480 ISLAND LIFE part ii 



of wingless birds. Our best anatomists, as we have seen, 

 agree that both Dinornis and Apteryx are more nearly 

 allied to the cassowaries and emus than to the ostriches 

 and rheas ; and we see that the form of the sea-bottom 

 suggests a former connection with North Australia and 

 New Guinea — the very region where these types most 

 abound, and where in all probability they originated. The 

 suggesti on that all the great wingless birds of the world sprang 

 from a common ancestor at no very remote period, and 

 that their existing distribution is due to direct land com- 

 munication between the countries they noiv inhabit, is one 

 utterly opposed to all sound principles of reasoning in 

 questions of geographical distribution. For it depends 

 upon two assumptions, both of which are at least doubtful, 

 if not certainly false — the first, that their distribution over 

 the globe has never in past ages been very different from 

 what it is now ; and the second, that the ancestral forms 

 of these birds never had the power of flight. As to the 

 first assumption, we have found in almost every case that 

 groups now scattered over two or more continents formerly 

 lived in intervening areas of existing land. Thus the 

 marsupials of South America and Australia are connected 

 by forms which lived in North America and Europe ; the 

 camels of Asia and the llamas of the Andes had many 

 extinct common ancestors in North America ; the lemurs 

 of Africa and Asia had their ancestors in Europe, as had 

 the trogons of South America, Africa, and tropical Asia. 

 But besides this general evidence we have direct proof that 

 the struthious birds had a wider range in past times than 

 now. Remains of extinct rheas have been found in 

 Central Brazil, and those of ostriches in North India ; while 

 remains, believed to be of struthious birds, are found in 

 the Eocene deposits of England ; and the Cretaceous rocks 

 of North America have yielded the extraordinary toothed 

 bird, Hesperornis, which Professor O. Marsh declares to 

 have been " a carnivorous swimming ostrich." 



As to the second point, we have the remarkable fact 

 that all known birds of this group have not only the rudi- 

 ments of wing-bones, but also the rudiments of wings, that 

 is, an external limb bearing rigid quills or largely-developed 



