would also be in relatively greater amount so far as
this cause is considered. For the planet would part, _caeteris paribus_,
with its lighter gases the quickest. Whence as regards both water-vapour
and carbon-dioxide we have reason to think them in relatively greater
quantity than in our own air at corresponding barometric pressure." I
cannot understand this passage except as implying that 'water-vapour and
carbon-dioxide' are among the heavier and not among the lighter gases of
the atmosphere--those which the planet 'parts with quickest.' But this
is just what water-vapour _is_, being a little less than two-thirds the
weight of air (0.6225), and one of those which the planet _would_ part
with the quickest, and which, according to Dr. Johnstone Stoney, it
loses altogether.
* * * * *

Note on Professor Lowell's article in the _Philosophical Magazine_; by
J.H. Poynting, F.R.S., Professor of Physics in the University of
Birmingham.

"I think Professor Lowell's results are erroneous through his neglect of
the heat stored in the air by its absorption of radiation both from the
sun and from the surface. The air thus heated radiates to the surface
and keeps up the temperature. I have sent to the _Philosophical
Magazine_ a paper in which I think it is shown that when the radiation
by the atmosphere is taken into account the results are entirely
changed. The temperature of Mars, with Professor Lowell's data, still
comes out far below the freezing-point--still further below than the
increased distance alone would make it. Indeed, the lower temperature on
elevated regions of the earth's surface would lead us to expect this. I
think it is impossible to raise the temperature of Mars to anything like
the value obtained by Professor Lowell, unless we assume some quality in
his atmosphere entirely different from any found in our own atmosphere."
J.H. POYNTING. October 19, 1907.

CHAPTER VI.

A NEW ESTIMATE OF THE TEMPERATURE OF MARS.

When we are presented with a complex problem depending on a great number
of imperfectly ascertained data, we may often check the results thus
obtained by the comparison of cases in which some of the more important
of these data are identical, while others are at a maximum or a minimum.
In the present case we can do this by a consideration of the Moon as
compared with the Earth and with Mars.

_Langley's Determination of the Moon's Temperature._

In the moon w