ON SEISMOLOGICAL INVESTIGATION. 197 



increase in depth of over 2,400 feet, and in another of 1,200 feet ; whilst 

 in the case of the 1889 disturbance, eleven and a half minutes later, 

 unfelt earth-waves of considerable magnitude were recorded at Wilhelms- 

 haven, 1,732 kilometres distant. Similar unfelt movements have also 

 been recorded at distant places at about the time when cable-interruptions 

 took place, in every instance where we have been able to make comparisons. 

 The conclusion, then, is that in this region earthquakes occur, producing 

 beneath the ocean what is equivalent to the landslips which similar move- 

 ments produce on land. 



Bay of Biscay. — About 1875 the Direct Spanish cable was broken 

 about 150 miles north of Bilbao by what seemed to be a submarine 

 landslip, which may have been produced by an undercurrent produced by 

 the piling up of the surface waters under the influence of a westerly gale. 

 The soundings showing the neighbourhood of the interruption indicate 

 slopes of 1 in 7 and even 1 in 3, and it is therefore a district in which 

 landslides and dislocations might be expected to occur. From Mr. R. 

 Kaye Gray I learn that the 1872 Bilbao cable broke down periodically — 

 usually in the month of March, with or after a heavy north-west gale. 

 This took place about 30 miles to the north of Bilbao, and, when repairing, 

 it was invariably found that 4 or 5 miles had been buried. The cause of 

 these interruptions was attributed to a heavy submarine current caused 

 by the piling-up of surface water, cutting the prolongation of a river-bed 

 with steep walls which, when undercut, fell in masses to bury the cable. 



St. Thome — St. Paul de Loanda. — Interruptions which have been 

 noted on this section wei*e as follows : — 



January 22, 1892; September 13, 1892*; November 24, 1892*; February 17, 

 1893*; April 11, 1893*; May 30, 1893*; February 5, 1894*; January 22, 1895* ; 

 January 15, 1896* ; May 2, 1896* ; June 15, 1896.* 



The dates on which unfelt earthquakes were recorded were as 

 follows : — 



September 13, 1892. At Strassburg a very large disturbance from 9'54h. to 

 13*31h. Origin unknown. 



February 16, 1893. At Strassburg at 008h. Origin possibly in Japan. 



April 11, 1893. At Strassburg and Nicolaiew, 18-58h. to 19h. Moderate. On 

 April 8 at these stations there was a heavy movement from l-871i. to i'lTh. Origin 

 unknown. 



May 30, 1893. At the above stations from 4 33h. to 5-32h. ; a great movement. 



February 5, 1894. At Charkow from 4h. 54m. to lOh. 34m. there was a strong 

 movement. 



January 18, 1895, 2h. 37m. At many places in Italy. 



January 15, 1896, 7h. 10m. At many places in Italy. 



May 2, 1896, lb. 20m. Strong through Europe. 



June 13, 1896, 14h. 54m. Strong through Italy. 



June 14, 1896, 22h. 46m. Strong through Italy and at Shide. Origin, Pacific 

 Ocean. 



We have therefore ten cases of interruptions on or near to the dates 

 of nine of which large earthquakes were recorded. It is difiicult to 

 imagine that this particular district should be characterised by any 

 seismic activity, but it seems possible that, if it is a district where 

 sediments rapidly accumulate to attain an unstable form, these might 

 from time to time give way under the influence of earth-waves originating 

 at a great distance. 



On this particular section Mr. R. Kaye Gray points out that, from 



