THE DISTRIBUTION OF ANIMALS AND PLANTS 413 



land. Their exact geological age is unknown, but there is no 

 improbability in supposing that they may have existed with 

 more or less of extension since the Secondary or Mesozoic 

 period. In any case their fauna is in some respects a survival 

 of that age. Lyell has truly remarked, " In the fauna of the 

 Galapagos Islands we have a state of things very analogous to 

 that of the Secondary period." 



Like other oceanic islands, the Galapagos have no indigenous 

 mammals, with the doubtful exception of a South American 

 mouse ; but, unlike most others, they are rich in reptiles. At 

 the head of these stand several species of gigantic tortoises. 

 This group of animals, so far as known, commenced its exist- 

 ence in the Eocene Tertiary ; and in this and the Miocene 

 period still more gigantic species existed on the continents. It 

 has been supposed that at some such early date they reached 

 the Galapagos from South America. Another group of Gala- 

 pagan reptiles, perhaps still more remarkable, is that of iguana- 

 like lizards of the genus Amblyrhyncus, which are vegetable 

 feeders, one of them browsing on marine weeds. They recall 

 the great iguana-like reptiles of the European Wealden, and 

 stand remote from all modern types. There are also snakes of 

 two species, but these are South American forms, and may 

 have drifted to the islands in comparatively recent times on 

 floating trees. The birds are a curious assemblage. A few are 

 common American species, like the rice bird. Others are 

 quaint and peculiar creatures, allied to South American birds, 

 but probably representing forms long since extinct on the 

 continent. The bird fauna, as Wallace remarks, indicates that 

 some of these animals are old residents, others more recent 

 arrivals ; and it is probable that they have arrived at various 

 times since the early Tertiary. He assumes that the earlier 

 arrivals have been modified in the islands " into a variety of 

 distinct types "; but the only evidence of this is that some of 

 the species arc closely related to each other. It is more likely 



