Potassium Hydroxide hy Dicranura vinula, 403 
dorsally a minute diverticulum {a r), apparently- 
springing from the anterior dorsal extremity of tlie 
mesenteron, but resembling rather the oesophagus in 
the character of its walls. The chief changes which 
take place concern this diverticulum, and the relative 
lengths of the three main sections of the canal. The 
latter may be dismissed in a few words — the oesophagus 
remains of the same length, the mesenteron becomes 
very much shorter and slightly narrower, while the rectum 
becomes correspondingly longer, until in the freshly- 
hatched imago it is rather longer than the two other 
sections combined. Figs. 2 and 3. There take place in the 
hinder region of the rectum other changes which I 
have not followed out in detail, the most remarkable of 
which is the existence, at the time of emergence, of a 
wide short tube (Fig. 3, d) opening into the body cavity 
at one end and into the posterior third of the rectum 
at the other; this tube is full of a brownish -red thick 
fluid (nitrogenous excretion ?), and appears to serve as an 
exit for the fluids so abundantly discharged per anum 
immediately after emergence. After four days of 
imaginal life the tube had shortened to a mere stump 
(Fig. OA, d), and six days later it could not be 
recognized. 
The mesenteric diverticulum gradually increases in 
size {a r, Figs. 2 and 3), and comes to lie in such a 
position that the only communication between the 
mesenteron and oesophagus is through its somewhat 
pointed ventral end. Eventually it assumes a pear- 
shaped appearance, with the broader end directed dor- 
sally (Fig. 3a, a r). 
At the time when the imago is ready to emerge this 
diverticulum contains a clear liquid with strongly 
alkaline reaction to red litmus paper ; by gentle pressure 
the liquid may be driven along the oesophagus and 
caused to exude from the mouth in drops. 
Now the contents of the larval mesenteron are 
intensely alkaline (as indeed is the case in every insect 
that I have examined), and it is probable that the 
digestive ferments need an alkaline medium for their 
action. Throughout pupal life the contents of the 
mesenteron continue strongly alkaline, though they 
undergo other alterations, viz., after fourteen days they 
are granular and blood-red in colour, and, as develop- 
