THE McINTOSH DROP 7 



Table 2 — Yield, Drop, and Percent Drop by Periods, Block G 



♦Standard error. 



program was followed which in general resulted in greatly increased preharvest 

 fruit drop. Four of the plots exhibited significantly higher percentages of drop 

 than the others. These are plot 6, in sod, which had liberal nitrogen applica- 

 tions (300-500 lbs. /A) during the entire period plus potassium since 1927; plot 1, 

 in cultivation, which had no fertilizer additions up to 1932 when a complete 

 program was initiated; plot 7, a duplicate of plot 1; and plot 5, in cultivation, 

 which was fertilized liberally with nitrogen since 1927. Of the plots with com- 

 paratively low drop, 2 and 4 are in sod fertilized with nitrogen only and 3 is a 

 check plot in cultivation. With regard to plots 2 and 4, it is generally recognized 

 that a good sod utilizes considerable nitrogen thus reducing the amount available 

 to the trees. Plot 3 has never been fertilized. From these data it appears in a 

 general way that as better fertilizing practices were followed, increases in drop 

 resulted. At the same time, there were definite crop increases. Table 4 shows 

 the relationship between yield and drop over the entire 16-year cropping period. 

 The plots are listed in descending order according to the approximate percentage 

 of drop. In comparing the percentage of drop with the yield, it is evident that 

 a correlation exists. Dickson (4) likewise found evidence that severity of dropping 

 was associated with increased yield induced by better cultural conditions. 



In order to test the significance of the association between yield and drop in 

 the separate plots, correlation coefficients were derived using individual tree 

 yields (above 50 pounds) from 1927 through 1937. These values also are given 

 in Table 4. The three highest coefficients — for plots 1, 3, and 6 — are sig- 

 nificant but rather small, giving coefficients of determination: .28, .22, and .15. 

 Hence not more than 15 to 28 percent of the variance in drop could be attributed 

 to the effect of yield. The other values range below the usual levels of signi- 

 ficance. This indicates a variable association between yield and drop within 

 the plots. Data from other plots tend to substantiate this conclusion. 



It is interesting to note that the trend of preharvest dropping severity was in a 

 general way similar to that of Block E. Plots 1, 6, and 7 have shown a steadily 



