Annals Entomological Society of America [Vol. VIII, 
oo 
OO 
CO 
the mutationist that a sudden and pronounced deepening of color 
of the cocoon would be necessary to insure the survival of the 
pupa will have little force, because even slight differences in 
color in the direction of greater opacity would be conceivably 
useful. But no variations would be significant unless the 
workers developed concomitantly the instinct to expose the 
cocoons to the heat and hght on the surface of the nest. The 
neovitalist might claim, therefore, that the whole phenomenon 
is merely another instance of the prospective potency of the 
colonial psychoid. The modest naturalist who does not 
habitually abide in this tenuous atmosphere of supreme specula- 
tion will be satisfied to regard the melanism of the cocoons as a 
detail of the larger subject of the coloration of ants in general. 
As almost nothing has been published on this subject I may be 
permitted to consider it briefly in these concluding paragraphs. 
Adult ants range in color from pale yellow (the primitive 
color of chitin) through various tones of red and brown to black 
and, through interference colors superadded to pigmentation, 
from bronze or zneous, through metallic green, blue and violet 
to metallic crimson. As here enumerated the shades evidently 
represent the order of the phylogenetic development of insect 
coloration generally, as indicated both by the ontogenetic 
sequence and the comparative study of the various.species and 
genera. It is not so generally known that both the pigment and 
interference colors may be shown to depend intimately on the 
amount of light to which the ants are subjected and therefore 
on their mode of life. 
The primitive, ancestral colors of .the Formicide were 
probably red, brown or black, precisely those still prevalent in 
the majority of species inhabiting temperate regions or the 
forested portions of the tropics, where the workers, while 
foraging on the surface of the earth or of the vegetation, are 
exposed to a moderate insolation. From this condition we can 
trace three divergent lines of development, each terminating in 
a peculiar type of coloration, as follows: 
1. A number of species have greatly exaggerated the neg- 
ative phototropism, which is characteristic of most ants, and have 
therefore taken to a nocturnal or hypogeeic mode of life. In all 
of these forms pigment has been lost, and the workers of the 
hypogeeic species, which never come to the surface of the ground, 
