REGULATION OF ENVIRONMENT 67 



extremely sensitive to the slightest changes in the 

 concentrations of inorganic salts in their environment. 



Cell-secretion, cell-respiration, and cell-nutrition are 

 clearly only different aspects of the same whirl of 

 molecular activity. Where secretion or nutrition seems 

 to be stationary, there is in reality only a balance 

 between ingoing and outcoming molecular streams. 

 Instances of this occur when the kidney is not secret- 

 ing chlorides, or when no oxygen is passing into or 

 out of the swim bladder, or when all external activity 

 is latent, as in a dry seed. The apparent stand-still is 

 similar to that in a blood corpuscle in a test tube of 

 blood half saturated with oxygen, when the stream of 

 oxygen molecules entering the corpuscle is balanced 

 by the stream leaving it. The unstable oxyhaemo- 

 globin molecules in the blood corpuscle are constantly 

 losing oxygen molecules and as constantly regaining 

 others, so that the half saturation of the blood cor- 

 puscle with oxygen represents the average of the 

 gains and losses of the haemoglobin molecules. This 

 we can understand. But what conception can we form 

 of the molecular streaming in the living cell and the 

 strange co-ordination which the different molecular 

 streams exhibit? I have tried to indicate how this 

 problem, which will be followed up in the next lecture, 

 rises directly out of the fact of oxygen secretion. But 

 meanwhile we must follow further various other facts 

 relating to respiration. 



The evidence existing at present is strongly against 

 the theory of active secretion of GO z outwards by the 

 lung epithelium. Krogh's experiments gave very defi- 



