352 SEXUAL SELECTION. [Part II. 



settled in numbers on thelblue float of a fishing-line ; while 

 two other species were attracted by shining white colors. 



It is an interesting fact, first observed by Schelver, 

 that the males, in several genera belonging to two sub- 

 families, when they first emerge from the pupal state are 

 colored exactly like the females ; but that their bodies in 

 a short time assume a conspicuous milky-blue tint, owing 

 to the exudation of a kind of oil, soluble in ether and alco- 

 hol. Mr. MacLachlan believes that in the male of Libel- 

 lula depressa this change of color does not occur until 

 nearly a fortnight after the metamorphosis, when the sexes 

 are ready to pair. 



Certain species of Neurothemis present, according to 

 Brauer, 61 a curious case of dimorphism, some of the females 

 having their wings netted in the usual manner; while 

 other females have them " very richly netted as in the 

 males of the same species." Brauer " explains the phe- 

 nomenon on Darwinian principles by the supposition that 

 the close netting of the veins is a secondary sexual char- 

 acter in the males." This latter character is generally de- 

 veloped in the males alone, but being, like every other 

 masculine character, latent in the female, is occasionally 

 developed in them. We have here an illustration of the 

 manner in which the two sexes of many animals have 

 probably come to resemble each other, namely, by varia- 

 tions first appearing in the males, being preserved in 

 them, and then transmitted to and developed in the fe- 

 males ; but in this particular genus a complete transference 

 is occasionally and abruptly effected. Mr. MacLachlan 

 informs me of another case of dimorphism occurring in 

 several species of Agrion in which a certain number of 

 individuals are found of an orange-color, and these are in- 

 variably females. This is probably a case of reversion, 

 for in the true Libellulse, when the sexes differ in color, 



61 See abstract in the 'Zoological Record' for 186*7, p. 450 



