We performed two experiments to examine how temperature and nutrients interact to control dinitrogen (N2) fixation, chlorophyll a (Chl a) biomass, and community composition of periphyton in subalpine oligotrophic streams in the Sawtooth Mountains of Idaho. We grew periphyton on nutrient-diffusing substrata (NDS) in a cold lake inlet (7uC) and a warm lake outlet (18uC). We then switched substrata between the two stream sites to test the effect of incubation temperature on N2-fixation rates. Periphyton on substrata grown at both sites exhibited greater N2-fixation rates when incubated in the warm outlet, which indicates physiologic temperature control. Periphyton on P-enriched NDS grown in the warm outlet had the greatest N2-fixation rates, largest Chl a biomass, and largest percentage of N2-fixing taxa of any treatment, which indicates that temperature and P interact to influence the community. In the second experiment, colonized rocks and uncolonized NDS were placed in cold (13uC) and warm (18uC) mesocosms. Within 2 days, warm temperature stimulated N2 fixation by the rock periphyton community two times above cold temperatures, which indicates physiologic temperature control. After 45 days, warm temperatures and P enrichment led to Anabaena sp. in the periphyton community and the greatest rates of N2 fixation observed in the experiment, which also indicates temperature and nutrient control at the community level. This study indicates that N2 fixation and periphyton community composition in oligotrophic streams are controlled by both temperature and P supply, with temperature modulating the response to P. A dominant paradigm of aquatic ecology is that phosphorus (P) most often limits algal growth and production in freshwater ecosystems (Schindler 1977). This paradigm is rooted in the hypothesis that dinitrogen (N2) fixation by cyanobacteria should contribute sufficient nitrogen (N) to aquatic ecosystems, which, thus, perpetuates P limitation (Redfield 1958; Schindler 1977). However, other studies have indicated that growth of stream algal communities is as frequently limited by N as by P, and many streams are colimited by N and P (Francoeur 2001), which indicates that some factors must be limiting N2 fixation and maintaining N limitation in some streams. N2-fixation rates in streams are potentially controlled by availability of macronutrients and micronutrients. N additions have been widely demonstrated to suppress N2 fixation by cyanobacteria because N2 fixation is a costly process that will not occur when cyanobacteria have access to environmental N (Howarth et al. 1988). Stimulation of N2 fixation by P additions has been demonstrated in many marine and lake systems (Howarth et al. 1988). Cyanobacterial N2 fixation can also be limited by trace metals such as iron or molybdenum (Wurtsbaugh and Horne 1983; Howarth et al. 1988). No study has directly examined the effects of nutrients on periphyton N2-fixation rates in streams but several have identified responses of cyanobacteria abundance to nutrient availability. For example, cyanobacteria or diatoms with N2-fixing endosymbionts appear to be at a competitive disadvantage compared with other algae, and, thus, have lower abundances at high stream N concentrations in both experimental enrichments (Peterson and Grimm 1992) and under natural conditions (Henry and Fisher 2003). Temperature can control N2-fixation rates in a variety of settings by affecting enzymatic activity and, in turn, the physiologic ability of cells to fix N2 (DeNicola 1996). Reuter et al. (1983) found that N2-fixation by benthic periphyton from an oligotrophic lake increased linearly between 5uC and 25uC in short-term laboratory measurements. More recently, Staal et al. (2003) showed that N2fixation rates of four marine cyanobacteria increased linearly as temperatures increased from 10uC to 35uC in laboratory cultures. Temperature also has profound influences on periphyton community composition. Cairns (1956) demonstrated that algal communities from a nonpolluted temperate stream shifted from dominance by diatoms at 20uC, to green algae between 30uC and 35uC, and finally to cyanobacteria at temperatures above 35uC. In experimental streams, Wilde and Tilly (1981) found that the cyanobacteria Schizothrix Acknowledgments P. Brown, J. Anderson, and J. Garrett provided design and field assistance that was instrumental to the success of the study. B. Snyder and the staff of the Sawtooth Valley Fish Hatchery provided invaluable logistical support. A. Chartier performed chemistry analyses. S. Durham assisted with statistical analyses. R. Lowe kindly identified the species of Rhopalodiales in our samples. Discussion with and comments by M. Baker, R. Hall, C. Arp, K. Nydick, G. Burkart, H. Van Miegroet, and two anonymous reviewers greatly improved the manuscript. This work was supported by NSF grants DEB 01-32983 to W.W. and DEB 04-12081 (Doctoral Dissertation Improvement Grant) to A.M. and W.W. A.M. was also supported by the College of Natural Resources and Ecology Center at Utah State University. Limnol. Oceanogr., 51(5), 2006, 2278–2289 E 2006, by the American Society of Limnology and Oceanography, Inc.