am sorry I had not space to refer more fully to your
interesting work.[27] The most important point on which I think your
views require emendation is on _instinct_. I see you quote Spalding's
experiments, but these have been quite superseded and shown to be
seriously incorrect by Prof. Lloyd Morgan. A paper by him in the
_Fortnightly Review_ of August, 1893, gives an account of his
experiments, and he read a paper on the same subject at the British
Association last year. He is now preparing a volume on the subject
which will contain the most valuable series of observations yet made on
this question. Another point of some importance where I cannot agree
with you is your treating dipsomania as a disease, only to be eliminated
by drunkenness and its effects. It appears to me to be only a vicious
habit or indulgence which would cease to exist in a state of society in
which the habit were almost universally reprobated, and the means for
its indulgence almost absent. But this is a matter of comparatively
small importance.--Believe me yours very truly,

ALFRED R. WALLACE.

* * * * *

TO DR. ARCHDALL REID

_Parkstone. April 28, 1896._

Dear Sir,--"We can but reason from the facts we know." We know a good
deal of the senses of the higher animals, very little of those of
insects. If we find--as I think we do--that all cases of supposed
"instinctive knowledge" in the former turn out to be merely intuitive
reactions to various kinds of stimulus, combined with very rapidly
acquired experience, we shall be justified in thinking that the actions
of the latter will some day be similarly explained. When Lloyd Morgan's
book is published we shall have much information on this question.
(_See_ "Natural Selection and Tropical Nature," pp. 91-7.)--Yours truly,
ALFRED R. WALLACE.

* * * * *

TO PROF. MELDOLA

_Parkstone, Dorset. October 12, 1896._

My dear Meldola,--I got Weismann's "Germinal Selection" two or three
months back and read it very carefully, and on the whole I admire it
very much, and think it does complete the work of ordinary variation and
selection. Of course it is a pure hypothesis, and can never perhaps be
directly proved, but it seems to me a reasonable one, and it enables us
to understand two groups of facts which I have never been able to work
out satisfactorily by the old method. These two facts are: (1) the
total, or almost total, disappeara