This proposal addresses a fundamental problem limiting progress in microbial ecology--the poor correlation between cultivated and naturally occurring microorganisms. This paradox has been revealed using small-subunit ribosomal RNA (SSU rRNA) sequences to characterize a model hot spring cyanobacterial mat community. Because the isolates come mainly from the model community, the model is particularly well-suited for investigating the basis for this discrepancy. rRNA-based hybridization probe studies of cultivated and uncultivated phototrophic microorganisms of the mat provide a more focused view of the boundaries of the model community, and permit formulation of several hypotheses that are likely to explain the failure to detect SSU rRNAs of cultivated species. Hypotheses are based on how trophic structure affects abundance and how culture procedures select for microorganisms not adapted hypotheses, (a) a more streamlined search for SSU rRNAs of natural community members, (b) a comprehensive probe study to determine the distributions and numeric relevance of all cultivated and uncultivated species, and (c) the cultivation of more relevant species using more ecologically sound approaches will be undertaken. The results are essential for future studies as well as parallel studies of the mat as a model for ancient microbial communities.