from which nature (supposing no exception taken to her
personification) can select. The bottles have the same labels, and
they are of the same colour, but the one holds brandy, and the other
toast and water. Nature can, by a figure of speech, be said to
select from variations that are mainly functional or from variations
that are mainly accidental; in the first case she will eventually
get an accumulation of variation, and widely different types will
come into existence; in the second, the variations will not occur
with sufficient steadiness for accumulation to be possible. In the
body of Mr. Darwin's book the variations are supposed to be mainly
due to accident, and function, though not denied all efficacy, is
declared to be the greatly subordinate factor; natural selection,
therefore, has been hitherto throughout tantamount to luck; in the
peroration the position is reversed in toto; the selection is now
made from variations into which luck has entered so little that it
may be neglected, the greatly preponderating factor being function;
here, then, natural selection is tantamount to cunning. We are such
slaves of words that, seeing the words "natural selection" employed-
-and forgetting that the results ensuing on natural selection will
depend entirely on what it is that is selected from, so that the
gist of the matter lies in this and not in the words "natural
selection"--it escaped us that a change of front had been made, and
a conclusion entirely alien to the tenor of the whole book smuggled
into the last paragraph as the one which it had been written to
support; the book preached luck, the peroration cunning.

And there can be no doubt Mr. Darwin intended that the change of
front should escape us; for it cannot be believed that he did not
perfectly well know what he had done. Mr. Darwin edited and re-
edited with such minuteness of revision that it may be said no