736 
ZOOLOGY: W. H. LONGLEY 
a way that, upon the average, their conspicuousness would apparently 
be increased if their pigments were very different from those they 
display. 
Additional research demonstrates that in so far as this class of ani- 
mals is concerned there is no ground for the belief that bright color is 
correlated in any way with armament or distastefulness. If a list of 
species possessing organs capable of inflicting painful bodily injury 
be compiled, their colors and ability to change them are found to dif- 
fer in no essential respect from those of any other group of the same 
size selected at random. That no unpalatability is correlated with 
gaudy coloration is proved by the fact that forms unsurpassed in bril- 
liancy constitute an important part of the food, and may be recovered 
regularly from the stomachs of snappers, which are among the com- 
monest of the predaceous fishes of the Tortugas. Hence, in view of all 
the evidence, it seems improbable, to say the least, that the bright colors 
of some fishes differ in function from the dull hues of their more modest 
congeners. But since the most highly colored of these creatures vie 
with birds and butterflies in vividness of coloration, it becomes necessary 
to revise all hypotheses which postulate conspicuousness. 
It is interesting to find that mimicry among insects may be explained 
much more consistently than is otherwise possible upon the assumption 
that even the colors of the most gaudy tend to reduce their visibility. 
Every authenticated fact adduced by the supporters of the mimicry 
hypotheses may be accounted for, and most, if not all of the criti- 
cism levelled against them may be met, if the matter be set forth as 
follows : 
Mimicry has arisen through bionomic pressure applied first by in- 
discriminate feeders, which have forced upon their accustomed prey 
color combinations which most eft'ectually conceal it in its normal envir- 
onment. In addition, by chance, in a few of many thousands of cases 
in which colors appropriate to the surroundings and habits of their pos- 
sessors have been evolved, patterns have appeared, sufficiently like one 
another to deceive enemies that exercise discrimination in their choice 
of food. From this point onward the evolution of resemblance has 
proceeded according to accepted formulae, without conspicuousness 
being involved in the process. 
We may assume that the Pieridae and Heliconidae, for example, are 
usually distinctly different in habit, and that the coloration of typical 
members of each family is a combination of hues well suited upon the 
average to render them inconspicuous in such places as they frequent. 
If this be so, the initial step toward mimicry might be any one of many 
