852 THE AMERICAN NATURALIST. [VoL. XXXIII. 
air it is easy to see everywhere anastomoses between the distal 
ends of the tracheoles. This is not shown in our figure ; indeed, 
it is difficult to see it in sections anywhere. The walls of the 
tracheoles are of extreme tenuity; the best of chitine stains 
leave them untouched; they are probably protoplasmic tubes 
at their extremities, such as would best subserve the respiratory 
process. Tracheoles of this type are everywhere relatively short. 
In the much discussed tracheation of the lepidopterous larval 
wing there are the two well-known systems: the temporary 
Fic. 86. — Wings of 7, KPRP I3-punctata, two stages later than those 
shown in Fig. 85: Z, £, trachez; Z7, £Z, tracheoles 
system of tracheoles, which enter the wing in the penultimate 
larval stage and which are functional in the last larval stage ; 
and the permanent wing trachez, which grow out into the 
wings in the last larval stage, but do not become functional 
until pupation, when they have acquired terminal and lateral 
tracheoles of their own. In each case the developing air- 
vessels arise from the epithelium of the lateral tracheal trunk ; 
but they do not communicate with the lumen of this trachea 
