view video (5126.1K MPG file)
[video details here]

View distribution map using BerkeleyMapper.

	Conservation Status (definitions)
IUCN (Red List) Status	Least Concern (LC) See threat category on the IUCN web site.
CITES	Appendix II
Other International Status	None
National Status	None
Regional Status	None

Description

Agalychnis callidryas, the Red-Eyed Tree Frog, is a slender, colorful, medium-sized frog. Females measure up to 77 mm, and males to 59 mm (Savage 2002). This frog has leaf-green to dark green dorsal surfaces; dark blue, purple, or brownish flanks, with cream-colored or yellow vertical or diagonal bars; blue or orange upper arms; thighs that are blue or orange on the anterior, posterior, and ventral surfaces; orange hands and feet, except for the outermost digits on each; a white ventrum; and protuberant red eyes, with vertical pupils (Savage 2002; Leenders 2001; Duellman 2001). The back is sometimes marked with faint tranverse darker green lines (especially in individuals from Nicaragua or Costa Rica) or small white dots (Duellman 2001). The average number of bars on the flanks increases in populations from north to south, with a mean of 5.0 bars in Mexico and a mean of 9.0 bars in Panama (Duellman 2001). In some populations from the middle part of the range (Nicaragua and Costa Rica, on the Caribbean side), there is often a continuous, longitudinal yellowish stripe connecting the upper ends of the vertical bars and separating the blue flanks from the green dorsum (Duellman 2001).

The skin is smooth both dorsally and ventrally (Savage 2002). Agalychnis callidryas has a rounded head and a truncated snout when viewed from above (Duellman 2001). Eyes are large and directed sideways (Leenders 2001). When this frog closes its eyes, transparent lower eyelids marked with a network of gold are apparent (Leenders 2001). It has distinct tympana (Savage 2002). The body is slender and somewhat flattened (Leenders 2001). Fingers are short, about one-half webbed, and have moderately large discs (Duellman 2001). The toes are short, about two-thirds webbed, and also have moderately large discs that are nearly as large as those on the fingers, with a narrow fold running from the heel to the disc of the fifth toe (Duellman 2001; Savage 2002). Adult males have paired vocal slits and a single internal median subgular vocal sac, as well as a grayish brown spinose nuptial pad at the base of each thumb (Savage 2002).

Young froglets (at least from Panama) are able to change color; they are green by day and change to purplish or reddish brown at night (Pyburn 1963). In addition, young froglets have yellow rather than red eyes, and have lighter-colored flanks lacking whitish bars (Pyburn 1963). The red eye coloration appears first at the periphery of the eye at about two weeks post-metamorphosis, and over a period of several days spreads inward to make the iris wholly red (Starrett 1960).

Agalychnis callidryas tadpoles are large, with a robust body that can measure 48 mm in total length at stage 34 (Savage 2002). The tail and caudal fins are moderately sized, with the tail tip narrowing to a thin flagellum (Savage 2002). The spiracle is sinistral and lateroventral, while the vent is dextral (Savage 2002). Eyes and nostrils are dorsolateral (Savage 2002). The mouth is anteroventral, with a small and complete oral disk, serrated beaks, and two upper plus three lower rows of denticles (Savage 2002; Duellman 2001). The row of denticles just above the mouth has a small gap medially (Savage 2002). Labial papillae are present on the lower lip in one to three rows lateral and ventral to the mouth, and on the upper lip in one to two rows lateral to the mouth, but lacking directly above the mouth (Duellman 2001). The tadpole body is olive gray dorsally, shading into a bluish gray speckled with olive-brown on the sides and undersides (Duellman 2001). Larval caudal musculature is a light grayish brown while the caudal fins are transparent, but both are speckled with dark gray (Duellman 2001).

Distribution and Habitat

Country distribution from AmphibiaWeb's database: Belize, Colombia, Costa Rica, Guatemala, Honduras, Mexico, Nicaragua, Panama

View distribution map using BerkeleyMapper.

Red-Eyed Tree Frogs are distributed from Mexico (Yucatán) to Panama, with an isolated report from the Cartagena Botanic Gardens in Colombia (Savage 2002; Ruiz-Carranza et al. 1996). They inhabit humid forests, primarily in lowlands and sometimes on premontane slopes up to 1,250 m (Leenders 2001).

Life History, Abundance, Activity, and Special Behaviors
Agalychnis callidryas is nocturnally active. This frog is arboreal and shelters on the underside of a broad leaf during the day and during the dry season, with limbs folded underneath its body (Leenders 2001). It has also been found in bromeliads, though this appears to be rare (Duellman 2001). This species is abundant in suitable habitat (Stuart et al. 2008).

Breeding occurs during the wet season (late May to November), beginning with the first rains (Pyburn 1970). Mating takes place throughout the rainy season but is particularly frequent in June, and occasionally peaks again in October (Donnelly and Guyer 1994). Agalychnis callidryas generally prefers quiet pools of water with overhanging vegetation as breeding sites (but see the paragraph below for other types of sites used); the pools may be permanent, seasonal but long-lasting, or temporary (Warkentin et al. 2001; Pyburn 1970). Males make aggressive calls, to deter other males from intruding onto their territories, and advertisement calls, to attract females (Gray and Rand 1997). The aggressive call sounds like a soft chuckle, and the advertisement call is a "chack" or "chack-chack", repeated at intervals of 8-60 seconds (Duellman 1967; Pyburn 1970; Duellman 2001). The dominant call frequency ranges from 1.5-2.5 kHz (Gray and Rand 1997). Calling begins at dusk and is most frequent in the evening, especially during rains, as males advertise for mates (Pyburn 1970). However, Agalychnis callidryas have also been reported to participate in a brief daybreak chorus of advertisement calls, at a much higher call rate than evening or morning advertisement choruses (Gray and Rand 1997; Duellman 2001). On dry nights, males call from higher perches in the tree canopy (Pyburn 1970). On wetter nights, or when the ponds are full, males begin calling from the ground and from small trees and bushes near the edges of ponds or backwaters (Pyburn 1970). Calls are generally made from horizontal perches on leaves or branches, although vertical perches on stems are occasionally used (Pyburn 1970). Males move about frequently while calling, changing both their positions and the direction of calling (Pyburn 1970). If precipitation or water level conditions are sufficient, males then descend to the breeding sites at the water's edge and continue calling (Savage 2002). Females descend from the canopy, approach selected calling males in a straight line, and allow amplexus to take place (Pyburn 1970). Descent generally occurs slowly hand-over-hand, but parachuting has also been observed in this species, both in the wild (Pyburn 1964), and in experimental trials (Roberts 1994).

Once amplexed, a female will carry the amplectant male into the water and remain there for about ten minutes (Pyburn 1970; D'Orgeix and Turner 1995). Pyburn (1970) carried out experiments showing that the purpose of this behavior is to allow the female to absorb water through her skin, into her bladder, in order to make the jelly mass surrounding the clutch of eggs (Pyburn 1970). The amplectant pair then moves up into the trees, as the female searches for an appropriate egg deposition site on vegetation overhanging the water (Pyburn 1970; D'Orgeix and Turner 1995). Eggs are generally deposited on either the upper side (Pyburn 1970) or lower side (Fouquette 1968) of a broad leaf, as high as 12 feet over the water (Fouquette 1968). Sturdy plants and trees at the water's edge are preferred, but oviposition may also take place on emergent vegetation (Pyburn 1970). Egg clutches may also occasionally be deposited on substrates such as branches or fence wire (Pyburn 1970). Very rarely eggs will be deposited on the ground (Pyburn 1970). This species has also been reported to sometimes make use of a wide variety of other types of sites, such as water-filled depressions made by human and pack animals, completely lacking overhanging vegetation; vines and small trees overhanging cavities of water in fallen trees; passionflower vines growing above a trickle of water from a pig pen; or by attaching eggs to the inside surface of water collection devices (McCranie et al. 2003). Occasionally clutches are found with leaves partially or completely folded over the eggs (Duellman 2001). If the deposition site is on top of a leaf, the parents may be protecting the clutch from sun and predators by folding the leaf edge over, where it adheres to the sticky jelly mass surrounding the eggs (Leenders 2001). Other leaf-breeding frogs (Phyllomedusa iheringi, P. hypochondrialis, and P. sauvagii) protect their clutches within folded leaves (Pyburn 1970). However, Duellman (2001) cautions that for Agalychnis callidryas, the occasional folding may be due to attachment of the eggs to particular types of leaves more prone to curling, as neither he nor others (e.g. Pyburn) have observed this species actively curling leaves during oviposition. Pyburn (1970), in fact, specifically mentions that Agalychnis callidryas lays its eggs on open leaves, in contrast to other phyllomedusine frogs.

Fertilization occurs immediately after egg deposition (Leenders 2001). Occasionally amplectant pairs are attacked by single males, as the intruder attempts to dislodge the first male or attach himself alongside or on the first male's back (Pyburn 1970; D'Orgeix and Turner 1995). This strategy sometimes succeeds, as molecular analysis has confirmed that multiple paternity does occur and that the second male can contribute significantly to clutch fertilization in the wild (D'Orgeix and Turner 1995). Clutches consist of about 40 green eggs, each surrounded by a clear jelly coat, with each egg having a diameter of about 3.7 mm at oviposition and 5.2 mm when mature enough to hatch (Pyburn 1963; Warkentin 2000a; Warkentin 2002). The entire clutch is itself surrounded in more jelly, which is sticky and serves to adhere the clutch to the substrate as well as to prevent desiccation (Pyburn 1963; Pyburn 1970; Warkentin et al. 2006). Yolk color is consistent within clutches, but varies between clutches, from cream to gold to turquoise to lime green (Warkentin et al. 2006); as the clutches get older, the yolk changes color from pale green to yellow (Duellman 2001). Females may lay multiple clutches (up to five) in a single night (Pyburn 1970). In between clutches, the female carries the amplectant male down into the water again as she rehydrates (D'Orgeix and Turner 1995). Ovulation is likely to occur twice during the reproductive season, given that gravid females containing ovulated eggs were also found to have equal numbers of immature half-size eggs still within the ovary, (Duellman 1963; Duellman 2001). This observation also fits with the second peak in mating sometimes observed near the end of the rainy season (Donnelly and Guyer 1994).

Since the egg clutches are attached to vegetation overhanging the water, hatching tadpoles generally fall into the water below as soon as they hatch (Pyburn 1963). Thus hatching involves a shift in habitat, from aerial to aquatic, with a concomitant change in the suite of potential predators and selection pressures (Warkentin 1995). Occasionally, hatching tadpoles may also fall onto the ground (Pyburn 1970). Tadpoles are able to live out of water for up to 20 hours (Valerio 1971). Those that fall onto the ground or remain stuck to the leaf may still survive if a later rain washes them into standing water, or if they are able to flip themselves into the water by thrashing about with their tails (Pyburn 1963; Pyburn 1970).

Agalychnis callidryas shows remarkable adaptive plasticity in the timing of hatching (Warkentin 1995). In undisturbed clutches, embryos develop essentially synchronously but hatch asynchronously, from six to ten days post-oviposition (Warkentin 1995; Warkentin 2005). Most undisturbed embryos will hatch at about seven days, with the study population in Panama hatching on average at about 6-7 days and the study population in Costa Rica hatching at about 7-8 days (Warkentin 2005). However, if disturbed, Agalychnis callidryas embryos can undergo early, relatively synchronous hatching, as early as 4 days in Panama (Warkentin 2000b) or 5 days in Costa Rica (Warkentin 1995). Accelerated hatching can occur in entire clutches or small groups of eggs within a clutch, in response to at least four different natural risks: snake attack (Warkentin 1995), wasp attack (Warkentin 2000b), fungus infestation (Warkentin et al. 2001), and flooding or submersion (Pyburn 1963; Warkentin 2002). Hypoxia from exposure to hypoxic gas mixtures also induces early hatching (Warkentin 2002). Fungal attack, flooding,and hypoxia result in more gradual early hatching than immediate threat of predation (Warkentin 2000b). The decision to hatch is a behavioral decision since highly energetic movement is required; embryos which remain motionless do not hatch, even though they have reached an appropriate developmental stage (Warkentin 1995; Warkentin et al. 2001). Mechanical disturbance is both sufficient and necessary to induce rapid, synchronous hatching, since simultaneous touching and jiggling of the eggs by poking forceps into the surrounding jelly results in rapid hatching (Warkentin 1995). However, Agalychnis callidryas embryos do not respond to all mechanical disturbances by rapid hatching (Warkentin 1995). Five-day-old embryos are insensitive to movements of the leaf substrate caused by simply touching the egg jelly coats, or collecting and transporting egg clutches, or even high winds and heavy rain (Warkentin 1995). The plasticity in timing and synchronization of hatching, particularly early hatching, is thus not simply a response to mechanical stimulus intensity (Warkentin 2000b), nor is it a response to visual or chemical cues from snake or wasp predators, or cues such as wetting of the clutch from precipitation (Warkentin 2005). Rather, experiments have shown that red-eyed tree frog embryos assess the temporal pattern of substrate-borne vibrations when making the decision to hatch (Warkentin 2005; Warkentin et al. 2006; Warkentin et al. 2007). Further, they do not hatch at the earliest possible moment following stimulation, but wait some seconds to minutes to evaluate the information before deciding to hatch (Warkentin et al. 2007).

A video clip is available on Karen Warkentin's website of Agalychnis callidryas embryos undergoing accelerated hatching resulting from snake attack. Developing Agalychnis callidryas embryos are commonly preyed on by snakes (e.g., Leptodeira septentrionalis) (Warkentin 1995). Snake attack induces embryos to hatch up to 30% early (as early as five days post-oviposition), rapidly (within seconds to minutes), and relatively synchronously, as the entire clutch hatches within a few minutes rather than over several days (Warkentin 1995). Embryos are able to distinguish between the potentially lethal event of a snake attack and the relatively non-dangerous event of a rainstorm by the duration and timing of resulting vibrations (Warkentin et al. 2006). When vibrational recordings produced by a snake attack or by a rainstorm are edited and played back, such that the timing of vibrations more closely resembles the other stimulus, the hatching response shifts in the expected direction: higher for more snakelike vibrations, lower for more rainlike vibrations (Warkentin et al. 2006). Tadpoles first enter the water on hatching, shifting their habitat from aerial to aquatic as they fall from from the overhanging leaves to which the egg clutch was attached (Pyburn 1963). If undisturbed by snakes, embryos tend to delay hatching until they are about seven days old, because more developed larvae are less vulnerable to aquatic predators such as shrimp and fish (Warkentin 1995; Warkentin 2005).

Developing Agalychnis callidryas embryos are also frequently vulnerable to predation by several different species of polybid wasps (Warkentin 2000b; Warkentin et al. 2006). Wasp predation, like snake predation, induces early and rapid hatching of Agalychnis callidryas embryos (Warkentin 2000b). Interestingly, the number of embryos hatching is different, depending on the predation scenario. Warkentin (2000b) reports that embryos hatch individually or in small groups (the ones grasped by a wasp, plus sometimes the nearest neighbors) in response to wasps, which consume a single embryo at a time, whereas whole clutches hatch in response to snakes, which eat entire clutches.

Fungal infection of Agalychnis callidryas egg clutches by a filamentous ascomycete (family Phaeosphaeriaceae) also induces early hatching (Warkentin et al. 2001). While eggs in non-infected clutches hatch in a spatially random order, hatching in infected clutches is non-random (Warkentin et al. 2001). In infected clutches, those eggs which are specifically in direct contact with fungal hyphae will hatch early if they are at least five days old (Warkentin et al. 2001). Adjacent empty eggs can form a "barrier" which the fungus does not typically cross, allowing the remaining embryos to continue developing within the clutch (Warkentin et al. 2001). In contrast to the immediate, synchronous hatching in response to snake or wasp attack, fungus-infected clutches that are at a sufficient stage of development hatch over a period of days, but at an average younger age than healthy clutches (Warkentin et al. 2001).

Agalychnis callidryas larvae have unusually large, elaborate external gills for anurans, consisting of a gill trunk on each side of the body, with one row of branched filaments (Pyburn 1963). The presence of large, elaborate gills is likely to indicate oxygen limitation throughout development, especially since the embryos of this species develop in large, closely packed eggs under quite warm conditions (Warkentin 2000a). Larvae show strikingly rapid external gill loss on hatching, with substantial gill regression occurring as fast as three minutes post-hatching (Warkentin 2000a). Newly hatched tadpoles respire via internal gills, as well as cutaneously (Warkentin 2000a). The increase in environmentally available oxygen upon leaving the egg has been experimentally demonstrated to be the primary stimulus for external gill loss in this species, with the process of hatching itself playing a secondary role in triggering gill loss (Warkentin 2000a). Since the presence of external gills creates drag in the water (Dudley et al. 1991), rapid gill loss at the time these tadpoles first enter the water may enhance swimming performance, which could enable survival on exposure to aquatic predators (Warkentin 2000a). In this species (and probably other anurans with external gills), the loss of external gills at hatching appears to be mediated by prostaglandins of the E family, or PGEs (Warkentin and Wassersug 2001). In contrast, the loss of internal gills is mediated by thyroid hormone at the time of metamorphosis (Shi 2000).

Tadpoles of Red-Eyed Tree Frogs suspend themselves vertically in the water column to feed, with heads near the water surface (Savage 2002). They are primarily water-column filter feeders (Vonesh and Warkentin 2006). Gray and Nishikawa (1995) reported that in the laboratory, adult Agalychnis callidryas preferred crickets and moths as prey. To our knowledge, no field study has yet been published on the diet of adults of this species.

Eggs of Agalychnis callidryas are vulnerable to predation by arboreal snakes (such as the cat-eyed snake, Leptodeira septentrionalis), several species of polybid wasps (particularly Polybia rejecta), monkeys, and fly larvae (Hirtodrosophila batracida) (Warkentin 1995; Warkentin 2000b; Warkentin et al. 2001; Warkentin et al. 2006). In addition, eggs are susceptible to mortality from fungal infection by a filamentous ascomycete (family Phaeosphaeriaceae) (Warkentin et al. 2001). Tadpoles may be preyed on by aquatic predators such as shrimp (Macrobrachium americanum) and fish (Brachyraphis rhabdophora) (Warkentin 1995), as well as giant water bugs foraging below the water surface (Belostoma sp.) (Vonesh and Warkentin 2006). Newly metamorphosed froglets may be consumed by aquatic spiders foraging at or above the water surface (Thaumasia sp.) (Vonesh and Warkentin 2006). Predators on juvenile and adult Red-Eyed Tree Frogs include snakes (Warkentin 1995), as well as birds and bats (Leenders 2001).

Although phyllomedusine (leaf-breeding) frogs do not have the same sort of highly toxic compounds as other brightly colored frogs such as the dendrobatids, their skin does contain high levels of biologically active peptides (Mignogna et al. 1997). The skin of Agalychnis callidryas has been shown to contain five different families of biologically active peptides: tachykinins, bradykinins, caerulein, opioid peptides (dermorphin and [Hyp6]dermorphin), and sauvagine (Mignogna et al. 1997).

In addition to larvae being able to sense vibrations, it has now been shown that adult male Red-Eyed Tree Frogs also use vibrational signaling, in male-male aggressive interactions (Caldwell et al., 2010). The authors point out that substrate vibrations may be far more important in communication by arboreal vertebrates than had previously been realized. During contests over females, male Red-Eyed Tree Frogs emit "chuckle" calls and perform tremulation displays by shaking a branch rapidly with their hind legs. Tremulations only occurred when two males were within 2 m of each other or were on the same plant, and were often immediately followed by wrestling. To test whether the vibrations were important, various combinations of displays were made using a robotic frog on a wooden frame near a branch occupied by a live frog, plus a shaker on the branch where the live frog sat. In some tests, the robotic frog moved up and down, for a visual display, but no branch shaking was done; in others, the branch was vibrated but the robotic frog did not move; and in some tests, both visual and vibrational displays were made. The live males responded aggressively to all displays, both visual and vibrational, but only shook their own branch when vibrations were felt from the shaker (Caldwell et al. 2010).

Trends and Threats
This species can tolerate some habitat modification, such as in areas where selective logging has been done, but is thought not to be able to live where the forest has been heavily degraded (Stuart et al. 2008). However, McCranie et al. (2003) report finding Agalychnis callidryas eggs and tadpoles in a wide variety of breeding sites in Honduras, including some in heavily deforested areas (see Life History section above for a description).

All five Agalychnis species (A. annae, A. callidryas, A. moreletii, A. saltator, and A. spurrelli) have just received CITES protection, under Appendix II (as of March 21, 2010). Within the past decade the U.S. alone has imported 221,960 Agalychnis frogs, according to the Species Survival Network (SSN).

Relation to Humans
Red-Eyed Tree Frogs are popular in the pet trade, and can be bred in captivity under suitable conditions.

Comments

The diploid number of chromosomes is 26 for Agalychnis callidryas (Duellman and Cole 1965).

A Spanish-language species account can be found at the website of Instituto Nacional de Biodiversidad (INBio).

References
 

Caldwell, M. S., Johnston, G. R., McDaniel, J. G., and Warkentin, K. M. (2010). ''Vibrational signaling in the agonistic interactions of red-eyed treefrogs.'' Current Biology, doi:10.1016/j.cub.2010.03.069.  

Cope, E. D. (1862). ''Catalogue of the reptiles obtained during the explorations of the Parana, Paraguay, Vermejo, and Uruguay Rivers, by Captain Thos. J. Page, U.S.N., and of those procured by Lieut. N. Michler, U.S. Top. Eng., Commander of the expedition conducting the survey.'' Proceedings of the Academy of Natural Sciences of Philadelphia, 14, 346-359.  

D'Orgeix, C. A., and Turner, B. J. (1995). ''Multiple paternity in the red-eyed treefrog Agalychnis callidryas (Cope).'' Molecular Ecology, 4, 505-508.  

Donnelly, M., and Guyer, C. (1994). ''Patterns of reproduction and habitat use in an assemblage of Neotropical hylid frogs.'' Oecologia, 98, 291-302.  

Dudley, R., King, V. A., and Wassersug, R. J. (1991). ''The implications of shape and metamorphosis for drag forces on a generalized pond tadpole (Rana catesbeiana).'' Copeia, 1991, 252-257.  

Duellman, W. E. (1967). ''Courtship isolating mechanisms in Costa Rican hylid frogs.'' Herpetologica, 23(3), 169-183.  

Duellman, W. E. (1963). ''Amphibians and reptiles of the rainforest of southern El Petén, Guatemala.'' University of Kansas Publications, Museum of Natural History, 15, 205-249.  

Duellman, W. E. (2001). The Hylid Frogs of Middle America. Society for the Study of Amphibians and Reptiles, Ithaca, New York.  

Duellman, W. E., and Cole, C. J. (1965). ''Studies of chromosomes of some anuran amphibians.'' Systematic Zoology, 14(2), 139-143.  

Fouquette, M. J., Jr. (1968). ''Some hylid frogs of the Canal Zone, with special reference to call structure.'' Caribbean Journal of Science, 6(3-4), 167-172.  

Gray, L. A., and Nishikawa, K.C. (1995). ''Feeding kinematics of phyllomedusine tree frogs.'' The Journal of Experimental Biology, 198, 457-463.  

Gray, L. A., and Rand, A. S. (1997). ''A daybreak chorus in the frog, Agalychnis callidryas.'' Journal of Herpetology, 31(3), 440-441.  

Leenders, T. (2001). A Guide to Amphibians And Reptiles of Costa Rica. Zona Tropical, Miami.  

McCranie, J. R., Wilson, L. D., and Townsend, J. H. (2003). ''Agalychnis callidryas (Red-eyed Treefrog). Reproduction.'' Herpetological Review, 34(1), 43.  

Mignogna, G., Severina, C., Erspamer, G. F., Siciliano, R., Kreil, G., and Barra, D. (1997). ''Tachykinins and other biologically active peptides from the skin of the Costa Rican phyllomedusid frog Agalychnis callidryas.'' Peptides, 18(3), 367-372.  

Pyburn, W. F. (1963). ''Observations on the life history of the treefrog, Phyllomedusa callidryas (Cope).'' Texas Journal of Science, 15, 155-170.  

Pyburn, W. F. (1964). ''Breeding behavior of the leaf-frog, Phyllomedusa callidryas, in southern Veracruz.'' The American Philosophical Society Yearbook, 1964, 291-294.  

Pyburn, W. F. (1970). ''Breeding behavior of the leaf-frogs Phyllomedusa callidryas and Phyllomedusa dacnicolor in Mexico.'' Copeia, 2, 209-218.  

Roberts, W. E. (1994). ''Explosive breeding aggregations and parachuting in a neotropical frog, Agalychnis saltator (Hylidae).'' Journal of Herpetology, 28(2), 193-199.  

Ruiz-Carranza, P.M., Ardila-Robayo, M.C., and Lynch, J.D (1996). ''Lista actualizada de la fauna de Amphibia de Colombia.'' Revista de la Academia Colombiana de Ciencias Exactas, Físicas y Naturales, 20(77), 365-415.  

Savage, J. M. (2002). The Amphibians and Reptiles of Costa Rica. University of Chicago Press, Chicago and London.  

Shi, Y-B. (2000). Amphibian Metamorphosis: From Morphology to Molecular Biology. Wiley-Liss, Inc., New York.  

Starrett, P. (1960). ''Descriptions of tadpoles of Middle American frogs.'' Miscellaneous Publications Museum of Zoology, University of Michigan, 110, 5-37.  

Stuart, S., Hoffmann, M., Chanson, J., Cox, N., Berridge, R., Ramani, P., and Young, B. (eds.) (2008). Threatened Amphibians of the World. Lynx Edicions, IUCN, and Conservation International, Barcelona, Spain; Gland, Switzerland; and Arlington, Virginia, USA.  

Valerio, C.E. (1971). ''Ability of some tropical tadpoles to survive without water.'' Copeia, 1971(2), 364-365.  

Vonesh, J. R., and Warkentin, K. M. (2006). ''Opposite shifts in size at metamorphosis in response to larval and metamorph predators.'' Ecology, 87(3), 556-562.  

Warkentin, K. M. (1995). ''Adaptive plasticity in hatching age: A response to predation risk trade-offs.'' Proceedings of the National Academy of Sciences, 92, 3507-3510.  

Warkentin, K. M. (2000). ''Environmental and developmental effects on external gill loss in the Red-Eyed Tree Frog, Agalychnis callidryas.'' Physiological and Biochemical Zoology, 73(5), 557-565.  

Warkentin, K. M. (2000). ''Wasp predation and wasp-induced hatching of red-eyed treefrog eggs.'' Animal Behavior, 60, 503-510.  

Warkentin, K. M. (2002). ''Hatching timing, O2 availability and external gill regression in the treefrog, Agalychnis callidryas.'' Physiological and Biochemical Zoology, 75, 155-164.  

Warkentin, K. M. (2005). ''How do embryos assess risk? Vibrational cues in predator-induced hatching of red-eyed treefrogs.'' Animal Behavior, 70, 59-71.  

Warkentin, K. M., Buckley, C. R., and Metcalf, K. A. (2006). ''Development of red-eyed treefrog eggs affects efficiency and choices of egg-foraging wasps.'' Animal Behavior, 71, 417-425.  

Warkentin, K. M., Caldwell, M. S., Siok, T. D., D'Amato, A. T., and McDaniel, J. G. (2007). ''Flexible information sampling in vibrational assessment of predation risk by red-eyed treefrog embryos.'' The Journal of Experimental Biology, 210, 614-619.  

Warkentin, K. M., Caldwell, M. S., and McDaniel, J. G. (2006). ''Temporal pattern cues in vibrational risk assessment by embryos of the red-eyed treefrog, Agalychnis callidryas.'' Journal of Experimental Biology, 209, 1376-1384.  

Warkentin, K. M., Currie, C. R., and Rehner, S. A. (2001). ''Egg-killing fungus induces early hatching of red-eyed treefrog eggs.'' Ecology, 82(10), 2860-2869.  

Warkentin, K. M., and Wassersug, R. J. (2001). ''Do prostaglandins regulate external gill regression in anurans?'' Journal of Experimental Zoology, 289, 366-373.

AmphibiaWeb's policy on data use.