368 Messrs. Milne and Gray's Experiments 



It is to be observed that, as the record was written on a 

 strip of smoked paper wound round a drum which was started 

 automatically by the earthquake, the \ery beginning of the 

 earthquake is not shown. The waving line ABC shows 

 the successive motions of the earth in a direction N. 15° E. 

 magnified about ten times; while the line abc shows the 

 motions at right angles to that direction magnified to the 

 same extent. The spaces inclosed between the cross lines 

 were moved over in five seconds. The extremely gradual 

 manner in which the motion dies out is well illustrated in this 

 record. 



In Yedo, earthquakes showing east-and-west directions 

 of vibration have been proved by accurate time-observations 

 to have originated in the south. This would seem to indicate 

 that the normal vibration did not appear prominently in the 

 record. It is possible that in some cases the normal vibra- 

 tions are never very prominent, the original strain on the 

 material being more of a shear than a compression or extension. 

 If this supposition holds, it follows that we cannot with any 

 certainty make a deduction with regard to the direction of 

 propagation from the direction of motion. The direction of 

 original shear might be inclined at any angle to the direction 

 of propagation. 



Evidence that the motion is in some cases of a nature 

 similar to that here suggested has been obtained in the follow- 

 ing manner. Instruments were set up at several stations, 

 about 20 miles apart, round Yedo Bay ; and in conjunction 

 with the observations of these instruments the time of arrival 

 of the shock was observed at Yedo and Yokohama. From 

 the latter observations, it was found that the hyperboloid 

 which forms the locus of possible positions of the origin lies 

 between Yokohama and Yedo, and also that it is concave 

 towards Yokohama. These observations are corroborated by 

 the much greater intensity, greater verticality, and greater 

 number of the shocks as felt at Yokohama than at Yedo. In 

 some cases, however, the directions of motion at Yedo and at 

 stations situated at points in lines at right angles to the line 

 between Yedo and the probable origin, are found to be nearly 

 the same — thus indicating a disturbance which is propagated 

 in one direction as a normal vibration, while at right angles 

 to that it is almost a pure transverse or distortional vibration. 



The existence of such an irregular kind of disturbance is 

 rendered still more probable when we take into consideration 

 the highly irregular nature of the vibrations which are felt 

 both by our senses and by our instruments, and combine these 

 with the observation that the rocky masses in the vicinity of 



