The application of in-situ produced cosmogenic isotopes to problems in geomorphology has increased rapidly over the past decade. At least 57 papers and numerous abstracts have been published since the mid-1980s when the first mass-spectrometric measurements of terrestrially produced cosmogenic isotopes were made. Taken at face value, these studies provide quantitative information about rates of landscape evolution and landform age; however, the significance of calculated erosion rates and exposure ages depends strongly on the models used to interpret isotopic data, the validity of assumptions inherent to these models, and the geologic surroundings in which the samples were collected. This paper attempts to place cosmogenic isotope studies in a geomorphic context by reviewing fundamentals of the method and evaluating the validity of assumptions under which these data have been interpreted. At present, the establishment of high-precision, cosmogenically based glacial and alluvial chronologies is stymied by the evolution of geomorphic surfaces, the erosion of rock from sampled boulders, the potential for isotope inheritance from previous exposure, and the uncertainty of isotopic measurements. Uncertainties in isotope production rates and the observed variability of exposure ages on individual geomorphic surfaces limit the confidence with which cosmogenic ages can be correlated reliably with those obtained by other techniques. Estimation of erosion rates at single points on the landscape gives useful small-scale information. Extrapolation of these rates over longer time and larger spatial scales is less sure and most likely biased toward lower erosion rates by the inadvertent selection of resistant sample sites. However, because erosion rates are so poorly constrained at present, even estimates to within a factor of 2 may be of significant value to geomorphologists and tectonicists. ¿ American Geophysical Union 1994