Today there exists in the United States a large volume of nuclear wastes that result from both military and commercial activities. The United States has to date placed major emphasis on disposal in only one rock type--salt--whereas other nations have considered other rock types, such as granite in England and Sweden and clays in Belgium. No comprehensive evaluation of isolation in rocks other than salt has been made in the United States, and it is most appropriate that other rock types be evaluated both for constructing disposal sites in areas devoid of salt and also for having alternative waste management plans in case substantial problems are encountered in using salt as a disposal medium. To evaluate the state-of-the-art, research needs, and research priorities related to waste disposal in largely-impermeable rocks, scientists and engineers working on geologic aspects of nuclear waste disposal were brought together. The Geotechnical Assessment and Instrumentation Needs (GAIN) Symposium for Nuclear Waste Isolation in Crystalline and Argillaceous Rocks was held July 16-20, 1978 in Berkeley. This report and recommendations are the proceedings from that symposium. The location, design, and testing of a potential nuclear waste disposal site are both a geologic and an engineering problem. Disposal requires isolating the wastes from the surface and subsurface of the earth for a period of time of ten to hundreds of thousands of years. Engineers have never before been called upon to predict the behavior of structures or the flow of groundwater so far into the future.