402 



Journal of Agricultural Research 



Vol. VI, No. 11 



presence of B. rossicus as in its absence, and when arsenic is added to the 

 two there is an even greater fixation. This is also the case with P. 

 effusa; measured in terms of the increased nitrogen fixed by A. chroo- 

 coccum, it may therefore be safely concluded that both of the cellulose 

 ferments are stimulated by lead arsenate. 



The Azotobacter B differs from the A. chroococcum in that it is 

 directly stimulated by the arsenic, but is not as greatly helped by the 

 cellulose ferment. In this case the lead arsenate greatly stimulates the 

 activity of the cellulose ferments, and the stimulating influence is much 

 greater with P. effusa, the normal habitat of which is this soil, than 

 it is with B. rossicus. Hence, from this work it is safe to conclude that 

 the cellulose organisms, so far as arsenic is concerned, obey the same 

 laws as do the ammonifying, nitrifying, and nitrogen-fixing organisms 

 of the soil. 



It has been noted throughout all of this work that the soil taken direct 

 from the field was stimulated to a much greater extent by the arsenical 

 compounds than was the air-dried soil. Furthermore, it was noted that 

 the soil which had stood in the laboratory for a great length of time was 

 stimulated only very slightly by arsenic. For these reasons a series of 

 experiments was planned to throw more light upon this substance or 

 organism which disappears on drying. 



Fred (191 1) has suggested the use of filter paper for the separation of 

 the protozoa. Later this has been shown by Kopeloff and others (191 5) 

 to be quite effective. Using this suggestion, loo-gm. portions of soil 

 were placed in tumblers. To half of them was added 0.0728 gm. of lead 

 arsenate, and the mixture was autoclaved until free from bacterial life. 

 They were all inoculated with 10 c. c. of a solution obtained by shaking 

 100 gm. of soil in 1,000 c. c. of sterile water and then filtering through 

 three thicknesses of a fine grade of quantitative filter paper, after which 

 they v/ere incubated and nitrogen determined as in the previous set. 

 The results are given in Table X as milligrams of nitrogen per 100 gm. 

 of soil. All results are averages of six determinations made on that 

 number of incubated samples. 



Table X. — Quantity of nitrogen (in milligra^ns) fixed in loo gm. of sterile soil inocti- 

 lated with filtered soil extract, with and without arsenic 



