246 



BACTERIAL ORIGIN OF GUMS OF ARABIX GROUP, 



From the results it was clear that common salt or potassium 

 nitrate were the most promising for further experiment. Vary- 

 ing amounts of these were therefore added in a subsequent trial. 

 Owing to pressure of other work this was done six vreeks after- 

 wards, and during the interval the bacteria had been resting and 

 had partly lost their slime-producing faculty, as will be seen from 

 a comparison of the amounts formed on the control media. 

 During the interval the laboratory temperature had risen to about 

 22° in Experiment i. and to about 20° in Experiment ii. 



The depressing influence of Sodium Chloride and Potassium Nitrate. 



Slime in grms. from 100 c.c. of 

 media.* 



From the experiments it is unlikely that anything would be 

 gained by manuring the trees with salts such as common salt, for 

 the quantity which would affect the bacterium would also injure 

 the plant. But possibly something might be gained by the use 

 of small quantities of potassium nitrate, which would act as a 

 plant stimulant and as a bacterium depressant. Field experi- 

 ments would alone determine this question. 



Infection experimeyit with jyeach trees. — It is customary to 

 identify the relation of an organism with a particular disease, 

 whether of plants or of animals, by introducing a pure culture of 



* Levulose 2*0, tannin 0-1, asparagine 0-1, potassium citrate O'l, agar 2-0, water 100. 



