106 PROCEEDINGS OF THE ACADEMY OF [1898. 



Another fact on which I insist is the distribution of certain forms 

 from northern Brazil to the La Plata River, with the exclusion of 

 the Brazilian littoral. The genus Boa is an example of this. Ameiva 

 surinamensis does not exist in Rio Grande do Sul, but is said to live 

 in the La Plata region, also in Bahia and in Paraguay. In Sao 

 Paulo, however, this species is rare, and represented in the western 

 zone only. Here we have also Tupinambis rufescens, extending from 

 Mendoza to western Sao Paulo and Mabuia frenata, a Paraguayan 

 species, which we have received from St. Rita and Paracicaba. 



The species of Mabuia have an interesting distribution. As we 

 mentioned above, the Paraguayan species 31. frenata exists in the 

 western parts of Sao Paulo. From Santos I have 31. agilis Raddi, 

 known hitherto only from Rio and north of Rio, but on the central 

 highland we have 31. dorsivittata Cope, extending from Rio Grande 

 do Sul to Paraguay and to Sao Paulo. Tropidurus torquatus and 

 Polychrus acutirostris are Brazilian species, distributed from Bahia 

 to Sao Paulo and extending to Paraguay. It should be possible by 

 this time, to determine analytically the various regional compo- 

 nents of the diverse faunas. 



I have elsewhere discussed these problems in distribution, having 

 been the first to direct attention to them. In a paper on the distri- 

 bution of Ampullaria I have mentioned, that of the species common 

 to the Amazonian and San Francisco region, A. canaliculate does not 

 occur in St. Catharina, A. sordida being substituted for it. Also at 

 Rio Janeiro and in Sao Paulo there are other species of Ampullaria. 

 This is only one example from a great list of species. Glabaris 

 riograndensis, etc. are not represented in the River Parana system, 

 but are common in Rio Grande do Sul. Besides the species common 

 to the Parana system and to Rio Grande do Sul, there are others 

 which reached Rio Grande and the La Plata States by the Paraguay 

 system. This is now separated from the Amazonian system, but the 

 faunal identity demonstrates this separation to be of very recent 

 date. 



The distribution of neither the fresh-water nor of the land faunas 

 of Brazil can be at all understood without reference to these facts. 

 For instance there is Bulimus (Borus) oblongus Mull., a form 

 common in Rio Grande do Sul and the La Plata region, but also in 

 Bolivia, Guiana and Venezuela. From St. Catharina to Sao Paulo 

 and Rio where this species does not occur, B. ovatus Mull, is sub- 

 stituted for it, and north of Rio B. cantagallanus Rang takes its 



