9220005 Tullis This work is a laboratory experimental study of the frictional behavior of rocks with significance both for understanding fundamental frictional processes and for learning how to apply the laboratory results to faults in the Earth's crust that produce earthquakes. The focus is on two important problems. The first is to understand the processes responsible for the observed behavior, so that constitutive equations can be based on known processes. This is important for extrapolating laboratory results over large time and distance scales to earthquakes. The second problem is to understand the reason for differences in large displacement experiments between velocity weakening behavior of friction samples using initially bare surfaces and velocity strengthening behavior of samples using simulated gouge. Understanding the correlations between frictional behavior and structures in these samples will help ascertain what behavior may be associated with structures observed in natural faults. Experiments will be conducted using a unique high pressure rotary shear apparatus well suited for this purpose. Experiments will be done on rock types of geological importance with known flow and fracture properties. Pore fluid chemistry, sliding rate and temperature will be varied to determine how frictional behavior depends on these variables and to help identify operative processes. ***