teeth, the like of which of course exist in no living bird. The skeleton is for the most part reptilian; and, though the legs are bird-like, they are not much more so than those of compsognathus, an unmixed reptile. Even the wings are more like the fore-legs, and could only be used for flight by the aid of a side membrane. Accordingly, we may say that we have lithographed for us in archæopteryx a specimen of the intermediate state, when reptiles were just in the very act of passing into birds. The scales and protuberances on the body had already developed into feathers; the fore-legs had already developed into rude and imperfect wings, and the feet had become decidedly bird-like; but as yet there was only a very small breast-bone, the tail remained in internal structure like that of a lizard, the jaws still contained pointed teeth, and the wing ended in a three-toed hand, while flight was probably as rudimentary as in the flying-lemur and the flying-squirrel. Nowhere in the organic series has geology supplied us with a better missing link than this uncouth and half-formed creature, Nature's first tentative rough draft of the beautiful and exquisitely adapted modern birds.
Such an animal, once introduced, was sure to undergo further modification, to fit it more perfectly for its new sphere of action. In the first place, the tail was sure to grow shorter and shorter, by stress of natural selection, because a more fan-like organ would act better as a rudder to steer the flight than the long lizard-like tail of archæopteryx. In the second place, the general bony structure was sure to grow better adapted for flight, by the development of some such feature as the keeled breast-bone, and the general modification of the other parts (especially the wing) into better correspondence with their new function. At the same time, it must not be supposed that all intermediate birds would lose their reptilian features equally and symmetrically. Some for a time might retain one lizard-like peculiarity, say the teeth, and some might retain another, say sundry anatomical points in the structure of the skeleton. It was long indeed before the whole tribe of birds acquired the entire set of traits which we now regard as characteristic of their class. During the intervening period they kept varying in all directions, tentatively, if one may say so, and thus the early forms of birds differ far more among themselves than do any modern members of the feathered kingdom. In other words, when the full bird type was finally evolved, it proved so much better adapted to its airy mode of life than any other and earlier creature that it lived down not only the rude reptilian pterodactyls but also the simpler primeval forms of birds themselves: exactly as civilized European man is now living down not only the elephants and buffaloes but the red Indian and the Australian black fellow as well.
Some of the varying primeval forms have been preserved for us as fossils in the chalk deposits of the Western States, which are of course later in date than the oölitic slates of Solenhofen, where we find the
