358 
Transactions of the Society. 
ges. Hydrobiologie, 1908, Bd. 11 — that Corethra has been taken at 
a depth of 30 metres, or an equivalent of 6 atmospheres. 
As Krogh has demonstrated that this creature usually fails to 
survive a greater pressure than +70 to +80 cm., we have strong 
evidence that there are deep and shallow water species. 
A remarkable change in the shape of the air sacs commences to 
take place a few days prior to pupation, the cause of which we 
have failed to establish. From that of a kidney, they begin to 
assume a conchoidal form, this being brought about by the fact 
that the sacs, owing to their construction, can only expand “ con- 
certina wise.” (Plate XVIII., fig. 20.) The hoops occupying the 
centre portion become separated, the space between each being 
about two and a half to three times their width. As the two ends 
of the sac are approached the distance between the hoops becomes 
less, and at the extremities separation is hardly perceptible. It is 
easily observed at this stage that more than every other hoop is 
forked, the bifurcation commencing mid-way to the dorsal and 
ventral sides — in other instances near the dorsal — thus providing 
an increased number of supporting threads at the periphery, the 
larger outer circumference demanding this. 
The expansion is permanent, that is, no ultimate contraction 
takes place during the remainder of the larval existence. 
We have dissected out and mounted the sacs in camphor water 
plus 1 p.c. corrosive sublimate, and not the slightest indication of 
expansion has shown itself during a period of over two months, 
which appears to exclude the idea they are hygroscopic. 
With such an increase of the four sacs we should expect the 
larva to rise and remain near the surface. This consistently takes 
place when the larva is in captivity, and only a limited food 
supply available. In two instances, however, on examining 
freshly caught larvse, we found the conchoidal shaped sacs. When 
placed in the tank the creature took up a position near the bottom 
and remained there. Possibly the larva, on approaching pupation, 
gorges itself with food, increasing body weight, thereby resisting 
additional buoyancy of the enlarged sacs. In the pharyngeal tube 
of one of the above larvse we found three partly digested Cladocera 
( Chydorus sphericus), and a fourth recently taken was almost 
intact. Curiously enough, the fourth C. sphericus, by a definite 
constriction of the pharyngeal tube, was shut off from its partly 
digested friends. 
With this increase in size of the sacs the pigment spots are not 
confined to the dorsal half, but are distributed, and fairly evenly 
spaced over the whole area of the sacs. (PL XVIII., fig. 20.) 
This may be due to an increased number of spots or to the 
stretching of the outer gelatinous coat over a larger area, carrying 
the spots with it. 
We think the former suggestion is correct; the expansion of 
