So much for the respiratory system of insects as a group. We have seen how air is admitted into the body, how the entire bulk of the blood is continuously aerated, and how every particle of needless weight has carefully been dispensed with. There are many species, however, whose mode of life renders necessary certain further developments, in order that respiration may be carried on under circumstances which would otherwise render it impossible. Such, for example, are the various aquatic insects, which, while spending the greater part of their existence beneath the surface of the water, must yet be enabled to command a continual supply of atmospheric air. They are not, as a rule, furnished with gills like the fish, for it is necessary that they should be able to leave their ponds and streams at will, and become for the time terrestrial or aerial beings, subject to the same conditions as others of their class. But they are, nevertheless, provided with certain modifications of structure, which enable them to breathe with equal ease, whether submerged in the water, crawling upon the ground, or flying through the air. Even in these modifications there is considerable variety, dependent in all cases upon the requirements of the individual species. The Water-beetles, for instance, which must be able to lurk concealed among the weeds, &c., until a victim comes within their reach, and then to pursue and overtake it, carry down with them a supply of air in a kind of reservoir, situated between the body and the wing-cases. The former of these is concave and the latter convex, so that a chamber of considerable size is formed, containing sufficient for their requirements during a tolerably long period of time. And in these insects the spiracles, instead of being situated along the sides of the body, are placed upon the upper surface of the abdomen, so that they open into the air-chamber itself, and allow the respiration to be carried on without the slightest difficulty or inconvenience. There is only one drawback to this arrangement, and that is, that the increased buoyance prevents the insect from remaining beneath the water excepting at the expense of active exertion, unless it can find some submerged object to which to cling. Even this disadvantage, however, is more apparent than real, for, while on the watch for prey, it is necessary for the insect to remain as motionless as possible, and, when engaged in swimming, the peculiar action of the oar-like limbs neutralizes the tendency to rise towards the surface.