THE EARTHQUAKE PROBLEM HECK 369 



The actual use of the complex phases is made possible through 

 diagrams with distances plotted as ordinates and intervals in seconds 

 from time of origin of the earthquake as abscissas. Tables are also 

 used in which time intervals from time of the earthquake to arrival 

 of the given phase, or intervals between successive phases, are given 

 for each distance. Perliaps the most convenient tables that have 

 been developed are those of Rev. James B. Macelwane, S. J., of St. 

 Louis University. However, not all tables are in full agreement, and 

 seismologists are still working on the best values to adopt in some 

 cases. 



An important use of these diagrams and tables is to obtain distance 

 of station from earthquake in order that the earthquake epicenter 

 may be known. The epicenter is the point on the surface directly 

 beneath which the earthquake occurred. This can be located in 

 several ways but the graphical method of plotting on a large globe 

 need alone be mentioned. The positions of seismological stations 

 are marked on the globe and, by means of compasses, arcs are 

 swung from each for the given distance of the earthquake. The 

 intersection of the arcs gives the position of the earthquake (pi. 5). 

 Some years ago little use was made except of records giving 

 clear P and S phases. Now earthquakes can be located when these 

 phases are lacking on all available records. Such a location was 

 made of an earthquake in China, which later was found to agree with 

 the determination made from all available data, including those of 

 near stations. An adopted epicenter must meet the requirement of 

 reasonably good intersection of the arcs, and the times of origin of 

 the earthquake as computed from the several records must be in good 

 agreement. 



The tables and curves have been derived from study of records and 

 from theory. Certain assumptions are made, and if these agree with 

 a large mass of observational results they are held to be valid. 



In this way a solution can be found for the distribution of density 

 within the earth. Knowing the surface density and the mean density 

 as given by astronomy, a distribution can be worked out which fits 

 the velocity of seismic waves at all distances. 



If the observations fit adopted curves through a wide range and 

 then the agreement suddenly disappears, an explanation must be 

 found; in this w^ay the existence of certain layers of discontinuity has 

 been postulated. ' We do not know exactly what the significance of a 

 layer of discontinuity is, that is, whether there is a change in physical 

 conditions or in chemical composition, though there is evidence in 

 favor of both. We know from seismic evidence that such layers exist 

 because the characteristics of recorded earthquake waves can not 

 otherwise be explained. The best known and most definite of these is 



