Chap, vi.] AIR SACS OF FISHES. 235 



accessory sac is produced into a number of convoluted 

 folds (Fig. 99), which retain their moisture, and are 

 able to take up oxygen direct from the air, during the 

 comparatively long periods that the fish lives out of 

 the water. Its ally Ophiocephalus, Cobitis, and 

 various fishes with spongy air-bladders, such as Suclis 

 and Erythrinus, swallow air directly ; so that it is not 

 among the Dipnoi only that oxygen dissolved in 

 nitrogen (atmospheric air), is used for the necessary 

 oxvdation of the tissues of fishes. 



*/ 



It is, however, in certain Ganoids and in the 

 Dipnoi that we get the most certain proofs of aerial 

 respiration ; Lepidosteus has been observed to pro- 

 trude its head from the water, to emit a bubble of 

 air, and to make a swallowing movement, and a 

 similar phenomenon has been seen in Amia (B. G. 

 Wilder) ; the noise made by Ceratodus is explained as 

 being due to the swallowing of air, and the streams of 

 Australia in which it lives are known to become 

 liquid mud in the dry seasons of the year. Protopterus 

 has been brought from West Africa to this country 

 embedded in the mud balls in which it lives during 

 the droughts, and has been revived by being placed in 

 warm water. 



Fishes differ considerablv in the extent to which 



/ 



they are able to live on land ; thus, an eel will live 

 much longer than a gudgeon when taken out of the 

 water. The careful experiments of Bert show that 

 this difference is due not to a difference in gill arrange- 

 ment, but to a difference in the demand made by the 

 tissues of the body for their supply of oxygen. Here, 

 again, we have an example of the danger of arguing 

 from anatomical peculiarities where our hypotheses 

 are not controlled by experiments. In any discussion 

 of respiratory phenomena in animals it is neces- 

 sary to bear in mind the fact that all living 

 tissues are capable of absorbing oxygen, and that 



