EXAMINATION OF GKAHAM FLOUB. 45 



that the chop from the second break only had been used, thus elimi- 

 nating to a large extent the low-class products. The percentages of 

 fiber and pentosans in the original sample are in accordance with the 

 percentage of ash and indicate the removal of most of the bran in the 

 preparation of the flours. 



No. 9097 is made from soft winter wheat on a 4-set stand of corru- 

 gated rolls. The bran is scalped off after each grinding, removing 

 about 25 per cent altogether. Some of the middlings are also 

 removed. This imitation Graham flour is really made from the 

 unreduced product that makes up a cut-straight. The analysis shows 

 only 2.3 per cent of bran. The small amounts of ash, fiber, and 

 pentosans indicate that only a small part of the bran had been 

 allowed to remain in the sample. The gliadin ratios of the shorts 

 and coarse and fine middlings are very high. This is due to the fact 

 that so much of the coarser particles which tend to give low gliadin 

 ratios had been removed. It has been shown in an early discussion 

 that the more bran particles that are present in these products the 

 lower is the gliadin ratio, and vice versa. 



No. 9109 is made by mixing 83 parts of an 8 per cent clear flour and 

 17 parts of bran, the dealer claiming that this imitation Graham is 

 thus made in order to compete with the commercial that is, imitation 

 Grahams on the market. For the purpose of making this imitation 

 Graham both the bran and the clear flour are bought separately. The 

 separation shows 16.6 per cent of combined bran and shorts, only 1.1 

 per cent of combined middlings, the rest, or over 82 per cent, passing 

 through the 109 sieve. The analysis practically corroborates there- 

 fore the method of preparation, as stated by the dealer. Several 

 peculiarities are apparent in examining the result of analysis of this 

 sample. First, there is a high percentage of ash in the sample and 

 also in that part passing through the 109 sieve, showing that the 

 latter is due to the low-grade quality of the flour used. Another 

 peculiarity is that the percentage of nitrogen in the flour passing 

 through the 109 sieve is far higher than that of the bran and shorts. 

 In such case, where the bran and the clear flour are bought separately, 

 it may very easily happen that the bran may have been obtained by 

 the milling of one kind of wheat and a clear flour by the milling of 

 another kind. This sample may therefore be said to be characteristic 

 of an imitation Graham flour as made by the jobber. The gliadin 

 ratio of the coarse middlings is very low, and an examination of this 

 product shows it to contain a large proportion of branny particles. 



No. 9139 is made from soft winter wheat ground on steel rolls, some 

 of the bran being scalped. Both the separation and the analysis of 

 the sample indicate that only a small amount of the bran could have 

 been removed in the process of preparing this product, for we find 

 14.5 per cent of bran, 17.7 per cent of shorts, and the ash content is 



