185 
1918-19.] The Propagation of Earthquake Waves. 
march with the density, a change which first appears in the rigidity n. 
Since in round numbers lc is about double n, any change occurring in n 
will produce in Jc + 4<n/3 a less proportional change. Hence we infer that 
the elastic changes which appear at the depths mentioned affect first the 
rigidity n, and thereafter the incompressibility k. In other words, the 
changing constitution of the earth’s material at these critical depths 
discloses itself most markedly in that elastic characteristic which belongs 
to solids as distinguished from liquids or non-rigid substances. 
Bearing in mind the fact that the data do not carry us further down 
than about half the earth’s radius (corresponding to a ray of arcual range 
110°), may we not conclude that at this depth the rigidity becomes small 
or even zero, so that the distortional wave practically ceases to exist? 
What becomes of it ? It may either be lost in virtue of viscosity in the 
growingly plastic material, or its energy may pass into the compressional 
wave form. It is reasonable to regard the incompressibility of this plastic 
nucleus to be of much the same magnitude as that of the encompassing 
elastic solid shell. The speed of the compressional wave will therefore be 
less in the nucleus than in the shell. These suppositions are in accord 
with the facts that the speed of the Primary wave distinctly diminishes 
as the depth approaches the position of the inner nucleus, and that the 
records of the distortional wave as a distinct phenomenon which can be 
identified are very uncertain at arcual distances greater than 110°. To 
guard against the possibility of misunderstanding, I should state that the 
non-existence of the distortional wave beyond this arcual distance does 
not mean that no disturbances are recorded at times that might be 
identified with the Secondary waves assumed to emerge there, but that 
these observed disturbances have not the characteristics of the Secondary- 
wave disturbances at smaller distances and cannot be distinctly identified 
as caused by distortional waves passing continuously as such from the 
original source of the earthquake disturbance. 
The material of the earth may then be regarded as essentially an 
elastic solid down to a depth of half the earth’s radius. Throughout about 
half the thickness of this elastic shell the two elastic constants increase 
with depth at a greater rate than the density, so that the speeds of pro- 
pagation of the compressional and distortional waves increase steadily up 
to values respectively 80 per cent, and 73 per cent, greater than near the 
surface. Throughout the lower half of the elastic shell the speeds of 
propagation show a tendency to decrease slowly after reaching their 
maximum values, and this tendency first declares itself in the distortional 
wave. There is relatively a fall-off in the rigidity. The suggestion is 
