64 FORAGE PLANTS AND THEIR CULTURE 



accurate for other kinds of animals and other purposes of feeding. 

 Nevertheless, there seems to be reason for believing that they also 

 represent fairly well the relative values of feeding stuffs for sheep 

 at least, and probably for horses, and for growth and milk pro- 

 duction as well as for fattening. At any rate, there can be little 

 doubt that they are decidedly more accurate than the figures 

 which have been commonly used, and we are quite justified in 

 using them tentatively and subject to correction by the results 

 of later experiments. 



"As regards swine, the matter is far less certain, and it may 

 perhaps be questioned whether the values given in the table are 

 any more satisfactory for this animal than the older ones." 



54. Starch values. The unit of this system proposed 

 by Kellner is one pound of digestible starch for beef pro- 

 duction. Kellner found that one pound of digestible 

 starch in excess to a maintenance ration would form ap- 

 proximately one-fourth pound of fat. On this basis 

 1 pound of digestible protein is equal to .94 pound of 

 digestible starch, and 1 pound of oil in seeds equals 2.41 

 pounds of digestible starch in fattening value. 



These values are in excess of what the animal actually 

 gets from the feed, so that arbitrary deductions have to be 

 made to compensate for the work of mastication and 

 digestion. 



55. Comparison of feeding values. Woll has en- 

 deavored to reduce to a common basis the relative values 

 of various feeds as determined by the feed unit system, by 

 Kellner's starch values and by Armsby's therms. The 

 average of the net energy values of corn, wheat, rye, barley 

 and wheat middlings is approximately 80 therms, which is 

 considered equal to 1 feed unit. By the same method 

 83 starch values is equivalent to 1 feed unit. In general, 

 the corresponding values by the three methods are close, 

 but there are some marked exceptions : 



