28 HUMBLE BEE-WASP. Kraepelin has described at the end of the proboscis in the humble bee (Bombus), besides the hairs of touch, certain peculiar club-shaped hairs, which he believed were perforated at the end, and which he considered to be taste-hairs; and Haller has ascribed the same function to some very similar hairs which he found on the under lip of the Hydrachna. Fig. 27 represents the under side of the left maxilla of a wasp (Vespa vulgaris), after Will, magnified 55 times. Gm are the taste-cups; Shm, the protecting hairs; Tb, the tactile hairs. Fig. 28 represents a section through one of the tastecups, Sk is the taste-cone contained in the cup; it is perforated and continuous at the base with a nerve-fibre. Similarly, in the wonderfully beautiful and complex proboscis of the hive bee there is, between each of the trachea-like ducts, a row of minute pits (Fig. 29, Gs), with a central papilla, which have been described by Leydig, Meinert, Lowne, Kraepelin, and others, and are probably organs of taste. Kraepelin distinguishes four kinds of hairs on the proboscis of the fly: 1. Ordinary hairs, which are not hollow, and do not stand in connection with a nerve. 2. Hairs of touch. These are principally situated on the upper side. They are delicate, hollow, pointed organs, situated on a ring of the integument, and connected with a nerve. 3. Glandular hairs. These are larger than the former, and the chitinous ring is sometimes so much developed as to form a short cylin Gs 6 GS A -нь Gs Gs GS Sh L cis in the hive bee (Apis), Sh, guard-hairs; Hb, der surrounding the base of the Fig. 29.-Tip of the proboshair. The principal characteristic is, however, that the hair presents along one surface a deep furrow, and is connected at the base with a cellular organ. Kraepelin therefore considers that this is a gland, and that the secretion passes outwards along the furrow. Kunckel and Gazagnaire however, regard these also as sense-hairs. The supposed gland they consider to be a ganglion. 4. Taste-organs (Fig. 30). These lie in a row between * Kraepelin, "Zur Anat. und Phys. des Rüssels von Musca," Zeit. für Wiss. Zool., 1883. the trachea-like channels, and correspond to the similar organs in the bee (Fig. 29, Gs). Each of these resembles a double circle, which scarcely projects, if at all, beyond the general surface, and which he regards as a metamorphosed, hollow, perforated hair. At the base of each organ is a nerve, which at some little distance forms a multicellular ganglion, and the sheath of which, immediately below the skin, forms a delicate and short, but wellmarked, chitinous, cylinder. "99 J Fig. 30. gc gk Organ of taste of fly (Musca vomitoria); after Kraepelin. gn, Nerve; gg, ganglion; ax, axecylinder; gc, terminal cylinder gk, terminal cone. It may also be observed, at any rate in most insects, that while they are feeding the palpi hang down motionless, and evidently take no part in the operation. In reference to the sense of taste, I may also mention that an additional complexity arises from the fact that many insects possess more than one kind of salivary gland, and it is possible, as Wolff suggests, that the secretions may have different properties. In addition to this, Wolff thinks he has proved that the character of the secretion differs at different ages; that for many days after the bee has arrived at its imago condition, the glands are still imperfect and gradually increase to their full size. In old bees, again, according to him, the secretion diminishes in quantity. This, perhaps, throws some light on the division of labour. Forel has observed among ants that they remain for some days engaged in indoor * "Das Riechorgan der Biene." INDIVIDUAL DIFFERENCES. 31 duties, and do not leave the nest till some time after they have arrived at maturity. I have noticed, also, that some individuals seem to possess a finer sense of taste than others, and some light seems to be thrown on this difference by the fact that the number of the taste-pits is not the same in all individuals. Thus Will observed that the number on the tongue of Lasius flavus (our common yellow ant) varies from twenty to twenty-four, and in Atta from forty to fifty-two. The number of pits on the maxillæ is subject to still greater variations, and is not even always the same on the two sides of the same insect. On the whole, then, we may conclude that the organs of taste in insects are certain modified hairs situated either in the mouth itself or on the organs immediately surrounding it. CHAPTER III. THE SENSE OF SMELL. THE organ of smell is, in vertebrate animals, embedded in the mucous membrane of the nostrils, and in mammalia can generally be distinguished by its yellow or brownish colour. In birds, on the contrary, it presents hardly any peculiarity to the naked eye. For our knowledge of the minuter structure we are mainly indebted to Max Schultze. The cylindrical epithelial cells in the olfactory organs of man (Fig. 31) terminate in broad flat ends. Between them are rod-like filaments, which are supposed to expand into a ganglionic cell, terminating in a nerve-fibre. Schultze terms these olfactory cells. In other cases, as in birds, Amphibia (Fig. 32), etc., the olfactory cells terminate in fine ciliæ, or olfactory hairs, either one or many to each cell. These hairs are sometimes motionless, sometimes have a slight movement of their own. It is obvious that no one from the structure alone could have predicated the function; nor can we, I think, form to ourselves any satisfactory conception how such a structure conveys the impression of smell, or in what consist the differences between different odours. If, then, we know really so little as to the mode, or organs, by which the sense of smell is induced among |