roup have not only the rudiments of wing-bones, but also the
rudiments of wings, that is, an external limb bearing rigid quills or
largely-developed {481} plumes. In the cassowary these wing-feathers are
reduced to long spines like porcupine-quills, while even in the Apteryx,
the minute external wing bears a series of nearly twenty stiff quill-like
feathers.[123] These facts render it almost certain that the struthious
birds do not owe their imperfect wings to a direct evolution from a
reptilian type, but to a retrograde development from some low form of
winged birds, analogous to that which has produced the dodo and the
solitaire from the more highly-developed pigeon-type. Professor Marsh has
proved, that so far back as the Cretaceous period, the two great forms of
birds--those with a keeled sternum and fairly-developed wings, and those
with a convex keel-less sternum and rudimentary wings--already existed side
by side; while in the still earlier Archaeopteryx of the Jurassic period we
have a bird with well-developed wings, and therefore probably with a keeled
sternum. We are evidently, therefore, very far from a knowledge of the
earliest stages of bird life, and our acquaintance with the various forms
that have existed is scanty in the extreme; but we may be sure that birds
acquired wings, and feathers, and some power of flight, before they
developed a keeled sternum, since we see that bats with no such keel fly
very well. Since, therefore, the struthious birds all have perfect
feathers, and all have rudimentary wings, which are anatomically those of
true birds, not the rudimentary fore-legs of reptiles, and since we know
that in many higher groups of birds--as the pigeons and the rails--the
wings have become more or less aborted, and the keel of the sternum greatly
reduced in size by disuse, it seems probable that the very remote ancestors
of the rhea, the cassowary, and the apteryx, were true flying birds,
although not perhaps provided with a keeled sternum, or possessing very
great powers of flight. But in addition to the possible ancestral power of
flight, we have the undoubted fact that the rhea and the emu both swim
freely, the former having been seen swimming from island to island off the
coast of Patagonia. This, taken in connection with the wonderful aquatic
ostrich of the Cretaceous period discovered by Professor Marsh, opens {482}
up fresh possibilities of migration; while the immense antiquity thus given
to t