Endemic infection of the amphibian chytrid fungus in a frog community post-decline

Abstract

The chytrid fungus Batrachochytrium dendrobatidis has been implicated in the decline and extinction of numerous frog species worldwide. In Queensland, Australia, it has been proposed as the cause of the decline or apparent extinction of at least 14 high-elevation rainforest frog species. One of these, Taudactylus eungellensis, disappeared from rainforest streams in Eungella National Park in 1985-1986, but a few remnant populations were subsequently discovered. Here, we report the analysis of B. dendrobatidis infections in toe tips of T. eungellensis and sympatric species collected in a mark-recapture study between 1994 and 1998. This longitudinal study of the fungus in individually marked frogs sheds new light on the effect of this threatening infectious process in field, as distinct from laboratory, conditions. We found a seasonal peak of infection in the cooler months, with no evidence of interannual variation. The overall prevalence of infection was 18% in T. eungellensis and 28% in Litoria wilcoxii/jungguy, a sympatric frog that appeared not to decline in 1985-1986. No infection was found in any of the other sympatric species. Most importantly, we found no consistent evidence of lower survival in T. eungellensis that were infected at the time of first capture, compared with uninfected individuals. These results refute the hypothesis that remnant populations of T. eungellensis recovered after a B. dendrobatidis epidemic because the pathogen had disappeared. They show that populations of T. eungellensis now persist with stable, endemic infections of B. dendrobatidis.

Figure 1. Location of Study Sites

1, Mount David Creek; 2, Mount William Creek; 3, Dooloomai Falls; 4, Rawson Creek; 5, Picnic Ground Creek; and 6, Tree Fern Creek.

Figure 2. Seasonal Patterns of Prevalence of B. dendrobatidis in T. eungellensis

Solid circles show the observed prevalence, with binomial 95% confidence limits, in frogs pooled over years and age/sex class. The dashed line shows the prevalence predicted from the best-fitting logistic model. Numbers in brackets above each error bar are the sample sizes.

Figure 3. Estimated Quarterly Survival of Age/Sex Classes of Taudactylus eungellensis at Three Sites

Survival for uninfected frogs is shown with circles, and survival for infected frogs is shown with squares; 95% confidence limits around each point are also shown. Where there is no survival estimate for infected frogs, there were insufficient data to estimate the parameter. Sample sizes are given in Table 2.