Reviews — J. R. L. Vogt — Rock-Magmas as Solutions. 133 



but it is only in comparatively recent years that the modern theory 

 of solutions, as developed by Ostwald, Van t-Hoff, Nernst, Rooze- 

 boom, and others, has enabled the question to be treated in a precise 

 manner. In his "British Petrography" (1888) Dr. Teall suggested 

 that the lowering of melting-point in the mixture of two or more 

 constituents might be applicable to rock-magmas, and that the order 

 of crystallization of rock-forming minerals might depend upon the 

 proportions in which they are present as compared with certain 

 determinate (eutectic) proportions. He instanced micropegmatite 

 as probably of the nature of a eutectic of felspar and quartz. The 

 experimental data for putting the theory to the test were at that 

 time lacking. They have now been in great measure supplied by 

 Professor Vogt of Christiania. The starting-point of his work is 

 the chemical and microscopical study of a large series of silicate- 

 slags, more or less closely comparable in composition with igneous 

 rocks. The results of such study, and the conclusions based on 

 them, afford a striking illustration of the service which may be 

 rendered to pure science by technological researches. 



The first part of the memoir is essentially a comprehensive inquiry 

 into the order of crystallization of minerals from magmas of known 

 composition. Taking, for instance, slags consisting mainly of 

 silicates of calcium, magnesium, and iron, without much alumina, 

 the results are plotted on a diagram, ordinates representing acidity 

 (i.e. the oxygen-ratio of silica to bases) and abscissas representing 

 the ratio Ca : Mg, Fe. Each slag corresponds thus with some point 

 on the diagram, and the mineral which first crystallized out is noted 

 (augite, enstatite, meliiite, olivine, etc.). It is found that the points 

 belonging to the several minerals arrange themselves in well-defined 

 fields on the diagram. The boundaries {Individualisationsgrenze) 

 between these fields represent eutectic proportions between the 

 several constituents, and from them the percentage composition of 

 the eutectics can be calculated. Thus, for augite and olivine the 

 eutectic proportions are 70 : 30. From a molten slag consisting 

 essentially of these two minerals in any other relative proportions, 

 the one which is in excess, as compared with the eutectic, will first 

 begin to separate. 



In the second part of the memoir the lowering of melting-point 

 in such mixtures is investigated, which gives ^ another way of 

 approaching the question. Here use is made of Akerman's deter- 

 minations of total heat of fusion (in which the melting-point is the 

 chief variable element) ; and the author has also made a series of 

 experiments by the method of cooling, noting the temperatures at 

 which latent heat is evolved. It is found that mixtures of two 

 given minerals show a very notable lowering of melting-point, the 

 greatest lowering being reached for certain definite relative pro- 

 portions. These are, of course, the eutectic proportions ; and the 

 close agreement of the results with those already obtained inde- 

 pendently by the ' individualisation ' method afford a strong proof 

 of the validity of the author's reasoning. The maximum lowering 

 of melting-point is sometimes measured by hundreds of degrees. 



