THE EXTERNAL STRUCTURE OF THE LEPIDOPTEROUS LARVA. 35 



first thoracic and eighth abdominal segments are sometimes larger 

 than those on the other segments. This is supposed to be due to the 

 fact that the air- tubes from these spiracles ramify over a greater area 

 of the body than do those from the others. 



The tubes or tracheae which branch from the spiracles, carry air 

 to almost all parts of the body. The tracheal tube, or atrium, which 

 leads into the body from each spiracle, is provided with a muscular 

 apparatus for excluding foreign bodies, and for the control of the 

 admission of air into the tracheal system. These are, as it were, 

 muscular valves, and Landois describes the mechanism for this pur- 

 pose as consisting of four principal parts the bow, the lever, the band, 

 and the muscle. The contraction of the latter, acting on the lever, 

 causes the band and bow to meet and thus to close the passage. When 

 the muscle relaxes, the natural elasticity of the parts causes them to 

 separate again, and thus leave the tracheal tube open. The spiracle, 

 then, leads into the atrium, which passes, by means of a muscular 

 valve, into another chamber or vestibule, which, by means of another 

 valvular arrangement, leads into the tracheal tubes proper. Lowne con- 

 siders that the vestibule acts as a pump to force air into the tracheae. 



The true legs of insects are prolongations of the body wall, and 

 consist of : (1) The tarsus (or foot). (2) The tibia (or shank). 

 (3) The femur (or thigh). (4) The trochanter. (5) The coxa (or 

 base). The lepidopterous caterpillar has three pairs of true legs, one 

 pair being attached to each of the thoracic segments. They are five- 

 jointed, the two basal joints being, usually, larger than those which 

 follow ; these joints are of a fleshy structure, whilst the three beyond 

 are leathery or horny. The terminal joint is armed with a small, 

 usually curved, simple unguis or claw. Packard states that, besides 

 the terminal claw on the larval foot, there is apparently a second 

 rudimentary one at the base, which he calls a spine-like " tenant 

 hair," and sometimes also flattened lamellate set. The use of the 

 claw and tenant-hair, as grappling organs, is quite apparent ; the use 

 of the set (which may be identical with Chapman's " battledore 

 palpus ") is not known. 



The prolegs are also extensions of the integument, and consist, 

 usually, of two large, stout, fleshy joints, which are generally retrac- 

 tile within each other and the body-wall. The character of the pro- 

 legs is very important, and the arrangement of the hooks which termi- 

 nate them has recently been shown to have a distinct bearing on the 

 relationships of the various super-families of the Lepidoptera, and to 

 give important clues to their lines of evolution. In butterfly larvae 

 there is usually to be found on the inner side of the tip of the prolegs 

 a pair of thickened pads, which move laterally. These usually bear 

 a row of minute, but in some instances, very powerful hooks. 



The prolegs are found in most lepidopterous larvae on the third, 

 fourth, fifth, sixth and terminal abdominal segments, the last or anal, 

 pair, passing both downward and backward, and being, sometimes, 

 more plentifully supplied with little hooks than the other prolegs. 

 These hooks are embedded in the skin, and are arranged usually in 

 three rows, of which, however, sometimes only one and sometimes 

 two are developed. The hooks can be apparently extended at will, and 

 the tip of the foot, between the pads, may be so inflated in some 

 butterfly larvae as to bring the rows of hooks outside, and then the 



