70 



REPORTS ON THE STATE OP SCIENCE. — ^1915. 



S-P=662s. and A=90-8° (94-8°). The time for PR, would then be 

 12 h. 10 m. 35 s. (12 h. 10 m. 56 s.). For SR, we slioiild get 12 h. 24 m. 31 s. 



W"", 15'^ 16 



/3^ /3^ 



Eskdalerauir (Galitzin) N.S. Component, November 24, 1914. 



and there is a notable disturbance on the trace here (though it has 

 not been considered necessary to reproduce that part of it in the diagram). 



This argument has been much strengthened since the records of other 

 stations became available. These were collated by Mr. Burgess in the 

 ordinary course of the work at Shide in July 1915 without any knowledge 

 of the suggestion here made. The epicentre selected by him is 21 -5° N. 

 141° E. distant 97-3° from Eskdalemuir ; and the MS. sheet prepared 

 for the Shide Bulletin with this epicentre shows residuals in S of over 

 100 s. not only for Eskdalemuir and Dyce (Aberdeen), but for Padova, 

 Paris, and other stations. It would appear that y has been taken instead 

 of S at such stations, for the above epicentre, though possibly a few 

 degrees in error owing to the conflicting information, suits the near stations 

 (Zi-ka-wei, Batavia, and the Japanese stations), as well as Pulkovo, too 

 well to be so far wrong as must be the case if y were S. 



But if 8 is S, what is y ? Taking A=(94-8°) as above, the time 

 for the polychord is 94 •8x14-9 seconds=1413 s., following P by (606 s.), 

 and thus affecting the trace at (12 h. 16 m. 49 s.). The Eskdalemuir 

 reading for S (i.e., for y) is 12 h. 16 m. 56 s. 



The full details for this earthquake, which is not at present fully 

 discussed, will be printed in the Shide Bulletin ; but the following figures 

 will show approximately the nature of the results : 



Epicentre 24" N., 141° E. Time, 1914, Nov. 24, 11 h. 53 m. 16 s. 



