HYPOTHESES BEARING ON CLIMATIC CHANGES 671 
size of the moon a definite problem was presented of which the 
present moon stands as a possible representative and invites 
computation. If in its loose state of aggregation the mass had 
a specific gravity of 2. and if it shrank by self-condensation to 
3.4, the average specific gravity of the moon, the possible heat gen- 
erated by the gravitative fall would have equaled 3900° C. for the 
whole mass, the specific heat being assumed to be .2, which is 
very prudent. I owe the computation to Mr. Moulton. For 
convenience of computation the condensation was assumed to be 
uniform and the distribution of heat uniform. The original 
distribution of internal heat would perhaps have varied with the 
square root of the pressure, according to Laplace’s formula. 
As the computed temperature is more than twice the melting 
temperature of average rock not under pressure it seems ample 
for all igneous phenomena indicated on the moon with a large 
residue for secular loss. 
Assuming that the exterior temperature remained below zero 
during the pre-atmospheric stages of growth, the hypothetical 
structure of the planet when it reached the size of the moon may 
be pictured as embracing (1) a dense central portion raised to a 
high temperature by compression, giving a potentiality of lique- 
faction under relief of pressure; (2) a zone of declining temper- 
ature and less compressed structure, graduating toward a porous 
condition, and (3) at the surface the still unconsolidated open 
aggregation. The low average specific gravity of the moon 
(3.4) encourages the belief that the outer porous zone was deep 
and open. The notion is entertained that the central heat and 
compression would lead to the expulsion of a part of the cen- 
trally entrapped gases and vapors, and that these would be 
driven outward into the exterior porous portion, which having a 
low temperature, like that of the moon today, condensed the 
aqueous vapor in the spaces of the open texture and the whole 
became bound together more or less completely with a matrix 
of frost and ice. It is assumed that the internal condensation 
would be attended by readjustments of matter of the nature of 
diffusions, differentiations and concentrations, and that there 
