HYPOTHESES BEARING ON CLIMATIC CHANGES 665 
escape because they would not be projected away from the earth. 
Besides, the escape of molecules projected outwards is seriously 
limited by the interference of thé particles above. This inter- 
ference is practically prohibitory for the molecules in the base 
of the atmosphere. The problem, therefore, involves the extent 
to which the high velocities of the lower hot atmosphere would 
be communicated to the upper atmosphere whence escape would 
be possible. The interpretation of this is beset with great diffi- 
culties. The molecular velocities of the higher parts of an 
atmosphere surrounding a molten earth involve factors which 
cannot be safely estimated from the phenomena of a cold earth. 
It must of course be assumed that the molecular velocities of 
the molecules of the rising air would be lowered in proportion to 
the work done by them or the energy lost, but in convectional 
movements certain parts of the air are recipients of motion 
rather than generators of it, and do not lose the energy their 
movements might seem to imply. It is probable that the inter- 
change of lower and upper air about a molten earth would be 
extremely violent. It is not unlikely that explosive convection 
like that of the sun would be the customary mode of action. If 
hot bodies of vapor were shot violently into the outer limits of 
the atmosphere, molecular discharge would seem to be probable 
if not inevitable, whatever might be true of the more quiet mode 
of action. Besides this the current nebular hypothesis appar- 
ently involves the passage of even the outer atmosphere through 
very hot stages during the early period when the refractory 
gases of the now solid material were condensing and separating 
themselves from the atmospheric gases. 
The case in this form seems at present indeterminate. There 
is an apparent probability that a large loss would be suffered 
while the temperatures ranged from 3000° to 4000°. At the 
same time there is a possibility that a residue would remain if 
the period of this high temperature were not prolonged, and a 
probability that a large part of the atmosphere would be retained 
if it survived until the temperatures were near the melting point 
of rock. 
5 
