Ambystoma barbouri Kraus and Petranka, 1989
Mark B. Watson
Thomas K. Pauley
1. Historical versus Current Distribution. The current range of streamside
salamanders (Ambystoma barbouri) includes central and western Kentucky,
central Tennessee, southeastern Indiana, southwestern Ohio, and western West Virginia
along the Ohio River (Kraus and Petranka, 1989; Longbine et al., 1991; Petranka,
1998). Kraus and Petranka (1989) first described streamside salamanders as a
sibling species of small-mouthed salamanders (Ambystoma texanum).
Streamside salamanders and small-mouthed salamanders are allopatric throughout most of
their range, but parapatric or sympatric in portions of Kentucky and Tennessee (Petranka,
1982b). Streamside salamanders can be distinguished from small-mouthed salamanders
by differences in range, ecology, and slight morphological characters such as different
dentition. For example, small-mouthed salamanders have maxillary and premaxillary
teeth with long pointed cusps, whereas streamside salamanders have more rounded cusps
similar to other ambystomatids. All reports prior to 1989 considered streamside
salamanders to be stream-breeding small-mouthed salamanders. Therefore, changes in
historical distribution may not be apparent. New populations of streamside
salamanders may be discovered in the states within their range after a thorough
examination of small-mouthed salamander specimens.
2. Historical versus Current Abundance. Reports of stream-breeding small-mouthed
salamanders prior to the description of streamside salamanders as a separate species were
probably of streamside salamanders. For example, Green (1955) stated that the only
known population of small-mouthed salamanders in West Virginia breeds in streams.
Subsequently, Longbine et al. (1991) showed that this population consists of streamside
3. Life History Features.
Reproduction is aquatic.
i. Breeding migrations. Streamside salamanders have an extended breeding season
that lasts from late fall to early spring. Breeding migrations occur on rainy nights
and individuals usually move from forests to first- and second-order streams, although
streamside salamanders have been also found in ponds (Petranka, 1984a). Streamside
salamanders do not migrate en masse as seen in congeneric species such as spotted
salamanders (Ambystoma maculatum). Breeding in streamside
salamanders begins 4–5 wk earlier than in small-mouthed salamanders where the
species occur sympatrically. Additionally, streamside salamanders migrate to their
breeding streams in the fall and undergo low levels of mating activity from early winter
to early spring (Kraus and Petranka, 1989).
ii. Breeding habitat. Streamside salamanders were first described as a
stream-breeding type of small-mouthed salamanders (Petranka, 1982a). They breed in
first- and second-order streams that are usually devoid of fishes (Petranka, 1983,
1984d; Kats and Sih, 1992). Petranka (1982a) described the courtship behavior of
stream-breeding small-mouthed salamanders in the laboratory. He described
courtship in four phases. During Phase 1, males exhibited exaggerated undulations
of their bodies and tails. They also circled in tight groups for 18–35
min. Phase 2 was described as spermatophore deposition, which lasted about 10
s. The males grasped the substrate with their legs and deposited 1–5
spermatophores on the top of gravel particles or other substrate. He also observed
that males might try to dislodge spermatophores of conspecifics. In Phase 3,
females entered a field of spermatophores. Females mounted 18–27
spermatophores, some more than once. During Phase 4, females became dormant or
motionless and laid eggs within 48 hr. McWilliams (1992) observed that breeding
activities of streamside salamanders and small-mouthed salamanders are similar and that
the location for breeding is the major difference. McWilliams (1992) also noted
that unlike small-mouthed salamanders, breeding streamside salamander males are less
likely to interfere with other males.
i. Egg deposition sites. Eggs are attached in a single row on the lower surface of
flat rocks in flowing streams (Petranka, 1984d). Subsequent rows of eggs may be
deposited, which may form clumps. Eggs are usually deposited in hidden or cryptic
sites but may be in exposed locations (Sih and Maurer, 1992). Egg deposition occurs
from January to early April. Oviposition generally occurs in pools, as opposed to
riffles, in streams (Petranka, 1984a; Holomuzki, 1991).
ii. Clutch size. While clutch size varies from < 10 to > 1,000 eggs, the
average number of ova in gravid females is around 260, suggesting communal egg deposition
Larvae/Metamorphosis. Petranka (1984d) and Petranka and Sih (1986) observed that
eggs hatch during late April, and larvae transform 6–9 wk after hatching.
Brandon (1961) described the larval development of stream-breeding small-mouthed
salamanders from central Kentucky. Newly hatched larvae are 12 mm in length.
Individuals begin to metamorphose at 37–41 mm. He also observed two
individuals in that size range that had completely lost their gills. Petranka
(1984c,d) noted that temperature and food availability could influence larval growth
rates. At warm temperatures (24 ˚C) larvae metamorphosed in 27 d on average,
but at cooler temperatures (15 ˚C) metamorphosis was delayed to an average of 72
d. He also noted that larvae grew to a larger size when metamorphosis occurred at
cooler temperatures. Maurer and Sih (1996) compared larval growth rates of
streamside salamanders and small-mouthed salamanders. Streamside salamanders showed
increased feeding and activity levels and developed faster than small-mouthed
salamanders. Streamside salamanders also reduced their activity in response to food
deprivation. These differences were attributed to differences in habitat duration
between the stream-dwelling and pond-dwelling species.
In some streams,
competition for food may increase the length of the metamorphic period (Petranka and Sih,
1986). Chemical cues from green sunfish (Lepomis cyanellus)
could delay hatching of streamside salamanders (Moore et al., 1996). Similarly, Sih
and Moore (1993) reported experimental evidence to show that flatworm (Pagocotus
gracilis) predation can induce streamside salamander eggs to delay
hatching. Increased size of hatchlings decreased the likelihood of predation by
Habitat. Juvenile streamside salamanders utilize the same habitat as larvae.
Streamside salamander juveniles feed on various macroinvertebrates, including insect
larvae, zooplankton, isopods, and amphipods (Huang and Sih, 1990; Sparks, 1996).
Habitat. Adult streamside salamanders are extremely fossorial and above ground
activity is mostly observed during breeding migrations (Petranka, 1982a; Holomuzki,
1991). They are usually found in upland deciduous forests and are most common in
regions with exposed limestone (Petranka, 1998).
F. Home Range
Aestivation/Avoiding Dessication. Aestivation is unknown; animals may seek shelter
under cover objects or burrow when facing dessicating conditions.
Migrations. Petranka (1984a) found that migrations to breeding sites in central
Kentucky occur from late October to March. Males reach breeding sites before females
Associations/Exclusions. Streamside salamanders breed in first- and second-order
streams and may come into contact with individuals in the genera Desmognathus,
Eurycea, Pseudotriton, and Gyrinophilus.
L. Age/Size at
Reproductive Maturity. Unknown.
Behavior. Streamside salamander larvae can have substantial effects on the density
of benthic isopods. Huang and Sih (1991a) observed a large negative relationship
between density of salamander larvae and isopods in stream experiments. Sih and
Petranka (1988) showed that small-mouthed salamander larvae were not selective when prey
items were in low abundance, but would release previously acquired smaller prey in order
to acquire larger prey.
Fishes are the major predators of streamside salamander larvae. Petranka (1983)
showed that fish predation or the presence of fish restricted stream-breeding
small-mouthed salamander larvae to upper regions of streams that are devoid of
fishes. Kats et al. (1988) demonstrated that stream-breeding small-mouthed
salamanders were palatable to fish. Sih et al. (1992) reported that fish preyed
heavily upon larvae of streamside salamanders. They observed that 30–40% of
larvae drifted into a pool with fish, and of those only 6–8% survived to drift
out. Sih (1992) suggested that ineffective predator defense against sunfish
(Centrarchidae) may be important in the evolution of small-mouthed salamanders.
Sunfish in stream pools could be a barrier to gene flow by preventing movement of larvae
between pools (Storfer, 1999a). Populations of streamside salamanders separated by
fish within the same stream could be as genetically different as those from different
streams (Storfer, 1999b). Flatworms prey on small streamside salamander larvae in
streams (Petranka et al., 1987a; Sih and Moore, 1993). Three small-mouthed
salamander larvae and one mud salamander (Pseudotriton montanus) larva
were regurgitated by a northern water snake (Nerodia s.
sipedon) that was found foraging in a small stream in central Kentucky (Kats,
Mechanisms. Adult small-mouthed salamanders, a close relative of streamside
salamanders, use biting and immobility as anti-predator defenses. They also elevate
the tail and roll their bodies toward a grasped limb, right themselves, and coil upon
release (Brodie et al., 1974a; Brodie, 1977). Members of the genus
Ambystoma have many anti-predator adaptations including tail movements, coiling
and elevating the body, aposematic coloration, and noxious skin secretions (Brodie,
1977). Huang and Sih (1990) found stream-dwelling small-mouthed salamander larvae
increased the use of refugia in the presence of predatory sunfish.
In laboratory studies
that simulated fish predation, streamside salamanders took refuge under artificial
substrates (Huang and Sih, 1991b). Several studies have shown that streamside
salamander larvae respond to chemical cues from predatory fish. In these studies,
larvae exhibited anti-predator behavior when fish were kept isolated and not visible to
the larvae (Petranka et al., 1987b; Sih and Kats, 1991, 1994). Kats et al. (1988)
observed that stream-dwelling small-mouthed salamander larvae responded to chemical cues
from green sunfish by spending less time in the open and more time under cover.
They also observed that small-mouthed salamanders that breed in ponds normally devoid of
fish did not exhibit this behavior. Other studies have shown larvae that increase
time spent under substrates, such as rocks and algae, become immobile or shift to more
nocturnal feeding behavior in the presence of fish (Kats et al., 1988; Holomuzki, 1989a;
Sih et al., 1992; Sih and Kats, 1994). Storfer and Sih (1998) noted that larvae
from populations most isolated from fishless populations showed stronger anti-predator
behavior. Storfer et al. (1999) observed that streamside salamander larvae from
streams with fish were a more cryptic, lighter color than those found in shallow
ephemeral streams without fish.
Nematodes (Cosmoceroides dukae or C. variabilis) are
known to occur in the digestive tract (Baker, 1987).
4. Conservation. Streamside salamanders usually are not found in streams where
surrounding forest land has been timbered (Petranka, 1998). This suggests that
deforestation and development around streams and ravines within their range will be
detrimental to this species.
Only two populations
are known in West Virginia, and one of these may have been destroyed recently by
development. Mitchell et al. (1999) lists streamside salamanders as a species that
may need to be monitored in West Virginia because of few verified populations and lack of
data on the status of the known populations.
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: Information on
amphibian biology and conservation. [web application]. 2015. Berkeley, California:
(Accessed: May 6, 2015).
AmphibiaWeb's policy on data use.