222 T. GRAY ON A SEISMOGRAPHTC APPARATUS. 



Fig. 4. — Diagram of Continuous-motion Governor. 



Magazine' Nov. 1881, p. 358) to close an electric circuit which, ac- 

 tuating electromagnets, causes a mark to be made on the drum 1), 

 and at the same time the dial of the time- keeper T to move suddenly 

 forward against ink-pads fixed to the ends of the hands. In this 

 way the hour, minute, and second at which the mark was made on 

 the drum D is recorded ; and this, combined with the record of the 

 motion which is being written on the drum, is sufficient to determine 

 the time at which any particularly prominent feature of the shock 

 took place. 



The instruments just described have been made to suit the earth- 

 quakes which are commonly experienced in Japan, and are conse- 

 quently not adapted for use where the displacement of the earth is 

 great. A similar set of apparatus could, however, be made to record 

 earthquakes of considerable magnitude. The utility of having such 

 records of earthquakes will be readily admitted ; and I should call 

 special attention to the great value of the combination of the record 

 of the time of the occurrence with that of the motions. When the 

 time of occurrence of any of the chief features of the shock is known 

 for a number of places, it becomes a comparatively easy problem to 

 determine the origin of the shock and its rate of propagation. It has 

 been the usual practice in earthquake-observations, to take an approx- 

 imate measurement of the magnitude and direction of the movement 

 at any place and to deduce angles of emergence and direction of pro- 

 pagation. Now it is hardly necessary to point out that the direction 

 of movement at any point is very much influenced by the presence 

 or absence of surface undulations, and of refractions and reflections 

 from strata of different elastic moduli. I am of opinion that the 



