Summary:Forty strains ofA. chroococcumfrom Danish soils showed a rather uniform power of nitrogen fixation in mannitol solution where the mean gain of nitrogen after 12 days corresponded to 17.1 mg./g. of mannitol, with a standard deviation of ±9.1%. In agreement with the statement of Fischer (1949) the strains could be divided into two groups with different rates of nitrogen fixation. Ten strains fixed in 3 days less than 30% and 30 strains from 45 to 100% of the final amount of nitrogen after 12 days, which was almost the same in both groups. In glucose medium a rapid nitrogen fixation was shown by some of the strains that grew slowly with mannitol. Strains which fixed rapidly with mannitol did so with glucose also.Ten strains ofA. beijerinckiigrew slowly with glucose as well as mannitol, and two strains apparently failed to utilize the latter compound.In media sterilized by autoclaving, the gain of nitrogen was approximately one‐third higher with mannitol than with glucose, owing to deterioration of the glucose on heating. When glucose and mannitol were sterilized by filtration, the difference in nitrogen fixation corresponded closely to the difference in heat of combustion of the two compounds.Some strains of bothA. chroococcumandA. beijerinkiicould grow on nitrogen‐free agar containing sodium benzoate as the sole source of carbon, while others could