INTRODUCTION

There is world-wide concern about declines of some amphibian species (Blaustein and Wake 1990; Wake 1991; Grump et al. 1992; Blaustein et al. 1994a). Amphibian declines may be a consequence of human-related impacts, including chemical pollution, acid precipitation, ozone depletion, habitat destruction and introductions of exotic species (Hayes and Jennings 1986; Blaustein and Wake 1990; Wake 1991; Wissinger and Whiteman 1992; Blaustein et al. 1994a).

There is particular concern over amphibian declines in areas that have been relatively undisturbed by human activity, such as high-elevation locations in western North America (Blaustein and Wake 1990) and U.S. national parks (Bradford 1989; Bradford et al. 1993). Fish introductions have been implicated in altering amphibian distribution and abundance in high-elevation lakes (Bradford 1989; Bradford et al. 1993; Fellers and Drost 1993; Blaustein et al. 1994a, 1994b). However, fish are not indigenous to many high-elevation lakes in the western U.S. Bahls (1992) reported that 95% of mountain lakes in the western U.S. may have been naturally fishless. Presently, nearly 60% of all high-elevation lakes and about 95% of the larger, deeper lakes now support fish (Bahls 1992).

Although ambystomatid salamander larvae are important native aquatic predators in high- elevation lakes in western North America (Dodson 1970, 1974; Dodson and Dodson 1971; Sprules 1972; Taylor 1983), only a few studies have assessed the impact of fish on their abundance and behavior. Introduced trout can reduce abundance and cause a shift toward nocturnal behavior in the northwestern salamander (Ambystoma gracile) (Sprules 1972; Taylor 1983). Studies conducted on ambystomatid species from the eastern U.S. have shown that fish can reduce or eliminate ambystomatid larvae from lakes and streams (Petranka 1983; Semlitsch 1988; Sih et al. 1992), inhibit growth (Semlitsch 1987; Figiel and Semlitsch 1990) and reduce survival (Semlitsch 1987; Sih et al. 1988) of larvae in artificial ponds. Fish may also cause larvae to shift toward nocturnal activity (Sih et al. 1992), restrict activity (Semlitsch 1987; Stangel and Semlitsch 1987; Figiel and Semlitsch 1990), and increase refuge use (Sih et al. 1988, 1992).

Salamander distribution and abundance is influenced by abiotic factors and biotic factors other than fish. Abiotic factors that can influence salamander distributions include elevation (Snyder 1956; Howard and Wallace 1985; Leonard et al. 1993), lake area and depth (Kezer and Farner 1955; Sprules 1974a, 1974b), water temperature (Snyder 1956; Anderson 1968), and conditions in the terrestrial habitat (Sprules 1974a, 1974b). Availability of suitable food resources, such as benthic macroinvertebrates and zooplankton, especially cladocerans and copepods (Anderson 1968; Dodson 1970; Dodson and Dodson 1971; Licht 1975; Freda 1983), also may influence larval salamander abundance and distribution.

The objectives of this research were to: 1) determine relationships between lake chemical and physical characteristics and larval salamander (A. macrodactylum) density in each of three lake categories-fishless lakes, lakes with non-reproducing trout, and lakes with reproducing trout, 2) compare the relationships between lake categories, and 3) determine differences in percent larvae hidden in the bottom substrate between lake categories. The research was conducted in North Cascades National Park Service Complex, Washington, USA. Although A. macrodactylum is widespread in the Pacific Northwest (Nussbaum et al. 1983; Leonard et al. 1993), very little is known about its distribution within high-elevation areas, variation in larval abundance among lakes, and natural and human factors influencing salamander distribution and abundance.