In the late 1970s, Rana muscosa (mountain yellow-legged frog) was common in the Tableland area of Sequoia National Park, California where it was possible to find hundreds of tadpoles and adults around many of the ponds and lakes. Surveys in 1993-1995 demonstrated that R. muscosa was absent from more than half of all suitable habitat within the park, including the Tableland area. At that same time, R. muscosa was still common at Sixty Lake Basin, Kings Canyon National Park, 30 km to the northeast. To evaluate the potential causes for the extirpation, we repatriated R. muscosa eggs, tadpoles, subadults, and adult frogs from Sixty Lake Basin to four sites in the Tableland area in 1994 and 1995. We subsequently surveyed each release site and the surrounding area 2 - 3 times per week in 1994-1995, and intermittently in 1996-1997, to monitor the survival of all life history stages, and to detect dispersal of adults and subadults. We also monitored predation, water quality, weather, and water temperature. Our techniques for capturing, holding, transporting, and releasing R. muscosa were refined during the study, and during 1995 resulted in high initial survival rates of all life history stages. Adult frogs were anaesthetized, weighed, measured, tagged, and held in plastic boxes with wet paper towels. Tadpoles were collected and held in fiberglass screen cages set in the water at the edge of a pond. This resulted in relatively natural conditions with less crowding and good water circulation. Frogs, tadpoles, and eggs were placed in Ziploc bags for transport to the Tableland by helicopter. Short-term survival of tadpoles, subadults, and adults was high at all four release sites, tadpoles reached metamorphosis, and adult frogs were still present. However, we detected no evidence of reproduction at three sites (e.g., no new eggs or small tadpoles) and nearly all life history stages disappeared within 12 months. At the fourth site, there was limited reproduction, but it was insufficient to maintain a population. It appears that the causal factors for the demise of R. muscosa in the Tableland during the 1970s were still operating in the 1990s or that a new limiting factor has developed. Dispersal, weather, water quality, and predation do not appear to be causative agents; since fish have never been present in the portions of the watershed where we were working, they were not a factor. Observations and data are consistent with the hypotheses that chytridiomycosis, caused by the chytrid fungus Batrachochytrium dendrobatidis, and/or exposure to airborne pesticides caused both declines. However, at the time of our study, chytridiomycosis had not been described and the potentially significant role of contaminants was largely undocumented.