observations making yield on first erop a function of calendar datc^^ 

 The correlation coefficient value "r" of 0.8189 and S value of ± 0.6265 



was such that the function was considered acceptable for this analysis. 

 In other words, the function was considered an adequate representation 

 of yield response to time. 



A similar analysis was performed for second-crop yields. The 

 scatter diagrams of yield are pictured in Figure 5. The fitted first degree 

 equation was not a good fit; a correlation coefficient of only 0.26 was 

 obtained. This means that a linear regression equation provides an 

 estimator of second crop yield that is not significantly more accurate 

 than the average of the observations. For this reason, the yield of sec- 

 ond crop was taken to be a constant 0.75 tons per acre. 



The quality of first crop forage as related to time is described in 

 Figure 6. As the date of cut advances into the season, the quality per 

 unit decreases: for example, from 72.3 percent digestible dry matter 

 on .June 1 to 56.6 percent on July 1.^- Second crop forage does not have 

 as high a percentage of digestible dry matter as early cut first crop, 

 but neither does it decrease as fast. For the purposes of this study, sec- 

 ond crop digestibile dry matter is assumed to be the same as first crop 

 cut on June 20. 



U 

 < 



0£. 



to 



Z 



o 



>- 



1.5 _ 



1.0 



:7b 







" igure 



0.38487 + 0.00740476X 



.2636 

 ± .3094 



y.x 



± 



± 



10 20 30 40 50 60 70 

 DAYS BETWEEN FIRST AND SECOND CROP HARVEST 



5. Yield of Second Crop Forage as it is Related to the Lapse of Time 

 Between First and Second Crop Harvest 



11 The function is as follows: y = a + b + c x-. Where: y = yield of first 

 crop forage, x = date of cut. " 



1- Reid, J. T., "Nutrition of High-Producing Cows," a paper given at the winter 

 meeting of the N. H.-Vt. Breeding Association, 1963. 



15 



