RESPIRATIOX IN FISHES. 279 



extracted, either by the air-pump, or by boiling, is 

 to fishes what a vacuum is to a breathing terrestrial 

 animal. Humboldt and Provencal made a series of 

 experiments on the quantities of air which fishes 

 require for their respiration. They found that river- 

 water generally contains about one .36th of its bulk 

 of air ; of which quantity, one-third consists of 

 oxygen, being about one per cent, of the whole 

 volume. A tench is able to breathe when the quan- 

 tity of oxygen is reduced to the 5000th part of the 

 bulk of the water ; but soon becomes exceedingly 

 feeble by the privation of this necessary element. 

 The fact, however, shows the admirable perfection 

 of the organs of this fish, which can extract so 

 minute a quantity of air from water to which that 

 air adheres with great tenacity.* 



§ 3. Atmospheric Respiration. 



The next series of structures which are to come 

 under our review, comprehends all those adapted 

 to the respiration of atmospheric air in its gaseous 



* The swimming bladder of fishes is regarded by many of the 

 German naturalists as having some relations to the respiratory func- 

 tion, and as being the rudiment of the pulmonary cavity of land 

 animals; the passage of communication with the oesophagus being 

 conceived to represent the trachea. The air contained in the swim- 

 ming bladder of fishes has been examined by many chemists, but 

 although it is generally found to be a mixture of oxygen and nitro- 

 gen, the proportion in which these gases exist is observed to vary 

 considerably, Biot concluded from his experiments, that in the air- 

 bladder of fishes inhabiting the greatest depths of the ocean, the 

 quantity of oxygen is greater, while in those of fishes which come 

 often to the surface, the nitrogen is more abundant; and De la 

 Roche came to the same conclusion from his researches on the fishes 



