434 University of California Publications. [Geology 



east-side residents, and a fissure opened in the ground of great 



depth and unknown length The disturbance . . . was 



immediately followed by the appearance of the crack, which did 

 not open simultaneously its whole length, but gradually, from 

 its northern to its southern terminus." 



Professor Niles, describing the expansion of the rock in the 

 quarry at Monson, Mass., and the accompanying cracks, writes : 

 ' ' These cracks, or rents, are more commonly formed slowly, but 

 sometimes suddenly." 



It is interesting to note that Mallet believed the great 

 Neapolitan earthquake of 1857 to be due to a fracture in the 

 underlying rock, which began in a limited area and extended to 

 greater distances. 



Indeed, the progressive method of breaking is general, as it 

 depends upon the elastic properties of solids. Absolute rigidity 

 would be practically necessary to ensure simultaneous rupture 

 over a very large area. It may be objected that it is trivial to 

 emphasize the difference of a few seconds in the time of rupture 

 at different parts of the fault ; but the difference is not so very 

 small. If the rupture of the San Andreas fault began near its 

 middle point, it must have taken at least half a minute, and it 

 may have taken more than a minute, to reach the ends of the 

 fault ; and moreover, deductions based on the supposed simul- 

 taneity of fracture have led to conclusions regarding mass move- 

 ments, the place of origin of the vibrations, and the interpreta- 

 tion of instrumental records, which are quite out of harmony 

 with the conceptions advocated here. 



As the different parts of the same fault do not fracture 

 simultaneously, so there is no probability of neighboring faults 

 fracturing at the same time. If two faults are only a few miles 

 apart, it may happen that the relief of strain at one will increase 

 the strain at the other sufficiently to start a rupture there, if it 

 is already strained nearly to the limit. The vibrations from one 

 fracture, under the same conditions, might start the rupture of 

 a second. In all these cases the rupture begins in a very limited 

 area of a single fault, and extends along the same and perhaps 

 to other faults, but never at a greater rate than the velocity 

 of compressional elastic waves ; as this velocity may, in some 



