416 Messrs. Harden and Penfold. Chemical Action on [May 18, 



The medium employed in both cases contained 10 grin, peptone and 20 grin, 

 glucose per litre, 10 grm. of chalk being added per litre. The flasks were 

 connected with the gas-collecting apparatus described by Harden, Thompson, 

 and Young (2), and the air of the flask replaced by nitrogen before incubation. 

 The estimations were made as previously described (3). The results of two 

 such experiments are given in the table on p. 415, the products being 

 expressed both in percentage of the sugar used and as the number of carbon 

 atoms of the glucose molecule to which they correspond. 



These results must be regarded as preliminary, but are of sufficient interest 

 to justify some remark at the present stage of the work. 



In the first place, it is to be noted that, although the selected organism gave 

 no gas at all when tested by the Durham tube method, it yielded about 

 12 - 6 c.c. of hydrogen per gramme of sugar when grown anaerobically in 

 presence of chalk as against 80 - 6 given by the original organism. Similarly 

 the amount of gaseous C0 2 formed from the sugar was found to be 10 - 3 c.c. 

 j>er gramme, but as this number is the difference of two large volumes 

 (the total gaseous C0 2 + C0 2 dissolved in the medium — C0 2 liberated by 

 the acids formed) no great accuracy attaches to it. The reason of this 

 different behaviour is not understood and is at present being investigated. 



Apart from this, an examination of the results shows that the main 

 difference between the actions of the two bacteria is to be found in the 

 increased proportion of lactic acid, and the correspondingly diminished 

 proportion of the other products, formed by the selected organism. 



Harden (3) has previously shown that the action of organisms closely allied 

 to B. coli communis is roughly represented by the equation 



(1) 2C 6 H 12 6 + H 2 = 2C 3 H 6 3 + C 2 H 6 + C 2 H 4 2 -l-2C0 2 + 2H 2 , 

 which requires the proportions of products shown in the last column of 

 the table. Actually there is always less C0 2 than this produced, and with 

 the strain of normal B. coli communis here employed there is also some- 

 what more alcohol and acetic acid and less lactic acid formed. The 

 relations between the products of the selected organism appear, however, to 

 be quite incompatible with this equation. 



The explanation which suggests itself is that the products observed as 

 the result of action of the normal organism are formed by three 

 independent enzymes. One of these converts sugar into lactic acid : 

 (2) C 6 H 12 6 = 2C 3 H 6 3 . 



The other probably produces alcohol, acetic acid, and formic acid, and 

 the last of these is in all probability decomposed by a third enzyme into 

 carbon dioxide and hydrogen : 



