New York Agricultural Experiment Station. 39 



once checked is seldom so profitably resumed nor in its ultimate 

 attainment so satisfactory. This is known in milk flow and in egg 

 laying. It is easier to sustain the flood of production than to restore 

 it after any subsidence. 



Because of the importance of the mineral nutrients in poultry 

 feeding-, experiments in which they are considered were some time 

 ago undertaken at this station. The ordinary grain foods are nota- 

 bly low in lime content; but it was found- that laying hens could 

 readily obtain this necessary material, when demanded, from inor- 

 ganic sources such as oyster shells, etc. While the grain foods hold 

 considerable phospherous in organic combination, it appears that 

 more is needed by the growing chicks, for the benefit accompanying 

 the use of animal foods was found to be often very largely due to the 

 mineral matter which they contained. 



In a number of feeding trials with chicks reported in former bulle- 

 tins it was found that rations containing animal food gave much 

 better results than similar rations without animal food, and that when 

 the deficiency of mineral matter was supplied by the addition of bone 

 ash, certain rations, otherwise entirely of vegetable origin, were as 

 efticient as those containing much animal food. 



Although the chicks for these experiments were kept in pens which 

 had sand-covered floors, and were free to pick up all sand desired, 

 it was afterward thought that much of the benefit from the addition 

 of bone ash to the food might possibly be due to its mechanical use 

 as grit, for it was not all finely powdered. To a limited extent this 

 mechanical use probably did help ; for when sand was added to the 

 food of chicks, even those kept on sanded floors, better results fol- 

 lowed. 



To get testimony on this point and further intimation as to the 

 availability of inorganic lime and phosphorus a number of feeding 

 trials were made. Some of these were carried on about the time of 

 feeding experiments reported earlier (Bulletin 171), but so many 



