40 



REPORTS ON THE STATE OF SCIENCE. — 1916. 



a systematic run about the residuals for PR, which suggests a modifi- 

 cation of the tables in the neighbourhood of 60°-66° (the mid-points 

 of the arcs), but we shall not at present follow this thread. 



Earthquake of 1913 May 30^ ll'' 46'" 46«. 

 Adopted Epicentre 5°-0 S., 154O-0 E. 



Table III. 

 PEj recorded as P. 



Table IV. 

 PX recorded as P. 



Station 



Konigsberg 



Breslau . 



Hamburg 



Vienna 



Graz 



Sarajevo 



Zagreb 



Laibach 



Innsbruck 



Heidelberg 



Padova . 



Time 

 Observed 



1142 

 1169 

 1164 

 1158 

 1163 

 1158 

 1162 

 1162 

 1169 

 1196 

 1163 



The first group of stations have presumably recorded PE^ as P; 

 but the second group have recorded something else, which comes from 

 one to two minutes earlier. The records are so consistent as to suggest 

 a real phenomenon, which we may call PX for the present. More- 

 over, other earthquakes give similar results; and we may adopt, pro- 

 visionally, without giving further details here, 



s. 

 Time for PX at 120° = 1150 

 130=- = 1180 

 140°= 1190 



It is, however, probable that the adopted time at epicentre is in 

 en-or, in which case these are subject to a constant correction. 



Now, for reasons which need not be given here, it seemed possible 

 that PX might be an anomalous reflection of P by two very unequal 



