May, 1938] 



Experiments With Grass Hay 



The average recovery of nitrogen in the hay where nitrogen carriers 

 alone were used was 34.1%, while 65.8% of the nitrogen of the complete 

 fertilizers was returned in the hay. This is a matter of much impor- 

 tance to farmers. .Unless complete fertilizers are used frequently in the 

 rotation, it is quite possible that the top-dressing fertilizer for grass hay- 

 land should carry the three important plant food elements. The need 

 for a complete fertilizer is a matter that a rapid soil test should help to 

 determine. 



Protein Percentage 



The average percentages of protein in the forage of the different treat- 

 ments for the period studied are given here : 



Table V. 



These data indicate that all the nitrogen carriers increased the pro- 

 tein percentages of the forage but that the tendency of the complete 

 fertilizers was to decrease the percentage of protein, although they stimu- 

 lated dry matter production to a greater degree. This appears to sug- 

 gest that phosphoric acid and potash are limiting factors to the develop- 

 ment of the structure of the plants and that when nitrogen is balanced 

 with them, growth is able to proceed on a normal basis. 



This seems to be supported by rapid soil tests of untreated land 

 (check plots) on this field which show a ''very low" test for both phos- 

 phoric acid and potash, tests which are too low for maximum grass hay 

 production. 



Second Cutting, Mathes Field Plots 



Second cuttings were taken on the Mathes field plots annually in Sep- 

 tember of each year. The soil of this field is well supplied with moisture 

 and second cuttings are usually quite good. 



Because of the narrow range in yield among the various treatments, 

 the second cutting data are ail presented in Table VI. The interesting 

 thing about the data lies in the fact that the yield variations are so 

 slight. With the exception of calcium nitrate, all the hay yields are 

 above those of the check plots and the protein jdelds are all above those 

 of the untreated or check plot series. Except in the case of the dry 

 matter figure for the 10-20-20 series and the protein values for this 

 and the 8-6-6 home-mixed series, it is doubtful if any of the differ- 

 ences are significant. With respect to percentage protein, all the values 



