112 Dr. C. Davison — British Earthquakes, 1893-99. 



intersect, the intensities of both shocks were the same, and, if curves 

 (represented by dotted lines on the map) be drawn through the four 

 points of intersection on both the west and east sides of the disturbed 

 areas, the intensities of both shocks at any point of these curves 

 should be the same. These lines divide the whole area into three 

 districts. In the central district lie all the places where the first 

 shock was regarded as the stronger, and in the terminal districts 

 those where the second was the stronger, with one exception, that 

 of Wells. This city lies near the centres of both disturbed areas, 

 and, according to a newspaper report, the second shock was 

 severer than the first. Too much stress should not be laid upon 

 this exception, for there is always some variation in the estimates 

 of the relative intensities of two shocks where these do not differ 

 greatly. 



The explanation of the peculiarity is simple enough. The more 

 elongated form of the isoseismal lines of the second shock implies 

 that it had a longer focus. The overlapping towards the north and 

 south of the second disturbed area by the first shows that the initial 

 intensity of the first shock within a given area of the focus was 

 greater than that of the second. Thus, the second earthquake owes 

 its larger disturbed area simply to the greater length of its focus. 



Third Earthquake : Dec. 31, about 4 a.m. — A slight shock was felt 

 at Priddy, Street, Wells, and Wookey. At Street and Wells it was 

 accompanied by sound. So far as we can judge from the distribution 

 of these places it would seem that the epicentre lay to the west of 

 the epicentres of the two first shocks. 



Origin of the Earthquakes. — It has been suggested that the shocks 

 were not of seismic origin, but were caused by the fall of large 

 masses of rock in underground channels. If this were so, the 

 disturbed areas and isoseismal lines would have been approximately 

 circular and the sound would have been compared almost exclusively 

 to explosions or the fall of heavy bodies. The elongated forms of 

 the isoseismal lines and the general chai'acter of the sound are 

 therefore conclusive against such an explanation. 



On the other hand, all the phenomena known to me are in 

 complete accordance with the view that the shocks were caused 

 by slips along a fault, running W. 30° N. and E. 30° S., hading 

 towards the south-west, and passing through a point between one 

 and two miles north-east of Wells and between Wookey Hole and 

 Priddy. The Geological Survey map contains no fault in this 

 position, but it will be noticed that the direction assigned to the 

 earthquake-fault is roughly parallel to that of the Mendip Hills 

 in this district. The focus of the first shock must have been about 

 4 miles and that of the second about 6 or 7 miles in length. 



Fort Willia7)% Farihquahe : Jan. 12, 1894. 



Owing to its time of occurrence (about 11.50 or 11.55 p.m.) and to 

 the small number of villages within the disturbed area, I have only 

 been able to obtain 17 records from 14 places, in addition to 

 negative records from 5 places. I regret my failure especially in 



