VII] THE SENSE ORGANS 151 



in complexity. It is interesting to note that the smallness of 

 the larval eye in the Libellulidae is due to the fact that only the 

 smaller (pigmented) ommatidia are functional in the larva. These 

 form the definitely projecting eye of the larva. Between these 

 and the epicranium proper there is a rather flat unpigmented 

 area, in which the elements of the larger (unpigmented) ommatidia 

 are gradually formed during later larval life. At metamorphosis 

 the functional eye of the larva becomes the lower pigmented 

 portion of the imaginal eye, while the nearly flat unpigmented 

 part becomes convexly curved to form the upper portion. In the 

 larva, the optic ganglion is actually at first only in connection 

 with the pigmented part, the connection with the unpigmented 

 ommatidia being a late outgrowth of lateral nerve-elements from 

 the ganglion. 



The Antennae (figs. 2, 65). 



The external form of the antennae has already been dealt with 

 in chap. n. They are of small size, both in the larva and the 

 imago. They are, however, comparatively larger in the larva, 

 and appear also to be more functionally active than in the imago. 

 Their reduction is clearly correlated with the increase in power 

 of the compound eyes. The largest antennae are found in some of 

 the older Zygoptera, such as Synlestes (fig. 2 F, G), while the smallest 

 occur in the Libellulidae and Aeschninae, where the eyes reach 

 their maximum development. The sense of smell, which has 

 been proved to be located in the antennae of insects, has in the 

 Dragonflies become entirely subordinated to that of sight. It is 

 doubtful whether it was ever at all well-developed in the Order. 

 The scent of carrion, which so powerfully attracts many insects, 

 would be a matter of no moment to the Dragonfly, which feeds 

 exclusively on living insects. On the other hand, their well- 

 developed sense of taste is sufficient to enable them to reject any 

 obnoxious morsel caught on the wing. 



In the antennae of the larva of Aeschna, Zawarsin [2(M] has 

 demonstrated the existence of a Johnston's organ (fig. 65). This 

 is undoubtedly a sense-organ. It may have an auditory function, 

 but is more probably connected with testing the purity of the 

 water. It lies in the second segment or pedicel. It is not visible 



