Augite from Vesuvius and Etna. 25 



pagne to keep up a dry raisin, which would otherwise 

 sink. {''Forsan et liaec olim meminisse juvahit/') 



The interesting point is that, with portions of the 

 magma of almost identical composition, we get, in the one 

 case, a typically; granular rock, and in the other, well- 

 formed, loose crystals. We have, unfortunately, no 

 detailed petrographical description of Lacroix's pyroxen- 

 ites. Of some of them he says (page 210) that they 

 are composed of ^^a little leucite filling the interspaces 

 of plates of biotite which surround automorphic crystals 

 of augite or inclose them poikilitically. " Of the most 

 pyroxenic type, that of which an analysis has been cited, 

 nothing is said as to the form of the augites ; but their 

 bein^ spoken of as granular (^^grenues") leads one to 

 think that the augites are xenomorphic. We know many 

 pyroxenites from elsewhere of this granitoid type of 

 texture and, so far as my experience of them goes, they 

 do not show evidence of being built around automorphic 

 and euhedral crystal cores ; though conceivably evidence 

 of such an origin may well have been obliterated in the 

 process of growth, if this were not zonal. 



It will be e^ddent that pyroxenites of the first type of 

 Lacroix, with automorphic augites, would be quite in 

 harmony with Bowen's theory of the settling of the heavy 

 crystals in a magma.^ Bowen studied the sinking of 

 olivine and pyroxene crystals, and the rising of those of 

 trid^/mite, in artificial melts, and the striking way in 

 which the first two sank and the third rose was sufficient 

 proof of the actuality of the phenomenon and the rel- 

 ative densities of crystal and liquid. So far as I know, 

 we have few data, at least exact data, on the densities of 

 liquid lavas, but the point arises as to whether the augite 

 crystals are really heavier than the liquid in which they 

 occurred. 



The density of the augite crystals was determined as 

 3.242 at 23°. The average density of the solid leucite 

 tephrite of Vesuvius^ may be taken as about 2.8, while 



^^ N. L. Bowen, this Journal, 39, 175, 1915. The sinking of crystals of 

 feldspar and their accumulation at the bottom of flows of obsidian were 

 observed by Von Buch (Geogn. Beschr. Canar. Ins., 229, 1825), who men- 

 tions experiments made by a M. de Dree, in which feldspar crystals settled to 

 the bottom of a crucible. The matter is discussed by C. Darwin, about 

 1844 (Geological Observations, 2d ed., 132-140, 1891), who attempts thus to 

 account for the differentiation of trachyte and basalt. 



' Cf . Eoth, Abh. Berl. Akad. Wiss., 1877, 13 ; and Zirkel, Lehrb. Petrog., 

 3, 19, 1894. Eoth gives the average as 2.77. 



