THE CHEMISTRY OF RESPIRATION 1125 



26 per cent., at which figure it remained stationary. This would correspond to an 

 oxygen tension of about 25 per cent, of an atmosphere, whereas we have already seen 

 that the oxygen tension in the alveoli cannot be greater than 15 per cent. He therefore 

 concluded that the epithelial cells of the alveoli play an active part in the respiratory 

 interchange, taking up the oxygen on one side at a tension of 15 per cent, and piling it 

 up on the other until the pressure in the blood is much higher than that in the alveolar 

 air. Theoretically there is no reason to deny the possibility of such powers to the 

 pulmonary epithelium. We know that the secreting cells of the kidney take up urea 

 from the blood which contains only about *02 per cent, of this substance, and excrete 

 it into the renal tubule, into a medium containing about 2 per cent. ; and if the data 

 given by Haldane are correct we must ascribe an analogous function to the pulmonary 

 epithelium. These data however were obtained by a colorimetric method working 

 with very minute quantities of blood, and lacked the support of control experiments. 

 As a result of further experiments, Haldane has modified his position so far as to allow 

 that under normal conditions the absorption of oxygen from the alveolar air takes place 

 in accordance with the difference of pressure, i. e. by a process of diffusion. He is still 

 of opinion that under abnormal conditions, when the oxygen tension in the alveolar 

 air is very low, there is an active absorption and transference of oxygen to the blood 

 on the part of the pulmonary epithelium. Why animals should evolve a function 

 which can be brought into play only on climbing mountains seems difficult to under- 

 stand, and it does not seem probable that a reinvestigation of the tensions of oxygen 

 in the blood under such conditions by Krogh's method will lend any confirmation to 

 Haldane's conclusions. 



An analogy has been drawn between the processes of gas interchange in the lungs 

 and that hi the swim bladder of the fish. Bohr has shown that the gas obtained by 

 puncturing the bladder often contains considerable excess of oxygen. If the bladder 

 be punctured and the fish then left in the water, the gas rapidly reaccumulates, and it 

 is found, on tapping a second time, that the percentage of oxygen is largely increased, 

 and may amount to between 60 and 80 per cent, of the total gases. This reaccumulation 

 of the gases does not take place if both vagi are cut, and is therefore ascribed to a direct 

 secretory activity on the part of the epithelium lining the swim bladder under the 

 influence of the vagus nerves. Bohr, as the result of experiments by himself and some 

 of his pupils, is inclined to endow the vagus nerves in the higher vertebrates, including 

 mammals, with an analogous regulatory influence on the gaseous exchanges in the lungs. 

 As regards the evolution of carbon dioxide, the facts elucidated by Haldane himself 

 would make one hesitate in ascribing any special secretory activity to the pulmonary 

 epithelium. We find, namely, that the respiratory centre reacts immediately to the 

 slightest increase in the tension of the carbon dioxide in the alveolar air. Since this 

 behaviour of the respiratory centre is independent of any nervous connections between 

 the lungs and the. brain, it seems to indicate, as indeed Krogh has found, that the tension 

 of the carbon dioxide in the blood follows closely the tension of the carbon dioxide in 

 the alveolar air. If the carbon dioxide were secreted by the pulmonary epithelium, we 

 should expect the lungs to react to increased carbon dioxide in the alveoli by simply 

 increasing their work so as to maintain the tension of carbon dioxide in the blood at a 

 constant level. This at any rate is the way in which the kidney would behave under 

 analogous circumstances. Moreover there is no likeness between the thick typical 

 secreting cells of the ' red gland,' which is the gas-secreting part of the swim bladder, 

 and the thin structureless plates which separate the capillaries of the lungs from the 

 alveolar air. 



