110 THE LIVING WORLD. 



It seems probable that the remarkable social habit of the ants and 

 bees did not develop until later than the Cretaceous (8). 



VERTEBRATA. 



/ The earliest traces of the Vertebrata are in the rocks of the lower 

 Silurian (2). At that time there is no doubt that some forms of fishes 

 were in existence, though the remains are rather scanty. The verte- 

 brates then existing were low fishes whose skeleton was poorly adapt- 

 ed to preservation. Indeed, in the whole Silurian the traces of 

 vertebrates are rare, although there can be no doubt that they were in 

 existence. With the Devonian (3), they became very numerous. The 

 Devonian seas were filled with large numbers of fishes, chiefly 

 ganoids and elasmobranchs (related to the gar-pikes, and sharks). 

 During this period these orders of fishes became very abundant and 

 highly diversified. Late in the Devonian age some of the ganoids 

 would seem to have acquired the habit of living partly in the air, and 

 the habit thus acquired gave rise to the amphibians, which appeared 

 for the first time in the next age, the Carboniferous (4). During this 

 age the amphibians became abundant, and towards its close they 

 seem to have become more distinctly aerial and to have ceased to be 

 able to live in the water. This produced the true land vertebrates. 

 In the latter part of the Carboniferous (4) or the Permian (5), we find 

 that true reptiles had come into existence. Having once assumed a 

 terrestrial habit, the reptiles rapidly expanded to appropriate the large 

 field open to them, and in the next two ages (Triassic and Jurassic) 

 they became more and more abundant, and grew to immense size. 

 While these first land vertebrates were thus expanding, one side 

 branch of them gradually acquired wings and developed into birds, 

 which seem to have appeared first in the Jurassic (7) and Creta- 

 ceous (8). 



From the Carboniferous (4) the reptiles had been constantly expand- 

 ing in diversity, in number, and in size. But with the Cretaceous (S) 

 they were slowly, yet surely, giving way to another and better adapted 

 type of land vertebrates. Way back before the Jurassic (7) period, prob- 

 ably in the Permian (5), one of the early generalized types of reptiles 

 seem to have sent off a branch of descendants which produced their 

 young alive instead of laying eggs, and nourished them for a longer 

 or shorter time by secretions from the dermal glands of the mother. 

 These animals were at first small and probably weak. But they 

 continued to exist during the age of reptiles, as a comparatively unim- 



