[klotz] earthquake WAVES 39 



period of quiescence above referred to, but becomes a medley of 

 waves, in which we sometimes can distinguish the reflected waves 

 following P and S. As already intimated the velocity of the P and S 

 waves increases with distance due to the increased depth of their path. 

 P increases from about 6-6 km. per second to nearly 14 kilometres 

 per second at 12,000 km. arcual distance. For the S waves the velocity 

 would be -58 of the preceding on the assumption that Poisson 's ratio o 

 is "25, that is the ratio of transverse contraction to longitudinal ex- 

 pansion, a value which experience shows to be within narrow limits 

 of observation. 



From the preliminary remarks it may be inferred that the seismo- 

 grams present a rather complicated series of disturbances due to the. 

 motion of the earth particles. Added to this are the unavoidable 

 movements of the pendulum of the seismograph itself. Theoretically 

 the bob or heavy mass of the pendulum is supposed quiescent, the 

 reverse of the ordinary functions of a pendulum, and it is the earth — 

 the pier — that makes the oscillations. 



However, it is not the intention to present a general paper on 

 seismology, but rather to give the result of an investigation of a definite 

 problem carried out at the Dominion Observatory. 



In my paper, "Velocity of L waves," published in the Bulletin 

 of the Seismological Society of America last June, it was stated that 

 "it is not unreasonable to believe that with high-class seismographs 

 and expert readings of their records we would not only be able to 

 obtain the average position of the epicentre but also the position and 

 direction of the fault line itself." The problem then was the location 

 of a fault line along which the disturbance or readjustment of the 

 stresses and strains took place, for an earthquake is always such a 

 readjustment. No stresses, no strains, no earthquakes. The location 

 of an earthquake is evidently a matter of the utilization of the times 

 of "Transmission of Earthquake Waves," the title of this paper. 



The Dominion Astronomical Observatory publishes annually 

 the geographical position of epicentres for the preceding year, derived 

 from seismograms of the Observatory and those of other stations. 

 The records of every earthquake do not always lend themselves to a 

 definite location, mostly due to the lack of the phases of the PandS 

 waves, the important factors in determining the distance. If onl}^ 

 one phase of a quake is recorded it is always of the L or long waves, 

 but they never alone give satisfactory data for location of an epi- 

 centre. 



On May 1, 1915, a severe earthquake was universally recorded, 

 the position of its epicentre was determined principally from the 

 seismograms of Eskdalemuir and Ottawa and published as being in 



