154 EEPORT — 1894. 



The values of the velocity in this table have been obtained on the 

 supposition that all phases of the disturbance left the origin simul- 

 taneously at the initial times given above. This is of course in the 

 highest degree improbable, but it was so taken in the absence of any 

 certainty as to which phase corresponded to these times. It is evident 

 that the discordances between many of the above results must in a great 

 measure be due to tliis assumption. For instance, on April 27, the 

 beginning of the pulsations was registered at Charkow one minute sooner 

 than at Nicolaiew, though the former place is 460 km. further from the 

 epicentrum. Both stations being observatories, the explanation appears 

 to be that the pendulum at Charkow was affected by earlier pulsations of 

 smaller amplitude.' 



The most probable estimate of the velocity is, I believe, that derived 

 from the epochs of the beginning of the larger pulsations. Including 

 those obtained from magnetographs, which are not disturbed by the small 

 initial movements, we have the following results ^ for the mean velocity : — 



April 20. 2-08 ± -08 km. per sec. 

 „ 27. 3-21 ± -07 km. 



Future Work of the Committee. 



The grant of 50Z. awarded last year to the Committee has been spent 

 in providing for the foundation, &c., of the bifilar pendulum at Bir- 

 mingham, and (in part) for one of the improved pendulums with photo- 

 graphic recording apparatus, to be placed in that city under the charge of 

 the Secretary. 



The Committee consider that it would be desirable to test the working 

 of the pendulum by placing another of similar construction at a short 

 distance from it. The comparison would probably be made for a year, 

 and the second instrument would afterwards be available for use else- 

 where. They accordingly request that they be reappointed, with a grant 

 of 100^. 



it follows that either (1) the instrument there was only affected by the larger 

 oscillations, or (2) that the rapid vibrations which constitute the earthquake shock 

 were distinct from the pulsations, and that the former alone were registered. The 

 initial times above given were obtained on the supposition that the former alterna- 

 tive is correct. It seems possible, however, that the pulsations are not merely the 

 distant equivalent of the shock, but that they may travel with a diflferent, and pro- 

 bably greater, velocity. If this be the case, the estimates of the velocity may be 

 a little too great. 



' If the velocity of the pulsations is independent of their amplitude, these 

 small pulsations must have left the origin more than sis minutes before the larger 

 ones (as if the earth's crust slowly quivered before giving way), and might possibly 

 be utilised for giving earthquake warnings (see Professor Mikie's suggestion in 

 Seismol. Journ., vol. i. 1893, pp. 10-15). 



^ The first of these is calculated from six observations (Nos. 11, 14, 25, 27, 

 31, 33), the second from thirteen observations (Nos. 8, 11, 14, 26, 27, 29, 31, 32. 33, 

 36,37,38.41). 



