by means involving ideas, however vague in the first instance, of
applying it to its subsequent function.

If any one could be found so blind to obvious inferences as to
accept natural selection, "or the preservation of favoured
machines," as the main means of mechanical modification, we might
suppose him to argue much as follows:- "I can quite understand," he
would exclaim, "how any one who reflects upon the originally simple
form of the earliest jemmies, and observes the developments they
have since attained in the hands of our most accomplished
housebreakers, might at first be tempted to believe that the present
form of the instrument has been arrived at by long-continued
improvement in the hands of an almost infinite succession of
thieves; but may not this inference be somewhat too hastily drawn?
Have we any right to assume that burglars work by means analogous to
those employed by other people? If any thief happened to pick up
any crowbar which happened to be ever such a little better suited to
his purpose than the one he had been in the habit of using hitherto,
he would at once seize and carefully preserve it. If it got worn
out or broken he would begin searching for a crowbar as like as
possible to the one that he had lost; and when, with advancing
skill, and in default of being able to find the exact thing he
wanted, he took at length to making a jemmy for himself, he would
imitate the latest and most perfect adaptation, which would thus be
most likely to be preserved in the struggle of competitive forms.
Let this process go on for countless generations, among countless
burglars of all nations, and may we not suppose that a jemmy would
be in time arrived at, as superior to any that could have been
designed as the effect of the Niagara Falls is superior to the puny
efforts of the landscape gardener?"