114 BACTERIA IN RELATION TO PI^ANT DISEASES. 



In some cases motile stages appeared in cultures a week or more old. These never traveled far; 

 often one end remained stationary while the other oscillated to and fro. 



In plate-cultures of organisms obtained from nodules of different genera of Leguminosae no 

 radical differences were noticed. Tiny cream-like colonies become visible after two days' cultivation 

 on nutrient gelatin. 



"After four days a distinction [under the microscope] is obvious between the flat, pale yellowish- 

 brown emerged colonies and the submerged ones which show a manubrium either central or on the 

 margin, and are darker in color and more opaque than the former. The submerged colonies frequently 

 become confluent. No liquefaction of the gelatin had taken place after two months." 



On gelatin the surface colonies of the organism from nodules of Desmodium gyrans become 

 distinctly domed when the size of a pin's head. 



"In streak cultures slight differences in the mode of growth of the various t5rpes occur, but no 

 corresponding microscopic differences could be seen." 



In stab cultures no liquefaction of the gelatin occurred. As in streak cultures, the surface 

 growth in A was at first more pearly than in B and C. 



Regarding the systematic position of this organism as affected by her investigations. Miss 

 Dawson says : 



It may now be assumed as proved beyond question that we are dealing with independent 

 organisms which have become very specially adapted for life within the cells of leguminous plants — 

 a specialization which varies apparently with different hosts. * * * It is impossible, as yet, to 

 assign them to any other group, if not to the schizomycetes. 



She also says : 



"On the other hand, if branched forms are to be included among true bacteria, in accordance 

 with the views of the aforementioned investigators, the organisms (which it is convenient to continue 

 to describe as Rhizobium) could certainly claim to be classified with them, and in this case the 

 branched individuals would probably be regarded as involution forms. As, however, such a classi- 

 fication would involve a new definition of bacteria, it seems advisable to adhere to the view that the 

 organisms are in reality either very primitive or very degenerate fungi."* 



A further conclusion based on the experiments described in this paper, in addition to others not 

 recorded, is that there is only one organism capable of forming nodules upon the roots of leguminous 

 plants. 



In 1900, Hiltner published a long paper on the causes which affect the size, number, location 

 and activity of the root-nodules of the Leguminosae. 



He takes as a motto the statement of Beyerinck that a very subtle balance must exist between 

 the growth of the two organisms, the bacterium and the leguminous plant. This, he thinks, is not 

 a purely symbiotic relation, but a contest, in which a greater part of the penetrating bacteria are 

 destroyed, and in which the host plant also suffers evident injury, but a relation which, nevertheless, 

 is advantageous to both in the end. His experience covers a period of 10 years. The following is 

 an abstract of this paper. 



Prazmowski, he says, was the first to demonstrate the true method of infection through the 

 root-hairs. He showed in the pea that only young hairs near the root-tip are penetrated by the 

 bacteria. The attacked hairs curl at the tip, and in their interior there soon appears a shining button- 

 shaped body from which a filament filled with bacteria grows along the root-hair and penetrates into 

 the root, boring through cell-membranes. Some of Hiltner's own experiments on the subject were 

 as follows : 



He carefully washed and sterilized with absolute alcohol and corrosive sublimate, mature pea 

 nodules, and rubbed them up with a little water. The whole mass was then filtered repeatedly 

 through a Chamberland filter. The result was a clear liquid free as a rule from bacteria but containing 

 soluble secretions from the bacteroids. Young pea plants, growing in nitrogen-free nutrient solutions 

 were inoculated with a small amount of this filtrate. After a short time numerous young root-hairs 

 showed the same phenomena of curling as did hairs attacked by bacteria. As no nodules were formed, 

 however, it was evident the bacteria had been retained by the filter. 



When seedlings of Lathyrus sylvestris were inoculated with the same filtrate no curling of the 

 root-hairs occurred for a long time, i. e., not until after the supply of nitrates in the seed had been 

 exhausted. Then when the plants began to hunger the hairs on the youngest roots showed the 

 curling. Pea seedlings inoculated with a filtrate from Lathyrus nodules gave like results. These 

 experiments indicate that excretions from the nodule bacteria of one species are able to influence 

 only weakened plants of other species. 



*In opposition to this view it may be stated that branched forms are not peculiar to this organism but common to 

 many bacteria when growing under adverse conditions e. g., in the presence of acids, and these are to be regarded as 

 degeneration forms. Such forms are either dead, or if capable of further growth return to the original form when placed 

 on suitable media. 



