180 Scientific Intelligence. 



tangular axes in which the abscissas are the percentages of silica 

 and the ordinates are the ratios obtained by dividing the molecu- 

 lar proportions of the combined alkalies by the molecular propor- 

 tions of the silica. This point forms the locus of the small 

 diagram. 



In the first table are given the small colored diagrams of over 

 950 analyses of igneous rocks having over 28. per cent of silica, 

 selected to show all possible variations. The table is continued 

 and completed in plate viii, which shows the rocks with less than 

 28 per cent of silica. In another table are shown the loci of the 

 alkali-silica ratios of over 2000 analyses ; in this case the indi- 

 vidual diagrams are omitted and the loci shown by simple dots. 

 In other plates are shown the diagrams of analyses, disposed as 

 in plates i and viii, divided and arranged according to the new 

 quantitative chemical classification recently proposed by Cross, 

 Iddings, Pirsson and Washington. 



There are many interesting and instructive features shown by 

 these diagrams which must be seen to be appreciated. Perhaps 

 the most convincing thing they prove is the impossibility of any 

 "natural" classification of igneous rocks and that any division 

 lines must of necessity be wholly arbitrary ones. 



The work is certainly a tribute to the patience and industry of 

 the author, a reflection that is forced home when one thinks of 

 the vast number of analyses that have been computed and care- 

 fully arranged in these diagrams. Whatever evidence they pre- 

 sent certainly does not lack weight for want of facts to sustain 

 it. l. v. P. 



8. JPetrographisches PraMicum • von R. Rheiniscii. Part II, 

 Gesteine. Pp. 180, 8°. Berlin, 1904 (Gebr. Born traeger).— The 

 first part of this work has been previously noticed in this Journal 

 (vol. xiii, p. 243, 1902). The object of this elementary treatise is to 

 afford help to the beginner in the study of rocks ; it is not 

 intended as a complete text-book and its use demands some previ- 

 ous acquaintance with the elements of the science. The igneous 

 rocks, the sedimentary rocks and the crystalline schists are briefly 

 treated. The " dike rocks " are not assigned a separate position 

 but are mentioned under the abyssal granular types. The sys- 

 tem of classification followed is essentially that of Zirkel. Each 

 rock is briefly described, its mineral composition mentioned, a 

 few mass analyses given and the more important localities, espe- 

 cially the German ones, stated. The book is well gotten up and 

 appears carefully written and will no doubt prove useful in Ger- 

 manic countries. The thought that strikes one in connection 

 with it is, that since a knowledge of the subject and the use of 

 the microscope are implied, why not use one of the larger text- 

 books at once in which all phases of the subject are covered and 

 more information given. l. v. p. 



9. Pes Roches alcalines caracterisant. la Province petrogra- 

 phique cVAmpasindava y par A. Lacroix. Nouv. Archiv. du 

 Museum, 4 ser., v, 4°, pp. 171-254, pi. 14, 1903. — It will be 



