6 
4 DIGESTIBILITY OF FISH AND POULTRY. Tat 
It will be noted that the separate experiments of each series 
agree on the whole very closely. The greatest variations are 
found in the series with chicken, in which the extremes were 
98.18 and 93.15 per cent. for protein, and 86.98 and 82.11 per 
cent. for energy. In most of the other instances, the agree- 
ment was within 1 to 2 per cent. 
As regards the different series, the variations of the coeff- 
cients for fat are remarkably small, the extreme difference in the 
average being but.39 percent. Itis to be questioned whether 
such close agreement was not more or less accidental. Appar- 
ently, however, the presence of large amounts of fat did not 
appreciably affect its digestibility. 
The average availability of the protein of the different foods 
varied less than that of the same protein by the four subjects. 
For the cod and salmon the coefficient was about 96 per cent., 
for chicken a little higher, and for duck somewhat lower than 
this figure. ‘ 
The greatest variation will be noted as regards energy. For 
salmon and chicken, the coefficients are about 85 per cent., for 
cod 80 per cent. and for duck 91 per cent. Reference to the 
composition table shows that the heat of combustion and fat 
content were highest in the duck sample, about equal for the 
salmon and chicken, and least in the case of the cod. This 
may be an indication that the availability of the energy is 
highest in foods containing a large amount of fat, though the 
data at hand are not sufficient to warrant too sweeping deduc- 
tions. 
A comparison of the coefficients of availability of the poultry 
and fish with those of other common foods is given in Table 19, 
which follows. 
