Yol. 62.] OF THE INTERIOR OF THE EARTH. 457 



two phases, forming what are known as the preliminary tremors, 

 represented the cropping-out of mass-waves which had travelled 

 through the earth. It is these two phases alone with which I am 

 at present concerned, for the third-phase waves can obviously give- 

 no information regarding the interior of the earth, as their wave- 

 paths lie along its surface. 



The researches of Dr. C. G. Knott l and Dr. P. Rudzki 2 have 

 shown that no simple form of wave-motion can be transmitted 

 through the heterogeneous rocks forming the outermost crust of the 

 earth, and that the records from instruments situated near the origin 

 of an earthquake cannot show any sorting-out of different kinds of 

 wave-motion. It is only in more homogeneous material that this 

 sorting-out can take place, and it is only at a distance of 10 degrees 

 of arc, or about 700 miles, from the origin that the three-phase 

 character of the record begins to appear. The waves emerging at 

 this distance have evidently traversed more homogeneous material 

 for a part of their course ; in this part there has been a sorting-out 

 of the forms of wave-motion into which the disturbance has been 

 converted, and the fact that the sorting-out can be detected at so 

 comparatively small a distance from the origin shows that the outer 

 crust must be, comparatively, very thin. I have not been able to 

 collect sufficient data for an accurate estimate of its thickness, but 

 this cannot be more than about a score of miles, 3 and below it comes 

 material of a very different character, which not only allows a 

 sorting-out of different forms of wave-motion, but, as has been shown 

 by Prof. Milne, 4 transmits these at a velocity much greater than is 

 met with in the outer crust. If the figures given in the following 

 pages do not bear out in detail his further conclusions regarding the 

 homogeneity of the whole of the core and the rectilinear propagation 

 of the wave-motion, this must be ascribed to the accumulation of 

 more data than were available when he wrote. As will be seen, they 

 are confirmed in essentials, so far as the outer six-tenths of the radius 

 are concerned, and in the central four-tenths the first-phase waves, 

 which alone were dealt with by Prof. Milne, are so little affected 

 that the change might easily have escaped recognition but for the 

 clue given by those of the second phase. It is, therefore, desirable 

 to devote a little space to the demonstration of the reality of a 

 distinction in land between these two sets of waves. 



In my paper quoted above, I pointed out that the different rates 

 of propagation of the first and second phases showed that they must 

 be referred to different forms of wave-motion, which I interpreted 

 as being, probably, the two known forms — compressional and dis- 

 tortional — which can be transmitted b} r a homogeneous solid. As 



1 Trans. Seisrnol. Soc. Japan, vol. xii (1888) pp. 115-36. 



2 Beitrage zur Geophysik, vol. iii (1898) pp. 519-40. 



3 There is some, seismologies 1, indication of a want of uniformity in this 

 thickness, for earthquakes originating off the eastern coast of Japan exhibit a 

 three-phase character at less distances from the origin than appears to be the 

 case in Europe, indicating a lesser thickness of the outer crust in the former 

 region . 



4 'Nature' April 9th, 1903, vol. lxvii, pp. 538-39. 



