68 



Each of these tubes was inoculated February 12 with one loop from 

 a beef broth culture made Februar}^ 5. This culture was well clouded 

 and becoming moderately turbid from the presence of numerous 

 small zoogloeae. It also contained a moderate amount of yellow 

 precipitate. The tubes were very clear when inoculated, and perfecth^ 

 sterile, the third steaming havnng taken place some weeks previous. 

 The fluid in each tube was feebly alkaline, to litmus when inoculated. 

 The cultures were kept at living-room temperatures (20^ to 23^ C). 

 February 19 the cultures were first tested with litmus (the best neutral 

 litmus paper procurable). The fluid in each was plainh' alkaline 

 (much more so than on the start). March 1 the fluids were again tested. 

 All were alkaline to neutral litmus paper. ]\larch 12 the cultures were 

 again tested. Each was plainly alkaline, although not strongly so. 

 The blue color faded out when the paper dried. If anj- acid was 

 formed it was masked by the alkali originally present in the tubes 

 and by that produced during the growth of the organism. 



The above results were obtained in 1897. In 1899 additional fer- 

 mentation-tube experiments were instituted with the following results: 



(1) One of the fluids used was a 1:2 nonpeptonized beef bouillon 

 (stock 382) rendered neutral to phenolphthalein with sodium hydrate 

 and deprived of its muscle sugar by growing Bac'iUus coll in it over 

 night. It was then cleared by passing it through a Chamberland 

 filter. The following substances were tested in this bouillon: Grape 

 sugar, cane sugar, and galactose (3 tubes of each). Each tube con- 

 tained 5 per cent of the sugar to be tested except those with the 

 grape sugar which contained 2.8 per cent. The inoculations were 

 made February 2 and the experiment was closed March 4. The tubes 

 of grape sugar and cane sugar were all well clouded (in the bowl and 

 outer two-thirds of the U) on the fifth day, with exception of one of 

 the grape-sugar tubes which was then only ver}" feebly clouded, but 

 was well clouded 2 da3's later. In each case the closed end of the 

 tube and the inner one-third of the U remained clear until the end of 

 the experiment. The reaction to litmus was watched carefully. The 

 fluid in the bowl of each of the tubes was plainly alkaline to litmus 

 paper (wet or dried) on the ninth, fifteenth, and twenty-third days. 

 On the thirtieth da}' in each tube, whethe^i* of grape sugar or cane 

 sugar, the litmus reaction was distinctly different. The tests were 

 made with two freshly prepared sensitive litmus papers, the one 

 purplish red. the other pale lavender blue. The fluids how blued the 

 purplish red paper slightly and at the same time reddened the bluish 

 paper. The contrast in each case to inoculated check tubes of the 

 plain bouillon (which were now intensely alkaline and blued both 

 papers) was striking. The only conclusion I could come to was that a 

 definite but small amount of acid had been formed slowly from the 

 grape and the cane sugar. In comparison with plain bouillon these 



