behind a year. Transformation occurs as early as May in the 2-yr-old pre-metamorphic group, reaches its height in July, and continues into early fall. Laggards from this group may pass through a third winter. Elsewhere, I.W. Wilder (1924b) reported that two individuals, hatching at the same time in the laboratory and kept under the same conditions, metamorphosed at dates approximately 1 yr apart. My own observations lead me to believe that metamorphosis may occur more synchronously in some populations. In late summer and early fall, I often find a marked increase in the number of recently transformed juveniles under rocks along woodland streams.
v. Post-metamorphic migrations. Juveniles apparently stay close to streams through their first breeding season, which may occur in the spring following metamorphosis in the previous autumn (Bishop, 1941b). MacCulloch and Bider (1975) in Québec found that metamorphosing individuals in August remained closer to their natal streams than post-breeding adults.
vi. Neoteny. Neoteny is not known to occur. Cope (1889) noted, “It is one of those species whose metamorphoses are prolonged and which remains in the larval state until nearly fully grown.” Thus, large larvae often possess well-developed gonads, and, as proposed by Bishop (1941b), maturity may be attained within a comparatively short time after transformation.
D. Juvenile Habitat. As indicated above, juveniles rapidly mature into adults, and if metamorphosis occurred in a given stretch of a stream, juveniles occupy that area through their first breeding season. A brief post-larval migration into terrestrial habitats < 100 m from the stream may occur (MacCulloch and Bider, 1975). Thus, the requirements during this short juvenile stage are not different from adults.
E. Adult Habitat. Adults from various populations of northern two-lined salamanders are 28.9–40.9 mm mean SVL; the record individual is a female 53 mm SVL and 123 mm TL from Ohio (Sever, 1999a). Stewart (1968) reported that females from New York are significantly longer than males. Two-lined salamanders usually are the smallest species when they occur in communities containing other salamanders in the northeastern United States. The primary habitats are unpolluted bogs, springs, streams, or lakes in wooded areas. Most small streams in the wooded, mountainous areas of the range are rocky, which provides a good substrate for nesting, and lack large predatory fish. Occasionally, specimens are found along larger streams and rivers, but nesting has not been reported in such habitats. Although members of the northern two-lined salamander complex can co-exist in streams with predatory fishes (Petranka et al., 1987), predation by fish still is a factor that likely limits extensive exploitation of rocky areas along large streams. Lake Minnewaska, where Bahret (1996) found nesting at depths of 9–13.5 m and adults living year-round at 18 m, lacks fish. The disappearance of fish is attributed to atmospheric acid deposition in the lake during the early 1900s, resulting in a pH of 4.5 (Bahret, 1996). This high acidity, however, seemingly does not affect success of northern two-lined salamanders.
F. Home Range Size. Stewart (1968) found as many as 11 adults/m2 in New York populations, whereas Burton and Likens (1975b) reported only 0.02–0.04 individuals/m2 in the Hubbard Brook Experimental Forest of New Hampshire. These densities seem quite disparate in populations that presumably occupy favorable habitats. Density studies on terrestrial northern two-lined salamanders, however, can be confounded by influx of numerous newly metamorphosed juveniles in late summer and fall, aggregations in spring of adults at favorable mating/nesting sites, interspecific interactions, and/or other factors.
G. Territories. Grant (1955) reported territorial defense (by biting intruders) in captive northern two-lined salamanders kept in relatively close quarters within a terrarium; whether this occurs in the wild is uncertain.
H. Aestivation/Avoiding Dessication. I am unaware of any data.
I. Seasonal Migrations. Include breeding migrations and post-metamorphic migrations (see above).
J. Torpor (Hibernation). I am unaware of any data.
K. Interspecific Associations/Exclusions. As noted previously, in much of the Canadian range, especially north of the St. Lawrence Seaway and the Great Lakes, northern two-lined salamanders are the only stream-side salamanders. In more southern portions of its range, northern two-lined salamanders share stream habitats with one or more of the following species: long-tailed salamanders (Eurycea longicauda), northern dusky salamanders (Desmognathus fuscus), Allegheny Mountain dusky salamanders (Desmognathus ochrophaeus), red salamanders (Pseudotriton ruber), and/or spring salamanders (Gyrinophilus porphyriticus). The latter species is a known predator of northern two-lined salamanders (Wright and Haber, 1922; Bishop, 1941b). No detailed studies, however, have been done on ecological interactions between northern two-lined salamanders and these species. Studies are also necessary on the ecological and genetic relationships of northern two-lined salamanders and southern two-lined salamanders in areas of sympatry in Ohio, West Virginia, Virginia, and Maryland.
L. Age/Size at Reproductive Maturity. As mentioned earlier, metamorphosis generally occurs in the summer or fall of the second or third year, and these individuals may reach sexual maturity in time to participate in breeding activities the following spring. As a result, individuals are breeding at the beginning of their third or fourth year, depending upon length of the larval period. The smallest mature specimens I have examined were 30 mm SVL; however, I examined two males 25 mm SVL from Chittenden County, Vermont, that had evenly pigmented testes and may have been mature. Bishop (1941b) reported that sexual maturity is reached in males at 67 mm TL and some females when only 61 mm TL.
M. Longevity. In Québec, MacCulloch and Bider (1975) reported that only 25% of post-breeding adults who migrated > 100 m from the mating/nesting area returned for the subsequent breeding season. They attributed this loss of 75% of these migrants to mortality, which indicates a rather high turnover in the adult population each year. I am unaware of any literature on maximum life span either in nature or in captivity.
N. Feeding Behavior. Feeding habits of metamorphosed northern two-lined salamanders have been studied by Surface (1913) in Pennsylvania, Smallwood (1928) and Hamilton (1932) in New York, and Burton (1976) in New Hampshire. As expected, a wide range of invertebrates are eaten, although Burton (1976) noted that prey items are primarily terrestrial. Terrestrial items include various insects (beetles, roaches, springtails, dipterans, and hymenopterans), earthworms, snails, spiders, and isopods, whereas aquatic prey includes stonefly nymphs, caddisfly larvae, midges, and mayflies. Thus, northern two-lined salamanders are generalist feeders that are likely to eat any small invertebrate within the appropriate size range.
O. Predators. Wright and Haber (1922) included northern two-lined salamanders among other salamander species that are preyed upon by spring salamanders. Bishop (1941b) noted that a number of recently captured spring salamanders from New York disgorged partly digested specimens of northern two-lined salamanders. Other known predators include shrews, owls, blue jays, garter snakes, and trout (Brodie, 1977; Petranka, 1998). I have observed screech owls (Otus asio) capturing northern two-lined salamanders on blacktop roads on rainy nights.
P. Anti-Predator Mechanisms. Brodie (1977) reported that northern two-lined salamanders assume a defensive posture when attacked by shrews (Blarina brevicauda) and blue jays. The body is coiled with the head near the vent, and the tail is elevated and undulated. In a test chamber, 10 of 11 northern two-lined salamanders attacked by shrews were first bitten on the tail. The shrew backed away, wiping its mouth, which in nature could give the salamander time to escape (Brodie, 1977). In a later study in which northern two-lined salamanders from New York were exposed to shrews, only 26% of the salamanders survived attacks (Brodie et al., 1979). In some cases, the tail was contacted by the shrew and avoidance followed, but in other cases, the head was bitten before the tail was contacted. The shrews, however, did not eat 47.8% of the northern two-lined salamanders killed, indicating that shrews find the salamander relatively unpalatable (Brodie et al., 1979). Dowdey and Brodie (1989) studied anti-predator response of northern two-lined salamanders from New York to garter snakes (Thamnophis sp.). Garter snakes are predators that are not repulsed by skin secretions, and salamanders do not assume the defensive postures used against blue jays and shrews. Instead, northern two-lined salamanders run or remain immobile during encounters with garter snakes. In a testing chamber, most salamanders ran when flicked by a snake's tongue, and 90% of these with tails survived (tailed individuals ran faster than tailless ones). All salamanders that remained immobile after tongue flicks were eaten. The response to running was well developed only in a population subjected to heavy snake predation. However, when a salamander was merely contacted by the head or body of a garter snake, most remained immobile and 100% of these survived. Only 34% survived of those who ran. In an experiment using Pennsylvania specimens, Whiteman and Wissinger (1991) found that northern two-lined salamanders (59%) were almost twice as likely to autotomize their tails when attacked by garter snakes than northern dusky salamanders (24%) or Allegheny Mountain dusky salamanders (28%). Tail autotomy can increase the chance of the salamander escaping, although it makes the animal more susceptible to future predator attack (Dowdey and Brodie, 1989). Two-lined salamanders, therefore, have alternative anti-predator strategies, and the type and efficacy of each strategy varies with the type of predator and, in the case of garter snakes, the nature of contact.
Q. Diseases. I am not aware of any reports on diseases of northern two-lined salamanders.
R. Parasites. I am not aware of any reports on parasites of northern two-lined salamanders. Some literature on this subject (Burchett and Shoemaker, 1990) does exist for the sister taxon, southern two-lined salamanders.
S. Comments. A considerable literature exists on northern two-lined salamanders, and Sever (1999a) should be consulted for additional references, especially on topics not covered here such as morphology, physiology, and diagnostic characters. Note especially that several recent accounts (e.g., Petranka, 1998, Powell et al., 1998) continue to use characters like tail stripe length and number of costal grooves between toes of the adpressed limbs to separate species in the complex, even though such characters have long been discredited (Sever, 1972, 1989, 1999a,b,c). Because Petranka (1998) did not recognize southern two-lined salamanders and Blue Ridge two-lined salamanders as separate species, he combined data on these species with those on northern two-lined salamanders into one account; a careful reading, however, will reveal much material relevant to each taxon.
4. Conservation. Northern two-lined salamanders are listed as Protected by the State of New Jersey (Levell, 1997). This categorization applies to all indigenous non-game species in this state and does not necessarily indicate a conservation concern. In general, northern two-lined salamanders are widespread and suitable habitat is available across their range. Indeed, northern two-lined salamanders may be spreading their range north.
1David M. Sever
Literature references for Amphibian Declines: The Conservation Status of United States Species, edited by Michael Lannoo, are here.
Feedback or comments about this page.
Citation: AmphibiaWeb. 2018. <http://amphibiaweb.org> University of California, Berkeley, CA, USA. Accessed 21 Sep 2018.
AmphibiaWeb's policy on data use.