SPECIES 293 



infected by one variety, whereas with another variety infection is 

 accomplished with difficulty or not at all. Moreover, the sero- 

 logical test yielded by different varieties is specific. These facts have 

 led some observers to consider the types as quite distinct, whereas 

 others consider them as simply physiological varieties of the same 

 general species. On the whole, the consensus of opinion at the 

 present time seems to be decidedly in favor of this latter view. 



Some of Hellriegel's experiments indicated that bacteria from 

 clover could not produce tubercles on lupines and serradella. Simi- 

 lar results were obtained by Nobbe and his associates, yet they were 

 finally led to conclude that the root invasion of legumes is caused by 

 a single species. Long-continued growth of the organism on a 

 legume adapts it to that legume so it no longer invades the roots of 

 other legumes. But Petermann '(1893) considered it probable that 

 every genus of plant has its specific bacteria. Buhlert considered 

 that all of the organisms are forms of B. radicicola but that the 

 bacteria best adapted to a given species of leguminous plant are those 

 naturally found upon that plant. However, cross inoculation is 

 possible within certain limits. From the root tubercles of some 

 leguminous plants he obtained bacteria which seemed to be very 

 highly specialized, but he considers that this specialization does not 

 extend to differences that may be regarded as specific. 



As a result of a large number of experiments with different kinds 

 of legumes, Maassen and Miiller (1907) reached the conclusion that: 

 (1) The organisms of Pisum satium will inoculate Vicia faba, V. 

 saliva, V. villosa, Lens esculenta, Laihyrus sativus, L. odoralus, 

 and L. silvestris; (2) that of Trifolium incarnatum will inoculate 

 T. pratense; (3) that of Medicago saliva will inoculate M. lupulina 

 and Melilotus officinalis; and (4) that of Lupinus lutens will inoculate 

 L. angustifolius and Ornilhopus salivus. The organisms of Phaseolus 

 vulgaris, Soja hispida, and Robinia pseudacacia, according to 

 Maassen and Miiller, will apparently not inoculate any other plant. 

 Similar conclusions were reached regarding the organisms of Coronilla 

 varia, Onobrychis satava, Anthyllis vulneraria, Sarclhamnus scoparius, 

 Amorpha frnticosa, Caragana frutescens, and Acacia lophanta. 



De Rossi (1907) described a specific organism derived by him from 

 root tubercles of V. faba which produces root tubercles and which 

 he claimed is morphologically, biologically, and culturally widely 

 different from Bacillus radicicola Beijerinck. 



Nobbe and coworkers (1908) showed that pure- cultures of 

 bacteria from tubercles of one member of a genus are effective on 

 other members of the same, and, as a rule, only of the same genus. 

 They found, however, complete interchangeability in case of peas 

 and vetches and partial in case of lupines and serradella. 



Zipfel (1912), with the hope of throwing light upon the kinship 

 among the various nodule bacteria, made use of .the agglutination 



