78 



REPORTS ON THE STATE OF SCIENCE. — 'IQIO. 



For sudden motions we may neglect n in comparison with y, and we 

 have left 



So that the standard relation of signs is + ^ with — x. 



For very slow motions we may neglect ~ in comparison with n, and 



OX 



we have left 



Now let — cc = sin ft. 



n^d = — icjl 



then d = — xlnP = — ^„ sin vt. 



Thus 6 lags half a wave-length behind — x. The sign of either side of 

 equation (1) may be changed without affecting this result. Galitzin gives 

 a numerical table for this lag (' Seismometrische Tabellen,' Tabelle VI.), 

 as a function of the degree of damping and of the periods of the pendulum 

 and ground. But as he gives the lag as zero for waves of very long period, 

 instead of half-a-wave as found above, it is clear that he has adopted 

 a different definition of the direction on the paper, and that his definition 

 would give the reversed sign to sudden motions if it were applied to them, 



I 



■t- 



10 



Period of Earth Wove in seconds 



Fig. 3. 

 Lag of Milne-Shaw Seismograph, No. 3. July-August 1915. 



which probably is not intended. In preparing fig. 3, the lags in Galitzin's 

 Table VI. have therefore been increased by half the period of the earth- 

 wave. 



In order to make an observational comparison between the lag of the 

 Milne-Shaw instrument and that of the Galitzin instrument with galvano- 



