778 T. C. CHAMBERLIN 
A question of no small interest is the special kind of effect 
which an atmosphere reduced by coal formation would induce 
as distinguished from that induced by depletion through earth- 
contact. In the latter, as already set forth, bicarbonates were 
the chief product, and carbon dioxide, temporarily locked up, 
played a very important part. Through the peculiar agency of 
the ocean this ‘‘loose”’ carbon dioxide hastened the develop- 
ment of glaciation, prepared the conditions for strong reaction 
and accelerated the reaction when inaugurated. In the forma- 
tion of coal and like products no such effective temporary factor 
is produced. The carbon is, to be sure, temporarily held in the 
vegetation, but so much as decays goes directly back into the 
atmosphere, in the main, and the rest becomes permanently fixed. 
Neither part goes into an intermediate state of reserve, subject 
to being called forth by change of conditions as in the case of 
the second equivalent of the bicarbonates of the ocean. The 
action is somewhat analagous to the original carbonation of the 
silicates of the crystalline rocks in which the first equivalent of 
carbon dioxide when once united remains fixed (barring accidents 
to which coal is also liable), but in this case there is also a 
second equivalent of carbon dioxide temporarily locked up so 
long as a state of solution is maintained. Succinctly stated, 
without the unessential qualifications: (1) Depletion by coal 
formation is accompanied by no intensifying and reactive factor ; 
(2) Depletion by conversion of silicates into carbonates is 
accompanied by an intensifying and reactive factor; (3) Deple- 
tion by the solution of limestone is wholly temporary in nature 
and specially capable of promoting intensification and reaction. 
If, therefore, the impoverishment of the atmosphere as the 
prerequisite of glaciation in Permo-Carboniferous times was due, 
in the main, to the extraction of carbon in the form of coal and 
like deposits, there was absent, to that extent, the factor to which 
the hastening of glaciation and of reaction is assigned. Before 
glaciation could be affected in the measurable absence of this 
accelerating agent, it was necessary for the permanent depletion 
to go to greater lengths. Moreover the depletion when once 
