172 REPORT — 1895. 



investigated is not the transversal vibrations of a truly elastic solid, but it 

 is a quasi-elastic surface disturbance, the propagation of which is accele- 

 rated by the influence of gravity. 



The preliminary tremors have, however, yet to be explained. At 

 stations within 100 miles of an origin, as recorded by seismographs, these 

 outrace the main disturbance — with which, however, they are invariably 

 connected — and often overlap it, by perhaps ten seconds. At a distance 

 of 6,000 miles they seem to outrace it by half an hour. 



Dr. C. G. Knott suggests that they are due to the quasi-elastic dis- 

 turbances which accompany earthquakes. When the earth movement is 

 violent, and possibly accompanied by destruction, the material of the 

 earth's surface is either strained beyond its limit of elasticity, or at least 

 so far strained that the resulting movements are governed by coefficients 

 other than those due to rigidity and compressibility. As these quasi- 

 elastic waves pass through a region of discontinuity, or as they lose 

 energy, they may be suddenly or gradually transformed into a purely 

 elastic disturbance. 



Although changes of this description may take place as a disturbance 

 passes from medium to medium, inasmuch as it implies the creation of 

 tremors as the surface waves progress, much in the same way that a 

 trotting pony or a railway train creates the sound waves which run before 

 them, we are led to the conclusion that the preliminary tremors have a 

 velocity very much higher than those already calculated. Because this 

 cannot be accepted, the only explanation remaining is the assumption 

 that the preliminary tremors are movements originating at an earthquake 

 centrum and propagated possibly as condensational waves along paths yet 

 to be discussed through our earth. If this is made, then apparent velocities 

 of 11 or \'l km. per second, as observed for example on March 22, 1894, 

 may be reduced to actual velocities of about 9 km. per second. 



Should a more extended and systematic observation confirm this pro- 

 visional assumption, we shall then be in a position to discuss from a new 

 point of view the physical nature of the materials constituting the interior 

 of our globe which apparently transmits motion at a greater rate than 

 glass or steel. 



Points not yet touched upon are the increase of velocity with an 

 increase in the intensity of the initial disturbance, and a decrease in 

 velocity as a disturbance radiates, both of which phenomena are marked 

 near to the origin of artificial disturbances. The only explanation which 

 suggests itself for both these phenomena is that around the epicentrum 

 there is a region to which motion is communicated partly by elastic yield- 

 ing and partly as a push. The volume of ground which may be thus 

 disturbed is called by my late colleague, Professor T. Alexander, an earth- 

 quake core. In the case of an artificial disturbance originating near to 

 the surface, the distance to which this effect extends will depend upon the 

 suddenness and magnitude of the initial disturbance. With an earth- 

 quake originating underground, the distance to which a high epifocal 

 velocity may be noticeable will not only dei^end upon the above two con- 

 ditions, but also upon the depth of the focal origin. The greater this 

 latter quantity becomes, the greater will be the radius of the epicentral 

 area in which there may be not only a real increase in velocity but also a 

 high ajjparent velocity. 



The conclusion to which these considerations and the observations 

 which have been made lead is that an earthquake gives rise to at least 



