may be added. A young form may become adapted to a new mode of life, or it may escape competition by seizing upon a new field; its enemies, dangers, and means of defense may change, and with these changes of habits corresponding changes of structure may occur, so that the primitive or ancestral record may become completely obscured by secondary changes.
The examination of the various kinds of modification which may be brought about in this way falls within the scope of a treatise on comparative embryology, but it would be out of place here, although one or two examples of the more common sorts of modification may be of interest.
The chrysalis stage of butterflies is an instance of the secondary acquisition of a new stage of development, which forms no part of the ancestral record.
It is obvious that the inactive pupa, which takes no food, and exhibits few of the ordinary activities of animal life, can not possibly have existed in the past as an adult ancestor of the butterflies, nor is it conceivable that any of the remote ancestry of this group bore a general resemblance to a pupa. While it is impossible to believe that the pupa stage is ancestral, we have good evidence to show the manner in which it has been acquired as a secondary modification.
Lubbock has pointed out that the least specialized or most primitive insects have mouth-parts which are indifferently adapted for either cutting or sucking, and that these insects do not undergo a metamorphosis, but are gradually converted into the adult form by a simple process of gradual development. He also shows good ground for believing that the common ancestors of all the groups of insects were like these forms in these particulars; and he holds that, as the stock which led to our present butterflies was evolved from this ancient stem-form, the young became adapted to a sedentary creeping life, and their indifferent mouth-parts became gradually converted into cutting jaws, while the adults became adapted to quite a different mode of life, and the same indifferent mouth-parts became gradually modified into a sucking proboscis. While the caterpillar and butterfly were thus diverging in two directions from the original unspecialized form, and the structure and habits of the larva were becoming more and more different from those of the adult, it is plain that the metamorphosis must at the same time have become more and more violent; and, according to Lubbock, one of the periods of slight activity which, in most insects, accompany the periodical molts, was seized upon, and gradually extended into a long resting or chrysalis stage, in order to enable the animal to exist while the highly specialized organs of the caterpillar are changing into the equally specialized but very different organs of the winged insect. The growing butterfly now passes through a resting or pupa stage which connects the