568 



These quantities agTee pretty well with those of WolfTs standard. If 

 we allow that one fifth of the nitrogen of the feeding stuflFs is so-called 

 amide nitrogen, the quantities of crude protein corresponding to the 

 1.5-2.4 pounds of actual protein Kiihn calls for would be from 2 to 3 

 pounds. The mean of these — 2.5 pounds — is exactly the quantity in 

 Wolflfs standard. Kiihn's mean for non ])rotoin (taking one part by 

 weight of fats as equivalent to 2.L'.") parts of <arb(»liydrates) would be 14.25 

 pounds — half a pound more than Wolffs. The potential energy of Kiihn's 

 mean ration would be about 31,100 Calories. Of course no one takes 

 such a standard as a fixed measure for the amount of food the average 

 cow ought to have per 1,000 pounds live weight. For that matter, a 

 more logical way to compute tlie ration would be to reckon a certain 

 amount of maintenance food by live weight, and an additional amount 

 for pro<luction by tlie milk yielded or tlie fat to be laid on if the cow is 

 to go to tlie butcher when her milk falls below the point of pi'otitable 

 production. But such standards serve for comitarisims, and it is clear 

 that as compared with them the quantiti«'S in these experiments were 

 very generous. Using the energy as the measure of the t,<3tal digestible 

 nutrients Wolft's and Kiihn's standards call for 30,000 or 31,000 Calories. 

 The quantities here range from 28,000 to 30,000. 



In the second ])la<e the quantities of protein are large — from 2.3 to 

 4.1 pounds per 1,0(10 pounds live weight as c(»mi>ared with the 2.5 

 pounds which the standards cited call for. As a corollary, the rations 

 are narrow, the range in width being from 1 :3.6 to 1 :5.G,that of Wolff"s 

 standard being 1 :."t.4. 



Taking the series one by one there are jteculiarities worthy of spe- 

 cial note. In series C the starting ration. i)eriods I and V, was the 

 smallest of all, reckoned i>er 1.000 ]»ounds Vwr w«'ight. But the cows 

 were large, so that reckoned jxi- head the food was lilieral. The yield 

 of milk was very large. Increasing the i>rotein from 2.3 to 3.2 ptmnds 

 l)er 1,000 pounds live weight, increasing the energy from 27,800 to 

 30.000 Calories, and diminishing tlie width from 1:4.0 to 1:30 did not 

 materially aflect the milk yield. This, as Maercker and ]\Iorgen say 

 in their report, was not surprising, since the cows were ajiiiarently near 

 the maximum of their capacity for milk i)roducfion at the start. The 

 slight increase in the estimated i»ecuniaiy gain was due to tlu' ineieased 

 value of the manure. 



Series D tells a sonx'what different story. The quantifies of nutri- 

 ents per 1,000 ])oinids live weight were larger than in series C, luit the 

 cows were smaller an<l the ([uantitics per head were nearly the same; 

 the milk yield was only a little over half as large. There was more 

 gain in live weight. The pecuniary ])roflt was smaller with the smaller 

 milk yield. The milk increased distinctly with the increase of ]»rotein 

 and the consequent increase of total nutrients. Tlie pecuniary gain 

 increased also, although it is explained as before by the value of the 

 nitrogen and i)liosphoric acid in the juanure. 



