24 BULLETIN 1460, IT. S. DEPARTMENT OF AGRICULTURE 
fore, can be used with safet}^ only when the temperature of the distil- 
late is not greater than 35° C. It must also be titrated within 4 hours 
after the ammonia has been introduced. It is evidently not safe to 
allow it to stand overnight, contrary to the findings of Scales and 
Harrison. 15 
These data are also contrary to that recently submitted by Markley 
and Hann 16 who find no appreciable loss of ammonia at 50° C. 
MERCURY PRECIPITANTS 
The use of potassium and sodium sulphide seems to be general 
among collaborators using mercury as a catalytic agent. But four 
of the collaborators used sodium thiosulphate for this purpose. This 
reagent, providing it is as effective as the first two, has a decided ad- 
vantage over the sulphides, as it is less expensive and has no repulsive 
odor. Tests were made of its efficiency as a mercury precipitant and 
it was found that 1 gram of sodium thiosulphate is sufficient to pre- 
cipitate 0.9 gram of metallic mercury. It was also found, during the 
investigations, that this quantity of mercury will hold back approxi- 
mately 13 per cent of the nitrogen present. This is close to the fig- 
ures as given by Paul and Berry. 17 It is, therefore, suggested that 
this reagent be given larger use, especially where objection is made 
to the odor of the sulphides of potassium and sodium, as is often the 
case in office buildings. 
STANDARD SOLUTIONS 
To gather some information relative to the accuracy of standard 
solutions, 8 ounces of sulphuric acid were sent to each collaborator 
with the request that he standardize the solution and report its 
value. These figures are given in Table 17 and show that the nor- 
mality as submitted by 34 collaborators varied from 0.13003 to 
0.1256, or a difference of 0.00443. Expressed in terms of protein, 
this is equivalent to 0.35 per cent, a very appreciable amount when it 
is considered that the Kansas City Board of Trade requires a re- 
check analysis when the protein test varies more than 0.12 per cent. 
It is felt that special attention has been given to this standardization 
work and that the usual routine in this field would show even wider 
variables. 
As an added interest, the same sample of acid was sent to the 
United States Bureau of Standards for test and the reported nor- 
mality value is given under the caption X in Table IT. This value 
was obtained gravimetrically by weighing the sulphate as barium 
sulphate. This value was further checked volumetrically by titra- 
tion with a standard alkali solution. The two values were identical. 
Xo nonvolatile sulphates were found. 
15 Scales, F. M., and Harrison,, A. P. boric acid modification of the k.teldahl 
method for crop and soil axalysis. Jour. Ind. Eng. Chem ., vol. 12. No. 4, p. 350. 1920. 
16 Markley, K. S., and Haxn, R. M. a comparative study of the guxning-arxold 
AND WINKLER BORIC ACID MODIFICATIONS OF THE KJELDAHL METHOD FOR THE DETERMINA- 
TION of nitrogen. Jour. Assoc. Off. Agri. Chemists, No. 4, vol. S, pp. 455-467. 1925. 
1 7 See footnote 8. 
