

-402 moF. i\ MABSHALt on iitE SEQUENCE of [Dec. 19 14, 



For a long time it has been widely known that water- vapour 

 plays an important part in most volcanic action, and that it is a 

 component substance of most igneous magmas. Exception lias, 

 however, to some extent been made in the case of the lavas of 

 Hawaii, but it has lately been shown by Messrs. A. L. # Day & E. S. 

 Shepherd that water is a most important substance in that magma. 1 



At the present day we arc justified in saying that water is an 

 important constituent of all igneous magmas, except perhaps 

 the ultrabasic material. 



The kaiwekitc magma would therefore contain water, and in 

 virtue of the presence of this substance barkevikite might crystallize 

 in it. Little appears to be known at present regarding the rate of 

 resorption of such minerals, but there is no reason to suppose that 

 a period of rest at the level of the intermediate reservoir would not 

 allow of the complete resorption of all crystals formed under 

 plutonic conditions. This would be almost certain to happen, if 

 the distance of this reservoir from the surface, and other conditions, 

 allowed of the escape of water- vapour. 



After this complete refusion differentiation might take place. 

 This would be of a molecular nature, for no crystals would be 

 present in the magma. The low specific gravity of the phonolite 

 compared with that of the basalt, suggests that the force of gravity 

 would be the main agent in effecting this differentiation. To what 

 extent gravity might have been aided by local differences of tempe- 

 rature and by relative want of miscibility of the femic and salic 

 minerals, it is probably impossible to guess — at any rate, in this 

 instance. 



It thus appears that, although the mineralogical and textural 

 characteristics of the kaiwekitc and associated traclwdolerites 

 suggest at first sight that these rocks represent the final residue of 

 the magma after cooling and crystallization had proceeded almost 

 to their final limit, a closer consideration points to a conclusion 

 which is wholly opposed to that idea. It supports the belief that 

 these rocks in reality represent the original undifferentiated 

 magma, and that the phonolites of hue texture are actually the 

 products of molecular differentiation after the crystals that had 

 originally crystallized had been completely resorbed into the magma. 



The suggestion is thus made that the phonolites and basalts are 

 the products of a differentiation of an essexitic magma. This is 

 wholly opposed to the conclusion to which I came in 1900, when 

 discussing the nature and relations of the igneous rocks throughout 

 this district — for it was then stated that the trachydolerites (the 

 phonolitic type of Kosenbusch was then alone referred to this group) 

 and the basanite 'have resulted from the mixture of magmas 

 before ejection."' 3 The strong light that lias been shed on the 

 whole question from a prolonged study of the North Head causes 

 me to regard these intermediate types as resulting from the eruption 



1 Bull. Geol. Soc. Am. vol. xxiv (1913) pp. 573 GOO. 



2 Q. J. G. S. vol. lxii (190G) p. 421. 



