EXCHANGE OF COLD-BLOODED ANIMALS. 705 
recent experiments by Mengarini 1 show that the goldfish {Carassius auratus) 
and roach (Leuciscus), in which there is a ductus pneumaticus, the mullet 
(Mugil cephalus) and ruckling (Motella), in which there is no duct, do take up 
hydrogen from water saturated with that gas. 
Moreau - has shown that the withdrawal of the gas of the swimming-hladder 
by means of a trocar leads in a short time to the secretion of a gas richer in 
oxygen, and that by a repetition of this process the percentage of oxygen can be 
raised as high as 85 ; he also states that section of the sympathetic nerve hastens 
the process of secretion. These observations have been repeated and extended 
by Hiifner 3 and Bohr. 4 Cod-fish (Gadus ccdlarias) were caught in a net at a 
depth of about 14 metres (46 feet), and when they were drawn to the surface 
the gas in the air-bladder expanded so much that the fish swam with their backs 
downwards. The gas was found in one case to contain 52 per cent, of oxygen : 
but when the fish had been near the surface of the water for some time only 13 
per cent, of oxygen was obtained. After the removal of the gas from the bladder, 
its secretion begins again : within six hours a little gas has accumulated, and in 
twenty-four hours the bladder is again full. The rapidity of the process and the 
increase in the percentage of oxygen is shown by the following examples : — 
Number of 
Hours after the 
Percentage 
Percentage of 
Remarks. 
Experiment. 
First Puncture. 
of Oxygen. 
Carbon Dioxide. 
, 

15-0 
8 c.c. 
of gas in the air 
-bladder. 
24 . \ 
48 
78-5 
1-0 
7£ c.c 
;) 
.j 
( 
71 
83-7 
0-5 
j 7 c.c. 
., 
5 
n . \ 

15-0 
; 6 C.C. 
50 
78-4 
0-8 
5 c.c. 
1 
73 
72-7 
1-0 
6 c.c. 
u. ) 

16-3 
o-i 
5 
16-3 
0-2 
Bohr has also shown that after section of the branches (rami intestinales) 
of the vagus which supply the air-bladder, the secretion of the gas ceases en- 
tirely, but that no effect upon the secretion is observed after section of the 
rami cardiaci or the nervi laterales. These phenomena observed in the case of 
the swimming-bladder can at present be explained only as a process of secretion. 
The air is not swallowed, for the fishes with the greatest percentage of oxygen 
in their bladders are those which do not come to the surface, but live at great 
depths ; in some of them, moreover, such as the cod (Gadus ccdlarias), there is 
no communication between the bladder and the mouth. The phenomena can- 
not be accounted for by simple diffusion, for the water which surrounds the fish 
cannot have a higher tension of oxygen than 21 per cent, of an atmosphere. 5 
Further, Bohr has shown that the percentage of oxygen is not reduced by 
diffusion outwards, for when the secretion of fresh oxygen is prevented by 
section of the vagus, the high percentage of oxygen is maintained for two or 
three days. As long as the swimming-bladder is fresh, it is almost im- 
permeable to oxygen, even when the difference of pressures inside and outside 
the bladder amounts to one atmosphere. It is to be noted that the membrane 
lining the swimming-bladder has a peculiar glandular structure. 
1 Arch.f. Physiol., Leipzig, 1889, S. 54. 
2 Coni.pt. rend. Acad. d. sc., Paris, 1873, tome lvii. pp. 37, 816 ; " Recherches experi- 
mentales sur les functiones de la vessie natatoire, " Paris, 1870 ; " Mem. de physiol.," Paris, 
1877, pp. 69-86. 
3 Arch. f. Physiol., Leipzig, 1892, S. 54. 
* Journ. Physiol., Cambridge and London, 1894, vol. xv. p. 494 ; Compi. rend. Acad. 
d. sc, Paris, 1892, tome cxiv. p. 1560. 
5 Biot, Delaroche, Moreau, loc. cit. ; Jakobsen, Ann. d. Chem. u. Fharm., Bd. elxvii. 
S. 1 ; Hlifner, Arch. f. Physiol., Leipzig, 1897, S. 112. 
VOL. i.— 45 
