600 Prof. H. E. Armstrong and Dr. E. F. Armstrong. [Apr. 29, 



quent on the production during muscular exertion of diffusible materials 

 (carbonic acid, lactic acid, etc.), the presence of which will necessarily 

 determine the flow of water into the cells in which they are formed — such 

 substances may well be ranked as poisons, as their effect would be to cause 

 the separation of the neurons contemplated above. During sleep such 

 substances would not only be removed but sooner or later synthetic changes 

 would also set in and prevail, so that gradually the osmotic tension would be 

 relieved and the neurons would again be brought into normal relationship. 



A similar explanation may be applied to the regulation of respiration in 

 animals. As Haldane and Priestly* have shown, even a very slight rise or 

 fall in the partial pressure of the carbon dioxide in alveolar air causes 

 a great increase or diminution, as the case may be, in the alveolar ventila- 

 tion, the respiratory centre being exquisitely sensitive to any rise in the 

 pressure of the carbon dioxide. All the available evidence, it is said, points 

 to the conclusion that the arterial blood alone furnishes the connection with 

 the respiratory centre and is the carrier of the stimuli. While prepared to 

 agree that carbon dioxide is the normal stimulant, we are of opinion that 

 such substances as lactic acid — indeed, all the products of " fatigue " which 

 can function as hormones — must contribute to the effect. 



The physiological effect of alcohol, of drugs such as caffeine, indeed, of 

 most drugs, of flavours and odours as well as that of the presence of more or 

 less digested food in inducing the liberation of the required enzymes in the 

 several regions of the digestive tract, in fact, the effect of hormones generally, 

 is more or less completely accounted for by our hypothesis. Not a few drugs 

 and selective hormones such as adrenaline must exercise, in addition, special 

 effects in virtue of their affinity to certain regionsf and it may be that their 

 action is in part inhibitive — similar to that which is exercised by a carbo- 

 hydrate material " sympatlietic " with a given sucroclastic enzyme in 

 retarding the hydrolytic activity of the enzyme. But we prefer to postpone 

 the discussion of these problems until we are able to offer further evidence 

 bearing directly on the subject that experiments we liave in contemplation 

 may be expected to furnish. 



The activity of substances which function as hcvinodaMs (htismolysiiis) and 

 haderioclasts (bacteriolysins) is probably to be explained without difficulty 

 from our i)oiiit of vicnv. 



* '.Tonrnal of riiyHiolof.y,' 1005, vol. 32, p. 225. 



•+ The leaHoii wliy cliloroforin and alcohol, foi' example, affect certain regions 

 preferentially — brain and nerve cells — in probably not far to seek : solvent and solute 

 must always be recijjrocally attractive ; as the regions specified are rich in fatty matters, 

 of whicli alcohol and chloroform are si)ecial solvents, they are doubtless specifically 

 attractive of such substances, so that these tend to accumulate! in them. 



