Measurements of in situ stress orientation and magnitude at the site of the Cajon Pass research borehole have been made from depths of 0.9--3.5 km using the hydraulic fracturing technique and analysis of stress-induced well bore breakouts. The results of these measurements support two important conclusions about the mechanics of crustal faulting. First, the magnitudes of measured in situ stresses indicate ratios of shear to normal stress on favorably oriented fault planes that are consistent with predictions based on Mohr-Coulomb theory and laboratory-determined coefficients of friction in the range of 0.6--1.0 assuming hydrostatic pore pressure (this is commonly known as Byerlee's law). Thus the stress measurements indicate that the frictional strength of the crust adjacent to the San Andreas fault is high (i.e., consistent with laboratory-derived friction values) and that the level of shear stress in the crust adjacent to the San Andreas is principally controlled by its frictional strength. However, data on the orientation of maximum horizontal compression in the borehole from 1.75 to 3.5 km (N57¿E¿19¿) indicate that the San Andreas must be quite weak as a complete absence of right-lateral shear stress resolved on planes parallel to the ~N60¿W striking San Andreas fault is observed. The lack of right-lateral shear stress on planes parallel to the San Andreas fault at this site is especially surprising as Cajon Pass is located along a section of the San Andreas which has not had a major earthquake since 1812 and is thus presumably quite ''late'' in the earthquake cycle.

Nevertheless, both the orientation and magnitudes of stresses measured in the well are consistent with the style of active faulting in the area surrounding the drill site, most notably normal faulting and Quaternary age left-lateral slip on the Cleghorn fault that parallels the San Andreas in the vicinity of the drill site (Meisling and Weldon, 1982; Weldon, 1986; R. J. Weldon et al., unpublished report, 1981). We argue that the stress state (and Quaternary fault offsets) observed in the Cajon Pass area could exist only if the San Andreas move at low shear stresses comparable to seismic stress drops rather than the much higher values predicted by Byerlee's law, a conclusion consistent with the lack of frictionally generated heat flow along the San Andreas system (e.g., Brune et al., 1969; Henyey and Wasserburg, 1971; Lachenbruch and Sass, 1973, 1980). Taken together, the Cajon Pass in situ stress and heat flow measurements (Lachenbruch and Sass, this issue) support a conceptual model of the San Andreas system in which the San Andreas is extremely weak with respect to the surrounding crust. ¿ American Geophysical Union 1992