648 



rations for each series were so adjusted as to contain as nearly as prac- 

 ticable tlie same quantity of digestible nutrients in the different periods, 

 so that the only difference should be in the amount of water in the dif- 

 fusion residue. The feeding stuffs were analyzed and the quantities of 

 digestible nutrients determined or estimated by the station. Before the 

 beginning of the feeding tests each experimenter set aside a sufficient 

 quantity of coarse food — hay, straw, chaff, etc. — and samples were taken 

 for analysis. A supply of each of the concentrated foods — oil cake, meal, 

 etc. — sufficient for all the experiments, was obtained trom a single source, 

 a sami)le analyzed, and portions distributed among exi)crinienters. A 

 sample of the dry residue used by each experimenter was analyzed. 

 As the wet residue varied in composition special means were adopted 

 for frequent analyses of the material used in each case. This caused 

 some inconvenience and a little disturbance of the unity of the trials on 

 account of the difficulty in determining the composition before the 

 material was fed. 



The quantities of the materials fed were calculated from the percent- 

 ages of digestible nutrients. The mctliods of estimating the rations 

 and the financial results are given in detail in the report. They will 

 be found decidedly helpful in the planning of similar experiments else- 

 where. The estimates of cost of feeding stuffs and value of products 

 are based upon figures similar to those above qiu>tt'd for the experi 

 ments of series A-N. The details of the composition of the feeding 

 stuffs and of the results of the experiments are given in tabular form. 

 The results of the experinumt^s are summarized in tabular statements 

 which are given below. 



Experiments with milch cows. — From the figures in the tables it will 

 be seen that the rations in series O were liberal, and although the quan- 

 tities of nutrients in series P are less, they exceed those of Woltfs 

 standards, except in period III, in which, o^ving to variations in the 

 composition of the wet residue, they could not be determined in season 

 to make tlie proper allowance in the making up of the ration. On this 

 account the results for period III can iu)t be averaged with those of 

 period I to make up for the change in the composition of the milk with the 

 advance of the period of lactation. Of these two series the authors say: 



'' Unfortunat«»ly only <uie of the exi»erinients is available. Further- 

 more the duration was too short for satisfactory conclusions as to 

 increase in live weight. For estimating the effects upon milk produc 

 tion, likewise, the periods of 10 days each are far too short. A more 

 reliable estimate could be made from trials extending through longer 

 periods. The best plan, if practicable, would be to divide the herd in 

 the stable into two lots and feed one lot with wet and the other with 

 the dry residue."' Xevertheless the conclusions seem warranted that 

 in series O "the feeding with dry residue exercised an extraiudiuarily 

 favorable influence upon the bodily condition of the cows," and that in 

 series B the effect was hkewise very favorable. 



