569 



On the whole, both series C and D show a profit from highly nitro- 

 genous rations and liberal feeding. 



With the succeeding series, E, F, and G, in which the rations were 

 equally liberal but had less protein, the results are less favorable in 

 respect to milk j)roduction, change of live weight, and pecuniary gain or 

 loss, and this is true notwithstanding the narrow rations of these series 

 are much narrower and the wider rations but little wider than Wolffs 

 and Kiihn's standards. 



In series E and F the rations per head are nearly the same, but as 

 the cows in F were larger the ])roportions ])ev 1,000 pounds are smaller, 

 though decidedly liberal as compared with the feeding standards. In 

 each of these the starting ration was moderate in total amount and 

 rather narrow, and when the carbohydrates were increased without 

 increasing the i)rotein the milk yield remained very nearly the same, but 

 the live weight fell off. The pecuniary results were decidedly unfavor- 

 able to the use of the non-nitrogenous materials. 



With series G the case is different. In live weight of cows and 

 quantities of nutrients it corresponds very closely with E. But whereas 

 in E, as in F, there was a gain in live weight with the starting ration 

 and with the succeeding rations, which contained more non-protein, 

 there Avas little change in milk production and a falling off in live 

 weight in G. The live weight decreased during the feeding of the first 

 ration, and when carbohydrates were added in the second and third the 

 milk yield and live weight were both increased. The pecuniary result 

 in series G agrees with that of E and F in showing a loss with increase 

 of carbohydrates. 



A very probable explanation of the difference in outcome of this last 

 series as compared with the others is found by the authors in the bodily 

 condition of the cows, which they think were somewhat low in flesh at 

 the beginning, so that the carbohydrates in proportions such as not to 

 make the whole ration wider than in this case could make themselves 

 felt in increase of both flesh and milk. According to the data of the 

 exi^eriment, the cows lost half a pound per day with the starting ration, 

 which agreed very well with the standards of Wolff and Kiihn in total 

 quantity of nutrients, though it was rather narrower, i. e. richer in 

 protein. When non-protein was added, so as to make the ration about 

 equivalent to these standards in width and equal to Kiihn's most liberal 

 allowance and much larger than Wolff's standard in amount, both the 

 flesh and the milk increased. The explanation given by the authors 

 certainly accords with the statistics of the experiment. 



For those who are interested in feeding experiments there is a les- 

 son of decided value in series E, F, and G. These were made to test 

 the same question — the effect of increase of nitrogenous foods. They 

 were all conducted on the same plan. Series E with eight cows and 

 five feeding periods gave very definite results; there was a gain of 

 live weight with the wider rations and loss with the narrower rations, 



