405 



in Darien. It does not hold true for Bocas in western Panama or 

 for Cartago in Costa Rica, where for various reasons there is no 

 marked dry season at all, but two periods of maximum rains and 

 two periods of somewhat drier weather. Thus a year's record 

 for Bocas shows no month in which the fall was less than 5 inches, 

 July and August had ten or more inches and so did November, 

 December and January. This is for a year in which Bocas had 

 a total of 104 inches. Colon, with a three year average of 150 

 inches, had a marked dry season from January to April, the 

 average fall for these months being 2.08 inches. 



Within the region of the Zone there are marked differences 

 between the Pacific side and the Atlantic, with Barro Colorado 

 occupying an intermediate position, although it is on the At- 

 lantic slope and nearer Colon. Three yearly averages for Balboa, 

 Barro Colorado, and Colon give 77, 112, and 150 inches respec- 

 tively. A five year average for Barro Colorado gives 100 inches 

 yearly rainfall. 



By and large, as I shall show later, the amphibian distribution 

 reflects these differences in rainfall. 



There is a region of very elevated mountains in Costa Rica, 

 which extends into western Panama. This has its appropriate 

 effect on the fauna, and many species are known only from that 

 region. The Canal Zone lies in a region of especially low altitude, 

 the so-called Tanama fault.' No very elevated mountains are 

 to be found in Eastern Panama. 



Barro Colorado Island, then, is a locality of low elevation, of 

 medium rainfall with marked seasonal change. It thus lacks 

 species which need altitude, either more or less total rain, and 

 those which cannot stand a marked dry season. 



Several species seem confined to the immediate Atlantic coast 

 in the Canal Zone region, and a fair proportion of these seem 

 extensions from the Costa Rican region where the rains are not so 

 seasonal. Others again seem restricted to the Pacific side, while 

 a few seem to prefer the higher regions to the east and west of the 

 Zone. It is difficult to produce a list of these, since our knowledge 

 is so imperfect, but there are enough such cases to justify the 

 statement, and I would be rather surprised if any of these forms 

 were to be found on Barro Colorado. 



There are, however, forms found on both sides of the region, 



