ON SEISMOLOGICAL INVESTIGATION. 189 



of these have been attributed to the displacement of materials which have 

 been loosened by the submarine escape of fresh water. Examples of 

 springs of fresh water in bays and along coast-lines are numerous, whilst 

 there is abundant evidence of the absorption of rainfall and even of rivers 

 on continental areas, which in some instances it is suspected find an exit 

 in the sea bottom. Granted the existence of sub-oceanic springs, we see in 

 them at and near their exits a possible cause by which deposits may 

 be loosened and landslips take place. Under certain conditions such 

 dislocations might be expected to be periodical, following, for example, 

 the rainy seasons. Ocean currents which fluctuate in direction and 

 intensity, together with those of temporary character produced by the 

 backing up of water during gales in bays, estuaries, and coasts, may also 

 disturb the isostasy of submarine materials. 



For details of these and other operations producing sub-oceanic change 

 reference must be made to the writer's original paper. 



3. Cable Fracture. 



The fact that, on the level plains of ocean beds, cables lie for years. 

 and years without disturbance is another testimony to the facts brought 

 together by geologists to show that the flat plains of ocean beds are regions 

 where there is but little change. Directly, however, we approach sub- 

 oceanic banks or the margins of continental slopes, although the 

 depths may be abysmal, the fact that cables after interruption 

 have to be broken away from beneath materials which hold them 

 fast, indicates that regions of dislocation have been reached, and 

 what is true for these great depths is also true for localities nearer 

 land. Sometimes cables are bent and twisted, sometimes they are crushed. 

 Now and again sections are recovered which, from the growth of shells 

 and coral on all sides, show that they have been suspended. Others show 

 that fracture has apparently been the result of abrasions, whilst the ends 

 of wires, one of which is concave and the other convex, slightly drawn 

 out, indicate that yielding has been the result of tension. Needle-pointed 

 ends suggest electrolytic action ; ' but, although cable-interruption may 

 occur in these and other ways, the explanation which best accords with 

 the observations made during cable-recovery generally are those which 

 attribute their dislocation to sudden displacement of the bed in which 

 they are laid, or to their burial by the sliding down of materials from 

 some neighbouring slope. 



Sometimes it will be seen that earthquake movement and cable 

 fracture have been simultaneous, whilst many instances will be given 

 where an interruption has occurred at about the same time that an unfelt 

 movement has been recorded on land. These latter records, which in the 

 lists are marked with an asterisk, are unfortunately not numerous, and 

 only refer to days between the following dates : — 



1. Observations at Potsdam, Wilhelmshaven, Strassburg, Nicolaiew, Teneriffe, anrl 

 in Japan. These, which include many of the writer's observations, are published in 

 ' Beitriige zur Geophysik,' Band II., by Dr. E. von Rebeur-Paschwitz, March 27 to 

 October 5, 1880; January 4 to April 27, 1891; February 23, 1892, to August 31, 

 1893. . ^. ■„ 



2. Observations at Charkow by Prof. G. Lewitzky, August 4, 1893, to October s! 

 1894. 



' This may be due to electrolytic action between the zinc and iron of the 

 sheathing wires, or to the cable having rested on a mineral deposit. 



