MAYER: DEVELOPMENT OF WING SCALES. 221 
find from two to five spherical masses of chromatin within the forma- 
tive cell (Fig. 18, cl. frm’, and Figs. 23 and 24). 
It is quite evident that these cells, having finished the formation of 
the scales, and being of no further use in the economy of the insect, are 
undergoing degeneration. The amitotic division of the nucleus is prob- 
ably one of the signs of this degeneration. I have observed this amito- 
tic division only in the case of Danais plexippus; for although I have 
a very complete series of sections of Callosamia promethea, I have never 
observed it in this insect. 
It should be noted that at this stage in Danais plexippus a single 
leucocyte (Fig. 18, /ew'cy.) enters each of the scales situated either upon 
the nervures or near the outer edges of the wings. These leucocyte- 
bearing scales are about twice as large as the ordinary wing scales, which 
are situated between the nervures ; the latter, indeed, are far too small 
to admit the introduction of leucocytes. We shall discuss the signifi- 
cance of these facts under “ The Development of the Pigment within 
the Scales,” pages 224, 225. 
The manner in which the scales are inserted into the wing membrane 
will become apparent from an inspection of Figures 30 (Plate 5), 31 
(Plate 6), or, better still, 29 (Plate 5). The last figure, drawn from 
Danais plexippus, represents a cross section (i. e. perpendicular to the 
trend of the nervures) of a wing that is still in the ‘ 
5 
‘white stage.” 
The narrow cylindrical stalk of the scale is merely inserted into a mi- 
nute close-fitting socket, which perforates the wing membrane, as was 
first described by Semper in 1857; it is not set into a tube, as Landois 
(71, Taf. XXII. Fig. 10) imagined. 
I cannot find anything resembling the curious structure described by 
Spuler (95, p. 526, Taf. XXXVI. Figs. 2, 3, 4) as serving for the inser- 
, 
tion of the scales, and called by him the *Sehuppenbalg." Spuler 
describes this Schuppenbalg method of insertion in Galleria mellonella, 
Polyommatus phleas, and Lycæna alexis, and comes to the conclusion 
that it is general in the Lepidoptera. I believe this conclusion to be 
erroneous, for I am unable to substantiate it in any of the forms which 
I have examined, although my sections were only 6.6 y thick, and were 
made in all of the three chief planes of the wing. Sometimes, however, 
in oblique sections (such as make an angle of 452 
with the plane of the 
wing) one finds an appearance which might be imagined to represent 
the Sehuppenbalg of Spuler. But the appearance is entirely due to the 
wing membrane being cut obliquely, the section embracing portions of 
two sockets, For this reason I am inclined to think that Spuler may be 
mistaken in his interpretation of what he saw. 
