556 Wisconsin Academy of Sciences , Arts , and Letters . 
pect to find in case such a curving has taken place, even if fol¬ 
lowed by a fusion at the margin. The median part is made up 
of a club-shaped chitinized thickening, which bears numerous 
blade-like hairs, especially crowded at the apex of the club and 
extending more sparsely over the /surrounding membranous 
area. Posteriorly, this chitinized median strip is continued 
into a transverse band, which shows two large asymmetrical 
notches in which are received the molars of the mandibles; it 
then runs dorsad at the sides and forms a supporting arch over 
the epipharynx. The epipharynx does not lie loosely in the 
roof of the mouth, but is connected to the labrum by means of 
a membrane which extends along the sides and the caudal 
margin of the supporting arch. 
The epipharynx is a hollow structure consisting of a double 
membrane. The upper membrane contains the supporting 
arch, from the caudal edge of which it is turned back and con¬ 
nected with the membranous lining of labrum and clypeus; 
this shows that the origin of the epipharynx is in a fold of the 
upper lining of the mouth. The lower membrane contains thr 
sense papillae and fuses with the dorsal wall of the pharynx. 
Toward the caudal end of the epipharynx, where the median 
chitinized strip fuses with the transverse band, we find a row 
of papillae, broadening out laterally, decidedly different in 
appearance from the chitinous hairs which are scattered over 
the median part of the epipharynx. These papillae may be 
gustatory, the hairs just mentioned being probably tactile in 
function. 
The great extension of the epipharynx and its many sense 
papillae would seem to indicate a highly developed gustatory 
sense, the need of which, however, does not seem clear in these 
animals which live on excrementitious matter. 
Mandibles (PI. XXX, Fig. 2). These present a great 
deviation from the type we should expect in insects with bit¬ 
ing mouth-parts, and show that they are originally composed of 
many segments. As Smith (8) was the first to call attention 
to these facts, and as he also worked out the homology of the 
parts, I will here quote his statements, followed by corrections 
which proved to be necessary. 
