32 BULLETIN OF THE BUSSEY INSTITUTION. 
been smaller than in the other cases, but in every instance some ash 
seems to have been counted as cellulose. Now, the average amount 
of ash in the crude cellulose of American shorts was found to be equal 
to 0.33. per cent of the weight of the original bran; and if we 
deduct this quantity from the mean (9.00) of the fourteen foreign 
determinations of cellulose (excluding No. [V., which is exceptionally 
high), we obtain the number 8.67, which agrees very well with that 
(8.29) found for the American shorts. 
As regards ‘“ albuminoids,” it is not easy to obtain a mean which 
shall accord closely with all the analyses. It would almost seem as 
if there were two classes of bran, distinguishable by the different . 
amounts of albuminoids contained in them. Thus, the analyses II. 
to VII. show from twelve and a half to nearly fifteen per cent of albu- 
minoids, in the mean 13.44, while the analyses VIII. to XII. show 
from seventeen and one third to nearly twenty-one per cent. The 
number 13.44, which accords well enough with the mean percentage 
of albuminoids found in the American shorts, undoubtedly represents 
very nearly the average proportion of these substances contained in > 
commercial bran. I believe the higher numbers to be either excep- 
tional or unreliable. Since American bran is probably, as a general 
rule, less thoroughly freed: from flour than European bran, it is not 
surprising that it should contain a somewhat smaller proportion of 
albuminoids. | 
Some explanation of Wicke’s analysis may perhaps have been given 
in his original paper, but I am unable to refer to the journal which 
contains it. The analyses of Johnston are old, and although his de- 
terminations of water and fat are undoubtedly very near the truth, it 
is not at all improbable that those of albuminoids and ash were made 
by untrustworthy processes. The analysis of Dumas and Boussin- 
gault is likewise old, and it is not unlikely that the bran examined 
by these’ chemists was of exceptional character, as the large propor- 
tion of fat found ‘in it would tend to show. It may be, too, that their 
process of analysis was faulty. 
The determinations of Frapolli and of v. Bibra seem, on the other 
hand, to have been made with care and by approved processes. It is 
probable, however, that their brans were really of exceptional charac- 
ter, since it is to be inferred from their statements that the materials 
analyzed were prepared expressly for that purpose. It would not be 
