may say that at an early period the central reptilian stock, consisting of the long, lithe, four-legged forms like the lizards, still closely allied in shape to their primitive newt-like and eel-like ancestors, began to divide laterally into sundry important branches. Some of them lost their limbs and became serpents; others acquired bony body-coverings and became turtles; but the vast majority went off in one of two directions, either as fish-like sea-saurians, or as bird-like land-saurians. It is with this last division alone that we shall have largely to deal in tracing out the pedigree of our existing birds. Their fossil remains supply us with many connecting links which help us to bridge over the distance between the modern representatives of the two classes. It is true, none of these links can be said to occupy an exactly intermediate place between reptiles and birds; none of them can be regarded as forming an actual part of the ancestry of our own swallows and pigeons: they are rather closely related collateral members of the family than real factors in the central line of descent. But they at least serve to show that, at and before the period when true birds first appeared upon earth, many members of one great reptilian group had made immense advances in several distinct directions toward the perfected avian type.
Clearly, the first step toward the development of a bird must consist in acquiring a more or less upright habit: for the legs must be well differentiated into a large hind pair and a free fore pair, before the last can be further specialized into feathered wings; and the body must have acquired a forward poise before flying becomes a possible mode of locomotion. Such an upright habit is first foreshadowed in the larger-limbed and longer-legged lizards like the dinosaurians, which walked to some extent erect, and more particularly in some highly specialized reptiles like the iguanodon, which had large hind legs and small fore-legs, and could walk or hop on the hind-legs alone, much after the fashion of a kangaroo, or still more of a jerboa or a chinchilla. Now, it is noticeable that the tendency to acquire the most rudimentary form of flying is common among animals of this semi-erect habit, especially when they frequent forests and jump about much from tree to tree. For example, among our modern mammals, the squirrels are a race much given to sitting on their hind-legs and using their paws as hands; while they are also much accustomed to jumping lightly from bough to bough: and some among them, the flying squirrels, have developed a sort of parachute consisting of an extensible skin between the fore and hind legs, which they use to break their fall in descending to the ground. Again, among the lower monkey-like animals, the so-called flying lemur or galeopithecus has hit upon an exactly similar plan; while, in the bats, a membrane which may be fairly called a wing has been evolved to a very high degree of perfection. Everywhere, the habit of living among trees or jumping from rocks tends to produce either parachute or wing-like organs;