204 REPORT— 1897. 



The coasts taken are the east and west sides of South America and 

 Africa. The total length considered representing shores which are 

 steep and those which are gently inclined is about 11,000 miles. The 

 general result which is reached is that the dislocations per mile per year, 

 on the coast-lines considered, which may be taken as having on the 

 average a character similar to that of the coast-lines of the world, are 

 represented by the number 0"0023, that is to say, there is on the average 

 one dislocation for every 434 miles per year. If we increase this number 

 to 500 miles, and remember the character of the records and that of the 

 paths to which they refer, although we have attributed all the interrup- 

 tions to submarine change, we are inclined to the opinion that the 

 estimate is not too great. This being granted, then, as there are about 

 156,000 miles of coast-line in the world, if the same were surrounded by 

 loops of cal)les, although each section might be laid in the most favour- 

 able position, more than three hundred interruptions resulting from sub- 

 marine disturbance might be expected to occur every year. In deep 

 water on a level soft bottom experience shows that a cable may remain 

 undisturbed and unchanged for long periods of time, indicating, as we 

 have already pointed out, that geological change is proceeding with 

 extreme slowness. 



4. Conclusions and Suggestions for a Seismic Survey of the World. 



Because earthquake origins are more numerous beneath the sea than 

 ■upon the land, it is fair to assume that the bradyseismical operations 

 resulting in the folding, bending, crushing, faulting, and thrusting of rock 

 masses are more active in the recesses of the ocean than they are upon 

 our continents. Sub-oceanic volcanic activity, as, for example, that which 

 is met with in the mid- Atlantic, probably indicates the existence of 

 bradyseismic movement and a relief of strain. The concentration of de- 

 tritus derived from continental surfaces along coast-lines on tracts which 

 are comparatively small, indicates that beneath the sea the growth by 

 sedimentation is greater per unit area than the similarly estimated loss is 

 by denudation on the land. This rapid submarine growth, largely under 

 the influence of gravity, but modified by hydrodynamic action, leads to 

 the building up of steep contours, the stability of which may be destroyed 

 •by the shaking of an earthquake, the escape of water from submarine 

 springs, the change in direction or intensity of an ocean current, or by 

 other causes which have been enumerated. That submarine landslides of 

 great magnitude have had a real existence is proved for certain localities 

 fcy the fact that after an interval of a few years very great diSerences in 

 depth of water have been found at the same place, whilst sudden changes in 

 depth have taken place at the time of and near to the origin of submarine 

 •earthquakes (see pp. 193 and 197). Large ocean-waves unaccompanied by 

 volcanic action indicate that there have been very great and sudden dis- 

 placements of materials beneath the ocean. The most important evidence 

 of sub- oceanic change is, however, to be found amongst the archives of 

 the cable engineer. The routes chosen for cables are carefully selected as 

 being those where interruptions are least likely to occur ; and yet, as it 

 ihas been shown, something which is often of the nature of a submarine 

 landslip takes place and some miles of cable may be buried. Here we 

 seem to have proof positive, especially along the submerged continental 

 plateaus, of sudden sub- oceanic dislocation. Because these changes are 



