﻿32 DR. CHARLES DAVISON ON [Feb. I905, 



that of the stronger part ; in other words, that the weaker part of 

 the shock was connected with the deeper focus. 



Hade of the Fault within the Twin-Foci. 



On this point we have little knowledge. In the Leicester earth- 

 quake of 1893, the fault hades to the north-east within the north- 

 western focus, and in the opposite direction in the south-eastern 

 focus. In the Hereford earthquake of 1896, the hade within the 

 north-western focus is to the north-east, and within the south- 

 eastern focus probably in the opposite direction. There is, however, 

 no evidence of any such change of hade in the Carlisle earthquake 

 of 1901, or in either of the Derby earthquakes of 1903 and 1904. 



VII. Origin op Twin-Earthquakes. 



In a simple earthquake, the immediate consequence of the parent 

 fault-slip is a change of stress within and near the focus, especially 

 an increase of stress along its margins. The after-slips conse- 

 quently take place, either in the focal region or just beyond it. 

 Thus, the foci of successive earthquakes are not detached, but are 

 either coincident or overlapping. 



A twin-earthquake is clearly of a different and more complex 

 origin. Through a single effort, movements occur almost simul- 

 taneously in two distinct regions of the fault ; and these regions are 

 probably situated at different depths, and are separated by a portion 

 of the fault in which there is little or no sensible displacement. 

 Such a movement could hardly be due to an interrupted slip ; for 

 this would involve an interval of time between the component slips 

 long enough for the increased stress resulting from the first slip to 

 take effect in the second focus, and therefore longer than the time 

 of transit between the two foci. 1 It would, however, be caused by 

 the growth of a fold cut transversely by the originating fault ; a 

 growth that would render the anticline of the fold more anticlinal 

 and the syncline more synclinal, while the middle limb would 

 remain practically undisturbed. The seismic focus would thus 

 consist of two detached portions situated at different depths. 



Now, in recent British twin-earthquakes, the distance between 

 the epicentres ranges from 4 to 23 miles, and, on an average, is 

 about 10 or 11 miles. If, then, the above explanation be the 

 correct one, this average should not differ much from the mean 

 distance between successive anticlines and synclines. No series 

 of measurements have, I believe, been made of this distance for 

 British crust-folds ; but a rough estimate can be obtained from 

 Prof. Marcel Bertrand's map of the synclinal folds of France. 

 Along several different lines, varying in length from 127 to 442 

 miles, the average distance between successive synclines lies between 

 18 and 24 miles, or the average distance between successive anticlines 



1 The Charleston earthquake may have originated in this manner. 



