1906.] m MEXICAN LIZARDS. 287 



species, subspecies, races and varieties, I have proceeded to point out 

 those individuals which upset the diagnoses. 



Cope has rightly said. Report U.S" Nat. Mus. for 1898 (1900) 

 p. 569 : — " The discrimination of the North American species of 

 this genus is the most difficult problem in our herpetology. No- 

 where are subspecies more sharply defined than in Onemidophor^is, 

 that is geographical forms, which are not always true to their 

 chai-acters." He, however, practically left the Mexicans untouched, 

 confining himself to those of the United States, 



Most of the " species " are so plastic, so variable, that they may 

 well drive the systematist to despair. Not two authorities will, nor 

 can, possibly agree upon the number of admissible species. 



The Cnemidophori in their unsettled condition, are truly 

 delightful as an ideal object lesson in Nature's way of species- 

 making. 



It has been my ambition to find truly intermediate individuals, 

 real links between the groups and between the reasonably supposed 

 species. This was suprisingly difficult ! It was a reasonable 

 premiss that such links should occur at the same place, at least in 

 the same district, with the two forms to be linked. If the link 

 occurs somewhere else, the question enters a new field of inquiry. 



It is fairly certain that the three forms of the cZe/^/jei-group, 

 C. deppei, G. immutahilis, and G. guttatus, are closely allied to 

 each other ; and it may now be taken as proven that G. immiotahilis 

 tvums into G. guttatus in consequence of living in the Atlantic 

 Tierra caliente. These two forms actually run into each other, 

 but they are easily separable when in their respective typical 

 garbs, in which case, moreover, slight structural differences are 

 apparent. Result : C. guttatus is a terminus of evolution, as 

 being the spotted race of 0. immutabilis — scientifically expressed, 

 G. immutabilis, var. guttata ; but thanks to accident, priority of 

 naming, it has to stand as G. guttatus guttattcs, and the parental 

 stock form stands as G. guttatus, var. immutabilis ! No sense in 

 that, but justice is done to the fetish, although not to the lizards 

 to which these paraphernalia should be subservient. 



Further, the Salina Cruz, Tequesixtlan, &c., specimens of 

 Oaxaca are typical, intensified G. immictabilis ; they can hold no 

 intercourse with those of the Atlantic side, a point about which 

 I am positive, owing to the configuration of the country. 

 Mingling still occurs on the isthmus proper ; and in the forest- 

 lands of Guerrero G. immutabilis tends to assume the spotted garb. 

 Now let us assume that these woods were destroyed for ever, and 

 that the divide between the Atlantic and Pacific hot-lands is also 

 laid bare, then we should have the typical G. guttatus in the 

 Atlantic Tierra caliente, and the typical G. immutabilis on the 

 Pacific coast : two good species, because they are well defined and 

 geographically separated. They were considei-ed as good species 

 by Cope and by Boulenger ; but I found the intermediate forms 

 in districts of intermediate bionomic conditions, so that now at 



