57 



later by dry cows, the milk cows meantime receiving 3 

 pounds of cotton seed meal per day. 



On this pasturage the yield of milk was 15.8 per cent 

 greater and of butter 9.5 per cent greater than when 

 the cows with the same grain feed ran on a good pasture 

 of bermuda grass, carpet grass, lespedeza, etc. 



The value of the product of butter and of the increase 

 in live weight of the cows averaged |4.47 per acre of 

 corn and pea field grazed, after deducting the cost of the 

 cotton seed meal fed at the same time. 



Introductory. 



Under some conditions it is practicable for the dairy- 

 man to purchase a considerable proportion of the food 

 which his cows consume. However, the temptation is to 

 rely to too great an extent on purchased foods. These 

 can be profitably used to a certain extent but rather as 

 supplements to foods produced on the farm than as sub- 

 stitutes for farm-grown food. It is believed that any 

 marked development of dairying and of beef production 

 in the South is conditioned on the increased reliance on 

 the foods Avhich the fields and pastures produce. The 

 feeder who buys thin cattle at a low price and, after a 

 few months feeding, sells them at a higher price per 

 pound, relies almost wholly upon cotton seed meal and 

 hulls, but the -^tock raiser cannot afford to make the oil 

 mill his prinnd^al depot of supplies. 



Bearing in mind this necessity for avoiding large ex- 

 penditures for purchased foods, we have planned a line 

 of experiments intended to ascertain the extent to which 

 farm-grown foods can be relied on in the feeding of dairy 

 cows and the best crops for use as food in effecting this 

 end. 



The first experiments here reported are preliminary to 

 this investigation and involve a comparison of a ratioi> 



