feed and never see the daylight.

{465} This corresponds to the _Origin_, Ed. i. p. 441, vi. p. 607,
where, however, the example is taken from the Cirripedes.

_Attempt to explain the facts of embryology._

I think considerable light can be thrown by the theory of descent on
these wonderful embryological facts which are common in a greater or
less degree to the whole animal kingdom, and in some manner to the
vegetable kingdom: on the fact, for instance, of the arteries in the
embryonic mammal, bird, reptile and fish, running and branching in the
same courses and nearly in the same manner with the arteries in the
full-grown fish; on the fact I may add of the high importance to
systematic naturalists{466} of the characters and resemblances in the
embryonic state, in ascertaining the true position in the natural system
of mature organic beings. The following are the considerations which
throw light on these curious points.

{466} _Origin_, Ed. i. p. 449, vi. p. 617.

In the economy, we will say of a feline animal{467}, the feline
structure of the embryo or of the sucking kitten is of quite secondary
importance to it; hence, if a feline animal varied (assuming for the
time the possibility of this) and if some place in the economy of
nature favoured the selection of a longer-limbed variety, it would be
quite unimportant to the production by natural selection of a
long-limbed breed, whether the limbs of the embryo and kitten were
elongated if they _became_ so _as soon_ as the animal had to provide
food for itself. And if it were found after continued selection and the
production of several new breeds from one parent-stock, that the
successive variations had supervened, not very early in the youth or
embryonic life of each breed (and we have just seen that it is quite
unimportant whether it does so or not), then it obviously follows that
the young or embryos of the several breeds will continue resembling each
other more closely than their adult parents{468}. And again, if two of
these breeds became each the parent-stock of several other breeds,
forming two genera, the young and embryos of these would still retain a
greater resemblance to the one original stock than when in an adult
state. Therefore if it could be shown that the period of the slight
successive variations does not always supervene at a very early period
of life, the greater resemblance or closer unity in type of animals in
the young than