E. H. Richards 



301 



wholly excluded, for the power to fix nitrogen to a limited extent is 

 possessed by many organisms of the class which cause butyric acid 

 fermentation in carbohydrate media. 



Since these experiments were begun J. Hanzawa\ working with 

 mixed strains of Azotobacter, obtained greater fixation in mannite than 

 with the same in pure culture, and notes a point of special interest m 

 the present connection, viz. that the humus of stable manure is capable 

 of being utilized as a source of nitrogen fixation, while that of green 

 manure is not. Presumably by humus he means the organic residue 

 of rotted manure. The material used in the present experiments was 

 always fresh from the animals. Hanzawa states that the presence of 

 humus nitrogen or nitrates was almost without effect on fixation by 

 Azotobacter; but if the nitrogen to carbon ratio exceeded 1 : 40 fixation 

 was retarded. 



More recently Tottingham^, working with mixed cow and horse 

 manure (faeces?), both with and without straw, has observed gains of 

 nitrogen up to 5 % after four weeks' fermentation. He believes that 

 the gain is due to the straw but was unable to detect Azotobacter. 



A second series of experiments with horse faeces was set up and run 

 for 119 days at laboratory temperature (15° C). As before, the faeces 

 containing only natural moisture showed no change of total nitrogen, 

 but both the units to which tap- water was added gave an increase 

 of total nitrogen that could not possibly be ascribed to experimental 

 error. No attempt was made to collect the faeces aseptically, but to 

 ensure that the normal soil organisms should be present originally, 1 c.c. 

 of garden soil extract was added to one flask of each pair. The figures 

 obtained in this experiment are given in the following table. 



Horse faeces. 



Weight of substance taken (grams) 

 Original total nitrogen (grams) 

 Final total nitrogen 



Faeces 

 alone 



5032 



00147 



00145 



Faeces 



and 



tap-water 



5-028 



0-0147 



0-0188 



Faeces 

 and soil 

 extract 



5-028 



0-0147 



0-0150 



-0-0002 

 -11 



+ 00041 



+ 27-it 



+ 0-0003 

 + 2-3 



Difference 

 Per cent, of original total N.* 



* In the estimations of nitrogen the results were calculated to four fi^ 

 three are given in the tables. The original four figures have been used in 

 percentage difference. 



1 Centr.f. Bakt. 2 Abt. Bd. xli, 1914, p. 573. 



2 Journal Bio. Chem. (Baltimore), xxiv. No. 3, 191G, p 221 



Faeces, 



tap- water 



and soU 



extract 



5-028 

 0-0147 

 0205 



+ 0-0058 

 + 39-4 



gures but only 

 calculating the 



