SYSTEMATIC PETROGRAPHY 355 
appears to the reader that those of the present day are more 
severely criticised by the writer than those of earlier periods, it 
must be remembered that such a discrimination is natural and 
just. With greater knowledge of rocks, and with the experience 
of earlier systematists to guide him, the petrographer of the 
present day should be able to present a classification freer from 
defects than any proposed by his predecessors. It is the writer’s 
aim to avoid all personal animus in discussion of the existing 
systems of classification. 
Petrography may be concisely defined as the systematic 
classification of rocks. But if the principal handbooks of today 
be referred to, it will be found that it is a matter of supposed 
mutual understanding, rather than of precise statement, as to 
what constitutes a rock. It is commonly said that rocks are 
masses of geological importance in the constitution of the earth, 
but no very serious attempt 1s made to define what is of impor- 
tance, or to discuss what treatment shall be given to substances 
excluded under such a definition. For the purposes of this 
review, it will be sufficient to define rocks as the substances con- 
stituting masses cf geological importance, recognizing that gco- 
logical units of mass are not necessarily units in constitution, 
and may change in material character from place to place. It 
is the task of petrography to furnish a systematic arrangement 
of these natural objects and a corresponding nomenclature. 
The geologist may wish to consider rocks from many different 
standpoints, each requiring a more or less special grouping 
and appropriate terms to express the observed or assumed 
relationships, and such groupings may be apart from systematic 
petrography. 
VIEWS OF ROCKS BEFORE THE NINETEENTH CENTURY. 
The category of rocks as understood today is somewhat dif- 
ferent from the categories conceived to be the subject-matter 
of petrography in various epochs of the past. The early history 
of petrography is therefore in large degree a history of the 
differentiation of the sciences dealing with various categories of 
