398 THE ALIMENTARY CANAL OF THE IMAGO. 
injection of air into its interior. Gleichen wrote: ‘ The exten- 
sion of the proboscis probably arises from the air which is 
driven by the fly into the base (rostrum), and from this into the 
middle tube (haustellum), and finally into the lips.’ He further 
added : ‘ The fly can, indeed, drive air between the membranes 
of the proboscis into the lips, but he cannot take in air with 
them.’ Dimmock [68, p. 46], says: ‘That it is air, at least 
for the most part, and not fluid, which expands the inner 
surface of the labellz, is easily proved by carefully pressing the 
head of a fly between the fingers until the proboscis is fully 
distended, and the labellz fully inflated; then, putting the fly 
under water, and pricking the inner surfaces of the labellz 
with a needle, they at once collapse, bubbles of air escaping at 
the same moment from the opening made in their surfaces by 
the needle.’ Beside these notices, I have found nothing further 
on the subject; but in transverse sections of the oral sucker, 
made from a proboscis distended by dropping an insect into 
hot alcohol, the inflated air-channels are not equivocal. 
I have represented a portion of a section in Fig. 50, 7, and it 
will be seen that each of the pseudo-tracheal channels forms 
the outer wall of two air-channels or modified tracheze which 
separate the hypodermis from the cuticle of the disc. These 
channels all radiate from the azygos air-sac already men- 
tioned (see Fig. 51, ¢), and have the same distribution as the 
pseudo-trachez which they accompany. 
Even when sections are made of a flaccid proboscis, it is not 
difficult to distinguish the air-channels, although their size is 
greatly diminished. In the pupa when the proboscis has arrived 
at the stage of development represented in Fig. 32, the separa- 
tion of the hypoderm from the cuticle of the disc is very 
apparent; the intervening space, like the lumen of the trachez 
of the pro-imago generally, is filled with a watery fluid which 
is readily distinguished from the granular blood in the blood- 
sinuses beneath the hypoderm. 
The air in these channels, when the organ is erected, may 
be assumed to be compressed. My reasons for this assumption 
have been fully discussed in the section devoted to the tracheal 
