GRAVITY WITHIN iNFRA-COtSTTlNENTAL BELTS 235 



separated from the surface by only a few miles of the much lighter crust 

 (specific gravit}^, 2.7), then any considerable vertical adjustment of adja- 

 cent blocks should disturb the gravity constants for points located near 

 the bounding faults. In Italy especially has a fourfold correspondence 

 of earthquake districts, fault zones, zones of abnormal gravity, and ab- 

 normal earth magnetism been established.* There is in this an ade- 

 qua:te explanation of the abnormal values of gravity within the infra- 

 continental zones, if it be true that the ocean basins be bounded by 

 dislocations on which large movements have occurred, f An extended 

 questionnaire by Milne has further shown J that of the earthquake sta- 

 tions supplied with magnetometers whose work is correlated under the 

 British Association, magnetometers are chiefly disturbed where the 

 gravity constants show abnormal values. Since 1888 continuous ob- 

 servations with magnetic instruments have been made in Japan, and since 

 1898 at five stations well distributed over the country. On several occa- 

 sions magnetic disturbances seem to have preceded or accompanied earth- 

 quakes.§ It has been shown that the depth or height from the station of 

 the source of the disturbance is small compared to the distance between 

 the stations. 



Genehal Absence op deep-sea Deposits from continental Areas 



The Challenger observations have satisfactorily explained this by show- 

 ing that no deposits are now forming over large areas of the sea-floor, 

 save only the red clay, which is probably largely of meteoric origin. 

 Corroded teeth of an extinct species of shark have been dredged up from 

 great depths, and show that for ages no deposits have there formed of 

 sufficient thickness to entomb them. 



Significance of faunal and floral Distribution 



It is the fourth of the arguments urged in support of the antiquity of 

 the ocean basins, namely, the restriction of related faunas and floras to 

 the same continent, which especially needs revision for the Tertiary and 

 later periods. In the language of Blanford,|| "If we wish to know any- 

 thing about ancient distribution of land and sea, we must scrupulously 



* A. Rlcco : Comptes Rendus Acad. Sci., Paris, vol. 137, 1903, p. 827. 



t See On some principles of seismic geology. Beitrage zur Geophysik, vol. 8, 1907, 

 chap. viii. 



t Jolin Milne : The Geographical Journal, London, 1903, pp. 15-18. 



§ Kikuchi : Recent seismological investigations in .lapan. Pub. E. I. C, foreign lan- 

 guages, no. 19, 1904, pp. 80-81. 



II Loc. cit., p. 84. 



