Amphibians of North Carolina
Scientific Name:
Common Name:
Family (Alpha):
« »
Hylidae Members:
NC Records

Hyla versicolor - Gray Treefrog



caption
Taxonomy
Class: Amphibia Order: Anura Family: Hylidae Subfamily: Hylinae Other Common Name(s): Eastern Gray Treefrog
Taxonomic Comments: Duellman et al. (2016) elected to split treefrogs in the genus Hyla into two genera. Hyla (sensu stricto) refers to a group of species that are found in Eurasia, while a new genus Dryophytes was resurrected to include all of the North and Central American species and three other species that are found in eastern temperate Asia. Members of these proposed genera cannot be distinguished by any known morphological features, but represent two clades. Whether members of these clades should be placed in separate genera or treated as lesser taxa is arbitrary and dependent on one's taxonomic philosophy. Here, we retain Hyla for North American species to be consistent with current usage by the Society for the Study of Amphibians and Reptiles.

Cope's Gray Treefrog (Hyla chrysoscelis) and the Gray Treefrog (H. versicolor) are two morphologically indistinguishable species that are members of a polyploid complex. Hyla chrysoscelis is a diploid species (2N = 24 chromosomes), while H. versicolor is a tetraploid (4N = 48 chromosomes). The exact origin of H. versicolor has intrigued evolutionary biologists and generated several hypotheses that are supported to varying degrees by molecular and genetic analyses. Holloway et al. (2006) proposed that H. versicolor arose via several speciation events from H. chrysoscelis-like diploid ancestors and two other now extinct lineages of tree frogs. A recent analysis that used a much more comprehensive set of molecular data suggests that H. versicolor most likely formed via a single genome duplication event that involved only H. chrysoscelis and did not involve hybridization with other species (Booker et al. 2021). Since then, there has been significant hybridization between diploids and tetraploids where they co-occur. This has led to the reformation of distinct polyploid lineages following the initial whole genome duplication event. Despite evidence of a complex evolutionary history that involves the repeated re-formation of H. versicolor lineages through time, researchers treat the diploid and tetraploid forms of this complex as only two species. Since they are morphologically identical, the two species can only be differentiated using other evidence such as the chromosome number or voice characteristics. Fortunately, these species are rarely found in sympatry in North Carolina and can usually be identified based on locality. Hyla chrysoscelis occurs statewide, but H. versicolor is only known from a northern tier of counties in the Piedmont near the Virginia border, and from one record from Rowan County in the Piedmont.
Species Comments: Hyla versicolor has served as a model organism to study acoustic communications in anurans, with dozens of papers now published on this topic. As noted by Dodd (2013), many of the older publications concerning gray treefrogs are compromised because of the inability to determine if the species that was studied was H. versicolor or H. chrysoscelis. This is a particular problem in areas of sympatry where both species could potentially occur at a study site.
Identification
Description: Hyla versicolor and H. chrysoscelis are morphologically indistinguishable and the following description of H. versicolor applies equally to both species. The Gray Treefrog is a small to medium-sized frog with conspicuous toe pads, granular skin, and a lichen-like dorsal pattern. The dorsal patterning of the adults often consists of a grayish ground color that is overlain by darker patches with thin black margins. However, the ground color can vary from ashy white to pale brown or greenish, and individuals can change body color to better match their backgrounds (Beane et al. 2010, Dodd 2013). The juveniles generally tend to be greener than the adults, and often have rather poorly developed dark patches. There is a conspicuous whitish patch underneath the eye that is often lined on the sides with a thin black line. The upper surface of the legs have dark bands and the rear toes are partially webbed. The concealed surfaces of the back legs are mottled with bright orange and blackish pigmentation near the margins and washed with bright orange elsewhere. The belly, throat, and undersides of the front legs are whitish and unmarked, except for males which have dark vocal sacs during the breeding season.

Hyla versicolor is a slightly larger species than H. chrysoscelis but there is wide overlap in the size ranges (Dodd 2013). As with H. chrysoscelis, the males average slightly smaller than the females. In Connecticut, the males vary from 36–51 mm SUL (mean = 46 mm) versus 43–60 mm SUL (mean = 50 mm) for females (Klemens 1993). In Ohio, the means for males from several ponds ranged from 43-46 mm SUL versus 49–53 mm SUL for females (Gatz 1981b). Beane et al. (2010) reported a range of 32-62 mm SUL for populations in Virginia and North Carolina.

The tadpoles are indistinguishable from those of H. chrysoscelis (Altig 1970). The hatchlings vary from 4-7 mm TL and are brownish with black mottling on the tail. There is a pair of preorbital stripes, but they are not distinctly outlined and only become apparent after two weeks (Dodd 2013). As the tadpole grows, the venter becomes more cream colored with a heavy intrusion of gold flecks on the posterior end. The maximum size is around 42 mm TL (Dodd 2013). This species has inducible defenses against predators. Tadpoles in ponds with predatory dragonfly larvae respond by developing relatively large, brightly colored tail fins with blackish spots or heavy mottling along the margins (McCollum and Van Buskirk 1996). So, expect significant variation depending on local site conditions.

Hyla versicolor and H. chrysoscelis can only be distinguished in the field by the trill rates of calling males (pulses per second). We recommend making recordings of calling males for later analysis at any site where this species appears to be present. Mitchell and Pague (2011) conducted an extensive survey of populations in Virginia and found that there is no overlap in the trill rates of the two species when recorded in the field (i.e., prior to adjusting for temperature differences). The mean and standard deviation for trill rates for H. chrysoscelis was 45.8 (4.9) versus 24.5 (3.4) for H. versicolor. In essence, any field recording that shows a trill rate of < 30 pulses/sec can be assigned to H. versicolor with very high confidence.
Vocalizations: The advertisement call is a very loud slow trill. The trill of this species is lower pitched and occurs at a slower pulse rate (17–35 notes/sec) than that of H. versicolor (34–69 notes per second). The pulse rate varies depending on the ambient temperature and location, but averages between 24 and 29 pulses/sec at 24°C (Dodd 2013). Males that were recorded in the field in Virginia averaged 24.5 pulses/sec at an average air temperature of 22.5°C (Mitchell and Pague 2011). Each trill last less than a second, and sequential trills are typically issued every 2-7 seconds.

In addition to the advertisement call, other vocalizations are made including a “rain call” that sounds like a slow, creaking trill. It is often heard from high in the trees when a storm is approaching or conditions suddenly become cloudy. Rain calls are frequently issued outside of the breeding season and their function is unknown. Males also have a release call that is issued if a male is amplexed or comes into contact with another male. It consists of a series of high-pitched "erps" or "yips" (Pierce and Ralin 1972). An aggressive call that sounds like the alarm call of a hen turkey is often issued if two individuals come into close contact at the breeding sites.
Technical Reference: Dodd (2013)
Online Photos:    Google   iNaturalist
Observation Methods: Individuals are most easily observed around the breeding sites at night during rainy weather. They are also commonly seen crossing roads on rainy nights as they move to and from the breeding sites.

Download Video: "MP4"

AmphibiaWeb Account
Distribution in North Carolina
Distribution Comments: Hyla versicolor tends to have a more northern distribution than H. chrysoscelis, although some populations occur as far south as southeastern Texas and there are broad areas of sympatry (Dodd 2013). The range extends from western New Brunswick westward across southern Canada to western Manitoba and eastern Saskatchewan. From there is extends southward to central Virginia and north-central North Carolina in the east, and to northeastern West Virginia, south-central Ohio, south-central Indiana, and much of Illinois and Missouri to the west. A southern band continues from Missouri through southeastern Kansas, eastern Oklahoma, and northwestern Arkansas into eastern Texas and southwestern Louisiana. Apparent disjuncts are present in northern Kentucky and western Tennessee. As of 2022, populations have only been located in North Carolina in four counties along the Virginia border and at a site in Rowan County.
Distribution Reference: Beane et al. (2010), Dodd (2013)
County Map: Clicking on a county returns the records for the species in that county.
GBIF Global Distribution
Key Habitat Requirements
Habitat: This highly arboreal species is most commonly associated with deciduous hardwood forests where it blends in well with tree limbs in the forest canopy. Its habit preferences overlap strongly with that of H. chrysoscelis and the two species often occur syntopically. Adult H. versicolor frequent mesic and bottomland hardwood forests, mixed hardwood-conifer forests, as well as boreal forests in Canada. Breeding sites are either embedded within the forest proper or occur in more open habitats that typically are < 50 m or so from a forest tract.

The breeding sites are similar to those of H. chrysoscelis. The adults generally prefer fish-free seasonal or semi-permanent wetlands for breeding, but will use permanent ponds with predatory fish to a lesser extent (Dodd 2013). Commonly used habitats include vernal ponds in forests, marshes, flooded fields, floodplain overflow pools, red maple and shrub swamps, beaver ponds, roadside ditches, quarry ponds, flooded sand pits, farm ponds, and the shallow sections of lakes. Wright (1914) noted that males in New York frequently call from ponds with floating aquatics such as lily pads that served as resting or calling perches. Hocking and Semlitsch (2007) found that females preferred artificial pools set up in open habitats rather than in shaded forested sites. They most frequently used pools that were within 10 m of the forest edge.

We have very few records for this species in the state, but the habitats that are used in North Carolina are probably very similar to those of Piedmont populations of Cope's Gray Treefrog. These consist of a variety of hardwood and mixed pine-hardwood forests with breeding sites in close proximity. At one site in the western Piedmont, both species were observed chorusing in a slough located in the floodplain of a moderate-sized river where fish were present. The adjoining forests included both bottomland hardwoods and a stand of rich mesic hardwoods growing along a slope above the river. The absence of H. versicolor from the Coastal Plain of Virginia and North Carolina (Beane et al. 2010, Mitchell and Pague 2011) suggests that this species may be less tolerant of acidic, blackwater breeding habitats than H. chrysoscelis. However, the eggs and larvae are fairly acid tolerant, so the reason for their seeming absence from these regions is poorly understood.
Environmental and Physiological Tolerances: This species has a more northerly range than Cope's Gray Treefrog, which may reflect a greater degree of cold-tolerance. Hyla versicolor does not occur in most of the Atlantic Coastal Plain or the Gulf Coastal Plain which suggests that is may be less tolerant of acidic waters for breeding. However, the eggs and larvae appear to be acid tolerant, with the a critical pH level of 3.5-4.3 for larval and 3.8 for embryos (Dodd 2013, Gosner and Black 1957a, Grant and Licht 1993). Pehek (1995) found that exposure to a low pH (3.9) has no effect on survival, body mass, growth rates, length of the larval period for specimens from New Jersey.
Biotic Relationships: As with H. chrysoscelis, the larvae are palatable to fishes and a variety of other aquatic predators such as insect larvae, mole salamander larvae and newts (Adams et al. 2011, Gunzburger and Travis 2005, Van Buskirk and McCollum 2000, Walters 1975). The tadpoles are frequently killed and eaten by aquatic invertebrates and have evolved inducible defenses that reduce attack success. When confined with dragonfly larvae, the developing tadpoles develop relatively large, brightly colored tail fins with dark spots and blotches that are concentrated along the margins. These entice predators to strike the tail region rather than the more vulnerable head and body.

Larvae will also reduce their activity levels when confronted with predators. This lowers the number of encounters, but also compromises growth (Skelly 1992). The induced defenses in response to dragonfly larvae also enhances the ability of larvae to evade attacks from Ambystoma larvae (Van Buskirk and McCollum 2000). Schoeppner and Relyea (2005) examined how tadpoles altered their defenses when 10 different prey of a dragonfly predator were either crushed by hand to release alarm chemicals or consumed by the predator. They found that crushed and consumed prey both caused the tadpoles to decrease their activity and hide more, but only the consumed prey consistently induced deep tails and shorter bodies.

Predation on the juveniles and adults is poorly documented. Large spiders (Argiope sp.) occasionally eat the small metamorphs (Groves and Groves 1978), and the juveniles and adults are undoubtedly eaten by birds, aquatic snakes, bullfrogs, skunks, raccoons, and other mammals. The juveniles and adults blend in very well with their substrates and presumably reduce predation risk by remaining immobile in a crouched position. The sudden display of the bright coloration on the inside of the thighs as the frog jumps may serve to startle a predator long enough for the frog to escape (Dodd 2013). The adults have noxious skin secretions that can irritate the eyes and mucous membranes of human handlers.

As well-documented in H. chrysoscelis, the breeding adults can reduce the adverse impacts of predators and competitors on their offspring by selecting oviposition sites that are favorable. Fish will readily consume the larvae (Adams et al. 2011, Kurzava and Morin 1998, Smith et al. 1999) and the adults are less likely to use ponds with predatory fishes (Babbitt et al. 2003, Baber et al. 2004, Dodd 2013). Smith (2021) found that females are less likely to oviposit in artificial pools with the Western Mosquitofish (Gambusia affinis) and Shulse et al. (2013) found that mosquitofish greatly reduced the abundance of Gray Treefrog larvae in constructed ponds. At sites where clusters of ponds are present, pond use can shift between years (Brodman 2009, Skelly et al. 2003). This suggests that adults monitor risks to their offspring prior to breeding and frequently shift to alternate breeding sites that maximize the growth and survival of their future offspring.
See also Habitat Account for General Hardwood Forests
Life History and Autecology
Breeding and Courtship: The males typically begin calling after the spring leaf-out and may continue through August, although most breeding occurs during the early part of the breeding season. Calling often does not begin until late April or May -- or even later at northern latitudes -- but can begin in March in Texas (Dodd 2013). The males descend from trees and make their way to the breeding sites where they set up calling stations either at ground-level or in low vegetation. They often call from shrubby vegetation or small trees in or near the pond, or from emergent or floating vegetation (Bertram et al. 1996). Dodd (2013) noted that chorusing in many local populations tends to be of short duration. This likely reflects the short growing season at many northern locales and the need to complete the larval stage prior to the onset of cold weather. Examples include 12–23 days in Maine (Sullivan and Hinshaw, 1992) and 13–37 days in Ontario (Bertram and Berrill 1997). A population studied by Fellers (1979b) in Maryland bred for around 7 weeks.

Calling typically begins with the onset of darkness and continues for two to four hours. Wells and Taigen (1986) found that males in Connecticut emit 1,200–1,300 calls per hour over a 2–4 hour period. They have very loud trills that are energetically expensive to produce and exhausting (Schwartz and Rahmeyer 2006, Taigen and Wells 1985, Wells and Taigen 1986). Individuals can lose a significant amount of body weight after a few days of calling, and are forced to move back to terrestrial feeding sites to renew their energy reserves before returning to call at a later date (Dodd 2013, Fellers 1976, Mahan and Johnson 2007).

As with most pond-breeding amphibians, the males often greatly outnumber females on any given night and competition for mates is intense (e.g., Bertram et al. 1996). Fellers (1979b) found that only one-fourth of the males find a mate, and that certain males may mate as many as four times during the breeding season. The likelihood of finding a mate generally increases with the number of nights a male calls at a pond (Sullivan and Hinshaw 1992, Bertram 1996). Wells and Taigen (1986) found that males that are calling in large choruses produce trills that are about twice as long as single males, but decrease the number of trills per hour so that energy expenditures are similar to those of single males (Wells and Taigen 1986). Females general prefer males with long trills.

Fellers (1975, 1979a, 1979b) conducted detailed studies of mating behavior in Maryland and found that the calling males are territorial and generally space out at least 70 cm apart. Their advertisement calls are used to both attract females and advertise their presence to rival males. The dominant males select optimal perch sites with relatively little obstructive vegetation that could compromise the quality of their calls. Both the resident and intruder may issue encounter (aggressive) calls if an intruder approaches a territory holder. Fighting may also ensue that involves wrestling, shoving, kicking and the like (Dodd 2013). The loser (usually the intruder) will stop calling and either depart the area or remain as a non-calling subordinate.

The females are attracted to calling males and appear to select mates based on call characteristics such their pulse rate and call duration (Gerhardt and Brooks 2009, Gerhardt et al. 2000, Klump and Gerhardt 1987, Schwartz et al. 2001). They initiate amplexus by approaching the male and touching him. She then positions herself where the male can amplex her. Once amplexed, the pair may remain near the perch site for four hours or more before deporting to the pond to lay and fertilize the eggs (Fellers 1979a). Once the pair leaves, a subordinate male may move to the dominant male’s calling perch and begin calling.
Reproductive Mode: Egg laying has yet to be described in detail, but is presumed to be very similar to that of H. chrysoscelis. The female deposits several rafts of perhaps 20-100 eggs in a monolayer on the water surface, and will often swim away to a nearby location in the pond after ovipositing a group of eggs. Dodd (2013) noted that the entire complement of eggs can be laid in 5-10 minutes, but can take up to an hour in some instances.

Females usually leave the breeding site the same night they oviposit and return to their terrestrial habitats (Dodd 2013, Godwin and Roble 1983, Sullivan and Hinshaw 1992). A female may return one or more times during the breeding season and lay more eggs, which suggest that only a portion of her clutch is laid on any particular visit. Females in a local population may lay eggs throughout the entire calling season, but most are laid early in the season. The developing embryos presumably reach the hatching stages within a few days in southern populations, but may take 2–3 weeks in the Northeast depending on weather patterns (Gosner and Black 1957b).

Clutch sizes that have been reported in the literature and summarized by Dodd (2013) include 1,018 in Missouri (Johnson and Semlitsch 2003), a mean of 2,070 (range = 1,288-2,604) in Arkansas (Trauth et al. 1990), and 1,800–2,000 in New York (Wright 1914).
Aquatic Life History: Very little is known about the larval stage except for the well-documented phenomenon of inducible defenses. The larval period in natural populations is reported to last around 40–60 days (Gosner and Black 1957b), and recent metamorphs vary from 13.6–20 mm SUL (mean 16 mm) in New York (Wright 1914). Metamorphs examined by Munz (1920) in New York varied from 19-21 mm SUL.
Terrestrial Life History: Roble (1979) found that most metamorphs at a Wisconsin site dispersed away from breeding ponds within a week after metamorphosing, then remained within 125 m of their natal ponds. Most were found < 1 meter above ground on the top surfaces of the leaves of sedges, false nettle, reed canary grass, young swamp oak saplings and other low-growing vegetation. Frogs were commonly found from late July through late September, but appeared to move to overwintering retreats by early October. Individuals in this study did not appear to be arboreal during their first year of life.

The young metamorphs grow quickly and become sexually mature after two years (Dodd 2013). Babbitt (1937) reported that froglets in Connecticut reach 25 mm SUL or more before overwintering. Wright (1932) reported them to reach 20–30 mm SUL at 1 year of age and 30–41 mm SUL at 2 years of age. Growth slows markedly after reaching sexual maturity, with adults varying from 41–51 mm SUL.

The adults appear to be short-lived, with most individuals rarely surviving more than two or three years after reaching sexual maturity. Bertram and Berrill (1997) found that 83% of calling males in Ontario were first-year breeders, and that only 11–21% returned to breed the following year. Even fewer (7.8%) survived to breed for three years. Johnson et al. (2007) found one individual in Missouri for six consecutive years, which is very likely close to the maximum longevity in the wild.

Johnson et al. (2007) tracked marked adults and juveniles at three breeding sites in Missouri and monitored their use of artificial refugia (pipes strapped to trees). The adults exhibited strong site fidelity to the artificial refugia, with 99% of frogs returning to the same refugia (pipe) after leaving to overwinter or enter breeding ponds. In one case an individual used the same pipe for six consecutive seasons. Females moved farther from the ponds than males, and many were found over 200 m away (maximum distance = 330 m). Juveniles stayed closer to the ponds, with more than 80% found within 35 m of the water's edge. The adults overwintered on land, often 100-200 m from the ponds. The adults typically overwinter in loose soil beneath leaf litter on the forest floor and produce cryoprotectants (glucose and glycerol) that allows them to tolerate freezing to as low as to -5.5°C (Layne 1999).

This species appears to be a gape-limited generalist that will consume a wide variety of prey that are palatable (Sweetman 1944). Munz (1920) examined several recently metamorphosed froglets in New York and most had very small numbers of tiny prey such as flies, ants, lace bugs, and soil mites. Frogs in Missouri that were examined by Mahan and Johnson (2007) consumed a variety of prey, particularly ants and numerous taxa of beetles. However, many other prey were taken, including moths, roaches, hemipterans, orthopterans, isopods, millipedes, and arachnids. Ralin (1968) examined the gut contents of calling males in Texas and found that only about half had prey in their guts. Those that had eaten mostly fed on ants and beetles. Mahan and Johnson (2007) also found that adults collected from ponds had less prey than those from the nearby forest, which implies that the adults rarely feed when at the ponds.
General Ecology
Population Ecology: Local breeding populations often consist of only a few hundred adults or less (Dodd 2013). Where clusters of local ponds occur locally substantial gene flow likely occurs due to dispersal of juveniles and adults between ponds. In Missouri, Johnson (2005) found that there was significant variation in population genetic structuring and that populations > 30 km apart were identifiable. Under 3 km, however, significant structuring was usually not evident. Populations appeared to be structured as classic metapopulations at regional distances, but more as “patchy” metapopulations with substantial interpond movement when clusters of local ponds were in close proximity. Factors that regulate local populations are poorly resolved but may involve both competition among territorial males for mates and density-dependent growth and survivorship during the larval stage.
Community Ecology: The larvae usually share breeding sites with other anurans, but little data are available on competitive interactions. Fish and aquatic invertebrates are important predators on the larvae and mortality is minimized through a combination of adult choice of breeding sites, altered behavior, and inducible defenses (see biotic interactions above).
Adverse Environmental Impacts
Habitat Loss: Many local populations have undoubtedly been lost historically due to deforestation and urbanization. Roads present hazards to adults that are moving to and from breeding sites, and run-off from salted roads in northern locales can be toxic to the embryos and larvae of these and other pond-breeding amphibians if salt concentrations are high.
Habitat Fragmentation: Gray Treefrogs appear to tolerate habitat fragmentation well so long as forested areas remain that are large enough to support adult populations. Artificial aquatic habitats that are found in disturbed and urbanized landscapes such as flooded ditches, flooded fields and farm ponds with emergent vegetation are often used as breeding sites and have compensated for the loss of natural wetlands.
Status in North Carolina
NHP State Rank: S2
Global Rank: G5
Status in North Carolina: SC
Status Comments: Regional populations of this species appear to be stable, with no evidence of decline at a landscape scale (Dodd 2013). The status of populations in North Carolina is largely unknown. Populations here are at the southern limit of the range for areas east of the Mississippi River and are in need of additional study and monitoring.
Stewardship: Local populations are best maintained by having tracts of deciduous hardwoods next to fish-free breeding sites. Cluster of breeding ponds that are within a few hundred meters of each other allow adults the opportunity to shift to alternate breeding sites in response to fish invasions or other biotic stressors.

Recording Gallery for Hyla versicolor - Gray Treefrog

2012-05-22. Rowan Co. Stephen P. Hall, Jeffrey C. Beane, and Crystal Cockman - In floodplain pool along the South Yadkin River. Large, mixed chorus of versicolor and chrysoscelis recorded just before a thunderstorm. Two individuals of each species were collected and their karyotypes later analyzed. The chrysoscelis were definitely haploid but the results for the versicolor (identified by voice) were not as clearcut. While they were not haploid, they could not be confirmed as tetraploid (J. Beane, pers. comm.).

Photo Gallery for Hyla versicolor - Gray Treefrog

2 photos are shown.

Recorded by: Stephen P. Hall, Jeffrey C. Beane, and Crystal Cockman
Rowan Co.
Recorded by: Stephen P. Hall, Jeffrey C. Beane, and Crystal Cockman
Rowan Co.
Comment: In this spectrogram, note the conspicuous difference in pulse rates that help to distinguish H. versicolor from H. chrysoscelis.