138 CARNEGIE INSTITUTION OF WASHINGTON. 



distinct periods, with interesting lavas, which show very marked cycles, or 

 recurrent successions of characters, in their eruptions. These lavas are now 

 being studied, and are soon to be described, along with some forty analyses. 

 Taking these volcanoes as a text, some broad subjects of modern petrology 

 are discussed briefly. It is pointed out that, while no one sequence of types 

 is generally applicable, the sequence seems to vary with the magmatic 

 character and usually closes Avith basalts, though any generalizations must 

 be rather hazardous owing to the inherently accidental character of the 

 rock sequences observed by us. A change in the character of the magma 

 and in the volcanic cycles seems to be connected with a change in the mode 

 of volcanicity, the relation being possibly a causal one, but the inadequate- 

 ness of present data for such studies is pointed out. The so-called Atlantic 

 and Pacific tribes of rocks are briefly discussed and objections raised against 

 them, it being urged among other things that it is illogical and unjustifiable 

 to select only two types for contrast to the exclusion of others equally impor- 

 tant. In conclusion, the importance of further and more detailed sj^stematic 

 study of volcanoes, the need of numerous chemical analyses, the importance 

 of the application of physico-chemical research to petrological problems, and 

 the magnitude and complexity of such future investigations, are insisted on. 



(19) The hydrothermal formation of silicates, a review. George W. Morey and Paul 

 NiggU. J. Am. Chem. Soc, 35, 1086-1130. 1913. 



This is a discussion of the theoretical principles underlying the behavior 

 of water-silicate systems at temperatures ranging up to 500°, followed by an 

 annotated bibliography in which are assembled all of the data relating to 

 hydrothermal syntheses. These data unfortunately afford practically no 

 reliable quantitative information; qualitatively, even, they leave much to 

 be desired, for many minerals have been obtained but once by a given inves- 

 tigator, in a manner which was not reproducible and under conditions which 

 were not specified. The minerals which have been most commonly obtained 

 are chiefly those which are stable — or at any rate phanerostable — over a 

 wide range of conditions; for example, quartz and the feldspars. In all 

 cases the crystals obtained are very small, so that accurate chemical analysis 

 is usually out of the question; their identification by optical methods may 

 even be doubtful. 



Hydrothermal syntheses, like the paragenetic relationships investigated 

 by van't Hoff, are determined by the solubility relations of all the possible 

 solid phases which may be formed from the components present in the solu- 

 tion, even although the concentration of these components in the solution 

 is vanishingly small. Many of the reactions are, without doubt, practically 

 restricted to the solid phase, although they take place through the medium 

 of the solution. These solubility relationships are thus not simple; but 

 study of the question is further complicated by the frequent appearance of 

 metastable phases, a phenomenon which is coordinated with the rates of the 

 various possible reactions. Now, as is well known, rates of reaction are 

 often affected very greatly by factors which in other respects are of altogether 

 minor importance; hence slight differences (e. g., in the composition, or 

 even in the texture or fineness of grain, of the initial solid phase) may exert 

 considerable influence on the result. These considerations serve to show 

 that there may in certain cases be difficulties in the way of always being 

 able to reproduce a given result; in order to do this in any case, it is necessary 

 to control carefully the amount of water relative to the volume of the con- 

 taining vessel (the degree of filling), the temperature, and, if possible, the 



