es. The edges, as a rule, were very irregular: in some places
there were large areas of fallow land, and others were very sparsely
covered with vegetation.

John remarked that if any of these bare or sparsely-covered places were
large enough to be detected by our telescopes, in moments of extremely
good seeing, we should no doubt be told that they afforded absolute
proof that the canal lines are only disconnected markings, and the
canals a myth.

"Very probably," I replied; "yet it should be obvious that vegetation
would be sparse, or altogether absent, perhaps, for miles, in many
places along the thousands of miles over which the canals extend, and
also that it is quite likely, if we could use higher powers so as to get
a better view of the lines, the edges would appear irregular. Nature is
rarely symmetrical in her work, there is nearly always irregularity of
growth; and in artificial cultivation it is neither possible nor
desirable to fill up every acre of land simultaneously."

Merna then told us that, owing to extensions of their irrigation system,
laterally, and the consequent growth of vegetation, the width of many of
the canal lines would be seen to increase.

"Yes," said John, "and when that phenomenon is seen by our observers we
shall be informed that such increase in width is still another proof
that there are no canals upon Mars."

"Well, John," replied Merna, "it seems to me very strange that your
people should so misinterpret the meaning of such indications. Do you
really think such a contention would be put forward?"

"I'm quite sure of it," said John; "and we should be told that canals
could not increase in width! Don't you agree with me, Professor?"

"Yes, John," I answered; "I have seen and heard so many contentions and
arguments of a like nature that I cannot say your supposition is not
justified.

"I may, however, point out that it is only when the most ideal
conditions of seeing exist that we can ever hope to secure a view of the
canal lines showing the apparent breaks in their continuity. I have on a
previous occasion alluded to the drawbacks connected with the use of
very large telescopes, and it may be well to sound a note of warning,
for it would be very easy for an observer to be deceived by an illusory
appearance of the breaking up of the canal lines into a series of
scattered markings. This effect would undoubtedly occur in using a very
large telescope in any but ideally favourable a