526 Washington — Contributions to Sardinian Petrography: 



of olivine from the magma liberated a certain amount of silica, 

 which would otherwise have formed metasilicate rather than 

 orthosilicate with the magnesia and ferrous oxide. The quartz- 

 bagalts described many years ago by Iddings and Diller * are 

 well-known cases of the simultaneous presence of olivine and 

 quartz. Another instance may be found in the sheets of 

 diabase which form the Palisades in New Jersey, where the 

 olivine has apparently settled to the bottom of the liquid mass, 

 forming a layer of olivine diabase, while the bulk of diabase 

 " is somewhat quartzose. " f 



Though the quartz is not actually visible in these Ferru 

 basalts, possibly because of the fine-grained character of the 

 interstitial groundmass, to which final stage of consolidation it 

 would belong, yet the comparison of the norms and modes 

 show that it must be present — forming an occult mineral, to 

 use a term recently introduced.^: It may be remarked that 

 this matter of occult minerals is one which has been much neg- 

 lected, but which is of much theoretic importance in the inter- 

 pretation of igneous rocks, and one for the study of which the 

 idea of the quantitative classification will be of great utility. 



In recent papers Bo wen and Andersen§ have clearly shown, 

 by physico-chemical study of the systems : MgO-SiO a and 

 diopside-forsterite-silica, that at a high temperature, that is an 

 early stage of the crystallization, olivine separates out, due 

 to dissociation of the molecule MgSiO s into Mg 2 Si0 4 + SiO Q , 

 and its relative insolubility. If the cooling be slow enough 

 this olivine will be gradually resorbed, the equation being 

 reversible. But if it be too rapid, part of the olivine will 

 remain, leaving an excess of silica in the magma, which must 

 crystallize as quartz or form a hypersilicic glass according to 

 circumstances. The embayed, corroded, and skeletal forms 

 of the olivine phenocrysts in the basalts here described are 

 fully in harmony with this interpretation, as well as is the 

 gradual transition from olivine-rich to olivine-free basalts of 

 what must be, judging from the microscopical examination, 

 much the same chemical composition. The common presence 

 of enstatite in these basalts is, of course, to be expected on this 

 hypothesis. 



Analcite-basalt. (III. 5-6. 2-3. 5). 



Small flows of what was formerly believed to be leucite- 

 basalt, but in which the so called leucite has been shown to be 



* J. P. Iddings, this Journal, xxxvi, p. 208, 1888 ; U. S. G. S., Bull. 66, 

 1890. J. S. Diller, U. S. G. S., Bull. 79, 1891. 

 f J. V. Lewis, N. J. Geol. Surv., Ann. Eep. (1907), p. 731, 1908. 

 % J. P. Iddings, Igneous Eocks, II, p. 19, 1913. 



§N. L. Bowen and O. Andersen, this Journal, xxxvii, p. 487, 1914. 

 N. L. Bowen, this Journal, xxxviii, p. 207, 191 4 



