MAYER: DEVELOPMENT OF WING SCALES. 213 
agrees very well with what I find in the larvæ of Pieris rape and Danais 
plexippus. 
When the larva changes into a chrysalis, the wings expand to about 
sixty times their former area, and as a consequence the cells which com- 
pose the wall of the wing-pad, being no longer crowded together, lose 
their spindle shape and flatten out into a pavement epithelium. 
Figure 4 (Plate 1), and Figures 5, 6 (Plate 2), represent the condition 
found during the winter months in the pups of Samia cecropia, but this 
condition is also quite typical for the overwintering pupæ of Callosamia 
promethea, Pieris rape, Papilio turnus, and Papilio asterias. The same 
condition is likewise found in the young summer pupo of Vanessa 
antiopa. 
Figure 4 is a longitudinal section taken near the free lower edge of 
the chitinous wing sheaths of the chrysalis, and Figure 5 is a small por- 
tion of the same section more highly magnified. Figure 6 is a view look- 
ing down obliquely upon the epithelium of the wing, the outer chitinous 
euticula of the pupa having been removed. 
| The chitinous outer cuticula (cta.’) of the pupa encloses each wing i 
1 
-— 
a separate sheath, — as is shown in Figure 4, where the upper wing is 
| seen lying above the lower, — and exhibits a layered or stratified con- 
dition; it is deeply pigmented near. its outer surface. This is best 
| seen in Figure 5 (c.“ ). A delicate structureless membrans, the inner 
cuticula (cta.”, Figure 5), lies between the outer cuticula and the 
hypodermis. 
It is evident that in this stage each wing consists of a hollow bag, the 
wall of which is composed of a single layer of hypodermis cells (A'drm., 
Figs. 4, 5, and 6). These hypodermis cells contain large oval nuolei, 
which exhibit, chromatin granules arranged near the periphery. 
The middle membrane (mbr. m., Fig. 3, Plate 1) has disappeared as 
such, and in its place one finds a delicate membrane (mbr. pr.) lining 
the whole interior of the wing-baes. This is the “Grundmembran ”of 
Semper (57), who showed that it was produced by mesenchymatous 
cells, whieh applied themselves to the deep surface of the hypodermis, 
and sent out lateral processes, serving both to connect the cells with one 
another and to give them a stellate form. Semper found that these 
stellate cells secreted an intereellular substance, filling up the inter- 
stices of the network formed by them, and that this substance, to- 
gether with the metamorphosed cells that produced it, finally became 
the thin structureless membrane to which he gave the name Grund- 
membran. This explanation I believe to be entirely correct. 
