38 
the closing mechanism in function. When closing the apparatus 
will sink 1 m more, and it contained when taken up 40 larvæ to- 
gether with a small quantity of mud, thus having just touched the 
surface of the bottom. On the same station, however, when totally 
filled with mud it contained 931 larvæ. This seems to me to indi- 
cate that these larvæ, as distinet from the pond form, are not 
swimming freely in the water, but living in or immediately above 
the muddy surface of the bottom. 
II. 
The Function of the Air-Sacs. 
By Richard Ege. 
The function of the air-sacs has been examined by Krogh. They 
act as a hydrostatic apparatus and have surely no respiratory im¬ 
portance. 
The real function of the air-sacs is to diminish the specific 
gravity of the animal in order to keep the larva in equilibrium 
with the water. 
There is however the inconvenience by using air as a hydro¬ 
static apparatus that the volume of the air alters as soon as the 
pressure changes. 
Only under the supposition that the air is contained in a per- 
feet rigid reservoir is it possible for the animal to be in equili¬ 
brium with the water in different levels. 
Therefore if an air-sac shall act as a hydrostatic apparatus it 
is necessarv either to make its walls so rigid that the volume of 
the biadder does not alter when the pressure alters, or the air-sac 
must be provided with a regulating mechanism the function of 
which should be to counteract the alteration in volume of the air- 
sac caused by alteration in the pressure. 
We can of course also have a combination of these two prin¬ 
ciples. The more rigid the air-sac is the smaller is the use of the 
mechanism, and the better will the air-sac fill its function as a 
hydrostatic apparatus. 
In the swimming biadder of the fish the walls are not rigid 
at all, its volurne must therefore change in the inverse ratio with 
