pposed to be descended from one
parent-stock, and to have been altered by such slight steps as man
effects by the selection of chance domestic variations. Now we can see
according to this view that a foot might be selected with longer and
longer bones, and wider connecting membranes, till it became a swimming
organ, and so on till it became an organ by which to flap along the
surface or to glide over it, and lastly to fly through the air: but in
such changes there would be no tendency to alter the framework of the
internal inherited structure. Parts might become lost (as the tail in
dogs, or horns in cattle, or the pistils in plants), others might become
united together (as in the feet of the Lincolnshire breed of pigs{458},
and in the stamens of many garden flowers); parts of a similar nature
might become increased in number (as the vertebrae in the tails of pigs,
&c., &c. and the fingers and toes in six-fingered races of men and in
the Dorking fowls), but analogous differences are observed in nature and
are not considered by naturalists to destroy the uniformity of the
types. We can, however, conceive such changes to be carried to such
length that the unity of type might be obscured and finally be
undistinguishable, and the paddle of the Plesiosaurus has been advanced
as an instance in which the uniformity of type can hardly be
recognised{459}. If after long and gradual changes in the structure of
the co-descendants from any parent stock, evidence (either from
monstrosities or from a graduated series) could be still detected of the
function, which certain parts or organs played in the parent stock,
these parts or organs might be strictly determined by their former
function with the term "metamorphosed" appended. Naturalists have used
this term in the same metaphorical manner as they have been obliged to
use the terms of affinity and relation; and when they affirm, for
instance, that the jaws of a crab are metamorphosed legs, so that one
crab has more legs and fewer jaws than another, they are far from
meaning that the jaws, either during the life of the individual crab or
of its progenitors, were really legs. By our theory this term assumes
its literal meaning{460}; and this wonderful fact of the complex jaws of
an animal retaining numerous characters, which they would probably have
retained if they had really been metamorphosed during many successive
generations from true legs, is simply explained.

{458} The so