38 JOSEPH BARRELL 



stress-differences for this is shown in curve E, Fig. i8. But if the 

 rock has a density of 2.67 and the sea- water a density of 1.03, 

 this corresponds to an amplitude beneath the ocean surface of 

 6,513 m. of uncompensated rock. This is only about two-thirds of 

 the maximum reHef which is observed, so that it is well within the 

 limits of possibility. These few available figures suggest that the 

 sharp submarine ridges and deeps may not be more than one-third 

 or two- thirds compensated. 



The Niger delta. — Reverting to the discussion of the Niger delta 

 given in Part I, it is seen that there is no evidence of depression 

 around its margin. It may be taken then as the positive half of 

 a harmonic wave well within the Hmits of crustal strength. If the 

 section of the delta be taken as given in Figs. 3 and 4, pp. 31, 43, it 

 is seen that the load is disk-like in form, instead of being indefinitely 

 elongated at right angles to the section in accordance with the 

 form of a zonal harmonic. It seems likely, because of these two 

 departures from the nature of a harmonic series, that the stresses 

 beneath it are not more than half of those which would be given 

 by the completed harmonic curve. As it is merely the order of 

 magnitude of the stress-differences which we may hope to attain 

 we may proceed in accordance with these rough assumptions. 

 It is seen that the section of the delta shows a half wave-length 

 of about 300 km. and a maximum thickness equivalent to 1,650 m. 

 of rock upon land. This corresponds to the half-amphtude or 

 thickness above the mean plane. If half of this is taken as a 

 measure of the stress, it gives a wave-length of 600 km. and a 

 total amplitude of 1,650 m. The stress-curve for this harmonic 

 series is 60 per cent larger than the stresses due to the outstanding 

 masses of the same wave-length as given by the geodetic evidence 

 in the United States and shown in curve B, Fig. 18. As the esti- 

 mate from the Niger delta is very imperfect and unchecked by 

 pendulum observations reduced by the New Method, the stress- 

 curve is not plotted. 



The existing continental ice sheets. — Two ice sheets of sub- 

 continental proportions remain in existence, the Greenland and 

 Antarctic. They form great plateaus sloping upward from the 

 margins; the Greenland sheet reaching elevations at its center 



