228 REPORTS ON THE STATE OF SCIENCE, ETC. 
‘TRANSMISSION ‘TIMES. 
By Dr. Harold Jeffreys. 
The paper on ‘ Transmission Times,’ by Mr. K. E. Bullen and myself, 
has been published by the International Seismological Association, part of 
the cost being covered by a grant from the Gray-Milne Fund. Ihave now 
rediscussed the whole of the data, supplemented by some further earthquakes 
and special studies. The P times are found to have standard errors of 
about 0-3 sec. at most distances, rising to about'1 sec. at 105°. S seems 
also to be well determined between about 25° and 70°; but at shorter and 
greater distances different earthquakes giving good series of observations 
give times differing by about 10 secs. when the adjustable constant Z is 
chosen to fit the times from 25° to 70°, The core reflexions PcP and ScS 
have been tabulated from Scrase’s and Stechschulte’s studies of deep-focus 
earthquakes. On combining these with the observed times of PKP and 
SKS times of K for all distances in the core have been found. A number 
of other core phases have been computed. SKKS, where it is best deter- 
mined from observation, agrees well with the calculated times. The only 
noteworthy changes from the times given in the Jeffreys-Bullen tables are 
in SKKS beyond 170°, and in PKP,, which has to be reduced by about 
7 secs. 
In a further paper, five deep-focus shocks have been discussed. The 
effect of focal depth has been calculated and allowed for, so as to reduce 
the observations to a standard focal depth, namely for a focus at the top of 
the lower layer. ‘The intervals pP—P, sP—P, and sS—S, when adjusted in 
this way, give determinations of the times lost in passage through the upper 
layers. ‘The results are not quite consistent, but what seem to be the best 
determinations agree with thicknesses of 14 km. and 28 km. for the upper 
and intermediate layers respectively ; the standard error of the sum is 
about 3 km. It seems probable that this work may lead to a determination 
of Z for a known focal depth and to crucial tests of the doubtful parts of the 
S table, but this part of the analysis is not yet complete. 
THE LONG WaAvE PHASE OF EARTHQUAKES. 
By Dr. R. Stoneley. 
The presence of this phase of a seismogram is generally attributed to the 
passage of elastic waves over the surface of the earth. Two distinct types 
of wave are known to be theoretically possible : Rayleigh waves, in which 
the motion of the ground has both vertical and horizontal components but 
is always in a vertical plane through the line of travel, and Love waves, in 
which the motion is horizontal and perpendicular to the direction of pro- 
pagation. ‘The velocity of a surface wave on the earth depends on its period ; 
the first surface waves to appear on a seismogram are the Love waves of very 
long period, which should, and apparently do, travel faster than the longest 
Rayleigh waves. Some progress has been made in separating these two 
types by choosing earthquakes whose surface waves arrive in a N., E., S., or 
W. azimuth ; for instance, if waves arrive in an E. azimuth, a N. component 
seismograph will record the Love waves and the E. and Z components the 
Rayleigh waves. In this way the transmission times of waves of different 
periods can be found. The velocity of travel is not the wave velocity, but 
the group velocity ; the wave velocity can, however, be obtained from this 
by numerical integration when some assumption has been rade about the 
velocity of a wave of very long period. The integration has been carried 
