404 GEOGRAPHICAL ZOOLOGY. [part iv. 



liar ; and it has about 40 peculiar genera in ten families, about 

 half of these genera belonging to the Scincidse. Only 3 

 families of almost universal distribution are common to the 

 Australian and Neotropical regions, with one species of the 

 American Iguanidse in the Fiji Islands, so that, as far as this 

 order is concerned, these two regions have little resemblance. 



The Neotropical region has 15 families, 6 of which are peculiar 

 to it, and it possesses more than 50 peculiar genera. These ai-e 

 distributed among 12 families, but more than half belong to the 

 Iguanidse, and half the remainder to the Teidee, — the two families 

 especially characteristic of the Neotropical region. All the Ne- 

 arctic families which are not of almost universal distribution are 

 peculiarly Neotropical, showing that the Lacertilia of the former 

 region have probably been derived almost exclusively from the 

 latter. 



On the whole the distribution of the Lacertilia shows a 

 remarkable amount of specialization in each of the great tropical 

 regions, whence we may infer that Southern Asia, Tropical 

 Africa, Australia, and South America, each obtained their original 

 stock of this order at very remote periods, and that there has 

 since been little intercommunication between them. The peculiar 

 affinities indicated by such cases as the Lepidosternidse, found 

 only in the tropics of Africa and South America, and Tachydromiis 

 in Eastern Asia and West Africa, may be the results either of 

 once widely distributed families surviving only in isolated locali- 

 ties where the conditions are favourable, — or of some partial and 

 temporary geographical connection, allowing of a limited degree 

 of intermixture of faunas. The former appears to be the more 

 probable and generally efficient cause, but the latter may have 

 operated in exceptional cases. 



Fossil Lacertilia. 

 These date back to the Triassic period, and they are found in 

 most succeeding formations, but it is not till the Tertiary period 

 that forms allied to existing genera occur. These are at present 

 too rare and too ill-defined to throw much light on the geo- 

 graphical distribution of the order. 



