248 PROCEEDINGS OF THE AMERICAN ACADEMY 



It would be interesting to know exactly the rule for the change of 

 color in males and females. So far as I know, this change seems to be 

 between related colors, and not between complementary ones. But 

 my observations are far from having any conclusive importance ; in 

 some cases, as in Hetaerina, the change is effected by complementary 

 colors, red and green. The same investigation will have to be made 

 for hypodermal colors. As far as I know, even here the change seems 

 commonly and chiefly between related colors ; though some cases of 

 complementary colors have been observed. 



The dermal colors never change during lifetime ; the hypodermal 

 ones may be changed in some way, and are known to be altered in 

 some instances in a male or female during its lifetime, by sexual or 

 other influences. By sexual influences, for instance, yellow is changed 

 into orange, brown into red (in some Agrion females), milk-white 

 into blue in males, and into green in females (Platycnemis). By 

 other influences, for instance, by cold (Brauer) in hibernation, pale 

 yellow is changed partially into red (Chrysopa). But this change 

 is said to be produced by chromatophores by other naturalists 

 (Leydig). The hypodermal colors in one insect {Cassida aurickal- 

 cea) are stated to be changed by a voluntary act of the insect, and 

 the new color can again disappear (Harris). 



Sexual Selection. 



Everybody has studied with interest in Mr. Darwin's works (De- 

 scent of Man, vol. i. p. 374; Nature, vol. xxi. p. 237, No. 532) the ex- 

 position of his theory of sexual selection. The part concerning the 

 butterflies is a rather prominent one. But when we weigh the given 

 facts, they do not seem to be so conclusive as they are considered to 

 be. Comparatively very few species agree very well, or only well, 

 with Mr. Darwin's assertions. The fact that it is only peculiar to 

 butterflies, and does not occur in other families of Lepidoptera, 

 becomes more important when we find that the theory is exemplified 

 only in less than one per cent of the butterflies themselves. Why is 

 the whole main army so much behind ? — is a question nowhere found 

 answered. After all, the statement, that the male is so much more 

 beautifully colored for sexual selection, will apply only to exceptional 

 cases among Le[)idoptera. 



Papilio tunnis in North America offers a rather puzzling case. 

 Its southern dark-colored female, formerly Papilio glaucus, will be as 



