Chap. X.] NEUROPTERA. 351 



able that the males are thus rendered attractive to the 

 females. The Libellulidse or dragon-flies are ornamented 

 with splendid green, blue, yellow, and vermilion metallic 

 tints ; and the sexes often differ. Thus, the males of some 

 of the Agrionidoe, as Prof. Westwood remarks, 48 " are of 

 a rich blue with black wings, while the females are fine 

 green with colorle'ss wings." But in Agrion Mamburii 

 these colors are exactly reversed in the two sexes. 49 In the 

 extensive North American genus of Hetserina, the males 

 alone have a beautiful carmine spot at the base of each 

 wing. In Anax Junius the basal part of the abdomen in 

 the male is a vivid ultra-marine blue, and in the female 

 grass-green. In the allied genus Gomphus, on the other 

 hand, and in some other genera, the sexes differ but little 

 in color. Throughout the animal kingdom, similar cases 

 of the sexes of closely-allied forms either differing greatly, 

 or very little, or not at all, are of frequent occurrence. 

 Although with many Libellulidae there is so wide" a differ- 

 ence in color between the sexes, it is often difficult to say 

 which is the most brilliant ; and the ordinary coloration 

 of the two sexes is exactly reversed, as we have just seen, 

 in one species of Agrion. It is not probable that their 

 colors in any case have been gained as a protection. As 

 Mr. MacLachlan, who has closely attended to this family, 

 writes to me, dragon-flies — the tyrants of the insect-world 

 — are the least liable of any insect to be attacked by birds 

 or other enemies. He believes that their bright colors 

 serve as a sexual attraction. It deserves notice, as bearing 

 on this subject, that certain dragon-flies appear to be at- 

 tracted by particular colors : Mr. Patterson observed B0 that 

 the species of Agrionida?, of which the males are blue, 



48 ' Modern Class.' vol. ii. p. 37. 



49 Walsh, ibid. p. 381. I am indebted to this naturalist for the follow, 

 ing facts on Hetaerina, Anax, and Gomphus. 



60 'Transact. Ent. Soc' vol. i. 1836, p. lxxxL 



