EXAMINATION OF GRAHAM FLOUR. 15 



ance of the shorts and the middlings will show the general character 

 of the products used. 



Some of the imitation Graham flours found on the market are 

 made from products of the mill which are valuable as flour-producing 

 stock. Such is the case when purified middlings are used, and if 

 these represent any large proportion of the total amount of the imita- 

 tion Graham, it is quite evident that no attempt is made to dispose 

 of inferior products, since it is more expensive for the miller to grind 

 the wheat and make the separations into offal, middlings, and flour 

 and then mix the resulting products than it would be to grind the 

 wheat on one or two sets of rolls without bolting. 



When purified middlings are used in large amounts it is difficult 

 to detect mixtures. Examples of such flours will be indicated in 

 the general discussion of the results (see p. 39). 



CHEMICAL METHODS. 



Besides making these mechanical separations, the several por- 

 tions were analyzed for total nitrogen and alcohol-soluble nitrogen 

 from which results the gliadin ratio was calculated by dividing the 

 alcohol-soluble nitrogen by the total nitrogen. In many cases, also, 

 the ash, fiber, and pentosans were determined hi the original sam- 

 ple of Graham flour and in that portion passing through the 109 sieve. 



The analysis of each of the samples is given in Tables 1 to 9. In 

 Tables 10, 12, 14, 16, 18, and 20 are found the relative nitrogen con- 

 tent in the flour examined as compared with that in each product of 

 separation and the same relation between the products of separation 

 with one another. In Tables 11, 13, 15, 17, 19, and 21 are found the 

 relative gliadin numbers in the flours examined as compared with 

 those of the products of separation and the same relation between 

 the products of separation with one another. 



GENERAL DESCRIPTION OF SAMPLES. 



Of the Graham flour samples analyzed, 12 were purchased on the 

 market and designated as true Graham flours. The data obtained 

 are given in Table 2 (p. 17). Seven samples, the analytical data on 

 which are found in Table 3 (p. 17), were also collected by the same 

 representatives of the bureau and are described as imitation Grahams. 

 In Tables 4 and 5 (pp. 17 and 18) will be found the analyses of 12 

 samples, 6 of them being of the nature of pure Graham and 6 imita- 

 tion Graham. 



Thirteen samples were obtained under observation from the mills 

 and are true to name. The analyses of these will be found in Table 

 6 (p. 18). Thirteen samples were obtained from "stock" and desig- 

 nated by the millers "true Grahams." Their analyses are to be 

 found in Table 7 (p. 19). 



67800 Bull. 16413 3 



