ie. 
1900.] LOWELL—MARS ON GLACIAL EPOCHS. 645 
be kept warmer and the surface bare of it colder than each other- 
wise would be until carbonic acid was deposited over the whole. 
8. From the foregoing it is clear that the amount of moisture 
consonant with temperate conditions upon the planet does not re- 
quire to be great. We need, therefore, feel no surprise that spec- 
troscopy should as yet give us uncertain answer on the subject. 
Huggins found marks of the presence of water vapor in the planet’s 
spectrum ; Campbell could find none. And the latter thought he 
should have done so had the amount been so much as one-fourth of 
our own. 
But there is another drawback to this deduction. It is possible 
to point out a fallacy in the assumption of the data upon which the 
detection depends. Weare told that the light examined has passed 
twice through the Martian atmosphere. But this is not the fact. 
A part of it has done so indeed, but only a part. A considerable 
portion has never traversed that atmosphere at all. We should 
know @ friort that this could not but be the case. But we are not 
left to @ prior? reasoning in the matter. We have direct evidence 
of the fact. One of the peculiar details of the disk the planet 
shows us is the presence over a part of it of a veil which is not only 
unmistakable but pronounced. This veil is known as the limb- 
light. Extending in from the limb along its whole length is a 
brilliance strong enough to swamp all but the heaviest markings for 
a distance in of thirty degrees. Circumstances of position show 
that this can only be the effect of an atmosphere (4nua/s Lowell 
Observatory, Vol. 1). It must extend over the whole disk, but be- 
comes conspicuous only as we approach the limb, owing to the 
greater depth of it passed through as we increase the inclination. 
It should vary roughly, though not exactly, as the cosecant of the 
angle in from the limb. The effect might be due to anything sus- 
pended in the air—dust or water vapor. It does not seem possible 
as yet to evaluate it satisfactorily, but from its action it would 
appear to bear no inconsiderable ratio to the rest of the illumination. 
Suppose this ratio to be one of equality—and the amount of obscura- 
tion it effects show this to be no unseemly supposition—then its 
presence would halve the precision, and instead of being able to 
detect a quantity one-fourth of our own, we should only be able to 
perceive the double of that. 
g. Now, whatever moisture there be on Mars—and water there 
must be to some extent, since otherwise no seasonal change could 
