Colorado Foddkrs. 
79 
in feeding them, were wholly unexpected,in that the former gave 
better results than was anticipated and the latter very unfavorable 
ones, which was contrary to what I had expected. Even now, I 
am scarcely content to accept the results of our feeding experiment. 
As will be seen subsequently, there is nothing in the composition, 
judged either by the regulation fodder analysis, or by what I have 
called the proximate analysis, or in its fuel values to suggest an 
inferior value for this sorghum fodder, but each of the three sheep 
fed on this fodder lost very nearly three pounds in five days during 
which time they not only consumed but digested a very fair 
amount of food, three pounds per hundred weight of animal daily. 
§i86. Our object was primarily to study the fodders to find 
out how we might learn more about the reasons why fodders are 
so different in value. The groups into which we have been accus¬ 
tomed to divide fodders are not definite groups, but are the best 
that we have had and they have not yet been displaced. Students 
have long felt that the results left much to be desired and have 
adopted other methods of investigation, in order to find out the 
value of the fodders, their composition, the heat values of the com¬ 
ponents, and the energy actually furnished to the animal. 
§187. It occurred to me that an endeavor to study the car¬ 
bohydrates might add something to our present knowledge and I 
chose the line of work presented in Bulletin No. 39, but I realized 
that it was only an attempt and left much that was unsatisfactory. 
Some of the results were not concordant and the work was not car¬ 
ried far enough and was not accompanied by any results obtained 
by these lines of examination in combination with feeding tests, 
the results of which should serve as facts on which to base our in¬ 
terpretations. I have tried to obviate these faults in the investiga¬ 
tion now presented. 
§188. I have adhered to the general line of experimentation 
outlined in Bulletin No. 39, so that the present work may properly 
be considered a continuation and extension of that. The general 
method of examination was to exhaust the fodders with boiling 80 
per cent, alcohol, cold water, boiling water and after cooling with 
malt extract, boiling i per cent, hydric chlorid, boiling i per cent, 
sodic hydrate, and lastly, with chlorin, subsequently washing with 
I per cent, sodic hydrate and hydric sulfite. I have not yet seen 
any good reason for changing this order of procedure. The results 
show, however, that it does not effect a division of the fodder into 
any definite groups. By this I mean that the extracts consist of 
mixtures of substance whose composition cannot be established in 
this way. I hoped, for instance, to be able to determine the sucrose 
in the alcoholic extract, but I find from the determinations of the 
furfurol removed from the fodder, that it is quite probable that the 
