The absence of early Eocene remains prevents us from tracing the lines of our vast and varied bird-kingdom to their Mesozoic beginnings. And when we appeal to the zoologist to supply the missing links of relationship, by a comparison of the structures of living birds, we receive only uncertain and very general suggestions. [*] He tells us that the ostrich-group (especially the emus and cassowaries) are one of the most primitive stocks of the bird world, and that the ancient Dinornis group and the recently extinct moas seem to be offshoots of that stock. The remaining many thousand species of Carinate birds (or flying birds with a keel [carina]-shaped breast-bone for the attachment of the flying muscles) are then gathered into two great branches, which are "traceable to a common stock" (Pycraft), and branch in their turn along the later lines of development. One of these lines--the pelicans, cormorants, etc.--seems to be a continuation of the Ichthyornis type of the Cretaceous, with the Odontopteryx as an Eocene offshoot; the divers, penguins, grebes, and petrels represent another ancient stock, which may be related to the Hesperornis group of the Cretaceous. Dr. Chalmers Mitchell thinks that the "screamers" of South America are the nearest representatives of the common ancestor of the keel-breasted birds. But even to give the broader divisions of the 19,000 species of living birds would be of little interest to the general reader. * The best treatment of the subject will be found in W. P. Pycraft's History of Birds, 1910. The special problems of bird-evolution are as numerous and unsettled as those of the insects. There is the same dispute as to "protective colours" and "recognition marks", the same uncertainty as to the origin of such instinctive practices as migration and nesting. The general feeling is that the annual migration had its origin in the overcrowding of the regions in which birds could live all the year round. They therefore pushed northward in the spring and remained north until the winter impoverishment drove them south again. On this view each group would be returning to its ancestral home, led by the older birds, in the great migration flights. The curious paths they follow are believed by some authorities to mark the original lines of their spread, preserved from generation to generation through the annual lead of the older birds. If we recollect the Ice-Age which drove the vast majority of the birds south at the end of the Tertiary, and imagine them later following the northward retreat of the ice, from their narrowed and overcrowded southern territory, we may not be far from the secret of the annual migration.