[Illustration: FIG. 67.—Longitudinal section of portion of caudal appendage of _Acheta domestica_ (after Vom Rath): _ch_, chitin; _hyp_, hypodermis; _n_, nerve; _h^1_, integumental hairs, not sensitive; _h^2_, ordinary hair; _h^3_, sensory hair; _h^4_, bladder-like hair; _sz_, sense-cell.] [Illustration: FIG. 68.—Longitudinal section of apex of palpus of _Pieris brassicae_: _sch_, scales; _ch_, chitin; _hyp_, hypodermis; _n_, nerve; _sz_, sense cells; _sh_, sense hairs. (After Vom Rath.)] Various parts of the mouth are also the seats of sense organs of different kinds, some of them of a compound character; in such cases there may be a considerable number of hairs seated on branches of a common nerve as figured {123}by Vom Rath[40] on the apex of the maxillary palp of _Locusta viridissima_, or a compound organ such as we represent in Fig. 68 may be located in the interior of the apical portion of the palp. The functions of the various structures that have been detected are, as already remarked, very difficult to discover. Vom Rath thinks the cones he describes on the antennae and palpi are organs of smell, while he assigns to those on the maxillae, lower lip, epipharynx, and hypopharynx the _rôle_ of taste organs, but admits he cannot draw any absolute line of distinction between the two forms. The opinions of Kraepelin, Hauser, and Will, as well as those of various earlier writers, are considered in Sir John Lubbock's book on this subject.[41] ALIMENTARY AND NUTRITIVE SYSTEM. The alimentary canal occupies the median longitudinal axis of the body, being situated below the dorsal vessel, and above the ventral nervous chain; it extends from the mouth to the opposite extremity of the body. It varies greatly in the different kinds of Insects, but in all its forms it is recognised as consisting essentially of three divisions: anterior, middle, and posterior. The first and last of these divisions are considered to be of quite different morphological nature from the middle part, or true stomach, and to be, as it were, invaginations of the extremities of a closed bag; it is ascertained that in the embryo these invaginations have really blind extremities (see Fig. 82, p. 151), and only subsequently become connected with the middle part of the canal. There are even some larvae of Insects in which the posterior portion of the canal is not opened till near the close of the larval life; this is the case with many Hymenoptera, and it is probable, though not as frequently stated certain, that the occlusion marks the point of junction of the proctodaeum with the stomach. The anterior and posterior parts of the canal are formed by the ectoderm of the embryo, and in embryological and morphological language are called respectively the stomodaeum and proctodaeum; the true stomach is formed from the endoderm, {124}and the muscular layer of the whole canal from the mesoderm.