54 GRAHAM FLOUR. 



purified middlings and bran, it is evident that there is no desire to 

 sell to the consumer a cheap material since the miller believes that 

 the flour thus prepared is a better grade and that better results in 

 baking may be obtained than with the true Graham. 



The methods of milling Graham flour are varied. Stone burrs, 

 steel corrugated rolls, attrition mills, and French burrs are used, 

 and usually when the Graham flour was made on steel rolls, anywhere 

 from two to six sets of corrugated rolls were employed, and in some 

 instances, after the wheat had been run through five sets of corru- 

 gated rolls, it was then run through six sets of smooth rolls. This is 

 not at all necessary, for good Graham flour can be made on two sets 

 of corrugated rolls or on one set of stones. 



WHEAT MEALS, OB IMPROPERLY CALLED WHOLE-WHEAT FLOURS. 



In Table 23 are grouped all the samples which resemble bolted 

 wheat meal. Nos. 9133 and 9229 were sold as such, while Nos. 9157, 

 9163, and 9211 were sold as Grahams. The other samples in the 

 table were stated by the millers to be imitation Grahams. These 

 samples have already been described with the exception of Nos. 9133 

 and 9229 and the analyses placed in Tables 7-9. 



They are grouped here for convenience, in order to compare them 

 with the two samples of improperly called whole- wheat flour, Nos. 

 9133 and 9229, which are characterized by low bran and shorts 

 contents. Both have, however, considerable amounts of middlings. 

 The ash, fiber, and pentosan contents are low. It is evident that they 

 contain only a very small amount of bran material because the ash is 

 so low and also because the ash of the flour passing through the 109 

 sieve is but a trifle lower than the ash of the original " wheat meal." 

 The ash of all the other samples varies from 0.79 to 1.36, indicating 

 that the millers make " wheat meal" in a variety of ways. Some 

 millers may make it by removing practically all of the bran; others 

 may remove only the coarse particles, as what remains on the 14 or 

 20 mesh sieve. The fiber and pentosan contents of the two samples 

 of " wheat meal" are also very low, showing by another means that 

 there is an almost total absence of bran in these samples. The 

 gliadin ratios of the coarse middlings of all the samples, except two, 

 are high. These two (Nos. 7354 and 7746) were made by removing 

 the coarse bran and adding more or less germ bran. 



It is evident from the foregoing that it is often necessary not only 

 to make a mechanical separation of the flour and a macroscopic exam- 

 ination of the products, but to determine also the ash, fiber, pentosans, 

 and the nitrogen and gliadin ratios of the products of separation. 

 To know from what kind of wheat the product was produced and 

 whether milled between rolls or burrs tones, helps many times to 

 determine with a greater degree of accuracy whether the flour in 

 question is a genuine product or an imitation Graham. 



