556 C. Davison — The Exmoor Earthquake. 



that the originating fault must hade to the south. Moreover, the 



centre of the isoseismal iv lying half a mile south-west of Simons- 

 hath (or ahout 5f miles west of Exford), the fault-line must pass 

 a little to the north of this point, and thus its position is approxi- 

 mately determined. 



Origin of the Earthquake. — The geology of the district has been 

 studied recently by Dr. Hicks, though the complete results of his work 

 are not yet published. 1 On the map are shown by continuous lines 

 the north and south boundaries of the Morte Slates, so far as they 

 are given in the papers referred to. The discovery of organic 

 remains in these slates has led Dr. Hicks to regard them as the 

 oldest rocks of the district, the fossils in some of the beds indicating 

 " an horizon low down in the Silurian (Upper Silurian of Survey)," 

 while the rocks on either side, though belonging to varying 

 horizons, are assigned by him to the Devonian period. " A great 

 thrust fault extends continuously along the northern boundary of 

 the Morte Slates, from the coast near llfracombe to the Exe Valley. 

 On the south side there is evidence generally also of a well-marked 

 fault." Both the faults hade towards the south. 



Unfortunately, Dr, Hicks' map does not show the course of these 

 faults more than about two miles east of Challacombe, so that 

 they stop short of the epicentral area of the Exmoor earthquake. 

 It is evident, however, from its final trend, that the northern 

 boundary fault occupies very nearly the position required by the 

 seismic evidence. The dotted line on the earthquake map is drawn 

 so as to continue the line of fault in a direction parallel to the 

 axis of the disturbed area, and the true position of the fault-line 

 will, I believe, be found to be either along this line or, more 

 probably, a little to the north of it. 



If the two preliminary tremors referred to above wei - e of seismic 

 origin, the first movement seems to have been a slip along the 

 southern boundary fault in the neighbourhood of Arlington. This 

 was apparently followed, after less than eleven hours, by a slip 

 along the northern fault, somewhere near Berrynarbor, and about 

 twenty minutes later by the more marked but interrupted slip 

 which gave rise to the double series of vibrations of the earthquake 

 here described. That the length of fault over which the latter 

 slip occurred was considerable, is evident from the form and 

 dimensions of the isoseismal lines; possibly it may have been as 

 much as nine or ten miles. Whether the rocks on the north or 

 south side of the fault were displaced, is not quite clear; but, if we 

 may rely on the evidence of the two preparatory shocks, the more 

 probable alternative is, that the Morte Slates over a length of several 

 miles were forced bodily forward along the thrust-plane. 



1 "On the Morte Slates, and Associated Beds, in North Devon and "West 

 Somerset," Part I: Quart. Journ. Geol. Soc, lii, 1896, pp. 254 72. "The 

 Paheozoic Eocks of West Somerset and North Devon" : Proc. Geol. Assoc, xiv, 

 1896, pp. 257-70. 



