Biological soil crusts, a community of cyanobacteria, lichens, and mosses that live on the soil surface, occur in deserts throughout the world. They are a critical component of desert ecosystems, as they are important contributors to soil fertility and stability. Future climate scenarios predict alteration of the timing and amount of precipitation in desert environments. Because biological soil crust organisms are only metabolically active when wet, and as soil surfaces dry quickly in deserts during late spring, summer, and early fall, the amount and timing of precipitation is likely to have significant impacts on the physiological functioning of these communities. Using the three dominant soil crust types found in the western United States, we applied three levels of precipitation frequency (50% below-average, average, and 50% above-average) while maintaining average precipitation amount (therefore changing both timing and size of applied events). We measured the impact of these treatments on photosynthetic performance (as indicated by dark-adapted quantum yield and chlorophyll a concentrations), nitrogenase activity, and the ability of these organisms to maintain concentrations of radiation-protective pigments (scytonemin, beta-carotene, echinenone, xanthophylls, and canthaxanthin). Increased precipitation frequency produced little response after 2.5 months exposure during spring (1 April-15 June) or summer (15 June-31 August). In contrast, most of the above variables had a large, negative response after exposure to increased precipitation frequency for 6 months spring-fall (1 April-31 October) treatment. The crusts dominated by the soil lichen Collema, being dark and protruding above the surface, dried the most rapidly, followed by the dark surface cyanobacterial crusts (Nostoc-Scytonema-Microcoleus), and then by the light cyanobacterial crusts (Microcoleus). This order reflected the magnitude of the observed response: crusts dominated by the lichen Collema showed the largest decline in quantum yield, chlorophyll a, and protective pigments; crusts dominated by Nostoc-Scytonema-Microcoleus showed an intermediate decline in these variables; and the crusts dominated by Microcoleus showed the least negative response. Most previous studies of crust response to radiation stress have been short-term laboratory studies, where organisms were watered and kept under moderate temperatures. Such conditions would give crust organisms access to ample carbon to respond to imposed stresses (e.g., production of UV-protective pigments, replacement of degraded chlorophyll). In contrast, our longer-term study showed that under field conditions of high air temperatures and frequent, small precipitation events, crust organisms appear unable to produce protective pigments in response to radiation stress, as they likely dried more quickly than when they received larger, less frequent events. Reduced activity time likely resulted in less carbon available to produce or repair chlorophyll a and/or protective pigments. Our findings may partially explain the global observation that soil lichen cover and richness declines as the frequency of summer rainfall increases. ?? Springer-Verlag 2003.