48 ANNALS NEW YORK ACADEMY OF SCIENCES 



of little or no rainfall, cactus, spiny and thorny plants and scrubby trees, 

 dead insects, starving animals, hot dry air, and a gray dead appearance of 

 everything excepting the sun, moon and stars. 



When I think of Mattos Grosses, I think of forests and jungles which 

 are composed of an amazing number of kinds of plants all mixed up, 

 loaded with vines and orchids and harboring monkeys, parrots, sloths, 

 armadillos, humming birds, insects and other animals galore. In other 

 words, a region where life is at its present climax, a region where the con- 

 ditions for the existence and evolution of life are most favorable. 



In like manner, we can arrange the rivers of South America into a few 

 groups in which the sum total of the natural conditions are almost iden- 

 tical, excepting the volume of water. These groups of environmental 

 complexes are shown on Plate XV and are as follows : 



1. In the Campos of the State of Eio Grande do Sul and Uruguay is 

 Eio Uruguay on one side and Eio Grande do Sul on the other side of the 

 divide. 



2. Eio Parana and its affluents, Eio Tiete, Eio Grande and Eio Parana- 

 ponema with their cool, clear water rushing over many waterfalls, are on 

 the west slope of Serra do Mar, and Eio Bibeira de Iguape, Eio Parahyba 

 and Eio Doce on the east side of the same sierra. 



3. The northern part of Eio Sao Francisco belongs to the "secca" re- 

 gion, which includes the most of Eio Parahyba do Norte ; the western part 

 belongs to the highland region, which is also embraced by the Amazon, 

 and the southern part of Eio Sao Francisco belongs to the region of Alto 

 Eio Parana. 



4. The Amazon is so diverse that it includes several smaller regions; 

 and Eio Essequibo and perhaps Eio Orinoco on one side and the Alto Eio 

 Paraguay on the other side belong to the Amazonian realm in reference 

 to the sum total of their natural conditions. 



5. The Patagonian and the West Andean rivers each belong to quite 

 distinct natural regions. Perhaps the Patagonian should include the 

 southern portion of the West Andean, i. e., Chile. 



The profound importance of a clear conception of these regions will, I 

 hope, become more evident in the last part of this thesis. 23 



23 It is well known that the environment produces changes in both the individual and 

 the species. When these changes are produced in the gemiplasm, they are transmitted to 

 the offspring. Is it not to be expected then that at least some of the offspring from the 

 same species of plants and animals would become different after the parent species ar- 

 rived in the different environments? And may it not also be expected that at least 

 some of the offspring from the same species under identical external conditions, even if 

 widely separated, will evolve some identical species? 



