ON SEISMOLOGICAL INVESTIGATIONS. 67 



wings have been singed in the lamp, and which has fallen down the light 

 aperture and become wedged between the slit plate and the floating vane 

 on the boom. In either case the decrement curve and standard deflection 

 will show considerable deterioration. 



A piece of glass over the light aperture is a partial remedy ; but more 

 effectual is the addition of about 2 lb. of naphthalene (CioHs) well distri- 

 buted throughout the cases. 



Herewith are given illustrations of first-order trouble from St. Helena 

 and spider trouble from Bidston. The Bidston apparatus is a fully 

 damped Milne-Shaw type and has no decrement curve, but the standard 

 deflection fell from 27 mm. to 9 mm. as a result of the webbing of this 

 spider. 



X. The Identification of S : Suggestion of a Neiv Phenomenon Y. 



As shown in the last Keport, there are accidental deviations of 

 observation of Sfrom the times assigned l)y the tables. The mean of the 

 errors discussed is (+0.73 minutes or) +44 sec. for S, and (±0.31 minutes 

 or) ±19 sec. for P. This is the more remarkable since the amplitude 

 of S is usually much greater than that of P, so that there ought to be 

 less uncertainty in reading. The suspicion is aroused that there is some 

 other phenomenon liable to be mistaken for S ; and that many of the 

 errors are due to these mistakes. 



A suggestion of this kind is put forward by Dr. G. W. Walker in his 

 ' Modern Seismology,' but is apparently vitiated by an oversight. On 

 p. 41, after considering the first reflected wave PE], he next considers 

 a wave which travels as P to the point of reflection and as S subsequently ; 

 he points out that there is a lower limit to the possibility of such a wave 

 which he determines as A=110° or 12,000 km. ; and he proceeds : 



' Now, it has been observed that special difiiculty attaches to the 

 identification of S just when A is about 12,000 km. Thus with an earth- 

 quake in the northern Philippines, which are about 11,000 km. from this 

 country, S usually comes out very clearly, while in the case of an earth- 

 quake in the Caroline Islands, about 12,000 km. from us, S is most in- 

 distinct, and the tendency is to put it rather late. The result we have 

 obtained throws some light on the matter.' 



The oversight which vitiates this explanation is that Mr. Walker 

 is dealing at the moment with the hypothesis of a homogeneous earth, 

 which he soon shows to be quite untenable. His figures are those for 

 a homogeneous earth, and are quite inapplicable to the actual earth. 

 The lower limit he mentions does indeed exist, but instead of being at 

 A=110° it is about A=35°. It is readily found numerically by using 

 the existing tables printed on the last page of the Shide bulletins for 

 1914. Thus wlien ii=40° the times given for a wave to travel 



as P for 1° and S for 39°= 15 +832 =847 

 P for 2 and S for 38 = 31 +818 =849 

 P for 3 and S for 37 = 47 +804 =851 

 P for 4 and S for 36 = 62 +790 =852 

 P for 5 and S for 35 = 77 +775 =852 

 P for 6 and S for 34 = 92 +760 =852 

 P for 7 and S for 33 =106 +744 =850 

 P for 8 and S for 32 =121 +728 =849 

 P for 9 and S for 31 =136 +711 =847 

 &c. &c. 



F 2 



