42 
Journal of Agricultural Research 
Vol. XXVIII, No. 1 
The accuracy of the method is indicated by the results of the preliminary 
experiments reported in Table II. Almost complete recovery of the ammonia 
in ammonium sulphate was secured after aerating three hours. In the presence of 
soil it was more difficult to drive off the ammonia, 93.9 per cent being recovered 
in three hours and 98.5 per cent after six hours. When a dicyanodiamid solution 
was aerated the recovery as ammonia was 0.18 per cent. This small amount is 
easily within experimental error when dealing with such small quantities of 
nitrogen. 
DETERMINATION OF NITRATE NITROGEN 
Two methods of determining nitrates were used in this work, namely, the 
reduction and the colorimetric methods. 
REDUCTION METHOD 
The determination of nitrate nitrogen by the reduction method, using 
De varda’s alloy, is now generally considered as the most satisfactory for most 
purposes. However, in the presence of undecomposed cyanamid, dicyanodiamid, 
urea, or guanylurea the reduction method as ordinarily carried out gives high 
results. This is because these compounds, when boiled with weak alkali, slowly 
decompose with the evolution of ammonia. A modification of the method to 
avoid the difficulties was devised and has been reported elsbwhere (8). The 
modified procedure was used in this work wherever the reduction method was 
employed. 
COLORIMETRIC METHOD 
The phenol-disulphonic acid method of determining nitrates in soil solutions 
has been practically set aside in recent years because there are numerous salts 
and organic materials which interfere in the color production or particular shade 
of color. Regardless of these objections tc an indiscriminate use of the method 
it is nevertheless recognized that under certain conditions the method is quite 
accurate, and much shorter than the reduction method. 
A- comparison between the modified reduction method and the phenol-disul¬ 
phonic acid method was made during the course of these investigations on 
samples of soil containing cyanamid and its decomposition products. For the 
particular soil used the results were in quite close agreement and the compounds 
which require elimination in the case of the reduction method did not interfere 
in the colorimetric method. For small quantities of nitrate nitrogen the latter 
method is more accurate, and since it is shorter it was used in obtaining a por¬ 
tion of the results reported in this paper. Where this method was used the fact 
is mentioned, otherwise the modified reduction method was used. 
DETERMINATION OF COMBINED CYANAMID AND DICYANODIAMID 
NITROGEN 
The method of Brioux was used for the determination of cyanamid and dicy¬ 
anodiamid nitrogen. To 50 cc. aliquots of soil extract were added 10 cc. each of 
5 and 10 per cent solutions of silver nitrate and potassium hydroxid, respectively. 
It was found necessary to wash the precipitates 12 or 13 times with distilled water 
in order to remove all traces of ammonia. The total nitrogen in the precipitate 
was then determined by the usual Kjeldahl method. 
\ 
DETERMINATION OF UREA NITROGEN 
The short urease method as developed at this laboratory by Fox and Geldard (5) 
was used for the determination of urea. This consists in adding to a 50 cc. 
