1906.] IN MEXICAN LIZARDS. 289 



of 1200 miles asunder. C. dep-pei and C. sexlineaf,us, with regard 

 to each other, are two good species in the fullest sense, although 

 all their available characters may overlap, or intergi-ade, not only 

 singly, but conjointly. They are two old species, sprung from 

 one common stock, well and firmly established, representing each 

 other in widely separate and apparently very different countries, 

 one in the Tropics, the other in North America proper. Florida 

 and Texas have much of the type of the Tierra caliente, but it 

 would be hopeless to look for the tertkmi comjmrationis between 

 the more Northern States and the Tropics of Mexico and Central 

 America, unless we assume that the North- western Plateau, with 

 its ranges of mountains, fi-om the Western States right through 

 Mexico, has caused the evolution of the many other kinds of 

 Cnemidophori, which now separate and connect C. de2)pei and 

 C. sexlineatus * . 



Our problem is not to explain why these two species should 

 occasionally be so much alike each other in their widely different 

 habitats, but to investigate whether, how, and why the intervening 

 country, the bulk of Mexico, has turned its lizards into what they 

 are, namely the great gida7'is-gvo\xp. 



The family of the Tejidje is old. Of several dozen so-called 

 genera in South America, only two are found also in Mexico ; 

 namely, one species of Ameiva in the eastern and western hot- 

 lands, and the genus Cnemido2ihorus incl. Verticaria. This genus 

 is old, but not old enough to occur on the West Indian Islands, a 

 fact which limits it to the end of the Miocene epoch. Yucatan 

 was under water until the beginning of the Pliocene; it has 

 received its few Cnemidophori since that epoch, and the same 

 applies to the Atlantic lowlands along the Gulf to Florida. Only 

 C. guttatus and C. depj^ei have got into these parts of Mexico. 

 For Texas only C. sexlineatus and the little C. gidaris were 

 available immigrants. 



C mexicanus, a very distinct species, exists on the Tres Marias 

 Islands. Other species inhabit the small islands of Lower Cali- 

 fornia, both in the Gulf and to the west of the peninsula ; proofs 

 of the existence of the genus in Mexico in early Pliocene times. 



It is doubtful when the great central plateau between the 

 Eastern and Western Sierras Madres became dry land ; until late 

 Tertiary times it was an inland lake. Longest available for 

 terrestial creatures were Southern Mexico and the Pacific portion — 

 a great stretch of land from Central America to California, 

 including parts of the present Pacific Ocean. It is in this belt 

 that we have to look for the home of the Mexican and North- 

 American Cnemidophori. Their present distribution agrees well 

 with this hypothesis. There is an abundance of species in the 

 South and in the North- west, whilst towards the North and 

 East, across the plateau, occur far fewer forms. 

 The great TESSELLATUS-gvow^ is an illustration of a group centred 



* Lack of material has prevented me from corroborating Cope's statement tbat 

 C. sexlineatus and C. gularis absolutely merge into each other, of. p. 305. 



