424 TOWER, DARWINISM. 
cist considers these data to prove the perfection of. mimicry to a 
degree that enemies are no longer able to discriminate between models 
and mimics. However it would be of interest to have conclusive 
evidence that the postulated discrimination exists at all. There is 
no escape from the consequences arising from the fact that of all 
Lepidoptera killed over these plots, only 0.967% or about 21,464 
were killed by all vertebrate enemies, which is too slight an 
elimination to have any determining effect in any selective alterations. 
. Similar results were obtained between 1905 and 190? in the 
valley of the Rio Coatzacoalcos, and between 1911 and 1914 in 
the valley of the Rio Changuinola in northern Panama. In all of 
the locations specimens were taken of the living adults at the 
time each census was made, especially of the models and mimics, 
for comparison with the eliminated. These collections of the survivors 
were later- compared with the eliminated with the result that in 
the color patterns no differences were detected between the two 
groups. These color patterns are supposed to be the principal 
basis of elimination and it is reasonable to expect that if the 
activities postulated by the Hypothesis of Natural Selection were 
realities in nature, findings different from those obtained would 
have been discovered. 
The general conclusion from these observations is that it is 
chance position relative to the accidents of life that were operating in 
the elimination of these models and mimics, rather than any favorable 
or unfavorable characteristics possessed by either the survivors or 
the eliminated. Moreover the eliminated made so small a percentage 
of the adults that it did not indicate any intensity of struggle for 
existence against the selective predacious enemies. Of course it 
may be assumed that juvenile conditions are correlated with mimetic 
adult characters and that the selective action takes place in the 
approximately 900/) or more of elimination that occurs in each 
generation before the adult condition is reached. It is precisely 
this sort of contention which has been too commonly indulged in. 
I did not find what I expected to find, and since the selective 
effect was not found io be operating on the basis of the mimetic 
patterns, there is no a priori reason to suppose that the juvenile 


