292 FORMATION OF SLIME OR GUM BY BH1Z0BIUM LEGUMINOSARUM, 



of the various plants, we cannot tell, but they undoubtedly differ 

 after their isolation. Some form slime upon nutritive media and 

 others do not; some of the former produce slime upon composite 

 media and not upon synthetic media, while the majority form it 

 upon both. The bacteria from the nodules of the same plant may 

 or may not be similar physiologically. From the same kind of 

 plant isolated at different times of the year, they may be closely 

 related, and show fewer differences than bacteria from different 

 kinds of plants. The race from Robinia pseudacacia is quite 

 distinct from Krai's race from Lupinus luteus, and both of these 

 clearly differ from races obtained from Pisum. sativum (Krai's as 

 well as Australian races) and Phaseohis vulgaris (Australian). 

 The latter race approximates to those from Pisum, and that from 

 Viciafaba (Krai) approximates to the race Lupinus bj.leus (Krai)- 

 But that there can be no rule in the matter is shown by the 

 isolation of three physiologically different races from a single 

 nodule of the Blue Lupin. 



With regard to the function of the micro-organism in the 

 nodule, there is strong evidence that it is to produce slime which 

 may be closely related to the carbohydrate of the nucleoproteid 

 molecule. The slime, in all likelihood, is partly utilised in situ, 

 and the formation of the nodule is the result. Thus there is, 

 probably, a true symbiosis so far as the nodule is concerned. The 

 slime may be transported to other parts where it is utilised. 

 There is no evidence to show that the albuminoids of the micro- 

 organism are utilised by the growing plant, for, although the 

 majority of the bacteroids of the nodule are dead, they still 

 retain their chromatin. Living Bhizobia are found in other parts 

 of the plant, e.g., the stem, but in these places they are unable to 

 form slime, and are probably of no use to the host-plant. The 

 formation of slime is proportional to the nitrogen supplied, and 

 there appears to be no fixation of nitrogen in pure culture. The 

 formation of bacteroids depends upon the nature of the carbo- 

 hydrate, which is the principal constituent of the slime. Vibrio 

 denitrificans, which converts combined nitrogen, e.g., nitrate, 

 into free nitrogen gas, produces the same slime and gives the same 



