334 A - p - MATHEWS. 



toplasmic respiration, therefore, is in reality not the consumption 

 of gaseous oxygen and the liberation of carbon dioxide as ordi- 

 narily stated. We now know that the production of carbon 

 dioxide stands in no direct causal relation to the consumption of 

 oxygen. Respiration is in fact the dissociation of water with the 

 liberation of hydrogen, 



The evidences of the truth of this hypothesis are many and 

 have been collected in part by Hoppe-Seyler. Hydrogen is 

 produced and set free as such by a great many moulds and 

 bacteria, /. c., coli communis, penicillium, butyricus, etc. In the 

 case of facultative anaerobes like the colon bacillus, the hydrogen 

 appears only if no atmospheric oxygen is present ; it is burned 

 to water if this be present. In many cases where the hydrogen 

 is not set free as such it escapes as marsh gas or combines with the 

 protoplasm or some of the constituents, of the culture medium, 

 such as levulose or sulphur. Thus as Hoppe-Seyler showed, in 

 the fermentation of calcium acetate the hydrogen unites with the 

 methyl or methylene set free by fermentation and comes off as 



marsh gas. 



(CH,CO 2 ) 2 Ca + H 2 O = CaCO., + CO., + 2CH 4 



In the second place hydrogen is set free in the case of some 

 bacteria which are able to oxidize substances in the absence of 

 oxygen. In this case the oxygen can only have come from the 

 water. Such a case has been reported recently by Maze. 1 In 

 the case of a certain bacterium alcohol was oxidized to acetic acid 

 in the absence of air, if levulose was present. The levulose was 

 at the same time converted into mannite. The reaction was 

 probably as follows : 



I. C,H 5 OH + H 2 O = C 2 H 4 O 2 -j- 4 H 

 II. 4 H + 2C 6 H 12 6 = 2C 6 H 14 6 



Mannite. 



Reaction I. can only go on in a positive direction or from 

 left to right if some substance is present to remove the hydrogen 

 and thus prevent equilibrium from being established. 



It will be seen that as atmospheric oxygen acts the part only 

 of a depolarizer any other oxidizing agent, that is any other sub- 



1 Maze, Annales de /' Institut Pasteur, XVII. 



