162 MR. R. 1). OLDHAM ON THE PROPAGATION OF 



decrease, and within the outer area increase, as the distance from the epicentre 

 increases. 



The depth of the focus is always so small in comparison with the diameter, and the 

 size of the inner area so small in comparison with the surface, of the earth, that they 

 may be left oiit of consideration in a study of the propagation of earthquake motion 

 to great distances. Consequently we should find, if the disturbance is transmitted 

 through the earth, an increase in the apparent rate of transmission with an increase 

 in distance from the origin. This apparent rate of increase will be proportionate to 

 the ratio between arc and chord if the wave motion is propagated in straight lines, it 

 will be less if the rate of propagation diminishes with the depth and the wave paths 

 are concave towards the centre, and greater if they are convex, and the rate of 

 propagation increases with the depth. 



It must further be noticed that the regularity of increase of v and v n with the 

 distance from the origin only holds good if the increase of V is a constant function 

 of the distance from the centre of the earth. This may reasonably be expected so 

 long as there is no great change in the character of the medium traversed and the 

 change in the elastic constants is principally due to the increase of temperature and 

 pressure. It is, however, very probable that the central core of the earth is metallic, 

 composed principally of native iron, surrounded by an outer shell of magma, which 

 would be stony or glassy in a cooled and solidified form. If this be the case the wave 

 on passing from the one to the other would enter a medium in which the change of 

 elasticity due to temperature and pressure would be complicated by an initial 

 difference in the elastic constants. 



If, as is probable, there should be a sudden increase of V when the wave path 

 enters the central core, it is possible that v might become infinite along a circle 

 removed from the antipodes of the origin, and beyond that have a negative value. 

 That is, it is conceivable that the disturbance might emerge at the antipodes before 

 it reached the surface at some point nearer the origin. In any case a sudden change 

 in the rate of increase would exhibit itself as an interruption of the regular curvature 

 of the time curve, this being bent downwards if the change was an increase, upwards 

 if it was a decrease, of the regular rate of increase of velocity with depth below the 

 surface. 



Turning to the application of these general principles to the recorded observa- 

 tions, this seems to be most conveniently effected in the graphic manner employed in 

 the accompanying diagram. On this the recorded times of the first, second, and third 

 phases have been plotted against their distances from the origin in degrees of arc, and 

 smoothed curves drawn through them. 



It may not be amiss to repeat here that in obtaining these times no selection was 

 exercised. The times of commencement and of any marked change of amount or 

 character of movement were taken from the published record, and the only subse- 

 quent selection made was the rejection, in those cases only which have been specially 



