PACIFIC SLOPE OF ECUADOR, PERU AND CHILI. 513 



A possible route for the intermigration of genera from the At- 

 lantic to the Pacific side across the incipient Cordilleras may have 

 been provided in northern Peru long after the beginning of the up- 

 lift. East of the Piura River there is a saddle in the Andes with a 

 height of but 6,700 feet. At three thousand feet in the Magdalena 

 basin conditions are still possible for the principal genera inhabiting 

 the lowlands of the Guayas so that a lowering of the Andes of 

 4,000 to 6,000 feet would enable fishes to get across, if the condi- 

 tions were otherwise favorable for the transmigration.^^ 



The time when the segregation or translation took place can 

 only be given in terms of the lives of species. It happened before 

 the present species were differentiated, long enough ago to permit 

 many genera to develop on the Pacific slope but after the general 

 features of the tropical American fresh- water fish- fauna had ap- 

 peared. 



Summary. 



The fish-fauna of western Ecuador and western Peru has less 

 affinity for that of the Magdalena to the northeast of it than for 

 that of the Amazon with which it shows close relationship. Only 

 four of its species extend as far north as the Patia. 



The ancestors of but 2 of its genera certainly came from the 

 north (Panama), the ancestors of 14 of its genera certainly came 

 from the east, while the ancestors of the others may have come 

 from the north or east, probably from the east. The fauna can 

 most easily be explained on the assumption that it was segregated 

 from a fauna continuous from the Pacific to the Atlantic by the 

 elevation of the Andes into an insuperable barrier. The segregation 

 happened before the evolution of the present species and before 

 the evolution of many of the present genera. The segregated fishes 



1^ Prof. E. W. Berry expressed the opinion to the writer that the first 

 uplift in the region of the Andes was eroded to mature topography, the 

 present great height of the Andes is the result of a later uplift. The region 

 about Junin and to a less extent about Lake Titicaca shows the ancient 

 mature topography lifted to a height of 12,000 feet and more. Toward this 

 highlands the streams from the east and west have cut deep gorges. Bowman 

 according to Berry (1. c. LIV, 112) regards "the Andes as having undergone 

 progressive elevation throughout the Tertiary and he concludes that there has 

 been a change of elevation in the late Tertiary amounting to about 5,000 feet. 



