EXAMINATION OF GRAHAM FLOUR. 33 



such a small amount of bran, shorts, and middlings, and such a large 

 amount of flour passing through the 109 sieve and also because the 

 nitrogen of this latter product is greater than the middlings nitrogen. 

 The gliadin numbers of these products vary considerably, depending 

 upon the method of milling the wheat. This is due to the fact that 

 when a Graham is made, the wheat is simply run over corrugated rolls 

 or between stone burrs in such a way as to reduce the wheat without, 

 however, producing absolutely clean bran, shorts, or even clean mid- 

 dlings. The bran and shorts almost invariably contain a consider- 

 able amount of flour and hence the gliadin number is higher than in 

 the corresponding products of imitation Graham flour in which clean 

 bran, shorts, and low-grade flour are mixed together. The gliadin 

 number of the flour passing through the 109 sieve, obtained from 

 Graham flour, is as a rule higher than that obtained from imitation 

 Graham flours, this being due to the fact that the latter are generally 

 made from mixtures of low-grade flours and bran while Graham flour 

 contains the whole kernel of the wheat and hence a larger proportion 

 of good material. It has been found that the gliadin number of a 

 product increases with the quality of that product; that is to say, 

 the highest patent flour made from any particular variety of wheat 

 would have the highest gliadin number; therefore, when alleged 

 Graham flours give a product which passes the 109 sieve and have an 

 abnormally low gliadin number it is justifiable to assume that low- 

 grade flour has been used. No. 6167 of Table 3 and Nos. 6229, 6230, 

 and 6231 of Table 5 are types of this kind of flour. 



Samples Nos. 6229 to 6231 show another characteristic often found in 

 imitation Graham flours; that is, the percentage of nitrogen in the 

 flour passing through the 109 sieve is greater than that of any of the 

 other products and even of the original imitation Graham flour. In 

 these samples the nitrogen of the flour is also greater than the nitro- 

 gen of the bran. Occasionally this is also the case with true Graham 

 flour, but only when the amount which passes through the 109 sieve 

 is very small, indicating that the latter is of the nature of a break 

 flour; this generally occurs when the Graham flour is made from a 

 hard spring wheat. In these samples the high amount of flour clearly 

 indicates that it is not a question of break flour but that a low-grade 

 product has been used in making the alleged Graham flour. All 

 these samples in Table 5, except No. 6239, show the middlings nitrogen 

 to be greater than the bran nitrogen. This may be due to the fact 

 that in most cases the middlings are tailings from middlings which 

 are generally very high in nitrogen. 



Tables 10 and 11 (p. 21) show the relative amount of nitrogen and 

 the gliadin number, respectively, of the Graham flour and each pro- 

 duct of separation from Graham flour. Tables 12 and 13 give the 

 same relation with regard to imitation Graham flours. These tables 



