I70 STORRS AGRICULTURAL EXPERIMENT STATION. 
Long keeping reduces digestibility and is attended by loss of 
actual dry matter. | 
The chief thing to be observed in the use of concentrated feeds 
is to keep the ration narrow and not to supply too large a quantity 
of fat. Practically it is probably unwise to feed a ration whose 
nutritive ratio is greater than that of one to seven, or at the most 
one to eight. Some of the more recent experiments with milch 
cows seem to indicate that even a narrower ration than the 
ordinary standard, 1 to 5.4, can be used with profit. 
Reference has been made (p. 165) to the fact that there is mere 
or less uncertainty attending the use of coefficients of digestibil- 
ity, but as a good guess is a better guide than blind ignorance, so 
digestion factors help us to a much better understanding of the 
value of a food than is possible without an attempt to take diges- 
tibility into account. Because our facts are not sufficient to 
warrant absolutely definite conclusions, it does not follow that 
we should not live up to all the light we have. The coefficients 
of digestibility suggested in table 51, in which the average results 
obtained in European and American digestion experiments are 
compared, are perhaps as good as any we have as yet. Twenty 
years ago we had almost no knowledge of the composition of 
American feeding stuffs. The number of analyses were so few 
that no attempt was made to compile them for practical use. 
In the past twenty years nearly 5,000 specimens of American 
feeding stuffs have been analyzed, and we know within quite 
definite limits what the composition of any given feeding stuff is. 
It is to be hoped and expected that the next two decades will see 
such additions made to our knowledge of digestibility that we 
may then be able to speak as confidently of the coefficients of 
digestibility as we now do of percentages of total nutrients. 
In table 52 the attempt is made to calculate the weights of the 
different digestible nutrients and the calories of potential energy 
which would be supplied by a pound of each of the different 
kinds of feed most used in Connecticut. The figures here 
given were obtained by multiplying the averages of the analyses 
of different feeding stuffs given in table 48, p. 144, by the factors 
(digestion coefficients) given in table 51, p. 166. 

