ROOT-NODULES OF LEGUMINOSAE. I07 



in a medium deprived of combined nitrogen, obliging it to depend for nourishment from the beginning 

 upon the nitrogen of the atmosphere is to demand of it more than it is able to do. 



"We see that the dose of sugar can not fall much below 2 per cent, for the experimenters who 

 have worked with media containing only one per cent of sugar have not found any sensible increase 

 of the nitrogen." 



Easy access of air also exercises a very favorable influence on the fixation of nitrogen, and this 

 is easily comprehended, for the rapidity of the combustion of sugar stands in relation to the quantity 

 of oxygen furnished to the cultures. It is because he did not fulfil this condition of aeration that 

 Mr. Beyerinck has observed only a very limited fixation of nitrogen. 



Maz^ states that the plant must furnish the bacillus 100 grams of starch in order to receive 

 in exchange i gram of nitrogen. 



"The cultures of the bacillus of the L.eguminosae in bean broth, exhaled a strong odor, not 

 without analogy to that which is given off by soft cheeses (brie and camembert)." 



According to Maze's second paper (1898) Bad. radicicola does not grow in an atmosphere of 

 nitrogen, although it remains alive for some time. I^aurent's contradictory results are to be ascribed 

 to a defective experiment, i. e., to traces of oxygen left in his air. The organism is an aerobe. It is 

 greedy of oxygen. It is able to fix free nitrogen without the assistance of the plant. 



In fixing nitrogen in flask cultures Maze states the best result to be when the combined nitrogen 

 was I to 200 of the saccharose, the lower limit of the latter being 2 per cent and the upper limit 

 4 per cent. The minimum limit of combined nitrogen in bouillon cultures is 14 mgr. per 100 cc, 

 and the maximum about 30 mgr. per 100 cc. Maze's evidence in favor of the storage of nitrogen is 

 increased by another experiment. In three 50 cc. flask cultures there was more than twice as much 

 nitrogen at the end of the experiment as at the beginning, the gain being respectively 12.1 mgr., 12.8 

 mgr. and 15 mgr. Two other flasks in the same series, differing only slightly in nitrogen and sugar- 

 content, gave no increase of nitrogen and there was only a slight decrease in the amount of sugar. 

 Another experiment is mentioned but here the gains and losses are so slight as to seem within the 

 limits of experimental error (p. 133). 



The nitrogen is not all locked up in the organisms ; a portion is soluble and will dialyze (about 

 J in a flask culture of 100 cc. diluted to Boo cc. with distilled water, i. e., 8.5 mgr. out of 32.04 mgr.). 



In media containing very minute quantities of combined nitrogen the root-nodule organism 

 makes a feeble growth and does not fix free nitrogen. He got no increase of nitrogen in 50 cc. flasks 

 of bean broth containing as little as 3.3 mgr. of combined nitrogen. This agrees with Beyerinck's 

 results, and contradicts Frank's, Prazmowski's, and I^aurent's. 



Legumin is a good source of nitrogen. Nitrates are better foods than ammonia salts. In 

 ammoniacal bouillon cultures 30 days old there was no increase of nitrogen and very little diminution 

 of the saccharose. The nodule bacteria grew also in sterilized soil free from nitrates, but with no 

 increase of nitrogen (3 months) : One experiment only and believed to be insufficient. He states that 

 he did not succeed in isolating from the soil a bacillus capable by itself of producing nodules. 

 Saccharose and dextrose attract these bacteria. Water of germination repels them. They are sen- 

 sitive to acids. The only chemotactic substances emitted by the roots of legumes are carbohydrates. 



Laurent states that the maximum temperature for growth of Bacl. radicicola is 30° C, but Maze 

 found it grew very well on agar at 35° C, especially after a few transfers. 



The branched forms are due to vegetation under harmful conditions as shown by growth in acid 

 media and at 35° C. During the first few transfers at 35° C, and especially at the end of the first 

 24 hours, they are abundant. In successive transfers as the bacteria become accustomed to this 

 temperature the branched forms disappear entirely. If the branched cultures are diluted with bean 

 bouillon they give rise to unbranched rods. It is impossible to fix in the breeder's sense the branch^ed 

 forms by any method of culture. Growth in bean bouillon is prevented by the addition of a small 

 amount of tartaric acid (i : 1000). 



By sowing very copiously, growth was obtained on slightly alkaline agar to which i : 1000 tar- 

 taric or oxalic acid had been added and here pear-shaped forms were found. The pear-shaped and 

 branched forms found in the nodules are ascribed to the injurious action of the acid cell-sap of 

 the host. 



The bacteria as isolated from the nodules do not liquefy gelatin. I^ater Maz^ obtained from some 

 of his cultures round forms believed by him to be part of the life cycle, and these Uquefied gelatin 

 rapidly. 



Maz^ states that the round and rod-shaped forms, which he believed at first to be two species, but 

 later forms of one, inoculated separately do not give nodules. Those roots inoculated with mixtures 

 of the two organisms gave numerous nodules. Maze states that the active rod-shaped form is unable 

 to form nodules. In this he is clearly wrong. He is probably wrong also as to the relationship of the 

 round organism, and this throws more or less doubt on all of his paper. Many of his conclusions 



