EARTHQUAKES. 119 



axis a would reach the surface successfully ; while those portions which 

 struck against the strata of the flanks would be partially or wholly 

 quenched. The mode of outcrop on the surface is shown in the map- 

 view, Fig. 101, in which a is the epicentrum, b b the granite axis, and 

 c c the stratified flanks. It must be remembered also that the origin of 

 most earthquakes is by the formation or the readjustment of a fissure. 

 In such a case the shock will be simultaneous and severe all along the 

 fissure. It is probable that many so-called linear earthquakes are thus 

 accounted for. 



The velocity of the surface-waves, as observed in many cases of se- 

 vere earthquakes, is about twenty miles per minute. This accords well 

 with Mallet's experiments in granite. In some earthquakes the ve- 

 locity has been found to be twelve to fifteen miles (Mallet's results in 

 slate), and in some as high as thirty to thirty-five miles per minute. 

 In the Charleston earthquake of August, 1886, the surprising velocity 

 of eighty or even a hundred miles per minute was observed. In some 

 slight shocks, the velocity, as determined by telegraph, is estimated as 

 high as one hundred and forty miles per minute, or 12,000 feet per 

 second. 



This amazing difference may be thus explained : It will be remem- 

 bered that the velocity of the surface-wave is infinite at the epicentrum, 

 and diminishes, according to a law already discussed, until it reaches, 

 or nearly reaches, the velocity of the spherical wave. Now, if the 

 earthquake-focus be comparatively shallow, the initial velocity of the 

 surface-wave very rapidly approaches its minimum, and therefore the 

 observed velocity of the surface-wave may be taken as nearly the same 

 as that of the spherical wave ; but, if the earthquake be very deep, the 

 diminution, even on a plane surface, is far less rapid ; and when we 

 take into consideration the curvature of the earth-surface, it is evident 

 that the velocity of the surface- wave may be for all distances much 

 greater than that of the spherical wave. This would well account for 

 velocities of thirty to thirty-five miles, but not for velocities of one 

 hundred or one hundred and forty miles. This latter is accounted for 

 by another principle. 



These high velocities occur mostly in slight shocks. Now, while 

 heavy shocks (large and high waves) break the medium at every step of 

 their passage, and are therefore retarded, as already explained, slight 

 tremors (small and low waves) are successfully transmitted without 

 rupture, and therefore run with the natural velocity belonging to the 

 medium, i. e., the velocity of sound. Now, the velocity of sound in 

 granite is probably about 12,000 feet per second, or one hundred and 

 forty miles per minute. In the Charleston earthquake, however, 

 though a severe one, the waves seem to have been transmitted with 

 nearly the normal velocity belonging to the medium. 



