444 University of California Publications in Geology [Vol. 12 



time of the slip in 1906 the southwest side advanced and the northeast 

 side sprang back. It is highly probable, as Reid has pointed out, that 

 the fault traverses a shear zone, which was an antecedent condition to 

 the slip. But the shear zone could not have developed, nor the slip 

 have happened unless the strain had been greater to the west than 

 to the east. It is probable that somewhere to the east of the fault the 

 strain dies out. The locus of zero strain due to subcrustal flow is, 

 however, unknown. We do not even know positively whether this 

 locus is parallel to the San Andreas fault or not. It seems probable,, 

 however, that it is not, but that it is more nearly parallel to the 

 observed direction of strain creep. It follows from the foregoing that 

 the rate of strain creep must increase from nothing at the locus of 

 zero strain to the rate inferred from the geodetic observations near 

 the fault; and that this rate is greater on the west side of the fault 

 than on the east side. Beyond the fault an unknown distance the 

 rate doubtless diminishes again to zero, at the western border of the 

 subcrustal current. For points to the west of the easterly zero locus 

 the rates of strain creep, if plotted as ordinates from a straight line, 

 would doubtless terminate on a curve. But the curve is probably very 

 flat. The rates of creep inferred for Chaparral and Ross Mountain 

 indicate that the curve is nearly a straight line, and that the zero 

 locus must be far to the east. In so far as the curve approximates a 

 straight line the rate of creep is proportional to the distance from the 

 zero locus. In so far as the zero locus is remote the difference in rate 

 for points a few kilometers on either side of the fault is small. Now 

 all of the data that we have for the rate of creep of the region in the 

 vicinity of the fault is derived from stations on the east side of it. 

 In the discussion which follows the rates thus obtained are assumed 

 to hold for the stations on both sides of the fault for which we have 

 the measure of movement. This assumption is mathematically in- 

 correct; but it appears to be the best that can be done under the 

 circumstances. The error involved in the assumption is probably small 

 and does not invalidate the general nature of the results arrived at, 

 the purpose of the discussion being not to arrive at precise quantities, 

 but merely to interpret the process of earth movement in the light of 

 the rebound theory. 



It may be further observed from the general point of view, before 

 going into details, that the failure of slipping on the San Andreas fault 

 to relieve the longitudinal strain leaves us in the expectation of that 

 relief by other movements. These movements would affect the whole 



