466 
On the Composition of Oxen, Sheep, and Pigs, 
the estimated composition of the increase from the store to the 
fat condition of the single pig that was put up to fatten when it 
was, as nearly as could be judged, in a parallel state to the one 
analysed as " store," and was itself afterwards analysed as " fat," 
and whose composition, together with that of the store animal, 
provides the data for application to the other cases. 
Noticing first the composition of the increase of the oxen, it is 
probable that the estimate is the most nearly correct for the 
v36 animals that were under experiment for 26f weeks, and 
whose proportion of increase upon 100 original weight was the 
highest. The mean of all the 1)8 animals gives, for the compo- 
sition of the increase, 75*4 per cent, of total dry substance, of 
which 66"2 were fat, 7 '69 dry nitrogenous compounds, and 1"47 
mineral matter. These figures may perhaps be taken as pretty 
nearly representing the average composition of the increase, 
over the concluding period of half a year or more, of animals 
fed on good fattening food, and brought at last to a fair condi- 
tion of maturity and fatness. In passing a judgment as to the 
probable direction of their error, we should say that the fat and 
total dry matter may possibly be stated somewhat too high, and 
the nitrogenous matter somewhat too low. 
An examination of Table XIV., relating to sheep, will show 
that the several " Classes" of animals were fed upon different 
foods, and for different periods of time, and gave accordingly 
different proportions of increase upon 100 original weight, and 
also different proportions of carcass in fasted live-weight. These 
points are of course taken into consideration in adopting, from 
the data relating to the animals actually analysed, a composition 
for the original and final weights respectively of the different 
Classes. 
Thus, in Class I., large numbers of animals were fed for a con- 
siderable period of time from a fair store to a fair fat condition. 
Accordingly the composition of the analysed " store sheep " is 
applied to the original weights, and that of the analysed " fat 
sheep " to the final weights. 
In Class II. the animals were fed from nfat to a very fat con- 
dition — in fact, to the condition of w hat is called " Christmas 
mutton." In these cases, therefore, the original weights are 
calculated at the composition of " fat sheep," and the final 
weights at that of the "extra-fat sheep "' analysed. 
In Classes III. and IV. the animals were taken in a partially 
fattened condition, and fed to that of moderate fatness. In the 
case of Class III. the period of the feeding experiment was com- 
paratively short ; and in that of Class IV. the food was not well 
adapted for fattening. Hence in these cases the proportion of 
increase upon the original weights averaged only about half as 
