26 BULLETIN 1460, U. S. DEPARTMENT OF AGRICULTURE 
by Dodge 19 and their data is well worth studying. The use of 
potassium acid phthalate has been entirely satisfactory in this 
laboratory. 
METHOD RECOMMENDED FOR MAKING PROTEIN TESTS 
PROCURING THE SAMPLE 
The procuring of a representative sample is essential to the deter- 
mination of the true value of a given lot of grain. If the sample 
obtained is not representative, no amount of care in making deter- 
minations will establish the protein content of the grain samples. 
Consequently, great care should be taken in sampling and the rec- 
ommendations on the subject given in the Handbook of Official 
Grain Standards, Form 90, issued by the United States Department 
of Agriculture carefully considered. 
CLEANING THE SAMPLE 
Wheat samples should not be analyzed for protein content until 
the dockage has been removed from the sample. Dockage includes 
sand, dirt, weed seeds, weed stems, chaff, straw, grain other than 
wheat, and any other foreign material which can be removed readily 
from the wheat by the use of appropriate sieves, cleaning devices, 
or other practical means suited to separate the foreign material 
present. It also includes undeveloped, shriveled, and small pieces 
of wheat kernels which can be removed in separating the foreign 
material, and which can not be recovered by properly rescreening 
or recleaning. 
REMOVING THE DOCKAGE 
Such sieve or sieves should be used as will remove the foreign mate- 
rial with the least possible loss of wheat, including small, plump, 
or badly shriveled kernels, or large pieces of broken kernels. 
The details for the method for removing dockage are given in 
United States grain standards act, Handbook of Official Grain 
Standards, form 90, issued by the United States Department of 
Agriculture.. 
DIVIDING SAMPLE FOR TEST 
After a representative sample of the lot or parcel of grain is 
obtained, it is usually necessary to reduce its size considerably. To 
reduce the size of a sample of grain containing foreign substances 
of different specific gravity or size than of the grain with which they 
are mixed, and at the same time obtain a sample as representative 
as the original, is hardly possible except by mechanical means. 
The device, generally referred to as the Boerner sampler, which 
will divide a sample into small portions and still maintain the proper 
proportions for the various factors of the original sample, is the best 
for this purpose. The device is frilly described in United States 
Department of Agriculture Bulletins Nos. 287 and 857. 
19 Dodge, F. D. the standardization of alkalimetric solutions. Jour. Ind. Chem., 
vol. 7, p. 29. 1915. 
