EXAMINATION OF GRAHAM FLOUB. 39 



No. 9223 is made on burrs tones. The product is normal in every 

 respect. 



In only one of these samples does the percentage of nitrogen in 

 that product passing the 109 sieve even approximate the percentage 

 of nitrogen found in the bran and shorts of the same sample. This 

 is No. 9163, which, as already stated, is a sample of wheat meal and 

 not Graham. 



The bran content varied from 2.0 to 13.7, the shorts from 6.5 to 24.0, 

 the combined bran and shorts from 8.8 to 34.2, the coarse middlings 

 from 5.0 to 19.7, the fine middlings from 7.4 to 20.8, the combined 

 middlings from 14.-1 to 36.5, the flour from 29.0 to 71.7. Thus it is noted 

 that the samples milled under observation proved more regular than 

 those from stock. The variations in the ash, fiber, and pentosan 

 contents are far greater also among these samples than in the samples 

 of true Graham flour obtained under observation. This is no doubt 

 due to the fact that among the samples obtained from stock there are 

 two which are ordinary wheat meal, but which were designated as 

 Graham flour by the miller. If these two samples be eliminated, 

 the variations are not greater than were found in the case of Table 

 6 containing Graham flour samples. 



IMITATION GRAHAM FLOUR SAMPLES. 



Samples oltained under observation (Table 8). Table 8 contains 

 the results of analysis of imitation Graham flours obtained under 

 observation. The method of the manufacture of these samples and 

 a discussion of the results follow: 



No. 7721 is supposed to have been made by mixing 50 per cent of 

 purified middlings, 48 to 49 per cent of clear flour, and 1 to 2 per cent 

 of bran. The middlings are the stock used in producing patent flour. 

 They passed a 40-mesh cloth, but remained on No. 70. The flour is 

 the second clear from the breaks. The bran is that obtained from the 

 fourth break. The miller claimed that never more than 3 per cent 

 of bran is used, and they only mix in that much when customers want 

 very coarse Graham; but inasmuch as no measuring or weighing of 

 these products was made it is difficult to tell just how the miller 

 knows how much of each of these products he is using. The separa- 

 tion of the sample shows that the miller was guessing as to the 

 amounts used. In this sample the percentage of bran and shorts 

 combined amount to 11.2, an abnormally low amount to be found in 

 a Graham flour, but a much higher amount than the miller claimed to 

 have used. The percentage of ash is also much lower than normal, 

 being 1.2. These two facts go to show that a considerable amount 

 of bran must have been removed from the wheat in the production 

 of this sample, which is in agreement with the miller's statements." 



