2 BULLETIN 1<±60, U. S. DEPARTMENT OF AGRICULTURE 
at 13 to 16 cents over the May price ; 13 per cent protein was quoted 
at 14 to 15 cents over. At Kansas City, hard winter 12 per cent 
protein sold at 10 to 11 cents over the May price, 12y 2 per cent at 
11 to 12 cents over and 13 per cent liy 2 to I2V2 cents over." 
The practice of paying premiums for high-protein wheat is due, 
in a general way, to the fact that the value of flour for baking pur- 
poses is known to be related in part to the quantity of the gluten 
proteins present in the flour. 
As the gluten proteins constitute a fairly definite proportion of the 
total protein content of the wheat, 75 to 80 per cent, and as there 
is a close relationship between the gluten content of the wheat and 
that found in the flour milled therefrom, the quantity of crude pro- 
tein as determined by the Kjeldahl method or its modifications is a 
fairly dependable index of the quantity of the gluten protein present. 
In addition, since there has never been a single factor test to 
denote baking quality, and the crude protein test is capable of being 
expressed in definite mathematical terms, in percentage by weight, 
grain dealers, millers, and bakers have come to use the protein test 
as an aid in the valuation of wheat and flour. At present, the hard 
red spring and hard red winter wheats are marketed on a protein 
basis. 
On certain occasions when the supply of high-protein wheat has 
been less than the demand, premiums of as much as a cent have been 
paid for each added tenth of 1 per cent of protein over a given basic 
amount. To trade on so close a margin taxes the protein test to its 
uttermost and is the cause of many of the complaints as to inac- 
curacy of chemists in making their determinations. 
To determine how close different laboratories should be able to 
report results of protein tests on the same sample of wheat, the 
methods and technique used in making protein tests were studied in 
mill laboratories, commercial protein-testing laboratories and grain- 
inspection department laboratories in the Central Northwest and 
Southwest. The results of this survey are incorporated in this 
bulletin as well as the report of an exhaustive laboratory study re- 
lating to the value of many of the outstanding methods and tech- 
nique used by the different protein-testing laboratories. 
A standard method for making protein tests is described. Rec- 
ommendations are given for eliminating sources of error, so that 
concordant results between different laboratories can be obtained. 
PROTEIN TESTS BY VARIOUS LABORATORIES 
In an investigation to determine the reasons for non-uniformity 
or variations in protein results as made by different protein-testing 
laboratories 12 bulk samples of wheat representing the hard red 
spring and hard red winter crops were thoroughly cleaned, mixed, 
divided into approximately 50-gram portions and placed in sealed 
air-tight containers to insure uniformity of moisture on arrival at 
the various testing laboratories. Samples 1 to 6 represented wheats 
of the hard red winter class and samples 7 to 12 represented wheats 
of the hard red spring crop. 
Portions of each of these 12 samples were sent to 45 laboratories 
with a request for a report on their protein content. Four types of 
