362 CARNEGIE INSTITUTION OF WASHINGTON. 



temperature. This is shown by the increased velocity of the seismic waves, 

 both distortional and compressional. But the rate of this increase of velocity 

 falls off and practically ceases near the mid-depths. This has been interpreted 

 as implying a change from the elastico-rigid state to a viscous or some other 

 non-rigid state. But in drawing this inference the important factor of increas- 

 ing density seems to have been overlooked. Density tends to decrease wave 

 velocity, and hence to secure the velocity to be used in comparisons of the 

 elastico-rigid states, correction must be made for this increasing density. 

 When the observed velocities are thus corrected by the use of the Laplacian 

 law of density, or better still by the Roche formula which was devised especially 

 to meet the astronomical requirements, a continued increase of rigidity and 

 elasticity is indicated, though it is not so rapid an increase as that in the lesser 

 depths. There is probably a progressive change in the material of the earth 

 as held by Oldham. 



This correction brings the testimony of the seismic waves into general 

 harmony with the evidences of the body tides and the nutational oscillations, 

 both of which indicate that the earth as a whole is not only rigid and elastic 

 but that the mean rigidity and elasticity is much greater than that near the 

 surface. 



It has been noted that seismic waves of both distortional and compres- 

 sional types grow progressively imperfect as they penetrate to greater and 

 greater depths and distances, until finally they cease to give readable records 

 before the diameter of the earth is penetrated. This growing debility is a 

 feature of both sorts of waves. This implies that a common source of debility 

 affects both sorts of waves — doubtless the natural destructive and dispersive 

 effect of the increased penetration of an imperfect medium. As a rule, the 

 records of the distortional waves become illegible with less penetration than 

 the compressional waves, but this is not at variance with what might be 

 expected, for the compressional waves move about 50 per cent faster than the 

 distortional waves, a fact implying that they are actuated by a more effective 

 elasticity, elasticity of volume, the distortional waves being actuated by 

 elasticity of form. This difference in dying-out does not, therefore, seem to 

 warrant the inference that a viscous state replaces the elastico-rigid state. To 

 justify that, there should be a rather sudden cessation of the distortional waves, 

 while the compressional waves should continue to penetrate the greater depths 

 and distance even to the diameter, and give good records. Besides, as 

 already noted, the testimony of the velocities, when corrected, implies con- 

 tinued rigidity, as do also the independent evidences from the body tides and 

 from the earth's nutation. 



These several supplementary considerations seem to greatly strengthen the 

 conclusion that the body of the earth is essentially elastico-rigid. The 

 ground for doubt about even the most central portion seems to be vanishing. 



In the course of the year's studies some new computations, based on inde- 

 pendent ground, were made respecting the preponderance of the organizing 

 revolutional energies within the earth over the dissipative vibratory energies. 

 These gave results of the same high order of preponderance as the previous 

 ones. Some of the assumptions, however, are necessarily quite uncertain, 

 and it is scarcely worth while as yet to assume that the results are very close 

 to the true ratio. 



