M. H, Rathke on the Respiratory Process in Insects; 108 
mon cylindrical and elastic canal, which, however, is only of 
moderate diameter. 
The respiratory process in these larvee can only be effected by 
the above-mentioned movements, both general and partial. If 
the larva contracts longitudinally, when it does not gain so 
much in breadth as it loses in length, the contents of the body- 
cavity, and therefore the tracheary system, must be compressed; 
and the air contained therein driven out of the stigmata, in 
larger or smaller quantity according to the shortening. On the 
cessation of the activity of the muscles, the air-vessels, being 
freed from the previous pressure, must dilate again by virtue of 
their elasticity, and cause the air to enter through the stigmata. 
The same process of exspiration must also take place when the 
larva bends strongly to one side; for as the convex side then 
becomes smoother, and the structures contained in it are par- 
tially driven over into the opposite side, it is more than probable 
that its trachez also are then compressed and partly emptied of 
air; and as the concave side is shorter and moreover filled with 
a portion of the structures from the other side, its trachez also 
must be compressed and compelled to part with their air. 
§ 26. In the larve of the Scarabai, or at least in that of the 
May Bug (Melolontha majalis), the epidermis of each abdominal 
segment, except the last, forms two moderately firm plates, the 
upper a little larger than the lower, united at each side by a 
thinner and more flexible part of the epidermis. In these softer 
parts are the stigmata. Both the upper and lower plates of 
every two segments move upon each other by means of several 
muscular bundles, of which some, and these the innermost, im- 
mediately surrounding the abdominal cavity, run straight for- 
ward from the hinder margin of one segment to the same margin 
of the other, whilst others take an oblique direction from within 
outwards and forwards, and others from without inwards and for- 
wards. Of these oblique muscles some have the same insertions 
as the straight muscles, but others pass from the middle of one 
segment to the hinder margin of the preceding one. In each ab- 
dominal segment, except the last, there are also two, and in some 
even three pairs of pretty strong muscles, which all run from 
the ends of the upper plate, over the softer parts of the skin, 
inwards and downwards. Those of the inner pair pass straight 
to the lower plate of the same segment, and are attached to this ; 
those of the other one or two pairs run obliquely forward, and 
attach themselves to the lower plate of the preceding segment. 
The hindmost segment has only one pair of muscles, and these 
are oblique. 
By the muscles running obliquely from above downwards and 
