208 



BtTLLARD AND MASON 



[chap. 10 



this bank has yielded a variety of metamor|)hic and sedimentary rocks; many 

 of these are erratics, but it is beheved that some of the Hmestones are in place 

 and it seems probable that the bank is to be regarded as a part of the continent 

 separated from the mainland by faulting. Islands with great thicknesses of 

 coral also show less marked anomalies than those composed throughout of 

 basalt. Fig. 25 shows a map of Eniwetok Island which has a range of anomaly 

 of only a few hundred gammas. 



It is rare that sufficient is known about the form and susceptibility of the 

 rocks of a seamount for any meaningful comparison to be made between the 



30*40'N 



30*4tfN 



Fig. 22. Magnetic and topographic maps of Jasper Seamount. Magnetic anomalies in 

 gammas. Contour interval 50 y. Depths in fathoms. Contour interval 500 fm. (From 

 R. G. Mason, impublished.) 



observed and expected anomalies. The positive peak of the anomaly is fre- 

 quently displaced towards the magnetic equator, as it should be (see, for 

 example. Fig. 21), and a rough quantitative agreement with theory has been 

 obtained by Miller and Ewing (1956) for Caryn peak and by Keller et al. (1954) 

 for Bermuda. On the other hand, the positive peak is often displaced in quite 

 a different direction. Fig. 22 shows a plot of the field over Jasper seamount in 

 the Pacific in which the major positive peak is displaced 4 km to the ENE. In 

 such cases it is not easily decided whether this is due to discordant permanent 

 magnetization of the seamount as a whole or to inhomogeneity of its petrology, 

 and hence of its magnetic properties. 



In some cases the effect of the visible seamount may be overshadowed by the 

 effect of its roots on the normal crustal structure. Fig. 23 shows a plot of the 

 anomaly over a row of four seamounts in the north-east Pacific. It is dominated 

 by the magnetic lows to the south of the topographic peaks and the highs to 

 the north. At first sight it might appear that the seamounts are negatively 

 magnetized, but, when the observed anomaly is compared with the theoretical 

 anomaly computed for normal magnetization of the topography, it is seen that 



