252 Tlte Anici'iran (rrolor/isf. October, ism 
Prof. C. \\. Xan Hise spoke of the methoils of correlation employed for 
pre-Cambrian rocks, which occur in widely separated areas and are 
devoid of fossils. Physical data only are available for correlatins^ these 
formations. Experience has shown that among all physical methods, 
unconformity is by far the most important. Other physical criteria, such 
as the degree of induration, metamorphism, and relation to eruptives, 
are valuable for the subdivision of single areas, but cannot be safely 
used in identifying synchronous formations in widely-separated areas. 
The idea that lithological character is any direct proof of geological age 
has retarded tlie scientific subdivision of pre-Canibrian rocks. The re- 
searches of Puiiipelly and others in the eastern United States have 
demonstrated that Silurian, Devonion, and even Carboniferous deposits 
might become, under certain physical conditions, as highly crystalline as 
much more ancient rocks of the West. For this reason it has been found 
necessary to abandon such term as Huronidn a.ud Koneemncini. Evidences 
of life are not lacking in pre-Cambrian rocks, and it is to be hoped that the 
paleontologist will succeed in differentiating several separate formations 
below the Cambrian, as the Cambrian itself was differentiated from the 
base of tiie Silurian. 
SESSION- OF AUG. 29, 1891, 10 A. M. 
Prof. Albert Gaudry, Vice President, in the chair. 
3Ir. S. F. Emmons, General Secretary, made announcements. 
M. Alexis Delaire presented two communications in behalf of prince 
Roland Bonaparte relating to the phenomena of the Aletsch glacier and 
upon an excursion to Corsica. 
Prof. Chas. Barrois presented a communication on behalf of Prof. 
Michel Lt''vy upon the geologic history of the Auvergne volcanoes, con- 
taining a classification of eruptive rocks as represented by sjMubols. 
Prof. E. AV. Hilgard laid stress upon the importance of the abundance 
or scarcity of species in the correlation of strata. He thinks some 
quantitative estimation of the species should be made. He is of the 
opinion, also, that as compared with marine fauna, plants have but little 
value, for purposes of correlation owing to their local distribution, their 
accidental proximity to water, transportation, and preservation. Plants 
can be so used only after large areas are worked over. 
Prof. Zittel was called to the chair, and Prof, f^ester F. Ward then con- 
tinued the discussion. He developed two of the more general princi- 
ples of correlation by means of fossil plants, as follows: 
I. That the great types of vegetation are characteristic of the great 
epochs in geologj-. 
This principle is applicable in comparing deposits of widely different 
age when the stratigraphy is indecisive. For example, even a small 
fragment of a Carboniferous plant proves conclusively that the rocks in 
which it occurs are paleozoic, or a single dicotyledonous leaf proves 
that they must be as late as the Cretaceous. 
II. That for deposits not thus widely different in age, as for example, 
within the same geologic system or series, ample material is necessar}' 
to fix their position b}- means of fossil plants. 
