Abstract
A procedure is described for observation and enumeration of microscopic epilithic algae, directly on stone surfaces. The method involves the use of incident light fluorescence microscopy and provides rapid estimates of total numbers as well as information on the distribution of the organisms. Comparison with conventional procedures, which involve scraping the stone surface, suggest that these latter methods may produce serious underestimates of the algal population. The within-stone and between-stone errors of estimates of the diatom Cocconeis placentula are discussed in relation to a) a suitable transformation of the data before use of parametric statistical tests and b) the number of sampling units required to achieve given 95% confidence limits.