No.41] REVIEWS OF RECENT LITERATURE. 235 
Pieri and Dubois that it belongs to the enzymes lacking in support 
as yet. 
The observations which have just been briefly summarized add to 
the accumulating evidence that fertilization may be essentially a 
process of stimulation of the egg. This, of course, is true only if one 
leaves out of account the contribution of hereditary tendencies made 
by the spermatozoón, which add to the variability of the offspring. 
The egg, however, apparently contains everything necessary for the 
production of a new and complete organism like the mother, and 
needs only appropriate stimulation to start it on its course of develop- 
ment (see Koulagine, 1898, Zool. Anz., 21, p. 653 ; also Delage, 1900, 
Archives de Zool., Exp. (3), 7, p 525). In this direction point also 
the observations of Tichomirov, who showed that the unfertilized 
eggs of Bombyx mori can be made to develop, at least to an advanced 
embryonic stage, either by dipping them in sulphuric acid or by brush- 
ing them; likewise thé observations of Dewitz, who obtained cleavage 
of unfertilized frog's eggs by treating them with corrosive sublimate ; 
likewise the observations of Hertwig, who observed a similar result 
following the treatment of sea-urchin eggs with strychnine ; further, 
those of Koulagine (already cited), who has induced cleavage of 
certain fish and amphibian eggs by treating them with the antitoxin 
of diphtheria; and those of Klebs, who found that parthenospores 
are formed in conjugating filaments of Spirogyra, in the presence of 
salt or sugar solutions of appropriate density. Still more emphatically 
is this indicated by the recent work, already mentioned, of Morgan, 
and especially of Loeb. W. E. C. 
PETROGRAPHY. 
Graphic Representation of Rock Composition and Schemes of 
Rock Classification. — The recent great increase in our knowledge 
of rocks and the complexity of their chemical relationships is responsi- 
ble for the attempts now being made to indicate by graphic methods 
these relations and incidentally to classify rock magmas on some 
chemical basis. The differential hypothesis was intended to explain 
the cause of the chemical relations existing between rock masses. 
It cannot serve, or at any rate it has not served, as a basis for rock 
classification. Rocks as objects of study are but portions of differ- 
entiated masses. A rock classification must deal with these portions, 
and it must necessarily deal with well-characterized portions or 
