PUBLICATIONS. 857 
shows his appreciation of the broad field of the science, which, as he 
says, is not merely an attempt to discover the composition of a rock, 
but to unravel its history as well. His clear understanding of the sub- 
ject is also shown in his discussion of the classification of rocks, espe- 
cially those of igneous origin. In the present chaotic condition of 
the nomenclature of rocks it will be difficult for any one, who does 
not succeed in reforming the whole system, to classify rocks to his 
own complete satisfaction or to the satisfaction of anyone else. In 
his attempt at simplification Mr. Harker has shown his independence 
to a considerable extent, while following in the main the classification 
of igneous rock adopted by Rosenbusch, though under a different 
terminology. Thus massive igneous rocks are subdivided into plutonic, 
tntrusive and volcanic, corresponding closely to sefengesteine, gang- 
gesteine and vulkanischegesteine. In many other ways also the author 
follows the methods and principles of Rosenbusch. Under each of 
the three great divisions above named the rocks are arranged accord- 
ing to their mineralogical or chemical composition beginning with the 
most acid. The names used for varieties of rocks within different 
families are generally those expressing the mineralogical characteristics 
of the particular variety rather than those of a geographical character, 
which may already be in common use. But in most cases both names 
are given. The most noticeable instance of this is in the treatment 
of the peridotites. 
In substituting the term z#fruszve for that of ganggestemme and in 
maintaining an independent grouping for certain varieties of intrusive 
rocks the author has not improved on the presentation of the case as 
made by Rosenbusch, and his remarks in introduction of his zx¢ruszve 
division are in the nature of an apology. Nor does his use of the 
term, acid intrusives, in distinction to that of porphyries and porphy- 
rites, appear to be fortunate. Diabases are classed as intrusives. 
Under volcanic rocks no distinction is made between older and younger 
lavas, which certainly seems to be the only proper method of treatment. 
In this respect the classification follows the English usage. The frag- 
mental products of volcanic action are described in connection with 
sedimentary rocks. 
The descriptions of the various rocks embrace a general definition 
in mineralogical and structural terms, followed by an account of the 
constituent minerals and of the microstructure. Illustrative examples 
are chosen as far as possible from occurrences in Great Britain. The 
many references to the writings of British geologists and numerous 
