A fluid injection-induced seismicity experiment was conducted in the German Continental Deep Drilling Program (KTB) main borehole at 9.1 km depth (in situ temperature of 260 ¿C) to extend knowledge about stress magnitudes and brittle faulting to depths and temperatures approaching the brittle-ductile transition. Almost 400 microearthquakes were induced at an average depth of 8.8 km by injection of KBr/KCl brine into a ~70 m open hole section near the bottom of the borehole. Although most focal plane mechanisms were poorly constrained due to the very small size of the induced earthquakes, several different clusters of microearthquakes with distinct mechanisms were defined. Most of the microearthquakes for which focal plane mechanisms were determined were strike-slip events with a NNW trending P axis, essentially parallel to the direction of maximum horizontal compression observed in the borehole. The largest induced earthquake, M 1.2, occurred 18 hours after injection was started. This event was a strike-slip/reverse faulting event which also had a NNW trending P axis. Utilization of a precise relative location technique indicates that many of microearthquakes occurred relatively far (50--100 m) from the well bore. Modeling of the pore pressure disturbance caused by injection suggests that many of the earthquakes were induced by extremely small pore pressure perturbations (<1 MPa) less than 1% greater than the ambient, approximately hydrostatic pore pressure at depth. Thus it is apparent that there are critically stressed, permeable fault zones in the crust, even at great depth and temperature. A frictional analysis of the focal plane mechanisms of the induced microearthquakes indicates that fault slip is consistent with the stress magnitudes and orientations determined in situ at depths to 7.7 km in the borehole and relatively high coefficients of friction (~0.6--0.7) reported by Brudy et al. . This and the observation that very small pore pressure perturbations were able to trigger seismicity appear to confirm the hypothesis that Byerlee's law (i.e., that differential stresses in situ are limited by the frictional strength of well-oriented, preexisting faults) is valid to great crustal depth and that the crust is in brittle failure equilibrium at depths and temperatures approaching the brittle-ductile transition, even in this relatively stable intraplate area.¿ 1997 American Geophysical Union