550 Journal of Agricultural Research vol. xx.No. 7 



Cultures on beef gelatine and in beef broth differentiate clearly Radio- 

 bacter and nodule bacteria, while, as stated in Table I, the two groups 

 of nodule organisms grow very much alike on these substrates. Micro- 

 scopic tests, however, made from gelatine and broth furnish, in most 

 cases, especially characteristic pictures, provided that the growth has 

 not been altogether too poor to get a satisfactory preparate. 



The growth in milk and on potato, as described in Table I and illus- 

 trated on Plate 69, is very characteristic and can be used to great advan- 

 tage for diagnosis. It is not to be denied that with old stock cultures 

 atypical results may sometimes be obtained in this direction also. 

 Especially cultures rich in or entirely made up of the globular regenera- 

 tive bodies, which are produced by these as well as by all other bacteria, 

 furnish whitish, yellowish, or only slightly brownish growth on potato 

 in the case of Bacillus radiobacter and B. radicicola. But we have never 

 seen such atypical growth with new isolations. Here the coli-brown 

 color of the potato cultures separates Radiobacter sharply from the 

 nodule bacteria, and these in turn are equally sharply to be distinguished 

 by the behavior of their milk cultures. It is true that sometimes milk 

 cultures of the B. radicocola group also leave the milk unchanged, but 

 the microscopic test of such abnormal cases probably will always show, 

 as it did in the cases studied by us, that the abnormality was simply 

 caused by the fact that the bacteria which were inoculated did not multi- 

 ply at all. Furthermore, no alteration may be seen if milk is used which 

 has been kept for a long time and has become concentrated by evapora- 

 tion of part of its water. 



To determine on a larger scale whether this different behavior of the 

 two groups of nodule bacteria, when grown in milk, can be correctly 

 accepted as of real diagnostic value, all cultures of nodule bacteria at 

 our disposal were tested simultaneously with the following results: 



MILK WAS CHANGED AS TYPICAL FOR MILK WAS LEFT UNCHANGED BY THE 

 BACILLUS RADICICOLA BY THE FOLLOWING CULTURES: 



FOLLOWING CULTURES: 



5 from red clover. 10 from cowpea. 

 4 from sweet clover. 8 from soybean. 



6 from navy bean. 5 from peanut. 



1 from vetch. 4 from Japan clover. 



2 from lupine. 2 from beggar weed. 



3 from black locust. 2 from Cassia chamaecrista. 

 3 from Amorpha. 



2 from Strophostyles. 



If kept for longer than four weeks milk cultures of the cowpea-soybean 

 organisms usually become more or less transparent on account of partial 

 decomposition of the casein; but they never show the perfectly clear zone 

 characteristic of the other group. 



The bacteria were also tested on other media besides the standard 

 substrates, of which sterilized soil, moistened with 0.5 per cent mannite 



