showed the same general relationship - the later the date of harvest, the 

 lower was the crude protein content. The calculated weekly decline in crude 

 protein content averaged 0. 63 percent, or approximately two-thirds greater 

 than was found by Miller. Willhite's data were obtained from 13 different 

 ranches (30 observations). These hays were not all harvested in the same 

 way; the sward composition and irrigation practices differed; and the sampling 

 was done by 13 different people. Therefore, a greater variation in crude 

 protein content than occurred in Miller's experiment would be expected 

 from Willhite's observations. But it is likely that the decrease of 0. 38 per- 

 cent for unfertilized hay and 0. 94 percent for fertilized hay as found by 

 Miller would be more generally applicable than the decrease observed by 

 Willhite. 



Maturity of the species in the sward, rather than the calendar date, is 

 the major determinant of protein content. One method of obtaining high- 

 protein hay, at least in limited quantities, is to harvest at an early date. 

 Ranchers usually harvest their meadows only once and as late in the season 

 as possible, whether the meadows are fertilized or unfertilized. This prac- 

 tice produces the greatest volume of hay, although its protein content is 

 lower than for hay cut earlier. 



Fertilization 



Results from large numbers of fertilizer experiments have shown that 

 production responds to applications of fertilizer in a general pattern. When 

 soil is low in a particular nutrient or nutrients, the relationship between 

 nutrients added and crop yields usually follows a curve as illustrated in 

 figure 2. 



The first part of this curve (segment A) is an area in which the crop yield 

 increases with each additional unit of fertilizer but contributes less to the 

 yield than the preceding unit. Also, if enough fertilizer is applied, a point 

 may be reached at which an actual decrease in the total yield is associated, 

 with increased applications of fertilizer (segment B). In general, work on 

 prodvfction responses of hay to applications of nitrogen and phosphorus by 

 Willhite and Rouse has been with that segment of the response curve labeled 

 "A, " in which total yield was increased with each additional unit of fertilizer, 

 but with each unit contributing less to yield than the preceding unit. 



In 1952, Willhite and Rouse conducted an experiment on seven mountain 

 meadow ranches, in which 200 pounds of available phosphate and varying 

 quantities of nitrogen were applied to individual meadows in an effort to learn 

 the response of these meadows to nitrogen and phosphate under normal ranch 



