242 REPORT OF THE AMERICAN COMMITTEE. 
the Olean Conglomerate. Tin's interval i^ filled by barren shales 
and sandstones, sometimes red and green shales, and in places is 
but a very few hundred feet in thickness. 
In Eastern Ohio the Portage stage of the Chemung series is 
represented, but the Chemung fauna fails, and the black shale 
contains the last representative of the Devonian, followed by a 
Waverly fauna. This is substantially the case in Michigan. 
Farther west and south, in Indiana, Illinois, Missouri, Kentucky, 
and Tennessee, the black shale and its fauna may be regarded as 
terminating the Devonian, the next fauna above being the repre- 
sentative of the Ohio Waverly. Wherever, as in Iowa, the black 
siiale ceases, the Hamilton fauna is the last Devonian fauna, and 
above it comes the representative of the Waverly in yellow sili- 
ceous limestones or yellow calcareous sandstones, of slight thick- 
ness. In the sections in the extreme west, no distinct separation 
of the faunas at the top of the Devonian is reported, but there 
appears to be a gradual passage into a fauna which resembles our 
eastern Waverly. 
§ 5. The Subdivision of the Devonian. 
The subdivision of any great geological system must depend 
in great measure upon the differences in the faunas or floras con- 
tained, and, as I believe, this depends in no small degree upon 
the geological changes which modified and shifted the geograph- 
ical conditions under which the animals lived. A sandstone will 
not contain the same species as a following limestone, and even 
in arenaceous shales, a slight change in the fineness or coarseness 
or in the amount of argillaceous mud mixed with the sand, modi- 
fies the composition of the fauna contained in it. 
Again, we do not need to go outside of the United States to 
find four entirely distinct sections, each of which represents un- 
mistakably the Devonian system. These may be represented in 
the Gasp6 section, the New York, the Iowa, and the Nevada 
sections, no two of them presenting any features of resemblance 
in their lithological composition or order of sequence, and giving, 
in the organic remains they carry, evidence of entirely different 
biological conditions, but at the same time by various links of 
evidence they are believed to represent the same geological age. 
It is thus evident that all the divergent conditions (except those 
of separate language) which might call for heterogeneity of 
