For surveillance studies, comparison of activity levels in soil and sediment samples is usually difficult, particularly for 'grab' samples, because of the wide variability in composition of both soil and sediment samples. Thus, radioactivity levels of sediment samples, for example, are likely to be a function of sample composition rather than of the amount of radioactivity present in the ambient water. For this reason, care at each step of the procedure--sample collection, preparation, and analysis--is vital for quantifying the radionuclide content of geologic and hydrologic materials. These studies were conducted to determine whether some active portion of a 'grab' sample, easily separated from the inert part of the sample, would provide a sensitive and reproducible indicator of radiocontamination in the environment. Although the experimental material for these studies consisted of soil and sediments from eastern Massachusetts, it is believed that, when standardized, the method will provide a useful and sensitive indicator for monitoring bottom sediments in general. (Author)