View distribution map using BerkeleyMapper.

	Conservation Status (definitions)
IUCN (Red List) Status	Critically Endangered (CR) See IUCN account.
CITES	No CITES Listing
Other International Status	None
National Status	None
Regional Status	None

Description
Dorsal coloration is mottled light and dark, with ground coloration ranging from lime-green to yellow-orange and surface markings in black (from dots to uniform coloring). The ventral surface ranges from yellow-green to orange with the same type of black surface markings. The throat and venter can appear bright scarlet, and the groin can appear blue-green or green. The snout is short, and in profile acute or rounded. The pupil is horizontally elliptical, there is no tympanum, and there are no distinct glandular ridges. Paired elongate vocal slits run parallel and lateral to the tongue; males have an internal vocal sac. Males have webbing between the first and second finger, and the first finger also bears a nuptial pad. Toes are webbed except for IV, and there is no tarsal fold (all descriptions from Savage 1972). Snout-vent length of adult males ranges from 25 to 41 mm, and females are 33 to 60 mm in length (Savage 1972; Crump 1988). Both bufadienolide and tetrodotoxin skin toxins are present in wild A. varius (Daly et al. 1997).

Distribution and Habitat

Country distribution from AmphibiaWeb's database: Colombia, Costa Rica, Panama

View distribution map using BerkeleyMapper.

A. varius is found from western Panama to the Canal Zone, through Central Talamanca and Chiriqui, and on both the Atlantic and Pacific slopes of the Cordilleras de Tilaran in Costa Rica (Savage 1972). It has been found at elevations ranging from 16 to 2000 meters, in life zones Holdridge (1967) characterizes as Tropical Lowland, Premontane, and Lower Montane. This habitat includes both dry and wet forest following small stream courses, where the seasonal weather pattern is divided into a wet (late May to mid-November) and dry season (Savage 1972; Crump 1988).

Life History, Abundance, Activity, and Special Behaviors
The Harlequin Frog is diurnal. It is usually found sitting on rocks and logs near a stream or in rock crevices during the day, and within crevices or low vegetation at night (Crump and Pounds 1985). It is a poor swimmer that rarely enters streams, and appears to depend on wet surfaces in the splash zone for moisture (Pounds and Crump 1994). It feeds on small arthropods (generally no larger than 7 mm in size) which are more abundant during the dry season (Crump 1988). It is slow-moving, sedentary, and less wary than other local frogs; these characteristics appear to be adaptive behaviors related to aposematism (Crump and Pounds 1985). Some individuals are long-lived and have been recaptured over 3 consecutive years (Crump and Pounds 1985). Over the year, Harlequin Frogs tend to aggregate in waterfall splash zones with the progression of the dry season (Pounds and Crump 1987; Crump and Pounds 1989). The only known predator or parasite is the sarcophagid fly, Notochaeta bufonivora, (Crump and Pounds 1985). The fly larviposits on the posterior surface of the frog's thigh, and the larvae subsequently burrow and feed within it, killing the frog within a matter of days (Pounds and Crump 1987).

Males begin calling and defending territories at the onset of the wet season, and pairing begins 2 to 5 months later, from mid-August through early December (Pounds and Crump 1987; Crump 1988). The sex ratio is male-biased, and males often enter into amplexus early and remain with the female for as long as 32 days (Crump 1988). Females deposit eggs in the water from the end of the wet season to the early dry season, October to December (Pounds and Crump 1987).

Crump (1988) has documented the aggressive behavior of A. varius, which occurs within and between the sexes and is characterized by forelimb and foot wagging, chasing, pouncing, and squashing. Male-male antagonism is generally longer than female-female aggression and involves chirp-like advertisement calls and wrestling behavior. Male aggression is heightened both during the breeding season and in high density situations (Crump 1988). Female-female aggression lacks wrestling and calling, is generally less intense and less frequent, and seems to be related to the defense of foraging and/or shelter sites (Crump 1988). Intersexual aggression is almost always female to male; females will chase, pounce on, and squash males who have invaded their territory, and they will attack males whom they have dislodged from amplexus. The only hostile behavior of males towards females is holding the amplectant position despite a female's efforts to disengage. In a series of invasion experiments, Crump (1988) demonstrated that territorial residents have the advantage in combat and higher success despite size differences.

Trends and Threats
Declines of A. variushave been documented in the Cordillera de Tilaran, near Monteverde, Costa Rica. Census studies conducted between 1982 and 1983 reported an average of 751 adults along a 200 m transect (Crump and Pounds 1985), a number which appeared stable up to the late 1980's. Surveys conducted between 1990 and 1992, however, did not detect a single frog (Pounds and Crump 1994). It remains absent from the Monteverde area.

Several causal explanations for this disappearance have been proposed, most involving climatic factors in a direct or synergistic manner. Pounds and Crump (1994) examined weather data and noted the correlation between the adults' disappearance and the 1982-1983 and the 1986-1987 El Niņo/Southern Oscillation events. This led them to hypothesize several climate-based causes of the decline: moisture stress, temperature stress, a climate-linked epidemic, and a climate-linked contaminant pulse. Recent studies have linked this and other declines to more subtle but cumulative weather changes caused by global warming. Pounds et al. (1999) conducted a careful analysis of precipitation, air temperature, sea surface temperature, and stream flow patterns as they related to tropical anuran, avian, and anoline lizard communities. They concluded that declines like that of the Harlequin Frog were not simply an effect of the El Niņo/Southern Oscillation, but rather of a long-term warming trend which crossed a threshold in late 1980's and precipitated a broad tropical anuran decline. This decline was accompanied by a decline in anoline lizard communities and significant restructuring of tropical avian communities.

Of the proposed causes, the climate-linked epidemic hypothesis (Pounds and Crump 1994) appears particularly important in the decline of A. varius in light of the discovery of the parasitic fly N. bufonivora in 1982 (Crump and Pounds 1985; Pounds and Crump 1987). The fly seems to cause density-dependent mortality (Crump and Pounds 1985), as clumping of frogs during the dry season of 1982 (an El Niņo year) was correlated with increased fly parasitism. Parasitism is particularly intense near waterfall spray zones, where Harlequin Frogs aggregate when extreme dry conditions threaten their water balance (Pounds and Crump 1987).

An additional epidemic and density-dependent threat to A. varius has appeared in the form of a novel pathogen. In 1997, Berger et al (1998) documented three deceased A. varius specimens from Fortuna, Panama which had been killed by a chytridiomycete fungus. This fungal pathogen invades keratinized body surfaces and kills postmetamorphic anurans within several weeks. Tests show that the chytrid fungus can be spread by skin scrapings from other individuals, and may impair cutaneous respiration and osmoregulation when it attacks the hypervascularized pelvic patch (Berger et al 1998). It has been documented as fatal in adults from 18 other species, both in South America and Australia, and has been implicated as a widespread cause of anuran decline (Berger et al 1998).

Possible reasons for amphibian decline

Subtle changes to necessary specialized habitat
Predators (natural or introduced)
Disease
Climate change, increased UVB or increased sensitivity to it, etc.

Comments
This once common species was last seen in 1996 (Bolaņos pers com) and thought to be extinct. A new population was recently discovered in the Rainmaker project in 2003.

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

References
 

Berger, L., Speare, R., Daszak, P., Green, D. E., Cunningham, A. A., Goggin, C. L., Slocombe, R., Ragan, M. A., Hyatt, A. D., McDonald, K. R., Hines, H. B., Lips, K. R., Marantelli, G., and Parkes, H. (1998). "Chytridiomycosis causes amphibian mortality associated with population declines in the rain forests of Australia and Central America." Proceedings of the National Academy of Sciences of the United States of America, 95(15), 9031-9036.  

Crump, M. L., and Pounds, J. A. (1985). ''Lethal parasitism of an aposematic anuran (Atelopus varius) by Notochaeta bufonivora (Diptera: Sarcophagidae).'' Journal of Parasitology, 71(5), 588-591.  

Crump, M.L. (1988). ''Aggression in Harlequin Frogs: male-male competition and a possible conflict of interest between the sexes.'' Animal Behaviour, 36(4), 1064-1077.  

Daly, J.W., Padgett, W.L., Saunders, R.L., and Cover, J.F., Jr. (1997). ''Absence of tetrodotoxins in a captive-raised riparian frog, Atelopus varius.'' Toxicon, 35(5), 705-709.  

Holdridge, L. R. (1967). Life Zone Ecology. Tropical Science Center, San Jose, Costa Rica.  

Pounds, J. A., Fogden, M. P. L., and Campbell, J. H. (1999). ''Biological response to climate change on a tropical mountain.'' Nature, 398(6728), 611-615.  

Pounds, J. A., and Crump, M. L. (1987). ''Harlequin Frogs along a tropical montane stream: Aggregation and the risk of predation by frog-eating flies.'' Biotropica, 19(4), 306-309.  

Pounds, J. A., and Crump, M. L. (1989). ''Temporal variation in the dispersion of a tropical anuran.'' Copeia, 1989(1), 209-211.  

Pounds, J. A., and Crump, M. L. (1994). ''Amphibian declines and climate disturbance: The case of the Golden Toad and the Harlequin Frog.'' Conservation Biology, 8(1), 72-85.  

Savage, J.M. (1972). ''The harlequin frogs, genus Atelopus, of Costa Rica and western Panama.'' Herpetologica, 28(2), 77-94.

AmphibiaWeb's policy on data use.