OF WASHINGTON. 39 
that they were part of the tracheal system. In 1869 he further examined 
the " battledoor " scales of the Lyccenidce and other Lepidoptera, and enu- 
merates 131 species in which they occur. 
Fritz Miiller. in 1877, called attention to these scales and hair tufts, and 
first asserted their function as odor-giving glands. 
Aurivillius in his work on secondary sexual characters of northern di- 
urnals devotes a large space to the discussion of these scales. The author 
then describes the form of scales in some species, finding them of various 
shapes, usually very different from the other wing scales, and always striate. 
In Thanaos they are hollow. Many of the scales, especially in the Lycce- 
nidcz, are covered with rows of very minute globules set into the scale by 
a pedicel. In Pieris and some other genera the tip of the scale is furnished 
with a tuft of fine hair, while the shape is very different from that of the 
others on the wing. In some species of Satyrus the tip is densely set with 
fine hair giving a brush-like appearance. 
In some species of Hesperidce the scales are articulate, consisting of 
several joints, which are easily separated. 
The distribution of these scales is then treated of, and, while they are, 
in rare instances, scattered over the wing, they are usually massed into 
spots, often differing in color from the other parts of the wing. Often, 
also, they occur on the anterior margin of the secondaries, which is cov- 
ered by the primaries, or at the inner margin, which is usually more or 
less folded. 
Not only are the $ scales often placed where they are more or less 
shielded, but there are often special structures looking to their protection. 
In Pamphila comma the oblique black dash in the $ is composed of such 
scales, and they are set into a depression of the wing and overlapped by 
ordinary scales, while in many other Hesperids there is a costal fold, 
tightly closed, and in which these peculiar scales are massed. 
In Danais the peculiar raised spot of the secondaries so well known to 
all collectors forms a sac in which the $ scales are concealed. What are 
the functions of these scales, and why do they need such protection? 
Fritz Miiller says they are odorous, and the author proceeds to cite cases 
where the odor was intense enough to be perceived by the human sense 
of smell. In Callidryas argante a musky odor was perceptible when the 
$ scales were exposed, and this was observed in all specimens. In Pre- 
pona laertes the odor is like that of a bat, in Dircenna xantho, vanilla 
like ; in both only noticeable at the anterior margin of the secondaries, 
where the (J 1 scales are situated. 
In Papilio Grayi the odor is as agreeable and intense as in flowers. 
Didonis biblis has also a distinctly odorous spot on the secondaries. 
That we are unable to perceive a distinct odor in all cases the author 
does not consider as militating against the idea that the scales give out 
an odor, tor he considers the sense of smell much more highly developed 
in the Lepidoptera than in man. 
The fact that the scales are always more or less shielded is explained by 
