482 THE JOURNAL OF GEOLOGY. 
hemisphere during the Glacial period. Also the Bryozoan group 
Cyclostomata are, as fossils, widely distributed in temperate 
regions, but the living forms of these are decidedly Arctic. 
The present distribution of animals shows many anomalies, 
many groups being present where naturally, according to their 
habits elsewhere, they ought to be lacking, and being absent 
where, according to analogy, they ought to be present. 
It is clear then that a study of the present distribution and 
habits of marine animals can throw little light on their history. 
But the geologist does not confine his observations to the present ; 
geologic history is a book whose leaves are scattered in many 
parts of the earth, and each observer gathers some of these leaves. 
A study of the distribution of animals in past time ought to 
explain their present occurrence, and the laws governing their 
distribution; and incidentally light should be thrown on the 
extinction of faunas, and upon changes in physical geography 
that accompanied and caused the migrations and extinctions. 
ZOOLOGICAL PROVINCES. 
Modern Provinces—One of the first facts in the study of geo- 
graphic zodlogy that attracted the attention of naturalists was 
that animals are not distributed strictly according to climate, or 
other physical conditions. This was inexplicable, according to 
the views then held. The work of Edward Forbes upon the 
Natural Ffistory of the European Seas began the scientific study of the 
distribution of marine animals. Forbes formulated the theory of 
generic and specific centers. ach genus and each species was 
to hima reality, and by distribution of these from the centers he 
satisfactorily accounted for their occurrence in other places. 
Although reasoning upon the theory of special creation for each 
species, Forbes recognized the fact that species and genera are 
never intermittent except locally. His results therefore do not 
differ in any respect from those obtained on the basis of develop- 
ment of one species out of another. He showed too that while 
separated regions under the same conditions show great similarity 
of faunas, this similarity is due rather to representative than to 
