93 



6. On a Suspected Quadrantal Acceleration in Earthquake Speed. 



The earthquakes here referred to are those which have been recorded at stations situated at distances of 

 at least 90 from their origins. In well-defined seismograms these disturbances show three phases of 

 motion. The preliminary tremors, or PI, reach stations 60 to 180 distant from origins with average 

 chordal velocities increasing from 11 to 12 krns. per second. These may be compressional waves. 

 Following these, but with larger amplitudes, we find a second phase, P 2 . These, which are regarded as 

 distortional waves, have over paths from 30 to 160 in length average arcual velocities increasing from 

 4 -2. to 6-4 kms. per second. Lastly, there is the maximum motion, or P s , which has an approximately 

 constant arcual velocity of 3 kms. per second. 



For the commencement of this phase, which is apparently recorded as an undulating movement of the 

 surface of the earth, and may therefore be regarded as being partially gravitational in character,* the 

 velocity becomes 3'3 kms. per second. With regard to P 3 , this, however, is a general statement. 

 Within 10 of an origin, the value for P 3 appears to be less than 3 kms. per second, whilst in the 

 quadrantal region it may exceed 4 kms. per second. There are also indications of variation in 

 velocity in the antipodean regions. The values for P 2 also appear to be increased in the quadrantal region. 

 These velocity changes were first discussed in a British Association Report for 1900, p. 64 et srq., but the 

 data then at hand were not sufficient to sustain any definite conclusion. 



The observations made by the "Discovery," taken in conjunction with observations referring to the 

 same earthquakes made at other stations, have added to the material illustrating the phenomena here 

 considered, and it is for this reason that I again call attention to the same 



The speed acceleration, particularly for P 3 , is shown in the eleven time curves (Plate 1), six of which 

 refer to the " Discovery " register. The flattening in these curves indicates an increased speed. This 

 usually commences at a distance of from 40 to 70 from an origin. 



Something analogous to these movements recorded on the surface of the earth is seen in WUEWEF.L'S 

 Oceanic Cotidal Chart.t In the narrowest part of the Atlantic, between Africa and South America, the 

 lines representing the hourly change in the position of the tidal crest are crowded together. As these 

 travel northwards into the broader, and in places somewhat deeper, water, they are more widely separated. 

 In other words, the tidal wave travels more quickly in the broader and deeper portions of ocean than in 

 the narrower portions, where it is retarded. Although the chart may not be " perfectly trustworthy," J 

 it at least suggests that a seismic wave of the type PS may be less constrained, and therefore travel more 

 quickly in its quadrantal than in its polar region. This comparison is only intended to illustrate a form 

 of progress, and not to suggest that the factors governing the variations in speed of the tidal and seismic 

 waves are altogether identical. Further, the seismic wave at its antipodes shows an apparent increase in 

 its velocity, which is the reverse of that which would be expected by a tidal wave when approaching the 

 head of an oceanic inlet. 



It-might be assumed that the earthquake wave passes beneath a crust and over a nucleus, into which it 

 merges. The upper portion of such a wave would be more retarded than its lower portion. It may also 

 be imagined that the more swiftly moving lower portion on the first 90 of its path fails to give a surface 

 indication of its existence because its external boundaries are widening. In the quadrantal region the 

 periphery of the boundaries is fairly constant, and it is here that we find apparent acceleration in its 

 speed. Still farther on its journey excessive contraction of the boundaries results in retardation of the 

 waves. 



7. Antipodean Re-appearances. 



For some years past I have noticed that earthquakes which had their origin in the vicinity of New 

 Zealand, and were recorded in that country, have also been recorded in Britain, particularly at Bidstone, 

 but had not necessarily been recorded at intermediate stations. The " Discovery " records, taken in 



* The influence of gravitation has been discussed by BBOMWICH, in 'Proc. Lond. Math. Soc.' 

 t See 'The Tides,' by O. H. DABWIN, p. 172. 

 J Ibid., p. 173. 



