EXAMINATION OF GRAHAM FLOUR. 35 



nitrogen content of the products. In these tables, and likewise in 

 all subsequent ones, the separations of the samples represented 

 therein were made on four sets of sieves, that portion remaining on 

 No. 20 being called the bran, on No. 40 shorts, on No. 70 coarse 

 middlings, on No. 109 fine middlings, and through No. 109 flour. 



Samples obtained under observation (Table 6). No. 7714 is made 

 on a burrstone mill. 



No. 7728 is ground on three sets of corrugated rolls without 

 bolting, the wheat having been moistened with 1J pints of water to 

 every 4 bushels of wheat, in order to make the bran more flaky. 



No. 7783 is made from soft winter wheat without moistening and 

 ground on a French burr mill. This sample contains a very small 

 amount of bran and likewise a very smah 1 amount of coarse middlings. 

 The gliadin ratio of the coarse middlings is the lowest of all the 

 samples in this table. The macroscopic examination of the coarse 

 middlings showed that the product of separation contained a large 

 amount of bran, which would of course increase the nitrogen content 

 and therefore lower the gliadin ratio. The use of a burr mill has a 

 tendency to give a more finely ground product, which would explain 

 the presence of bran (finely ground) in the middlings and also the 

 large amount of flour passing the 109 sieve. 



No. 7837 is made from hard spring wheat and contains a very 

 small amount of bran. The gliadin ratio of the fine middlings is lower 

 than that of any of the other samples of the Graham flour, and the 

 gliadin ratio of the flour passing through the 109 sieve is the lowest 

 but one. The amount of material which passes through this sieve 

 is likewise low, the lowest found among the samples recorded in the 

 table, indicating that a large proportion of it may be of the nature 

 of a break flour, due to the fact that the wheat used was a hard spring 

 wheat. The percentage of nitrogen in the flour is higher than the 

 percentage of nitrogen in the middlings. The ash content of this 

 product is much higher than that of any of the other samples, which 

 fact, taken in connection with the high nitrogen content and with 

 the low amount of material passing through the 109 sieve, goes to 

 show that this product was to a large extent a break flour and is what 

 one might expect when a hard spring wheat is made by the ordinary 

 process into Graham flour. 



Nos. 9049, 9055, and 9103 are from the same miller and all from 

 hard spring wheat. The first was ground as for the ordinary trade, 

 the second medium ground, and the third as finely ground as possible 

 without injuring the product. No. 9103 contains a very small amount 

 of bran and of coarse middlings and over 60 per cent of material 

 passed through the 109 sieve. The gliadin ratio of the coarse mid- 

 dlings of this sample is low, due to the fact that a considerable portion 



