406 Mr. 0. H. Latter on the secretion of 
i). bifida, . 
Here the shape of the " shield is slightly different, 
inasmuch as the posterior lateral edges of the pupal 
eyes are bent inwards, so as to partly grip the corres- 
ponding portion of the eyes of the imago. One pair 
of hooks are present on the inner face of the shield/' 
corresponding to the larger pair in D. furcula, and to 
the only pair in D. vinula. The smaller pair present 
in D. /ttrci<?a here finds its homologue in a stout piece 
of chitin, which forms a low ridge merely. From the tips 
of the hooks long transparent chitinous threads pass 
upwards. The hooks fit into grooves and sockets in 
the head of the imago, as in the other species. This 
shield has also attached to it the chitinous lining 
of the mouth and oesophagus, which appear as trans- 
parent membranes. Traces of these organs can also be 
made out in the two other related species. The shield 
measures 2*2 mm. dorso-ventrally by 3'2 mm. trans- 
versely. 
The head of the imago bears a pair of grooves (Fig. 7, 
}ihg)i leading into sockets for the reception of the 
hooks, but the labral prongs are hardly represented at 
all, a pair of minute lumps being the only vestige to be 
recognized. The remains of the mandibles (Fig. 7, 
mn) are rather prominent at the junction of the 
gen^ with the inner border of the eye. The palps 
(Fig. 7, 'p)y in the only two specimens I was able to 
examine, are moderately clothed with hair, except at the 
apex, which resembles that of D. furcula. Each palp 
consists of two well-marked joints, with indications of 
a possible third (Fig. 7, p). I would suggest that 
these are the organs by means of which the moth 
breaks open the cocoon. 
Saturnia carpini. 
The red litmus paper artificial cocoons for this species 
were made into a loosely coiled paper cone, the apex 
of which was left open and directed towards the light ; 
the head of the pupa lay near the apex. I was greatly 
astonished to find an abundant alkaline secretion dis- 
charged prior to exit from the cocoon. Doubtless, 
the fluid softens the converging silk fibres at the neck 
