NOVEMEEK 7, 1913] 



SCIENCE 



671 



some suggestive comparisons. In each set of 

 results, the total gas produced by a normal 

 colon from dextrose equals that from galac- 

 tose, and the total gas from dulcite equals that 

 from mannite. In each set, the total gas from 

 dulcite and mannite exceeds that from galac- 

 tose and dextrose. In our results, the total 

 gas from the alcohols is one and one half 

 times that from the sugars. 



We have also found that the total gas pro- 

 duced by colon in a peptone water medium, 

 such as Penfold used, is but little more than 

 half that produced in our broth with phosphate. 



With these facts in mind let us assume that 

 we have to cultivate in peptone water a colon 

 whose physiological powers are identical with 

 those of a normal organism except that its 

 activity has been greatly weakened. If it 

 produces only enough gas from dextrose or 

 galactose to saturate the medium, none will 

 appear in a Durham tube, and it might be 

 said that the gas-producing power was nil. 

 If the same relative power to ferment alcohols 

 that a normal organism possesses, is still pre- 

 served, the weakened organism might show 

 some gas in a Durham tube in dulcite or 

 mannite medium. 



When grown in Durham tubes, Penfold's 

 selected strain showed no gas in dextrose or 

 galactose media, while it did in mannite and 

 dulcite media. Our results show that a normal 

 colon produces ranch more gas from these 

 alcohols, and it may therefore be suspected 

 that Penfold's strain shows gas from these 

 alcohols and not from the sugars simply be- 

 cause it produces from the alcohols a suffi- 

 ciently greater volume of gas to become mani- 

 fest. 



It is significant that Harden and Penfold^" 

 by applying the more exact method of 

 Harden,^^ found that the selected organism 

 instead of producing no gas from dextrose, as 

 Penfold found by the Durham tube method, 

 does produce both hydrogen and carbon dioxid. 

 The ratio of these gases was not accurately 



1° Harden, Arthur, and Penford, W. J., Pro- 

 ceedings Boyal Society, B. 85, p. 415, 1912. 



11 Harden, Arthur, London Journ. Chem. Soc, 

 1901, p. 610. 



determined, but the amount of hydrogen was 

 found to be only 15 per cent, of that obtained 

 from a normal colon. The other products, 

 with the exception of lactic acid, were also 

 greatly reduced. 



Consequently, instead of concluding, as Pen- 

 fold did, that his selected organism has had 

 its power to produce gas from dextrose de- 

 stroyed, and that its physiological character- 

 istics have been qualitatively altered, we may 

 just as reasonably conclude, so far as Pen- 

 fold's original data are concerned, that the 

 selected organism has merely been weakened. 

 In addition to this it should be noted that Pen- 

 fold has had difficulty in producing with B. 

 lactis aerogenes modification similar to that 

 obtained with B. coli communis. In view of 

 this f«ct it may be illuminating to recall that 

 Harden and Walpole^^ found that B. lactis 

 aerogenes furnishes much more gas than does 

 B. coli on the same medium. 



If Penfold's culture is in this essential 

 identical with that of Harden and Walpole, or 

 with one of those organisms which Rogers, 

 Clark and Davis have described as producing 

 both more gas and a higher gas ratio than 

 B. coli, then it may be that Penfold could not 

 " suppress " the evolution of gas from his 

 lactis aerogenes cultures, simply because he 

 could not weaTcen it enough to prevent the 

 formation of sufficient gas to more than satu- 

 rate the medium ; and not because it refused to 

 undergo that fundamental " variation " which 

 Penfold ascribes to coli. 



It is of course impossible to make any accu- 

 rate comparisons between our own exact deter- 

 minations and those of Penfold, for the pur- 

 pose of estimating the extent of his error. 

 The unreliability and general inconstancy of 

 gas determinations made with the Durham or 

 Smith tubes is, or should be, universally recog- 

 nized. Of special significance is the more 

 recent work of Keyes and Gillespie^^ in demon- 

 strating that in contrast to anaerobic growths 

 there is a marked variation in the gas ratio of 



12 Harden, Arthur, and Walpole, Proceedings 

 Royal Society, B. 77, p. 399, 1906. 



13 Keyes and Gillepsie, Journal Biological Chem- 

 istry, Vol. 13, No. 3, p. 305, 1912. 



