218 BACTERIAL ORIGIN OF GUMS OF ARABIN GROUP, 



might be neglected. But since saccharose was found to be useful 

 in the saccharose-potato-tannin medium, the reducing sugars of 

 the extract might also be neglected. There remains then aspara- 

 gine as being the essential nutrient of the potato-extract. The 

 salts of the extract would also doubtless assist the bacteria. 



With this information at my disposal I prepared a saccharose- 

 asparagine-tannin medium and infected plates of it with the 

 bacteria. At the same time a comparative test was made with a 

 medium identical with the other, excepting that it contained no 

 tannin. 



It is evident that tannin was of considerable importance, not 

 only in causing a production of slime, but also in making the 

 slime non-adherent, an important point in the artificial production 

 of the carbohydrate. The quantity of slime, however, was small 

 compared with what was obtained when potato-extract had been 

 employed. The absence of slime upon the plates infected with 

 Bad. metarabinum is peculiar, and shows that the nutrition of 

 this organism may differ from that of Bact. acacice, and that these 

 bacteria would need to be investigated separately. 



Believing that the salts of the last medium had had an influence 

 in lessening the yield of slime, an experiment was made in which 

 the medium in the tubes was treated with varying quantities of 

 potassium chloride, potassium monohydrogen phosphate, and 

 potassium citrate. The last-named salt was suggested by some 

 earlier experiments, which indicated increased slime-yields by 



* Saccharose 5, asparagineO'l, tannin 0*3, KHoPO^ O'l, KCl. 0*5, agar 2, 

 water 100 grms. Medium made neutral to litmus before the addition of the 

 tannin. 



