EXAMINATION OF GRAHAM FLOUB. 47 



product. The chemical analysis of the products of separation would 

 therefore quite properly place the sample among the imitation 

 Graham flours. 



Regarding No. 9247 no information is at hand. From the mechan- 

 ical separation alone this sample might also be a Graham flour. A 

 very high nitrogen content of the bran and of the shorts and the 

 correspondingly low gliadin ratios of these products would seem to 

 indicate that this sample contained a certain amount of added germ. 

 The low gliadin ratio of the coarse middlings leads to the same 

 conclusion. 



The samples of imitation Graham flour found in stock, but whose 

 method of manufacture was not observed, show greater differences 

 than have been noted in either the Graham flours or even in the 

 imitation Graham flours collected under observation. The bran 

 varied from to 14.5, the shorts from 6.6 to 37.2, the coarse middlings 

 from 0.0 to 24.0, the fine middlings from to 27.2, and the flour 

 from 22.9 to 83.4 per cent, the combined bran and shorts from 

 10 to 39.9, the combined middlings from 0.0 to 39.5 per cent. These 

 samples also show a greater irregularity in chemical composition 

 tha do even the imitation Graham samples collected under observa- 

 tion. The nitrogen content of the original sample varies from 2.22 

 to 4.13, the high percentage being due to the presence of germ in the 

 bran. This accounts for the varying and low gliadin ratio found in 

 these samples. The same applies to the shorts. The ash content of the 

 product passing through the 109 sieve likewise varies greatly namely, 

 from 0.39 to 1.95 per cent, the use of low-grade flour tending to give 

 a high ash content. Where a high ash was found in this product it 

 was thought that it might be due to the fact that the wheat was 

 milled on burrstones and that a part of the ash was in reality insolu- 

 ble silica due to the wearing away of the stones, but analyses of the 

 ash of these samples for insoluble material or silica showed that only 

 a trace (0.03 to 0.07 per cent) of this substance was present. Thus 

 the only explanation left is that these particular alleged Graham 

 flours having a high ash content in that portion passing through the 

 109 sieve were made by the use of very poor flour material, a grade 

 even lower than a clear being sometimes used. This is especially 

 true of No. 9109 and may be also true of No. 9241. 



