We investigated the importance of snowmelt as an organizing factor for epilithic microalgae in a high-altitude montane stream during 3 yr (1995)(1996)(1997) of differing melt characteristics. Changes in algal biomass and taxonomic structure in two stream reaches that differed in hydrologic characteristics were assessed relative to variation in stream-water nutrient content, temperature, surface-water/groundwater exchange, and algivorous macroinvertebrate assemblages. Melt-induced increases in stream discharge were large and abrupt in 1995, nearly undetectable in 1996, and protracted and of intermediate magnitude in 1997. In 1995, algal and grazer densities were significantly reduced by melt discharge. Postmelt grazer recovery was slow, and the percentage of live cells in the diatom assemblage increased abruptly; algal biomass initially increased and subsequently varied with stream-water N : P ratio. In 1996, snowmelt produced no evident proximate effects. Initial high grazer densities declined throughout the summer. In midsummer, an N : P ratio of 140 in the upstream reach corresponded to a more than fourfold increase in algal biovolume and a shift to dominance by large diatoms. Downstream, a lower N : P peak induced no algal response. In 1997, grazer densities declined during protracted melt runoff and increased sharply during melt recession, concurrent with decreases in live diatom percentage and algal biovolume. N : P declined in 1997 from ϳ16 : 1 to values indicative of N limitation and correlated with decreases in algal biovolume. Our results show that effects of snowmelt in montane streams reflect both initial melt-induced mortality that sets initial conditions for succession and melt-induced aquifer recharge that controls nutrient supply in the months following peak melt discharge. The influence of these two components extends beyond snowmelt recession into summer base flow, suggesting that variation in melt characteristics generates interannual differences in the functioning of these systems.Patterns of temporal and spatial variation in communities of primary producers are underlain by a template of physical, chemical, and biological factors. These factors act individually or interactively to dictate taxonomic structure, biomass, and community dynamics. This ''habitat template'' shifts within years in response to seasonal changes in the physical/ chemical environment (Duncan and Blinn 1989) and the phenology or timing of visitation by herbivores (Jacoby 1987). Shifts can also be induced by stochastic episodes of physical disturbance (Zimmerman et al. 1996) that modify the template on multiple levels or by pathogen outbreaks that alter the biological component (Peterson et al. 1993). Variation in global weather, such as the El Niño-Southern Oscillation phenomenon, induces differences among years in seasonal precipitation and temperature and in the frequency and magnitude of flood disturbance (Molles and Dahm 1990). Thus, change in a habitat template within a given