1891.] PUBLIC DOCUMENT — No. 33. 93 



with a periodical increase of corn meal, beginning with two 

 ounces of corn meal per quart of milk, has given us highly 

 satisfactory results. 



3. Change the character of the diet, at certain stages of 

 growth, from a rich nitrogenous diet to that of a wider ratio 

 between the digestible nitrogenous and non-nitrogenous food 

 constituents of the feed. Begin, for instance, with two 

 ounces of corn meal to one quart of skim-milk ; when the 

 animal has reached from sixty to seventy pounds, use four 

 ounces per quart ; and feed six ounces of meal per quart 

 after its live weight amounts to from one hundred and 

 twenty to one hundred and thirty pounds. The superior 

 feeding effect, noticed in case of one and the same diet dur- 

 ing the earlier stages of growth, will not infrequently be 

 found to decrease seriously during later stages. 



4. It is not good economy to raise pigs for the meat 

 market to an exceptionally high weight. To go beyond from 

 one hundred and seventy-five to one hundred and eighty 

 pounds is only advisable when exceptionally high market 

 prices for dressed pork can be secured. The quality of the 

 meat is also apt to be impaired by an increased deposition of 

 fat. The power of assimilating food, and of converting it in 

 an economical way into an increase of live weight, decreases 

 with the progress of age. 



5. It pays well, as far as the cost of feed is concerned, 

 to protect the animals against the extremes of the season. 

 Feeding experiments carried on during moderate seasons 

 are more profitable than those carried on, under otherwise 

 corresponding circumstances, during the winter season. 



Our experiments previous to 1890 have been carried on 

 with mixed breeds of a more or less doubtful parentage. 

 Within the past year we began to compare thoroughbreds of 

 distinct breeds with each other or with grades of known 

 origin. The subsequent pages contain a description of two 

 experiments, — one with medium Yorkshires, the other with 

 medium Yorkshires and grade Chester Whites. 



Another experiment with throughbreds — Berkshire, Tam- 

 worth, small Yorkshire and Poland China — is finished; the 

 results will be published in connection with a second one of 

 the same character as soon as the latter is closed. 



