Yol. 65.] OP THE CALIFORNIA]* EARTHQUAKE OE 1906. IS 



The growth, of our knowledge of earthquakes is making it 

 continuously more and more evident that, whether great or small, 

 they have little or no connexion with the faults which reach the 

 surface of the earth. Leaving out of account minor earthquakes, the 

 origin of which can seldom be determined with sufficient accuracy to 

 connect them, even by position, with — though they can frequently 

 be shown to be independent of — any known surface-faults, and 

 considering only great earthquakes, we have Col. Harboe ? s demon- 

 stration that the origin of the Mino-Owari earthquake was probably 

 much more extensive and complicated than the fault to which it is 

 commonly ascribed, and the certainty that the Assam earthquake 

 was neither solely nor mainly the result of movement along the 

 great structural flexure which separates the elevated area of the 

 Assam Hills from the depressed area of the Barak Valley. The 

 Kangra earthquake of 1905 was unaccompanied by any surface- 

 faulting, or conspicuous changes of surface-level, nor does its 

 origin seem to have been due to movement along the great boundary- 

 fault of the Himalayas. The Californian earthquake of 1868 was 

 unaccompanied by any surface-faulting, and, so far as can be judged 

 from the displacements of trigonometrical stations, was the result of 

 a set of strains very different from that of 1906, and this again was 

 evidently far from being localized in its origin to the displacements 

 along the San Andreas fault. The local centre of great violence at 

 Santa Rosa points to movement along an independent fracture or 

 fault, and the same may be said of the earthquake at Cape Mendo- 

 cino, which can only be attributed to a submarine extension of the 

 San Andreas fault by postulating an improbable change in the 

 direction of the course of that fault. 



Not only are the irregularities in distribution, both of the violence 

 of the shock and of the permanent displacements, inexplicable if 

 the origin of the earthquake is assumed to have been localized to a 

 single fissure : but they are also difficult, if not impossible, to account 

 for if we suppose the earth's crust to have been involved as a unity 

 in the strains which caused the earthquake. This difficulty very 

 largely disappears if we adopt the not improbable hypothesis that 

 the outer 30 miles or so of rocks (which we are in the habit of 

 designating the crust of the earth) includes an outer skin of a few — 

 probably very few — miles in thickness of more discontinuous and 

 fractured rock. 



It is not to be supposed that there is a hard and fast boundary 

 between the rocks constituting what I have called the outer skin, 

 and those forming the greater part of the thickness of the crust. 

 To some extent there must be a difference of composition, for the 

 former consists largely of clastic rocks, composed of the products of 

 weathering and denudation, while the latter is mainly composed of 

 matter which has not been exposed to the action of air and water at 

 the surface of the earth. A more material difference, however, is to 

 be looked for in the fact that the surface-rocks, being exposed to a 

 smaller pressure, still preserve in the main the characteristics that we 

 attach to solidity ; while at a greater depth the increase of pressure 

 causes matter, which must still be called solid, to take on the 



