of least squares was used to olitain estimates for each major factor ami 

 one estimate of all interaction. 



The hays were chopped, sampled and fed twice a day at 7 a.m. and 

 5 p.m. in equal amounts, at a level slightly ahove maintenance. A pre- 

 liminary period of 15 days preceded an eight-day collection. Total sep- 

 arate collection of feces and urine was made with hoth female and male 

 animals hy means of collection devices described previously (8). The 

 solid and liquid excreta were weighed, sampled daily at 9 a.m. and each 

 composited. The composite feces was frozen solid and the composite 

 urine kept in a refrigerator just ahove freezing until each was analyzed. 



Heat production measurements of at least 24-hour duration on each 

 feed were made at the end of the collection periods hy means of an open- 

 circuit, indirect calorimeter (2) to complete the energy' balance of the 

 ration. This was followed by a 24-hour measurement in the post-absorp- 

 tive state to determine the heat increment of the ration, according to 

 the method used by Colovos et al. (6). On a maintenance-plus level of 

 feeding the forage, the post absorptive state, as indicated by the absence 

 of respired methane and a respiratory quotient of about 0.7, was reached 

 after 48 hours of fasting. 



The chemical composition for dry matter, ash, crude protein, ether 

 extract, crude fiber and nitrogen-free extract were made using the 

 methods in the A. O. A. C. Manual (1). Gross energy determinations 

 were made using an adiabatic bomb calorimeter. 



The digestibility of energy, dry matter, protein, ether extract, fiber 

 and nitrogen-free extract was calculated. Methane, carbon dioxide and 

 oxygen determinations of samples of outdoor air and respiration cham- 

 ber air were carried out using Carpenter's modification of the Haldane 

 gas analysis apparatus (4) with slight changes made at New Hampshire 

 (8). 



Results and Discussion 



The forages harvested in 1960, 1961 and 1962 at both Northwood 

 and Groveton differed only slightly in chemical composition from year 

 to year, where the timothy had received the same rate of nitrogen and 

 had been harvested on the same dates. 



Figure 1 shows the comparative yields of Commercial timothy 

 grown under three levels of nitrogen, 50. 100 and 150 lb. per acre and cut 

 at three different dates, June 1, 15 and 30 and also the yield of the late- 

 maturing Essex variety grown under 100 11). nitrogen per acre in North- 

 wood, N. H. The yields of the same two varieties grown under 100 lb. 

 nitrogen fertilization and cut in Groveton, N. H. on the same dates as 

 above are also shown. There was not enough growth of the Essex timothy 

 to harvest on June 1 in Groveton. 



Table 2 shows the total yields for each year, first cutting and after- 

 math, and the mean for the three-year period. The study did not in- 

 clude Groveton in 1960. There was not enough growth of Essex at Grove- 

 ton to harvest on June 1 in either 1961 or 1962. 



Table 3 shows the average composition of the forages. 



1. Protein content of the forage increased as the level of nitrogen 

 fertilization increased. 



