15 



nite, might be expected to approach the latter in fixation, but 

 this did not prove to be the case, yet it differed from maltose 

 only in the second place of decimals. Erythrite fixed an 

 average, on soils 1 and 43 in twenty-one days, of 0.18 mg, 

 while on soils 10, 24, 34 and 41, in fifty-four days, an average 

 of 5.59 mg was fixed. The slow fermentation of this sugar 

 renders it useless for laboratory purposes. Dextrin and 

 inulin gave comparable results which were inconsiderable. 

 Probably twenty-one days is insufficient to develop the maxi- 

 mum efficiency of these polysaccharides. 



Glycerine in soils 1 and 43 fixed an average of 1.13 mg in 

 twenty-one days. Soils 7, 10, 24, 34, 41, 47, and 61, fixed an 

 average of 6.64 mg in thirty-nine days. Soil number 1 fixed 

 3. 58 mg in thirty-nine days and soil 43 fixed 4. 74 mg in the same 

 time, a gain in the first instance of 2.61 mg and in the second 

 of 2.45 mg in eighteen days. The slow fermentation of gly- 

 cerine relegates it to the class with erythrite. 



In the work on mannite solutions one is struck with the 

 great variety of odors, but perhaps the most characteristic is 

 that of butyric acid. This led me to conclude that butyric 

 acid or oxybutric was either one of the splitting products of 

 mannite, or that according to an early discovery, two mole- 

 cules of lactic acid were changed to one of butyric acid, giving 

 off two molecules of carbon dioxide and two molecules of hy- 

 drogen. After adding calcium butyrate to Ashby's medium 

 it was inoculated with soils 1, 10 and 43, these yielded an aver- 

 age fixation of 0.15 mg. The butyrate therefore seemed not 

 available for carbon supply. In place of calcium butyrate, 

 calcium lactate was next introduced using three grams to the 

 flask, an equivalent of 2.45 grams calculated as free lactic acid. 

 The average fixation for ten soils was 3.01 mg. The figures 

 on this compound do not show the uniformity of the others, 

 although soils 2, 7, 47 and 61 did remarkably well. The odor 

 of butyric acid was not so pronounced as had been expected, 

 but some cultures showed unmistakable evidence of its presence. 



In the fermentation of mannite considerable ethyl alcohol 

 is split off. An analysis of the total acidity revealed approx- 

 imately 30% acetic acid and 70% butyric acid. 



The following table shows the amount of nitrogen fixed 

 when grown in a medium containing the compounds listed: 



