Section III, 1918 [37] Trans. R.S.C. 



The Transmission of Earthquake Waves 



By Otto Klotz, LL.D., F.R.S.C. 



(Read May Meeting, 1918.) 



As this is the first time I beHeve that this subject has been brought 

 before our Section, a few preliminary remarks in the new science of 

 seismology before we present the result of an investigation, may be 

 permissible. 



In every earthquake are recognized three distinct types of waves, 

 which, in the order of their arrival at any point, are known as longi- 

 tudinal waves, waves of compression and rarefaction analogous to 

 sound waves; next transverse waves, waves of distortion and which 

 may be likened to light waves; and lastly surface or undulatory 

 waves. The first two are transmitted as spherical waves through the 

 earth, while the last is confined to the crust. 



From theoretical considerations it is obvious that the speed or 

 velocity of propagation is dependent upon the physical properties of 

 the material through which the propagation takes place, and it further 

 follows that the path of any ray for the longitudinal and transverse 

 waves will not be a straight line, but a curve concave to the surface, 

 due to the changing density, elasticity and rigidity, as we penetrate the 

 earth. The path or curve will be the shortest time path, or as techni- 

 cally known — the brachistochronic curve. We might have reasoned 

 à priori that with increased depth a greater velocity would prevail, 

 which, however, has been abundantly proved by direct observation for 

 distances up to or approaching a quadrant of the earth, which would 

 mean a depth of about 1,000 miles. This is about the limit of our 

 reliable information, for it must be remembered that the more distant 

 the quake, the less effective will be the horizontal component of the 

 longitude wave, the first to arrive; so that for distant quakes the 

 record of this wave is frequently absent. However, for quakes under 

 10,000 km. distant we have a goodly supply of seismograms, although 

 not all of the same quality — and for various reasons. 



From the known geographic position of quakes such as Messina, 

 California, India and Japan and the seismograms at distant stations, 

 velocity curves have been constructed and interpolated for 10 km. 

 intervals, so that inversely when an earthquake occurs in the ocean 

 or in uninhabited regions, from gi good seismogram showing the various 

 phases, the distance to the hearth or epicentre can be found. As was 



