693 



ing filaments are Rimi)ly colonies of bacteria inclosed in a membrane of 

 their own manufacture, for their protection against the injurious action 

 oi" the plant tissue. The filaments swell with the multiplication of the 

 bacteria until tbey burst. The bacteria then coming into contact with 

 the plant tissue and no longer being able to grow, owing to an injurious 

 inliuence of the [)lant plasma upon them, degenerate into the bacteroids. 

 They are subseipiently absorbed by (he plant and incorporated into 

 the substance, serving therefore as food. 



The view held by Frank differs from this essentially in its explana- 

 tion of the lilauients and bacteroids. The fihmients are said to be a 

 mixture of plant protoplasm and bacteria. They are i)roduced by the 

 plant and serve to conduct the infectious matter into the midst of the 

 root. The bacteroids are also products of the i)lant plasma, and not 

 distinct organisms. Their absorption does not, therefore, especially 

 help the plant. 



The third view, that of VV^ard and Laurent, regards the infecting 

 organism not as a bacterium, but as a low fungus, somewhat closely 

 related to the yeasts. The filament is really a mycelial growth of the 

 organism, and the bacteroids arise from it by budding. The bacteroids 

 are thus distinct organisms — not degenerate forms, but normal growtlis. 



None of these views would regard the tubercle organisms as true 

 l)arasites on the plant since the plant is not injured by them, but is 

 l)robably directly beuefited. The association is rather to be regarded 

 as an instance of 83'mbiosis, an association of two organisms together 

 in such a way that each receives benefit from the other. The plant is 

 probably benefited in gaining nitrugen, and the infecting organism is 

 beuefited in gaining a brood i>ouch for its develoi)meut. 



