COMPAKISON OF SAMPLES. 49 



The analytical figures obtained for the alleged Graham flour are 

 generally a mean between those obtained for the Graham and the 

 imitation Graham. Of the 44 samples of Graham or alleged Graham 

 flour only 4 yielded 60 per cent or over of material passing the 109 

 sieve, the average of all being 40.8. Two of these 4 were, however, 

 shown to be so-called whole- wheat flours and not Grahams. The 

 other 2 samples (Nos. 9103 and 9145) were obtained under observa- 

 tion, but were ground as finely as possible on burrs tones. Of the 

 39 samples of imitation Grahams, 27 gave 60 per cent or over of 

 material fine enough to pass through a 109 sieve, the variation being 

 61.7 to 86.3, while the average of all the 39 samples was 64.6 per 

 cent. Only 1 out of the 44 samples of Graham gave as low a per- 

 centage of middlings as 20, this being No. 9145, which has already 

 been referred to as having been ground very fine and as being 

 high in material passing the 109 sieve. Of the 39 samples of imita- 

 tion Graham, 25 gave less than 20 per cent middlings. It is almost 

 conclusive proof that the sample is an imitation Graham when it 

 contains only a small amount of middlings. Of the 13 samples of 

 Graham collected under observation, 4 contained less than 20 per 

 cent combined bran and shorts, while 25 out of the 39 samples of 

 imitation Graham showed less than 20 per cent of these materials. 



NITROGEN CONTENT AND GLIADIN NUMBER. 



Tables 14 to 21 give the ratio percentage between the nitrogen 

 content and gliadin number, respectively, of the Graham flour and 

 those of the products of separation and also between those of the 

 products of separation as compared with one another. Discussion 

 of similar figures in Tables 10 to 12 has already been given, and in the 

 main the points brought out in Tables 14 to 21 agree with those given 

 in the former tables. 



COMPARISON OF SAMPLES FROM THE SAME MILLERS AT DIF- 

 FERENT TIMES. 



Nos. 9047 and 9048 are samples of wheat and Graham flour, respec- 

 tively, the latter supposed to have been made from the former. 

 Both were obtained from a jobber. The wheat is of a hard spring 

 variety. The flour was so finely ground that it seemed .improbable 

 that it could have been produced from such wheat by grinding it on 

 rolls and without separation of the products. The following samples 

 of Graham flour, milled from the same hard spring wheat under 

 observation, were all obtained from a miller who supplied the jobber 

 with the Graham -flour No. 9048: 



No. 9049 was ground as for the ordinary trade. 



No. 9055 was ground as per orders from the jobber just mentioned. 



