188 REPORT UNITED STATES ENTOMOLOGICAL COMMISSION. 
numerous, much thicker, and shorter than the cutlcular hairs. They 
have rounded apices, and are inclined backward. They do not seem 
to correspond in any way to the hairs, for they do not rest over pores, 
nor have I seen any specially modified cells underlying them. As 
far as I have observed, they are mere local irregularities, each nodule 
being apparently supported by some four or six unmodified epidermal 
cells. 230 The cones in those specimens I have examined are entirely want- 
ing in the ventral arch, in the upper portion of the dorsal arch, but on 
the sides of the dorsal arches they are of considerably greater dimensions 
than upon the spiracular membrane, and finally they are larger and more 
numerous on the anterior than on the posterior abdominal segments. 
I have not, however, attempted to follow out the distribution of these 
structures in greater detail. 
Second, the whole of the cuticula except the cones just described and 
the hairs, is divided into numerous minute fields (Figs. 63 and 65), each 
of which corresponds to a single cell of the underlying epidermis. Each 
field is bounded by a distinct polygonal outline, and its surface is either 
covered by a large number of extremelj 7 minute projecting points (Fig. 
65), as on the dorsal arch, or is smooth as upon the articular membrane 
and ventral arch. Upon the sides of the dorsal arch and upon the spirac- 
ular membrane each held has a projecting spine or sometimes two or even 
three. Fig. 65 represents a surface view of part of the side of one of the 
dorsal arches. Upon the articular cuticula each spine springs from a 
short basal collar. Fig. 63 represents a surface view of the upper and 
anterior part of the dorsal arch. The fine scidpture is drawn only on 
a few of the fields, none of which have spines. The figure is intended 
to show that from the smooth circular area around the base of the hairs, 
/t, the fine points of the sculpturing appear to radiate, while elsewhere 
they are only irregularly distributed. I have been unable to determine 
how this radiating appearance is caused. 
The ventral arch has a quite smooth surface and but few hairs. The 
articular membrane has few hairs, a number of broad cones, and sculp- 
tured fields, bearing spines, which have a thick collar around their 
bases. The dorsal arch resembles this membrane generally, but differs 
from it by the simpler character of its spines, by their absence from its 
upper portions, and by its brownish tinge. The cuticula between the 
segments resembles the spiracular membrane. 
The cuticula forms also the stigmata or openings of the tracheal sys- 
tem. Immediately around each opening the cuticula is perfectly smooth, 
while to form the stigmata it undergoes various modifications, which I 
have not studied. The cuticula of the thorax, head, and limbs I have not 
examined. 
Epidermis. — The cuticula is secreted by an underlying layer of cells, 
the epidermis proper, often called the matrix or hypodermis, but inas- 
00 1 think it possible that the examination of sections, which the imperfect preservation of the parts 
prevented my making, will show that the cones are after all really produced by specialized cells. The 
surface views I have obtained are none of them thoroughly satisfactory in this respect. 
