192 
Experiments on the Feeding of Sheep. 
tion of fat? Or whether, on the other hand, it serves little other 
purpose than that of supplying bulk, and dilution, so to speak, 
of the other constituents of the food — thus aiding their digestion 
and assimilation, and then passing off, itself undigested and un- 
changed ? 
It is obviously necessary for the elucidation of the points in- 
volved in these questions to determine, as far as chemistry 
enables us to do so, not only the amount of cellulose consumed 
in the food, but also the quantity voided in the excrements. 
Hence, although a large portion of the analyses have already 
been made, the consideration of them will be reserved until we 
enter upon the general subject of the composition of the manure. 
In order, however, to relieve from extraneous matter as far as 
possible the subsequent report on the composition of the excre- 
ments of fattening animals, which will of itself involve the 
record of a vast amount of detail, it is proposed to give, on the 
present occasion, the results of the experiments referred to so far 
only as they relate to the amount of food consumed and of increase 
yielded. 
With a view to the special objects above stated it was neces- 
sary to employ foods in which the proportion of woody fibre, and 
of the other non-nitrogenous constituents, would be pretty constant, 
and be comparatively easily determined. It was further desirable 
that, in some cases at least, the animals should have a somewhat 
excessive proportion of woody fibre in their food ; that in others 
the proportion of the more easily digestible non-nitrogenous sub- 
stances (starch, fatty matter, &c.), should be more liberal ; and 
that the amount and character of these other non-nitrogenous 
constituents should vary in the different experiments. It was 
hence thought undesirable, at any rate in the first experiments 
on the point, to employ roots or other succulent food, the com- 
position of which would not only be more subject to change 
during the course of the experiment, but would be more difficult 
and uncertain of determination in a large bulk, even at any one 
given time. Various so-called " dry foods " only, and water, 
were therefore selected ; and although, in some cases, these were, 
as such, of good quality, the result was, as might be expected, 
that the rate of increase was comparatively small in the absence 
of a certain proportion of the more natural succulent food. 
A number of 3-year-old Hampshire Down wether sheep, in 
very poor condition, had some time previously been purchased 
for the purposes of experiment. From these, 20 were selected, 
and divided into 4 lots of 5 each, in such manner that, as far as 
possible, each sheep should have its representative in weight and 
other characters in each of the other j)ens. They were put up 
on rafters, imder cover, on November 30, 1860. 
As the staple of the food throughout the experiments was to be 
