393 
side are seen the large trunks, containing each a trachea, that 
on the left side branched, and also a nerve-fibre which accom- 
panies the trachea in all the larger vessels in all insects which 
I have examined, and is present in the Melolontha where the 
trachea is absent. The transverse vessel is seen to have a 
thick lining of elongated, nucleated, closely packed cells. 
This layer of cells is present also in the main trunks, but is 
there, of course, not so easy to investigate, owing to the pig- 
mentation and thickness of the outer chitinous investment. 
I have injected the vessels of the wing with silver solution, 
but have been able to find no other endothelial lining to the 
vessel than this thick stratum of cells. 
Fig. 5 gives a group of the various forms of corpuscles, 
and fig. 4 shows one of the main trunks during active circula- 
tion. The direction of the current is shown by the arrows. 
The corpuscles behave in much the same manner as in the 
capillaries of a frog. Thus they change their form readily, 
the spindle-shaped ones doubling up in order to pass cross- 
ways through a narrow aperture, as seen in the one coming 
to the main trunk from the small transverse vessels on the 
left hand side of the figure. Moreover the corpuscles attach 
themselves to the inner wall of the vessel, and even seem to 
bury themselves a short way in it; though, of course, the 
chitinous investment renders diapedesis impossibie. In the 
irregularly formed corpuscles, which seem to represent leuco- 
cytes, amceboid movements may be observed. Corpuscles 
pass freely above and under the trachez, showing that these 
latter lie free in the vessels. 
Thinking that the cells lining the blood-vessels might 
be contractile, and that the nerves accompanying the trachea 
might be to some extent vaso-motor, I introduced fine wires 
into the hinder legs of a Blatta, and applied stimulus by 
means of an induction coil, but could not produce any effect. 
However, on repeating this experiment with Melolontha 
vulgaris, | found that there was an elongated sac projecting 
free into the main artery of the hind wing, which contracted 
readily on very slight stimulus being applied to the legs. If 
the stimulus were prolonged this sac remained in a state of 
tetanic contraction, but expanded again immediately the cur- 
rent was broken. This sac is one of the contractile organs 
which haye been described as occurring in the limbs and 
wings of insects, and which assists the heart in maintaining 
the circulation. In the hind wing of some small hymeno- 
ptera occur three rhythmically contractile sacs, or rather dila- 
tations of the vessels. They are represented in fig. 6, the 
contractile spots being marked with a cross. If electrical 
