THE GAS IH THE SWIM-BLADDER OF FISHES. 
129 
would be fairly constant in all fish subjected to the same conditions. The abundance 
of squeteag'ue supplied by the Fish Commission Laboratory at Woods Hole furnished 
the opportunity for a large number of analyses under different conditions. Few 
analyses had been completed, however, before it was evident that the per cent of 
oxygen in the air-bladder was no indication of the processes taking place there. In 
the accompanying table the squeteague numbered 1 to 9 were allowed to become 
asphyxiated by dying in the open air on the wharf. The animals were very active 
before they were taken from the live-car. In each case the gas was analyzed at a 
fixed time after asphyxiation. The bladder was rapidly removed from the dead 
animal, and a silk ligature was tied a short distance from the small end. The end of 
the bladder was then amputated, and the 3-way stopper of the Bunte apparatus was 
inserted into the lumen and fastened by another ligature. The entire Bunte apparatus 
being filled with water, the first ligature was carefully cut, and the bottom stopper of 
the apparatus being opened, the gas from the bladder replaced the water in the 
instrument. The stoppers were now closed and the gas analyzed by the ordinary 
methods. In No. I there was found 17.51 per cent of oxygen, while in No. 6 there 
was only 6.5 per cent; yet the animals were under practically the same conditions. 
The amount of C0 3 , considered independently of the oxygen, also shows no particular 
characteristic, but when compared with the oxygen it is seen to vary somewhat pro- 
portionately with the oxygen ; so that, in the two above cases, the quotient of 
ca 
0 2 
became respectively 0.215 and 0.26. It is evident, therefore, that in all nine cases 
the proportion of CO ? to 0 2 is nearly constant, and thus asphyxia has produced a like 
effect on the gases in the bladder. 
This fact naturally suggested analyses of gas taken from normal live animals. 
The first series were upon animals taken from the live-car and, as far as could be 
judged, in normal condition. The animals were taken from the water and the gas 
collected as rapidly as possible. The result of the analyses of this gas is given in the 
table, Nos. 11 to 20. Here again there is a marked difference in absolute amounts of 
0., in bladder, and also in absolute quantities of CO., . The extreme differences 
are shown by comparing Nos. 11 and 20. The q ! coefficient, however, shows a close 
agreement throughout the seven specimens subjected to the same conditions. Even 
in these experiments the chances for partial asphyxiation were not entirely eliminated, 
for this would begin during the time that the gas was being collected. Another 
set of experiments was made on squeteague in which the gas was collected while the 
gills of the animals were irrigated with salt water, thus giving no possible chance 
for asphyxiation; the results are given in Nos. 21 to 23 of the table. In this 
experiment, as in the two previous ones, there are differences in absolute amounts of 
CO 
0 2 and CO, in the different specimens, but the q 2 quotients are very nearly the same. 
Another series of experiments is recorded in Nos. 10 to 13. These squeteague 
represented different stages of asphyxiation. None were dead, but none were very 
active. The result of the analysis showed a ^ ( 1 quotient, which is especially inter- 
esting, for it varied with the condition of the animal. 
F. C. B. 1901—9 
In No. 10 the 
CO 
0 3 
' quotient 
