1900.] LOWELL—MARS ON GLACIAL EPOCHS. 663 
like-tinted summer habiliment of green. Bottom-lands, not eleva- 
tions then, are the cause of the snow’s survival. 
We are next concerned, therefore, with the way in which they 
may bring such result about. Just as it is not height but depth that 
determines the deposition, so it does this in quite unlike a manner. 
For by its indirect effects, more than by its direct efficiency, does 
it prove potent. Moisture in the air, it is true, flows into these 
lower levels because they are such, being hollows in the atmos- 
pheric ocean-bed. And they are such for water vapor over and 
above what they are for air, inasmuch as water vapor, though the 
lighter gas, is found most copious near the surface, thinning out 
more rapidly than the rest of the air upward. Depression is not 
so slight a factor as we proved elevation to be, because of this 
greater thinning out of moisture upward. but, even so, difference 
of level does directly but a part of the business. Indirectly de- 
pressions do something more: they start vegetation. Vegetation 
itself then takesa hand in the matter. Due to the humidity 
originally, once on the ground it reinforces the latter’s action. 
For asa part of its life economy the plant is busied in pumping 
up water from the earth for the sake of the substances held there in 
solution which it absorbs, and setting the residue free allows it to 
evaporate away. Thus the moisture attracted to the spot is 
returned to the air about it, to be again deposited on provocation. 
The deposit would not probably take the shape it does on earth. 
So thin, undoubtedly, is the air upon the surface of Mars that a 
precipitation in the form of rain or snow would seem not so likely 
amethod of deposit as that other form dependent on contact which 
we know as dew or hoar-frost. In which case the deposit would 
occur nearer the place of generation, since it would offer less 
chance to be wafted to the winds. Being in a valley, the wind 
would get less sweep, and plant growth would still more hinder 
and hamper its course; and, secondly, what moisture was caught up 
and snatched away would be unlikely to be precipitated elsewhere. 
It would eventually return, to begin its plant work once more. Some- 
where not far to the northward, then, of the general reservoirs of 
humidity, the great blue-green regions, we should expect to find 
the greatest accumulation of hoar-frost. This is precisely where we 
observe the centre of the snow-cap. 
31. That the little midsummer remnant of the south polar cap 
lies at some distance from the pole, proves its survival due not to 
