One of the most notable details of the alfalfa coef- 

 ficients determined from both areas is the variation associated 

 with percentage of ground cover. It is important to point out 

 that the method of collecting data on percentage of ground cover 

 was subjective^ being based upon personal judgment, and that esti- 

 mates by individual observers differ by perhaps 5 to 15 percent. 

 There is, however, general agreement that following mowing the 

 ground cover is usually reduced to 5 to 10 percent, and that grour. 

 cover usually approaches 100 percent cover prior to mov/ing . Al- 

 though there are exceptions due to possible experimental error 

 and other factors, the coefficients are smaller when the ground 

 cover is low following mowing, and become larger as the ground 

 cover increases. Plate 4, Figures F, G, H, and I, illustrate 

 these relationships between coefficients and percent of ground 

 cover, plotted against time. 



A more direct comparison of pan and atmometer coefficien: 

 with percent of ground cover is sho\m in Plate 6, entitled "Rela- 

 tionship Between Pan and Atmometer Coefficients for Alfalfa and 

 Ground Cover." Figure A shows atmometer coefficients, and Figure 

 shows pan coefficients. The data for both figures were the same 

 utilized in Plate 4. As indicated in Plate 6, the Pittville co- 

 efficients appear to be higher than the Arvin coefficients. Two 

 linear regression lines have been fitted to the data. However, 

 it may be that additional data will indicate a somewhat curvi- 

 linear relation. It seems reasonable to assume that coefficients 

 at 100 percent of ground cover would not be proportionally higher 

 than coefficients at 80 percent of gi'ound cover, which, for prac- 

 tical purposes, also provide nearly complete shade, except near 

 noonday. 



-60- 



