7° THE POPULAR SCIENCE MONTHLY 
remembered, however, that a boreal climate probably did not then mean 
a frigid climate, and that the differences in temperature were probably 
not so great as at the present time. 
The conclusions justified by the evidence from fossil invertebrates 
are: 
1. In the Paleozoic there is practically no faunal evidence of cli- 
matic zones comparable with those that now exist. 
2. In the Mesozoic there is a more or less definite zonal distribution 
of faunas which may be in part due to differences in climate but this 
conclusion in each case should be checked by the study of the floras and 
all other available lines of evidence. 
3. From the middle of the Tertiary on through the Pleistocene trust- 
worthy conclusions as to climatic conditions and changes can be made 
by direct comparisons with the distribution of living faunas. 
THE MIGRATION AND SHIFTING OF DEVONIAN FAUNAS 
By Prormssor HENRY 8S. WILLIAMS 
CORNELL UNIVERSITY 
N the year 1881 I presented before the American Association for 
the Advancement of Science the first definite announcement of 
the theory of recurrent faunas, applying it to the fauna of the Mar- 
cellus, Genesee and Ithaca black shales of New York, which I then 
conceived to be represented by the continuous fauna of the black shales 
of Ohio, Indiana, Kentucky and Tennessee; and also in the same paper 
the theory of shifting of faunas was applied to the Hamilton and Che- 
mung faunas of central New York.* Since that time a large amount of 
evidence has been accumulated confirming these hypotheses. 
The two hypotheses are correlated. Recurrence, or the departure of 
a fauna, its replacement by another and its final reappearance in the 
same section at a higher level, become the facts upon which the hypoth- 
esis of shifting of the faunas is based; and only on the assumption of 
the continuance and shifting of a fauna without losing its character- 
istics can we satisfactorily explain its recurrence. 
The following facts are among the more important which have come 
to light in the course of my studies : 
$1. The Catskill sedimentation was shown to be thicker and to start 
lower down in the geological column in eastern New York than in 
middle and western New York. In eastern, New York it began while 
the Hamilton marine fauna was still present and cut it off, bringing in 
estuarian conditions with a brackish water and land fauna and flora. 
In middle New York no Catskill sedimentation is present until after 
1 Proc. American Association for the Adwancement of Science, Vol. XXX., 
p. 186, ete. 
