April, 194^] PASTURE TOP-DRESSING IN N. H. 15 



when compared with nitrate of potash on this soiL Unfortunately 

 the method of harvesting that has been practiced in this experiment 

 does not take palatability into account as would the cows under ac- 

 tual grazing trials, since with a lawn mower we clip everything, 

 palata1)Ie or unpalatable, and include it in the yields. 



Change of Treatment on Livingston Plots 



Six of the Livingston plots which for five years had been receiv- 

 ing nitrogen alone were divided in the spring of 1938 and since then 

 one end of each plot has had 200 pounds of an 0-20-20 annually, the 

 other 600 pounds of an 0-20-20 in what is designed to be a triennial 

 application. Thus, in Table 15. one series has had a total of 400 

 pounds of an 0-20-20 in a divided application while the other has re- 

 ceived 600 pounds at one treatment. The amounts of phosphoric 

 acid and potash applied are therefore not yet equal but the data are 

 presented to indicate how quickly this soil responds to these miner- 

 als. The nitrogen applications were continued during this period. 

 Other plots receving nitrogen alone during the same period are 

 shown for comparison so that all data are for the same years. 



Table 15. — Two year average yield, Livingston pasture 



In this table the two 0-20-20 treatments with nitrogen have been 

 compared directly with the nitrogen alone plots. The increases re- 

 corded for the annual application of 200 pounds of 0-20-20 are 48 per 

 cent higher than nitrogen alone for dry matter and 56 per cent high- 

 er for protein while the 600 pound application runs 67 and 65 per 

 cent higher for these two substances, respectively. 



Even more interesting than the data is the rapidity of the change 

 in vegetation which occurred in the plots themselves, after applying 

 the fertilizer containing phosphoric acid and potash. Within a few 

 Aveeks wild white clover appeared, and by midsummer of the first 

 season the vegetation there was mixed grass and clover where very 

 little or no clover had occurred before under the previous nitrogen 

 treatments. 



Effect of Rainfall on Pasture Production 



It has lieen interesting to note the variation in pasture yields from 

 season to season on these two experimental fields. Rainfall always 

 varies, of course, and since no attempt is made to water these pas- 

 tures artificially the diiTerences that occur in productivity in different 

 seasons for the same treatments are governed largely by the amoimt 

 and distribution of summer rainfall. 



The summer season of 1938 was very wet and that of 1939 was 

 quite dry. The rainfall data in Table 16 are taken from records se- 

 cured at the Livingston farm, and from the Ireland farm which is 

 less than two miles distant by air line from Seavey pasture. 



