THE MOON S FACE. 
287 
certain pressure definitely related to the temperature of the 
ice. The low temperatures ascribed to the moon by Langley 
would correspond to an atmosphere of aqueous vapor so 
tenuous as to be very difficult of recognition; so that the 
prevalent doubt as to the demonstration of a lunar atmos¬ 
phere need not bar speculation as to lunar ice. The atmos¬ 
pheric pressure which W. C. Pickering estimates as possible* 
would indicate a maximum ice temperature of about —40°. 
The question whether a moonlet could consist partly or 
wholly of ice is more vital and more difficult. 
Tin, silver, phosphorus, and sulphur are more easily fused 
than ice and their physical properties are perhaps equally 
adjusted to the requirements of the problem; but tin and 
silver are rare substances, while phosphorus, which is less 
rare, does not occur naturally uncombined, and sulphur, 
though abundant in combination, is rare in the free state. 
Perhaps the free iron and nickel of aerolites may stand 
sponsor for free sulphur or phosphorus in moonlets. 
The white bands grouped about Copernicus, though un¬ 
mistakably derived from that center, do not radiate directly 
from the crater, are not straight, and are not of even width. 
They appear also to be diverted by the crater Eratosthenes, 
passing beyond it on both sides, but leaving a free space in 
its lee. These characters, and the rill pits previously de¬ 
scribed, lead me to refer the bands to a swift liquid flow 
over the surface. The flow probably included two sub¬ 
stances, the darker of which, occupying interspaces between 
the pale bands, is not distinguishable in color from the sur¬ 
rounding maria. The straight feathery rays from other 
craters are referred, in contradistinction, to jets or sprays 
projected free from the surface. 
Retrospect .—In the preparation of this manuscript I have 
been gradually drawn from the attitude of the judge to the 
attitude of the advocate. This transformation is but an echo 
of the history of the investigation, for, starting with two 
working hypotheses, the impact and the volcanic, I soon 
* Astronomy and Astro-physics, 1892, p. 778. 
