MICHIGAN ACADEMY OF SCIENCE. 
107 
SOME CONSIDERATIONS CONCERNING THE PLACE AND 
ORIGIN OF LAVA RESERVOIRS. 
W. II. IIOHBS. 
The ideas which are generally held concerning the origin of lava have 
all gone out from the idea of a liquid interior to the earth. It lias, how- 
ever, been shown in recent years that the earth cannot have a fluid in- 
terior, but on the other hand must be as rigid as a ball of glass of the 
same size. Within the earth the temperatures of the rock would easily 
melt it under surface conditions, but the pressure from the super- 
incumbent load elevates the point of fusion and so keeps the rock rigid, 
or as we might say, “solid” though realizing that the condition may in 
many respects be quite unlike that of solid bodies at the surface of the 
earth. There must, however, be local, and probably temporary, reservoirs 
which supply the lava which exudes or is ejected from volcanoes. It is 
shown in the paper that the position of active volcanoes, particularly 
about the Pacific, suggests that the lava reservoirs which supply their 
lava have been due to local reliefs from pressure beneath arches of strong 
formations developed in the process of mountain making. The relative 
strengths of different sedimentary formations lead inevitably to the con- 
elusion that the type of formation which thus fuses and produces lava, 
is what is known as shale or slate. Studies of the chemical composition 
of igneous rocks, which compositions are limited in range, support this 
view ; and the study of the gases which are given off from lavas afford 
some further striking confirmations of the theory. 
THE AGE OF THE KEWEENAW AN SERIES. 
A. C. LANE. 
Van Hise and Leith's monograph on Lake Superior includes a valu- 
able chapter on the Keweenawan series, and shows that as between them 
and the Michigan survey there is now no essential difference in the 
recognition of the facts. It is fully recognized by all that, so far as the 
facts go, they do not definitely prove or disprove the Cambrian or Pre- 
Cambrian age of the Keweenawan series. They think the balance of 
probability to be in favor of the Pre-Cambrian, the writer, in favor of 
the Cambrian, largely because the Keweenawan is a formation of very 
rapidly formed rocks (sandstones, conglomerates and traps) immediately 
underlying the Upper Cambrian, and is the first volcanic epoch proceed- 
ing. As the decision of this question has some bearing on important 
questions of nomenclature, it is worth considering how it may finally 
be settled : 
1. Though the Keweenawan, has a “largely subaerial formation is 
not likely to be particularly fossil iferous, yet intercalations of acqueous 
