14 
Siphonophora Rosarum 
From the above account it will be seen that no very close com¬ 
parison can be made between the mouth organs of S. rosarum and 
those of the other Aphides which have already been figured. The only 
conclusion one can come to is that there are wide differences between 
the various groups of Aphides in this respect. Of course one would 
expect to find some variations, especially in those Aphides which form 
galls, but it is remarkable that such wide differences should exist in 
such points as the presence or absence of a protractor muscle to each 
seta, and also the point of origin of the retractor muscles. 
The pharynx —the first portion of the oesophagus—is modified, as in 
all the Hemiptera, to fulfil a suctorial function. It extends from the 
oral cavity to where the oesophagus turns backwards. The posterior 
wall of the pharynx is strengthened by a strong layer of chitin, while 
the anterior wall consists of a flexible but elastic membrane (see Plate I, 
Fig. 12). Originating from the front wall of the head, and inserted 
into the anterior wall of the pharynx, where they split up into a 
number of tendon-like slips, is a double series of muscles. These 
muscles, contracting, will pull the anterior wall outwards, in that way 
increasing the capacity of the pharynx, and the oesophageal valve 
preventing the return of fluid from the crop into the oesophagus, fluid, 
drawn up the groove in the beak by capillarity, will be sucked up 
through the mouth into the pharynx. 
The muscles then relaxing, the membrane, by virtue of its own 
elasticity, will return to its former position ; and this process com¬ 
mencing at the oral end of the pharynx, the excess of fluid will be 
gradually forced along the oesophagus into the crop. In this way the 
insect is able to suck up the sap of the plant which oozes from the 
wound made by the setae. 
The oesophagus presents no remarkable features, structurally, beyond 
the fact that nuclei and cell boundaries are almost entirely absent—-I 
do not find the series of nucleate cells figured by Witlaczil (1882). 
The wall is comparatively thick, and composed of fine granular material 
(Plate I, Fig. 8). The point for remark about the oesophagus, however, 
is its extremely small size, and, as a natural consequence, the almost 
ultra-capillarity of the lumen. Dufour (1833) was evidently struck by 
this fact, for he says in describing this portion of the alimentary canal: 
“ L’cesophage a line tenuite plus que capillaire.” 
The croj) consists of a thick-walled, cylindrical sac or bag for the 
reception of the fluid sucked up from the host-plant by the Aphis. The 
Avail is composed of rather large cells with large nuclei and nuclear 
bodies, and somewhat granular cell-contents (Plate I, Fig. 9). The 
