RESPIRATION IN AIR 79 



Crocodiles, turtles, and many tortoises, living in water, dive 

 regularly and many of them are stated to stay for hours or 

 even days under water (Vos, 1936). The concentration of 

 oxygen in the lungs of the fresh-water tortoise Emys orbicularis 

 under water falls to 2-3% in half an hour (Vos). The cu- 

 taneous uptake of oxygen which is normally of the order of 2% 

 of the total, according to Liidicke (1936), may increase to 8%, 

 but it is evident that the metabolism must be either greatly 

 reduced or become mainly anaerobic. No really quantita- 

 tive work on the diving of reptiles has been made so far, al- 

 though some of them should be very suitable for the purpose. 



An excellent study of diving birds was made by Christian 

 Bohr in 1895, but unfortunately the results were only published 

 in Danish (1897) and have remained unnoticed until now. A 

 rather full account of this paper will therefore not be out of 

 place. 



The experiments and determinations were made on guille- 

 mot (Uria troile) weighing about 700 g and puffin {Mormon 

 ratercula — 350 g). These will stand submersion only for 

 short periods (up to 6 min. in the case of the guillemot), but 

 they will go down, swimming vigorously, for a few minutes at 

 a time with breathing spells at the surface of a few seconds 

 only, and artificial submersions of the same type, each lasting 

 2 minutes, could be sustained at least 10 to 20 times. The 

 blood volume and oxygen capacity of both birds were found 

 to be high, namely respectively 12-13% of the body weight 

 and about 25 ml O 2 /100 ml. This means 30 ml 2 /kg animal 

 as against 16 in non-diving mammals. 



A number of varied and instructive respiration experiments 

 were made, comprising periods in which the trachea was ob- 

 structed. I give as an example the determinations on a 

 guillemot weighing 685 g., shown in Table 8. 



The oxygen debt incurred during a period of obstruction 

 is more than made good in the next 5 minutes, while an excess 

 of C0 2 is blown off. In other experiments a small air sample 

 (54 ml) was drawn every 2 minutes from the closed trachea and 

 replaced with atmospheric air. This raised the C0 2 percent- 



