MUSEUM OF COMPAKATIVE ZOOLOGY. 



19 



In considering the significance of these results, it is to be kept in mind 

 that there are two principal problems involved. I have not attempted 

 to find out whether nitrogen is consumed or produced in this process ; but 

 simply to ascertain the changes effected in the atmosphere respired as 

 far as regards (1) the oxygen and (2) the carbon dioxide. The difficulty 

 in drawing at once a satisfactory conclusion from the analyses rests pri- 

 marily on the fact that the emitted gas could not be collected under 

 conditions which allowed it to be assumed that at the time of analysis 

 it was in the same condition which it presents in the air-bladder of the 

 fish. For previous to analysis the gas had bubbled through the water 

 in which the fishes were living, and had remained exposed to a limited 

 portion of its surface. The reduction of the proportion of oxygen con- 

 tained in the atmospheric air is so great, — between one third and 

 one half, — that the influence of the water upon the composition of the 

 emitted gas in this respect would certainly have been too insignificant 

 to modify essentially the result. With the carbon dioxide, however, the 

 case is different. The coefficient of absorption in water for the latter 

 is immensely greater than for oxygen, and the total amount of carbon 

 dioxide in comparison with the volume of the water to which it was 

 exposed is so small (never having exceeded 1.7% of the volume of gas 

 collected) as to make immediate deductions from the analyses of little 

 value. 



In a series of very carefully conducted experiments upon the intestinal 

 respiration of Cobitis fossilis, Baumert ('53) arrived at the conclusion 

 that the gas from the intestine, in bubbling through the water and 

 during its subsequent exposure to that liquid, was not essentially altered 

 either by the water or the gases contained in it (p. 48). The condi- 

 tions under which his collections of gas were made l were so like 



1 Baumert ('53, p. 39) employed five or six fishes (Cobitis fossilis) of medium size 

 which were kept in a vessel containing about twelve litres of water from the river 

 Oder. The gas was collected by means of a large inverted funnel terminating 

 in a small-necked receiver, and usually about two hours were required for its 

 accumulation. Three analyses, made at intervals of three days without any 



