I see no improbability in
several reverting. As I do not believe any well-founded experiments or
facts are known, each must form his opinion from vague generalities.
I think you confound two rather distinct considerations; a variation
arises from any cause, and reversion is not opposed to this, but solely
to its inheritance. Not but what I believe what we must call perhaps
a dozen distinct laws are all struggling against each other in every
variation which ever arises. To give my impression, if I were forced
to bet whether or not, after a hundred generations of growth in a poor
sandy soil, a cauliflower and red cabbage would or would not revert to
the same form, I must say I would rather stake my money that they would.
But in such a case the conditions of life are changed (and here comes
the question of direct influence of condition), and there is to be no
selection, the comparatively sudden effect of man's selection are left
to the free play of reversion.

In short, I dare not come to any conclusion without comparing all facts
which I have collected, and I do not think there are many.

Please do not say to any one that I thought my book on species would be
fairly popular and have a fairly remunerative sale (which was the height
of my ambition), for if it prove a dead failure it would make me the
more ridiculous.

LETTER 73. TO W.H. MILLER. Down, June 5th [1859].

I thank you much for your letter. Had I seen the interest of my remark
I would have made many more measurements, though I did make several.
I stated the facts merely to give the general reader an idea of the
thickness of the walls. (73/1. The walls of bees' cells: see Letter
173.)

Especially if I had seen that the fact had any general bearing, I should
have stated that as far as I could measure, the walls are by no means
perfectly of the same thickness. Also I should have stated that the
chief difference is when the thickness of walls of the upper part of
the hexagon and of the pyramidal basal plates are contrasted. Will you
oblige me by looking with a strong lens at the bit of comb, brushing off
with a knife the upper thickened edges, and then compare, by eye alone,
the thickness of the walls there with the thickness of the basal plates,
as seen in any cross section. I should very much like to hear whether,
even in this way, the difference is not perceptible. It is generally
thus perceptible by comparing the thickness of the walls of the hexagon
(if