666 S. TABER SOME CRITERIA USED IX RECOGNIZING FAULTS 



The branch faults that leave the main San x\ndreas fault on the San 

 Francisco Peninsula, such as the Black Mountain fault, the San Mateo 

 Creek fault, and a fault located by the writer, which extends from the 

 western branch of San Andreas Lake and reaches the coast a half mile 

 south of Mussel Eock (see figure 1), are all readily explained on the 

 hypothesis of horizontal shear. Branch faults of this tj-pe are not a 

 characteristic of normal faulting or of reverse faulting. The numerous 

 small ponds between Mussel Rock and San Andreas Lake indicate that 

 there are several other branch faults in that section. 



Several displacements are known to have occurred along the San 

 Andreas fault since the settlement of California, though in most in- 

 stances the direction of movement is unknown. A^ertical displacements, 

 however, are much more likely to be recognized and reported than hori- 

 zontal ones, especially in thinly settled regions, and the scarps formed 

 are conspicuous for years. In 1857 the southern half of the fault was 

 ruptured, but the direction of displacement was not recorded. In 1890 

 a rupture is said to have occuiTed near San Juan, with a small displace- 

 ment in the same direction as in 1906. During the earthquake of 1901 

 a crack was formed along the fault in the Cholame Valley which could 

 be traced for several miles.' 



The earthquake of January 31, 1922, which originated under the 

 Pacific Ocean about 300 miles off the Oregon coast and approximately 

 in line with the San Andreas fault, was felt much farther from the origin 

 than the earthquake of 1906. It likewise was probably caused by a hori- 

 zontal displacement along the San Andreas or some parallel fault; for, 

 as there was no sea wave, there could have been no appreciable vertical 

 movement of the sea-floor. 



While it is outside the scope of the present paper, I wish to point out 

 that the forces necessary to produce the displacements now in progress 

 along the San Andreas fault must be regional and not local, and that 

 there are other faults parallel to the San Andreas fault, such as the Hay- 

 wards fault, on the opposite side of San Francisco Bay, which show the 

 same physiographic characteristics. 



• Origin of small Fault Troughs along active Faults 



Several hypotheses have been advanced to explain the formation of 

 the small fault troughs and depressions formed during earthquakes. 

 Fuller suggests that those formed at the time of the New Madrid earth- 

 quake were due to undermining caused by the creep of quicksand into 



^ A. C. Lawson and others : Op. cit., pp. 38 and 40, 



