442 



University of California PiMications in Geology [Vol. 12 



development of bending shear, conducive to shortening in the direction 

 normal to the fault, it is very improbable that the highly elastic granite 

 which underlies the region would lend itself to such a bending shear 

 as is commonly developed in more plastic rocks. There must doubtless 

 have been developed an elastic deformation somewhat analogous to the 

 bending in plastic rocks; but, so long as this kept within the elastic 

 limit of the rock, the shortening would practically be inappreciable. 

 We may therefore conclude that there was no shortening transverse 

 to the shear zone of the San Andreas rift due to bending, as exemplified 

 in thinly stratified rocks in similar situations, or due to elastic defor- 

 mation of the granite. There is, however, another case to be con- 

 sidered. The San Andreas fault is an old faiilt, on which slips have 

 repeatedly occurred. At each slip some of the granite on each side of 

 the fault was doubtless crushed and partially pulverized, thus develop- 

 ing a zone of fault breccia and gouge. This zone might easily be 

 100 meters or more in width. Such a zone of breccia and gouge was 

 observed by the engineers who drove the tunnel of the Los Angeles 

 aqueduct through granite, across the San Andreas fault, near Lake 

 Elizabeth. Such a breccia zone once developed would afford oppor- 

 timity for plastic shearing for every subsequent renewal of strain; 

 and such plastic shearing might involve a very appreciable shortening 

 of distances measured transverse to the fault. It is highly probable 

 that a wide breccia zone exists on the San Andreas fault for its entire 

 length. Thus while there is no probability of shortening in this case 

 due to bending, whether plastic or elastic, there is a possibility that 

 such a shortening might occur by plastic shearing of the breccia zone. 

 This possibility, however, would be realized in fact only if the shear 

 were a bending shear such as is exemplified in thinly stratified rocks. 

 But the shortening thus effected transverse to a long fault, such as the 

 San Andreas fault, implies the accommodation elsewhere of a very 

 large volume of rock matter, since the total content of the breccia zone 

 cannot be diminished. Thus a shortening of a meter on a fault breccia 

 100 km. in length and 10 km. deep Avould necessitate the accommoda- 

 tion, outside of the shortened region, of a billion cubic meters of rock. 

 It is difficult to imagine any way in which such accommodation could 

 be effected otherwise than by extrusion, and there is no suggestion 

 of any such extrusion in fact. It seems clear, therefore, that the shear 

 on the breccia zone of the San Andreas fault could not have been a 

 bending shear and could only have been a slicing shear such as is 

 illustrated in the diagram, figure 4. Here all lines transverse to the 



