REVIEWS 169 



represent unfamiliar scenes, such as the four views of Taal in Luzon, or the 

 superb cone of Mayon in the same island, or the fascinating details of Tengger 

 Crater, Java (fig. 79), or the beautifully moulded cliffs of dissected rhyolites in 

 Colorado (fig. 77). We are spared the formal repetition of examples known to 

 Scrope and Lyell, and even our old friend, the nebula in Andromeda (fig. 25), gains 

 in glory by its representation in a fine photograph from the Yerkes Observatory. 

 The massive cloud above Mount Pelee (fig. 4) is perhaps not correctly described 

 as consisting only of " heated gases " ; but this reproduction from the work of 

 Lacroix is very welcome. The book is a treatise for readers who have got beyond 

 the stage of regarding " earthquakes and volcanoes " as inseparable occurrences, 

 designed to provide "copy" for the newspapers. It takes us back to origins, and 

 far below the surface on which lavas are outpoured. The planetesimal hypothesis, 

 which has so justly appealed to the scientific imagination, is traced, in its varied 

 aspects, from Kant in 1785 to Lockyer in 1890 and T. C. Chamberlin in 1897. 

 The possibility that the primordial crust of the earth was never hot is faced 

 on p. 51, the accretion of nebular particles having perhaps progressed at such 

 a rate that the temperature of the surface of the growing sphere was not raised 

 by the process. 



The observable facts of volcanism are features of the " lithosphere," which is 

 defined as " so much of the outer portion of the earth as is composed of rock 

 material having the physical properties of the rocks exposed to view at its surface. 

 The author warns us (p. 90) that the fusibility of the surface -rocks, as determined 

 experimentally, is not a safe guide in estimating their behaviour underground. 

 The liquid and gaseous components that may be present " materially modify the 

 solubility of the whole system." We think that such considerations offer an 

 answer to the doubts of Iddings, Harker, and others as to the likelihood of large 

 areas of the lithosphere becoming melted up by attack from igneous cauldrons ; 

 but Prof. Iddings himself is not willing to be persuaded (pp. 113 and 213). He 

 repeats in two places the statement that the composition of granite is not altered 

 by its being intruded into different kinds of rocks, such as limestone, sandstone, 

 shale, etc. This argument has been met by Goodchild's suggestion of " stoping," 

 which is strongly developed by Daly, and of which there is extensive evidence in 

 the field ; but work along the margin of a granite mass that crosses strata of 

 varied nature surely reveals features of actual absorption, such as local enrichment 

 in quartz from quartzite or biotite from hornblende-schist, which suggest that 

 extensive diffusion may have gone on. 



This, however, is a by-path of volcanism, and we may return to the point 

 (p. 103) that a small amount of water in rock magmas renders "them fluid at 

 temperatures considerably below the melting point of the rock minerals ; and 

 slightly more water probably renders them highly liquid and capable of 

 penetrating extremely thin fissures in heated rocks." When anhydrous minerals 

 separate out in the cooling magma, water accumulates in still larger proportion 

 in the uncrystallised material that remains, retarding crystallisation until 

 comparatively low temperatures are reached. Melts of quartz, orthoclase, or 

 albite (p. 106) are so viscous near the temperature of crystallisation that some- 

 thing which increases molecular mobility is necessary, to avoid the production of 

 a mere glass. The liquidity of lime-soda and lime felspars, on the other hand 

 and of augite and olivine, at temperatures just above their melting points, allows 

 of their ready crystallisation. 



The heterogeneous composition of that part of the lithosphere from which we 

 draw our lavas is presumed from their local variety, and absorption of differen- 



