have not been as accurately described here as in the fore-wings. Easy of recognition, capable of being committed to paper, if not to memory, at once and without any dissection, unfailing in their accuracy of definition, it would be impossible to overrate the importance of these characters for this purpose. But it would be useless to try to convey a correct idea of the arrangement of all these markings without the aid of diagrams or properly arranged specimens. With the aid, however, of a correct drawing or of a wing spread out on a piece of white paper, we may learn most of this important lesson without difficulty. Fig. 4. Nervures of the wings of the hornet. The four cubital or submarginal cells are indicated by the numbers. 3 Thus, on the front edge of the fore-wing is a dark spot called the stigma. A line curving outwards and backwards from this point forms the posterior boundary of an elongated oval space. This line is the radial nervure, and this space is the radial or marginal cell. To the posterior convex side of this cell parts of four spaces fit. These are the cubital, called also the sub-marginal cells;* for nomenclature has been very busy, causing confusion *See an elaborate paper by Mr. Shuckard, whose nomenclature I have adopted side by side with that of Latreille, in the 'Transactions of the Entomological Society of London,' Vol. I, p. 208. 1836. "A description of the superior wing of the Hymenoptera.' There are ten systems of nomenclature, more or less different from each other. here, and these little spaces and wandering lines have more than one set of names. The second and third of these cells, counting from the root of the wing, are the most important; their shape, their proportions to each other, and their connection with the various nervures being points to be specially attended to. The third cell is irregularly quadrilateral. The second is irregularly hexagonal; it is in contact with the marginal cell already described, by a broad surface, not by a narrow neck; and its bounding line inosculates with two lines called the recurrent nervures, as well as with the line which bounds the first cubital or sub-marginal cell. To assure ourselves of the reality of these distinctions, and to fix all these details exactly in the memory, we should compare the wing of a wasp with that of a bee, or with that of any of the little wasp-like insects which we may find in flowers. These nervures are the frame-work over which the membrane of the wing is spread. They give strength and direction to its blow, and radiating backwards from the root of the wing, assist in its feathering. The nervures are, or were in their first development, essentially air-tubes, and, as such, are lined with a spiral thread, which may be traced in some of the fine ramifications in a modified form. At the root of the wing, where they are gathered together to form the articulating surface on which the wing moves, they are thickest, and from this point they fade away till they are lost in the membranous expansion near the edges. The root of the fore-wings is covered in front by a small membranous scale called the tegula or wing-cover. The hind wings have no tegulæ, and are not folded in repose like the fore-wings. The membrane of the wings which is spread out on this frame-work is twofold, the nervures lying between the two layers, like the sticks in a paper fan. The outer surface is set with small short hairs or bristles which are most largely developed on the front edges of the wings. On the front edge of the hind pair, these hairs are for a short space still more enlarged, and turned over so as to form a row of spiral hooks. These hooks vary much both in form and number in the different orders of Hymenopterous insects. In all the female Vespa which I have examined, both English and foreign, there were thirtytwo hooklets. But in the males, both of the hornet and of the smaller species of English wasps, the number was considerably smaller and very uncertain. One had thirty, one twenty-eight, one only fifteen, but the greater part had from eighteen to twenty hooks. In one Odynerus there were fourteen only, while in different species of sand-wasps from Penang they ranged from nineteen to fifty-seven. Mr. Newport* infers, from the occurrence of a smaller number of hooklets in the weaker flying of the two sexes, that their number has a direct relation to the power of flight. Thus, the feebly flying male of the humble bee has only eighteen, while his partner has twenty-five; and, conversely, the male of the honey-bee, which flies so strongly, has twenty-one, while the female and worker have only seventeen and nineteen respectively. I would not question the general inference, but I doubt whether the numbers are constant enough to build any accurate conclusion upon. The numbers in the female Vespæ have appeared to me constant. But I * Cyclopædia Anat. Phys. Insecta,' Vol. II, p. 930. make the number in the female humble-bee twentythree only, and in the Ligurian honey-bee, seventeen in the male, twenty-three in the worker, and eighteen and twenty on the two sides of the female in my cabinet. They are seen to greatest advantage in the humble bee, but they come out very beautifully, like a rope-moulding, in the hornet, as shewn in the accompanying sketch. Corresponding to the row of hooklets Fig. 5.-Diagram of the hooklets on the hind wing of the hornet, seen from the lower surface. on the hind wing is a crest on the posterior edge of the fore-wing, forming a ledge beneath which the ends of the hooks catch. The points of the hooks turn up, and the corresponding ledge or flange turns down to meet them. The locking of the wings in Lepidopterous insects is effected by a different mechanical arrangement to that which obtains in wasps and bees. Thus: parallel to the front edge of the hind wing, and springing from its root, is a long, strong bristle. Near the root of the fore-wing, a slight horny process, somewhat hooked, rises from one of the nervures. The bristle is readily caught by this, or by the tuft of hairs which in some species answer the same purpose, so that the locking or unlocking is effected in a moment. The apparatus varies in form in different species, it is found chiefly among the moths; my own observations were made on a humming-bird moth. Butterflies, whose up and down flight contrasts strongly with the smooth rapid course of the sphinx-moths are entirely destitute of it.* Other forms of locking-gear are found in the Hemiptera. The Notonecta, or boat-fly, which we know best as it sculls along on its back, locks its wings in flight by means of a little hook which plays along the front nervure of the hind wing. This hook springs from the under surface of the fore-wing, at the point of junction between the clavus and the membrane of the elytron. And the same form of hook is found in the tropical Belostoma, one of the Nepidæ, an insect whose large size adapts it admirably for the examination of the organs of flight. In the British representative of the Nepidæ, our water-scorpion, the hook is made by the edge of the clavus being turned down. The outer half of this edge hooks into the hind wing in flight, while the inner half locks the elytron, in repose, to the edge of the thorax. Unless we bend the thorax, and so loosen the catch on its posterior edge, just as the insect would do, in preparing for flight, by the action of the muscles, it requires a great deal of force to spread the wings of the larger species. In the locking-gear as displayed in these different instances, and probably many other similar ones might be found on more extended examination,† it is to be observed that the wings are not merely hooked together, but that provision is made to keep them parallel and in contact during motion. The connec * See Westwood, Modern Classification,' Vol. II, p. 317. |