Purchased Food, and of its Residue as Manure. 213 
onl}' from 54 to 60 per cent, of dry feeding-matter, consisting 
mainly of sugar. 
At present locust-beans can be bought at about 71. 10s. per ton, 
whilst treacle or molasses of good quality costs about 9/. per ton. 
Locust-beans are thus not only much cheaper than molasses, 
weight for weight, but they likewise possess a higher nutritive 
value, and are equally well adapted to the sweetening of unpalat- 
able bulky food. 
Nitrogenous Food-constituents. — In the next place we have to 
consider the nutritive value of the albuminous or nitrogenous 
constituents of food. It is admitted on all hands that a certain 
amount and proportion of nitrogenous matter is essential in the 
food of all animals. Foods, like locust-beans, or rice-meal, 
or dari-grain (a species of sorghum), which contain less than 
8 or 9 per cent, of albuminoids, are too poor in nitrogenous 
substances to suit the requirements of the animal. Hence 
these and a few other feeding materials equally poor in nitrogen 
should not be given to fattening stock in too large proportions, 
or without the addition of other meals, or of oilcakes, richer in 
nitrogenous compounds. 
In wheat, oats, and barley, however, the proportion of albumi- 
nous substances is sufficiently high to meet the requirements 
of fattening stock ; and in leguminous seeds (such as beans, 
lentils, and peas), and in oilcakes, the proportions of these 
compounds are considerably in excess of the requirements of the 
animal. 
According to the views of not a few writers on agricultural 
chemistry and physiology, it is chiefly the proportion of the 
nitrogenous, or so-called flesh-forming substances contained in 
different kinds of food, which determines their comparative 
value for feeding purposes. 
If I am not mistaken, it was Boussingault who made the first 
attempt to construct a theoretical table of the nutritive value of 
articles of food, based upon the amount of nitrogen they contain ; 
but it is due to this most careful observer to mention, that in 
testing the correctness of his own tables by actual feeding experi- 
ments, Boussingault frequently found the results of the ex- 
periments at variance with the theoretical indications of his 
tables ; and he frankly confessed that the amount of nitrogen in 
a feeding substance must be regarded as one factor only in 
estimating its nutritive value. 
Presuming that the proportion of nitrogenous substances in 
the food given to fattening-stock is about the same as that in which 
we find them to exist in cereal grains, it may be asked, what will 
be the effect upon the animal when it receives in addition 
feeding materials rich in nitrogen ; or, on the other hand, when 
