MAYER: DEVELOPMENT OF WING SCALES. 215 
general hypodermis of the body is lined on the inner side by a thin 
membrane, coincident in relative position with the Grundmembran of 
the wings, and where this membrane is stretched, as in Figure 25, Plate 
4 (mbr. ba.), — which represents a cross section of the mid-dorsal re- 
gion just back of the head, where the cuticula splits when moults occur, 
— we see that the hypodermis cells send out processes which are con- 
nected with the membrane. This reminds one of the condition of the 
processes (pre., Figs. 4, 5) in the pupa. 
But to return to the discussion of the condition of the wings in the 
over-wintering pup. There is one more point to be noticed. The 
wings are filled with blood, or more properly speaking hemolymph, and 
this fluid contains blood corpuscles, which exhibit several shapes (/ew’cy., 
lewey.’, lewcy.”, Figs. 4 and 5). Some of these corpuscles (/ewcy.”, Fig. 
5) are much elongated or spindle-shaped, and their nuclei are oval. At 
one or occasionally both ends they exhibit long tail-like projections. 
Others, however (ew cy.’, Figs. 5, 6, and 7), which are found only in the 
very young pupee, are usually rounded or only slightly angular, and are 
often so vacuolated that the nucleus is crowded to one side and assumes 
a crescentic form. These vacnolated cells appear to be blood corpuscles 
. which are degenerating, for it is certain that there are no vacuolated 
blood corpuscles to be met with in the larvee, or in the older pupz. 
It seems probable to me that these transitory vacuolated corpuscles 
are the “ Fettkérper ” of Semper (’57, p. 327), for I find no true fat 
cells in the hemolymph. Schiffer (89) has shown that the leucocytes 
found in Lepidopterous larvze are morphologically equivalent to fat cells 
that have remained in an embryonic condition. He finds that the leu- 
cocytes are chiefly derived from large masses of fat cells which lie near 
the Anlagen of the wings in the larva, or from those which constitute the 
matrix of the tracheze. Most of the cells composing these masses are true 
vacuolated fat cells. Some of them, however, remain in an embryonic 
condition, never becoming vacuolated, and, separating off from the mass, 
become free in the body cavity. The cells which are thus set free be- 
come leucocytes. 
The wing remains in the simple histological condition just described 
until about three weeks before the insect is destined to emerge from 
the chrysalis. Then (Plate 2, Fig. 7) certain of the hypodermis cells 
(cl. frm.), which occur at regular intervals, begin to be modified. They 
begin to increase slightly in size, to project a little above the level of the 
ordinary hypodermis cells, and, most remarkable of all, to acquire each 
a vacuole. The cells which have become thus modified are destined 
