118 PETROLOGY OF THE ALKALINE ROCKS 



Prior has shown that these highly differentiated ferriferous basalts and limburgites 

 abound in many Antarctic islands. Such rocks occur also in Greenland and 

 Spitzbergen, and on basaltic volcanoes arising out of deep ocean. 



It is not unlikely that the greater proximity of the Polar regions of the earth, and 

 of the magma caverns on the floor of the ocean, to the earth's centre may stimulate 

 gravitative differentiation. So evenly are the various forces producing differentiation 

 balanced that a comparatively slight change in the conditions under which cooling 

 takes place may cause one force to greatly predominate over all others. 



The same highly differentiated nature of Antarctic eruptive rocks is evidenced by the 

 alkaline, subalkaline, and basic lavas obtained by Borchgrevink at Cape A dare, and 

 described by David, Smeeth, and Schofield (cf. J. 1, and Proc. Roy. Soc. N.S.W., 1895). 



There seems to be but little difference, if any, between the alkaline volcanic rocks 

 of Ross Island and those of the mainland. 



The sequence of these lavas appears to have been the same as that which is so 

 characteristic of the Australian alkaline province — that is, from acid to basic. 



The fact that the crater on the summit of Erebus is still erupting kenyte, while at 

 Cape Barne basalts are the latest effusive products, I do not take to indicate that these 

 lavas come from different magmas, but rather that the vent which gave the Cape 

 Barne lavas was supplied from a deeper source than the Erebus crater. The molten 

 magma has differentiated in the magma reservoir, and probably tongues of differentiated 

 rocks have been thrust into surrounding formations. Fissures may tap one or all of the 

 differentiation products, but the more elevated vents situated over the centre of the 

 magma chamber will tap the lavas in the order of their specific gravity, and will continue 

 to pour out each type of lava much longer than the subsidiary vents and fissures. Thus 

 if Cape Barne represents lavas poured from the fissure which has tapped sill- like 

 offshoots of the main reservoir, the whole sequence might be completed in this locality 

 while the summit of Erebus is still in the kenyte phase of eruption. The summit may 

 become extinct before emerging from this stage, the forces being insufficient in the 

 final stages of eruption to raise the subjacent basic lavas to the summit, so that these 

 may never reach the surface except by side fissures. 



Yet the order of eruption would be from acid to basic wherever several lavas have 

 flowed from the same vent, and where flows from different vents have not intermingled 

 or overlapped. Such seems to have been the sequence of the lavas of Ross Island as 

 far as can be determined on present knowledge. 



THE VOLCANIC ROCKS OF CAPE BIRD 



The Cape Bird rocks of the Expedition's collection were, according to Mr. 

 Priestley and Professor David, mostly water- worn and ice-carried specimens collected 

 at the foot of Cape Bird, which is a seaward projection of Mount Bird. None were 

 obtained in situ.* 



Being derived from an independent focus of eruption, a special interest attaches 

 to them and to their comparison with the rocks of Mount Erebus. 



The Cape Bird rocks are intimately related to those of Erebus, being alkaline in 

 facies. Their distinguishing feature is, however, in all varieties, the abundance of 

 brown basaltic hornblende and alteration products after this mineral. 



The alteration products consist of magnetite, colourless augite, red hypersthene, 

 and apparently in some cases a red titaniferous hornblende, which may be a kaersuetite. 



* Several of the specimens collected were in situ from the agglomerate formation at the landing- 

 place.— D. M., 1916. 



