22 Journal of Agricultural Research voi. iv, no. i 



needs further description, and the reliabihty of the results obtained by 

 it has been fully demonstrated through the work of a large number 

 of investigators who have taken it up since it was first proposed by 

 Stevens and Withers (i8, 19, 20). Lipman and Brown (11), working 

 together at the New Jersey Experiment Station, have proposed the 

 same manner of experimentation, and this method of investigation has 

 been almost universally adopted. The same general procedure was 

 therefore considered to be suitable for the study of the availability 

 of the organic matter of the kelps in question. 



EXPERIMENTAL METHODS 



In order that the results should be comparable to those obtained 

 under field conditions, the soils used throughout all the series were fresh 

 field soils. The surface soil was removed to a depth of iK to 2 inches, 

 it having been shown by Lohnis (15, p. 571) and others that this portion 

 does not possess a typical bacterial flora. A large sample was then taken 

 from the next 3 or 4 inches of soil, carried immediately to the laboratory, 

 spread out in a darkened room, and dried to a point where it could be 

 readily sifted and thoroughly mixed. In this condition it was still 

 moist and, according to the work of Brown (3, 4), undoubtedly possessed 

 a bacterial flora which was representative of field conditions. Two- 

 hundred-gram portions of this soil were then placed in sterile tumblers, 

 which were covered with Petri dishes. The various amounts of material 

 to be investigated were then added, the whole being thoroughly mixed. 

 Sterile distilled water was added to make optimum moisture conditions, 

 which in most cases were found to be about 18 per cent. The samples 

 and duplicate blanks were then incubated at 28° to 30° C. for the 

 required time. At the end of the period they were transferred to copper 

 distilling flasks. About 400 c. c. of distilled water were added, together 

 with magnesium oxid and a small quantity of paraffin to prevent foam- 

 ing. The ammonia present was distilled into standard Nlio hydro- 

 chloric acid and the excess portion titrated with standard Njio ammonia 

 in the usual manner. 



Upon consideration of the analytical results obtained from the first 

 portion of Series I and the very slow rate of ammonification of Macro- 

 cystis pyrifera revealed by these results, it seemed desirable to have deter- 

 minations of the amount of nitrification which \vas probably taking place 

 concurrently. This would, it was felt, be a satisfactory check on the 

 progress of the transformation of the nitrogen through the cycle of 

 changes taking place in the soil. The residues remaining in the copper 

 distilling flasks after the determination of ammonia were immediately 

 transferred to i -liter volumetric flasks of the Giles pattern, or, where 

 much small gravel was present, to i -liter graduated cylinders. They 

 were then made up to i liter and sufficient additional distilled water 

 added to compensate for the displacement caused by the soil and ferti- 



