MAYER: DEVELOPMENT OF WING SCALES. 217 
now thrown up into a regular series of ridges, which run across the 
wing, that is to say, at right angles to the general trend of the nerv- 
ures. Each ridge corresponds in position with a row of formative cells, 
and each furrow with the interval between two adjacent rows. Indeed, 
the nature of the folding is such as to show clearly that its character 
depends on the growth and arrangement of the formative cells. Asa 
consequence of this arrangement, the scales always project from the tops 
of these ridges. The Grundmembran (mdr. pr., Fig. 8) has not partici- 
pated in this folding, and the deep processes (pre., Fig. 8) of the hypo- 
dermis that once extended to this membrane have largely disappeared. 
Figures 9 and 10 (Plate 2) represent a still more advanced stage of 
the pupal wings, drawn in this case from Danais plexippus, about eight 
days before emergence from the pupa; the condition is the same, how- 
. ever, in all the other forms examined by me, a similar condition occurring 
in Callosamia promethea about eleven days before the moth issues. Fig- 
ure 10 is a portion of a longitudinal section through the wing, which 
therefore cuts across the ridges. Figure 9 is a much more highly mag- 
nified view of one of these ridges. In Figure 10 cta.’ represents the outer 
’ is the inner cuticular mem- 
chitinous cuticula of the pupa, and cta.’ 
brane, which we saw in Figure 5 lying almost in contact with the 
hypodermis cells. Now, however, it has been pushed outward by the 
development of the scales (sg.). It is evident that in the stage repre- 
sented in Figures 9 and 10 all traces of the protoplasmic processes which 
bound the hypodermis to the Grundmembran have disappeared. The 
Grundmembran, indeed, is now nothing more than a simple homogeneous 
structure, the stellate cells which were so evident in Figure 6 (Plate 2) 
having almost entirely disappeared. In most places it has the appear- 
ance of a structureless membrane lying below the hypodermis, but here 
and there one finds that its outer surface is striated, as shown in Figures 
26 and 27 (Plate 4). The striz run, for the most part, parallel to one 
another, and always across the wing; that is to say, perpendicular to 
the trend of the nervures. In appearance they remind one of the striz 
which are found upon the scales, excepting that they are very much 
farther apart. Figure 26 is an edge view of this membrane ; Figure 27, 
a view of its outer surface. 
Figure 12 (Plate 3) is drawn from a slightly later stage of Danais 
plexippus. In this case the Grundmembran presents nearly the same 
appearance as in the stage at present under discussion, and though the 
specimen is a little older than the one last described, the membrane 
still exhibits traces of the nuclei (x/.) of the stellate cells which pro- 
