certainly be lighter and less closely compacted than the snow
and ice upon our earth; but it is quite clear that it could not melt to
any extent unless the temperature remained above freezing-point for a
considerable length of time.

"It has, however, seriously been contended that the Martian polar caps
are not snow at all, but frozen carbon dioxide--the poisonous dregs of
what once was an atmosphere. Carbon dioxide, however, melts and becomes
gaseous almost suddenly, but these polar snow-caps melt gradually,
exactly as frozen snow would; so this theory fails altogether to fit the
circumstances.

"Moreover, the water which accumulates all round the base of the melting
snow-cap has been carefully observed on many occasions, and in the early
stage of melting it appears blue in tint, but later on, as upper layers
of snow dissolve and those nearer the soil are reached, the water
presents a turbid and muddy appearance; exactly what might be expected
when water has been contaminated by the surface soil.

"Dr. Alfred Russel Wallace declines to accept the blue tint as any proof
that the liquid is water, and contends that shallow water would not
appear that colour when viewed from a distance. You will, however, have
observed that the water in all our shallow reservoirs appears intensely
blue when observed from any distant and elevated point of view. It seems
to me that when, as in the case of Mars, we have a very thin atmosphere
laden with sand particles, we have exactly the conditions which would
produce a very blue sky, and cause the water to appear a deep blue
colour when viewed from a distance.

"It is also contended that water cannot be present on Mars, because none
of our skilled spectroscopists has yet been able to demonstrate by the
spectroscope that there is any water vapour in the Martian atmosphere.

"This, however, is generally acknowledged to be a very difficult and
delicate operation; and, in any case, it is purely negative evidence,
and cannot be accepted as final. I feel quite confident that sooner or
later a means will be found of definitely proving the presence of water
vapour upon Mars by the aid of the usual lines in the spectrum. There
are too many evidences of its presence, such as clouds, hoarfrost, snow,
and seasonal changes in vegetation, to warrant the rejection of the idea
of its existence merely because it has not been detected by the
particular means hitherto used by the spectroscopists.

"Mr.