stand a little to the right, and the other half a little to the left. The distance of the two parallel lines thus formed is called the course of the saw, and it is only the portion of wood which lies in the course that is cut into sawdust by the action of the instrument. It will follow, that in proportion to the thinness of a saw there will be the less destruction of wood which may be sawed. When cabinet-makers have to divide valuable wood into very thin leaves, they accordingly employ saws with a narrow course; while sawyers who cut planks, use one with a broad course. The ovipositor-saw being extremely fine, does not require the teeth to diverge much, but from the manner in which they operate, it is requisite that they should not stand like those of our saws in a straight line. The greater portion of the edge of the instrument, on the contrary, is towards the point somewhat cancave, similar to a scythe, while towards the base it becomes a little convex, the whole edge being nearly the shape of an Italic f

The ovipositor-saw of the fly is put in motion in the same way as a carpenter's hand-saw, supposing the tendons attached to its base to form the handle, and the muscles which put it in motion to be the hand of the carpenter. But the carpenter can only work one saw at a time, whereas each of these flies is furnished with two, equal and similar, which it works at the same time-one being advanced and the other retracted alternately. The secret, indeed, of working more saws than one at once is not unknown to our mechanics; for two or three are sometimes fixed in the same frame. These, however, not only all move upwards and downwards simultaneously, but cut the wood in different places; while the two saws of the ovipositor work in the same cut, and, consequently, though the teeth are extremely fine, the effect is similar to a saw with a wide set.

It is important, seeing that the ovipositor-saws are so fine, that they be not bent or separated while in operation-and this, also, nature has provided for, by lodging the backs of the saws in a groove, formed by two membranous plates, similar to the structure of a clasp-knife. These plates are thickest at the base, becoming gradually thinner as they approach the point which the form of the saws require. According to Vallisnieri, it is not the only use of this apparatus to form a back for the saws, he having discovered, between the component membranes, two canals, which he supposes are employed to conduct the eggs of the insect into the grooves which it has hollowed out for them*.

The teeth of a carpenter's saw, it may be remarked, are simple, whereas the teeth of the ovipositor-saw are themselves denticulated with fine teeth. The latter, also, combines at the same time the properties of a saw and of a rasp or file. So far as we are aware, * Réaumur, Mem. des Insectes, v., p. iii.

these two properties have never been combined in any of the tools of our carpenters. The rasping part of the ovipositor, however, is not constructed like our rasps, with short teeth thickly studded together, but has teeth almost as long as those of the and placed contiguous to them on the back of the instrument, resembling in their form and setting the teeth of a comb, as may be seen in the figure. Of course, such observations are conducted with the aid of a microscope.

saw,

Portion of Saw-fly's comb-toothed rasp, and saw

When a female saw-fly has selected the branch of a rose-tree, or any other, in which to deposit her eggs, she may be seen bending the end of her belly inwards, in form of a crescent, and protruding her saw, at the same time, to penetrate the bark or wood. She maintains this recurved position so long as she works in deepening the groove; but when she has attained the depth required, she unbends her body into a straight line, and in this position works upon the place lengthways, by applying the saw more horizontally. When she has rendered the groove as large as she wishes, the motion of the tendons ceases, and an egg is placed in the cavity. The saw is then withdrawn into the sheath for about two-thirds of its length, and at the same moment, a sort of frothy liquid, similar to a lather made with soap, is dropped over the egg, either for the purpose of gluing it in its place, or sheathing it from the action of the juices of the tree. She proceeds in the same manner in sawing out a second groove, and so on in

succession till she has deposited all her eggs, sometimes to the number of twenty-four. The grooves are usually placed in a line, at a small distance from one another, on the same branch; but sometimes the mother-fly shifts to another, or to a different part of the branch, when she is either scared or finds it unsuitable. She commonly, also, takes more than one day to the work, notwithstanding the superiority of her tools. Réaumur has seen a saw-fly make six grooves in succession, which occupied her about ten hours and a half.

The grooves, when finished, have externally little elevation above the level of the bark, appearing like the puncture of a lancet in the human skin; but in the course of a day or two the part becomes first brown and then black, while it also becomes more and more elevated. This increased elevation is not owing to the growth of the bark, the fibres of which, indeed, have been destroyed by the ovipositor-saw, but to the actual growth of the egg; for, when a newlaid egg of the saw-fly is compared with one which has been several days enclosed in the groove, the latter will be found to be very considerably the larger. This growth of the egg is contrary to the analogy observable in the eggs of birds, and even of most other insects; but it has its advantages. As it continues to increase, it raises the bark more and more, and consequently widens, at the same time, the slit at the entrance; so that, when the grub is hatched, it finds a passage ready for its exit. mother-fly seems to be aware of this growth of her eggs, for she takes care to deposit them at such distances as may prevent their disturbing one another by their development.

The

Another species of saw-1 v-fly, with a yellow body and deep violet-coloured wings, which also selects the rose-tree, deposits her eggs in a different manner. Instead of making a groove for each egg, like the

preceding, she forms a large single groove, sufficient for about two dozen eggs. These eggs are all arranged in pairs, forming two straight lines parallel to the sides of the branch. The eggs, however, though thus deposited in a common groove, are carefully kept each in its place; for a ridge of the wood is left to prevent those on the right from touching those on the left-and not only so, but between each egg of a row a thin partition of wood is left, forming a shallow cell.

Nest of eggs of Saw-fly, in rose-tree.

The edges of this groove, it will be obvious, must be farther apart than those which only contain a single egg, and, in fact, the whole is open to inspection; but the eggs are kept from falling out, both by the frothy glue before mentioned, and by the walls of the cells containing them. They were observed also, by Vallisnieri, to increase in size like the preceding.