466 PROCEEDINGS OF THE ACADEMY OF [April, 



Atlantic subregion to the Pacific coast into a Lower and an Upper 

 Sonoran accords best with present knowledge, the Lower Sonoran 

 seems to have about 74 peculiar species, and the Upper Sonoran plus 

 his Transition has but about 13, with 35 common to both, of which 

 19 may be regarded as Lower Sonoran and only 6 as Upper. The 

 remainder belong to the Atlantic fauna. It should be explained that 

 these estimates are based upon a somewhat conservative view as to 

 species, but if the method of minute analysis characteristic of the pres- 

 ent time be followed, the predominance of Lower Sonoran forms will be 

 even greater. 



Both from the logical necessity of the case and from the application 

 of one of the main criteria in common use, it seems clear that the areas 

 which at the retreat of the ice contained the elements from which our 

 present reptilian fauna has developed, must be sought for in the south. 



The briefest examination of the character and the abounding rich- 

 ness of the reptiles inhabiting the dry plateau extending from Texas 

 to Arizona and south into Mexico, as well as the physical geography 

 of the whole Sonoran subregion, are enough to show that the dispersal 

 centre of that faura M-as hfre, in Cope's Chihuahuan. 



The Austroriparian is less clear without close examination of evi- 

 dence. Two localities in this subregion excel all others in variety of 

 species, one in the extreme southeast toward Georgia and Florida, cor- 

 responding to the fauna termed by Cope Ocmulgian; the other, or 

 Louisianan, on the lower Mississippi. Fifty-one species are common to 

 both, and in addition the Ocmulgian, including Cope's Floridan, 

 presents 21 not extending into the Louisianan, which in turn has 14, 

 one half of them, however, only entering its western border from the 

 Sonoran. Of the two the Ocmulgian shows a decidedly greater diver- 

 sity in species. 



A study in some detail of the genera points in the same direction. 

 Little argument is required to demonstrate the proposition that genera 

 now common to the Nearctic and the western Palsearctic must have 

 had that range established at a time when a Tertiary north Atlantic 

 land connection coincided with a w^aim climate in the north, for the 

 absence from all beyond the northern border of the Neotropical of 

 such vigorous Nearctic genera as Coluber, Zamenis, Tropidonotus, 

 Eutcenia, Ophiholns and others, which are identical with or nearly re- 

 lated to Holarctic genei'a, renders it impossible that such community 

 in northern regions could have come about by means of the Antarctic 

 continent whose former existence is now urged by many palaeontolo- 

 gists, however well this would serve to explain such anomalies as the 



