Dr. C. Davison— British Earthquakes, 1893-99. 167 



of the disturbed area. The centre is 1^ miles S. 12° E. of Launceston, 

 and, as will be seen from the map (llg. 3), the sound-area overlaps 

 the disturbed area by u miles towards the west and by 1;^ miles 

 towards the south. 



Very few observers describe the shock. It was clearly a mere 

 tremoi', a dull shiver according to one, and a movement such as 

 is pi'oduced in a building by a heavy peal of thunder according 

 to another. 



The sound was, as usual, a far more prominent feature of the 

 earthquake. Comparisons to well - known sounds are made in 

 30 cases, in 27 per cent, of these to passing waggons, etc., in 57 

 per cent, to thunder, and generally distant thunder, and in 17 per 

 cent, to explosions, which were, however, either prolonged or 

 followed by a rumbling noise. Thus, the sounds must in all parts 

 have lasted several seconds. 



Origin of the Earthquake. — The longer axis of the disturbed 

 area and sound-area agree in assigning an east and west direction 

 to the originating fault. Its hade must be to the north, for the 

 disturbed area is overlapped by the sound-area towards the south. 

 The fault-line must therefore pass a short distance to the south of 

 the centre of the disturbed area, possibly not very far from Dunterton, 

 Landue, and Lezant. 



The horizontal length of the seismic focus must have been about 

 5 miles, but the extension of the sound-area towards the west shows 

 that the western margin of the locus was much longer than the 

 eastern margin. The displacement throughout the whole focus was 

 very small, and not much greater than that which will produce the 

 sensation of sound. 



In this case, again, the Geological Survey map provides no fault in 

 the required position ; but the early date of the survey should 

 be borne in mind. To the south of the disturbed area there are, 

 however, numerous mineral veins and elvan dykes, all of which run 

 east and west, and therefore parallel to the earthquake-fault. If the 

 district were to be carefully re-surveyed, a fault would no doubt 

 be discovered that would satisfy all the seismic conditions. 



Annandale Earthquake : May 29, 1896. 



The disturbed area of this earthquake occupies a more populous 

 part of Annandale than those of the shocks felt on March 8 and 

 May 14, 1891. The number of records which I possess, is, however, 

 small, only 15 from 14 places ; but there are several other places 

 from which negative records have come. The time of occurrence, 

 as determined by a signalman at Castlemilk station, was 4.47 a.m. 



The shock was distinctly felt at 7 places which lie within an 

 area roughly circular in form and about G miles in diameter. The 

 centre of the area is in lat. 55° 6' N. and long. 3° 17' W., a,nd is 

 equally distant (3 miles) from Lockerbie, Ecclefechan, and Minsca. 

 The sound alone was observed at several places outside the boundary ; 

 at Craighouse close to the boundary on the south-east side, Dormont 

 1^ miles to the south-west, and Corsegreen 3 miles to the west. 



