VIII, B, 2 Butler: Carbohydrate Reactions 127 



incubating overnight and culling out any infected tubes, the 

 medium is ready for inoculation. The objections which may 

 be offered to this medium relate to (1) the low content of carbo- 

 hydrate, (2) the high temperature to which the carbohydrate 

 is submitted, (3) the high temperature to which the litmus 

 solution is submitted, and (4) the large amount of litmus. As to 

 the last three objections I will only say that in my hands 

 this method has given better results than any other. As to the 

 first objection, that regarding the low content of carbohydrate, 

 a few words of explanation are necessary. The effects which 

 bacteria may produce upon carbohydrates are (1) to ferment 

 with acid production, (2) to ferment with acid and gas pro- 

 duction, and (3), the negative effect, to let the carbohydrate 

 alone entirely. The intestinal bacteria and practically all rapidly 

 growing bacteria show what they are going to do to any par- 

 ticular carbohydrate within five days, and most of them within 

 forty-eight hours. This statement does not refer to slow-grow- 

 ing organisms such as the tubercle bacillus. With organisms 

 which produce gas this can be shown by accurately marking off at 

 6-hour intervals the amount of gas in the closed arm of the 

 fermentation tube. There comes a time within three days, pro- 

 vided we have not used too much carbohydrate, when the gas 

 production stops, and if we measure at that point we shall find, 

 at the next reading, that the fluid has risen slightly above our 

 last mark. The organism has done all it will ever do to the 

 carbohydrate introduced, and the medium has taken up its limit 

 of gas at the existing temperature and pressure. If we give 

 it more carbohydrate than it can ferment, the organism will stop 

 growing when a certain degree of acidity is reached. Bacteria 

 do not produce alkali from carbohydrates, but if we leave our 

 inoculated tubes for two or three weeks in the incubator, con- 

 taminations, concentration of the medium, or the splitting of 

 proteid may result in a change of our original picture. We are 

 in that case, however, not reading the effect of the bacterium 

 upon the carbohydrate, but upon some other constituent of the 

 medium, or perhaps the effect of chemical reactions not connected 

 with the organism at all. 



In this paper space does not permit a consideration of all the 

 qualities and peculiarities of the 8 or 10 carbohydrates with 

 which I have worked. My attention was so soon focused upon 

 the sources of error in the fermentation tests of the important 

 carbohydrate, maltose, that I have not attempted to study criti- 

 cally the other carbohydrates from a bacteriological standpoint. 



