抄録
In order to reexamine the cross-inoculation groups and to know the behavior of each type of rhizobial trains (classified in the previous papers) to legumes, cross-inoculation studies were performed. According to the results obtained, the so-called cross-inoculation groups may be admitted, though not without considerable irregularities. Such irregularities are particularly marked around bean (Phaseolus vulgaris) and cowpea (Vigna sinensis). It is also to be noticed that the relation between Medicago denticulata and Melilotus alba (or Medicago sativa) is hardly recognized to be reciprocal. Among the legumes used, Albizzia, Calopogonium, Centrosema, Crotalaria, Desmodium, Mucuna, and Tephrosia are all included in the cowpea group. Indigofera (and certainly Styphnolobium) will also come into the same group. Vicia nipponica may belong to the pea group. The results on Sesbania and Astragalus (sinicus) agree with the proposition of some investigators that each of them constitutes an independent group, respectively. From the reason that very few of rhizobial strains from already known inoculation groups are able to from nodules on Mimosa and Leucaena, and the cross-inoculation between the two is reciprocal, one new group, in which Mimosa and Leucaena are included may be proposed, though the relation between Mimosa and Dalea (or Sesbania) remains to be studied. Sophora seems to be a separate group. That the strains from one group or even from one plant are not always similar in the nodule production kinship is clearly proved in the subgroups of rhizobia from alfalfa, soybean, cowpea and Sesbania. But further careful studies will be necessary in this respect, because there are not few cases that the strains grouped in the same position do not always behave in like manuer. Finally, it is to be added that some plants (or strains of rhizobia) are able to symbiose with wide range of strains (or plants), while others do so only with very limited range of them.
