i8 9 6. THE CONSTANTINOPLE EARTHQUAKE. 29 



intensity, but was at no time strong enough to overturn or displace 

 loose objects. A slight pause, and then came the second and most 

 powerful shock, vertical and rotatory in direction, and accompanied 

 as before by a deep rumbling noise. It grew gradually stronger, and 

 it was during the eight or nine seconds of its duration that nearly all 

 the loss of life and damage to property occurred. After a similar 

 pause followed the third shock, lasting five seconds, with the same 

 underground noise, undulatory, and towards the end horizontal in 

 direction. The total duration of the earthquake was thus about 

 seventeen or eighteen seconds. 



At many places within the first isoseismal the direction of the 

 shock was determined, and this was always found to be nearly 

 parallel to the shorter axis of the isoseismal. At others the shock 

 was vertical or rotatory, and, apparently, this was only the case in 

 the immediate neighbourhood of the epicentre. The reason of this 

 appears to be that the first vibrations come from the nearest parts of 

 the focus, and, therefore, in most neighbouring places, in a direction 

 roughly at right angles to it, while the later vibrations come from the 

 more distant parts and in a different direction ; so that it is only near 

 the focus that the rotatory motion should be distinctly observed. 

 In Prinkipos, the largest of the Princes Islands, a chimney was split 

 horizontally into three pieces, which were all twisted from the north 

 towards the east. 



Most of the secondary phenomena which generally accompany a 

 great earthquake were present in a more or less marked degree. 

 Some springs ceased to flow for a few hours, and afterwards returned 

 to their original condition. Others increased in volume, or the water 

 was disturbed. In several, though not many, places the ground was 

 fissured. The most important crevice observed was three kilo- 

 metres long, and 8 cm. in maximum width. This was in alluvial 

 ground, at Hambarly, to the west of Constantinople, and 300 metres 

 from the sea. It was, as usual, parallel to the coast, . and was, no 

 doubt, caused by the sliding of the unsupported mass. About three 

 miles from Kartal, that is, close to the longer axis of the first 

 isoseismal, the Kartal-Dardanelles cable, belonging to the Eastern 

 Telegraph Company, was broken in several places. The fractures 

 were quite clean, as if the cable had been cut with a knife, showing, 

 as Mr. Eginitis remarks, that they were not the result of a great 

 tension. He suggests that they were probably caused by the fall of 

 rocks, but bearing in mind the position of the fractures, is it not more 

 likely that they were due to a sudden displacement of the ocean-bed — 

 to the formation, perhaps, of a fault-scarp which was but the 

 superficial continuation of a slip that may have caused the shock ? 

 If this were the case, there would be some change in the depth of 

 that part of the Sea of Marmora, though it might be too slight to be 

 perceptible. Soundings were, indeed, made in this district, and they 

 differed from those given by the English Admiralty Chart, but 



