THE McINTOSH DROP 13 



numbers of apples dropping at different times, the following figures are illustrative. 

 In the case chosen, the apples which dropped before September 21 contained on 

 the average 2.7 seeds; during September 25 to 28, 4.0 seeds; on October 4, 8.5 

 seeds; and on October 12, 11 seeds. The drop in 1939 came later than usual. 

 For instance, in 1937 with another tree in the same block, over 50 percent of the 

 crop had fallen by September 15. The seed contents of the apples dropping at 

 weekly intervals were as follows: September 2, 7.5 seeds; September 9, 8.3 seeds; 

 September 16, 10.6 seeds; September 23, 11.8 seeds. The correlation here amount- 

 ed to only +.302 (Table 7 — G-18, 1937). Mean drop dates among individual 

 trees have been found to vary as much as 20 days in one locality. But the seed 

 influence is manifest, nevertheless, and the apples with the most seeds tend to 

 hang on the longest. 



Whether seed content is merely an associated factor or whether it actually 

 plays a causative role in determining time of drop is difficult to prove. The 

 significance of embryo and seed development in the life of young fruits is quite 

 well founded. It would seem logical to suppose that the association found to 

 exist in maturing fruits is just as definitely one of cause and effect. It seems 

 plausible that the presence of many viable seeds tends to cause the build-up of 

 stronger tissues in the pedicel and helps to insure more adequate and constant 

 supplies of nutrients and moisture. Perhaps the beneficial effect on "sticking" 

 of apples just prior to and during the harvest period is related to the lowering 

 of the ratio of nitrates to carbohydrates due to the increased utilization of available 

 nitrates. The author studied starved trees in which dropping was retarded in a 

 similar way to that described by MacDaniels (16) for weak trees. There was no 

 significant correlation between seed content and time of drop with these abnormal 

 trees. This indicates again that the value of seeds in retarding drop may be over- 

 shadowed by other physiological factors. The author (29) found some evidence 

 that larger spurs, measured by weight, favorably influenced the hanging ability 

 of apples at harvest time. As measured by cluster base diameter also, in a single 

 correlation, a positive coefficient of low value was obtained. In two other cases, 

 no association could be detected. The larger spurs (in a single experiment) 

 likewise contained lower percentages of nitrogen which in itself may have been 

 the significant factor (29). Heinicke (7) found spur size to exert a similar influence 

 with 3'oung fruits. Often, large spurs compensated for low seed values in assisting 

 fruits to stick on. 



An analysis of the dropping behavior of individual limbs of a tree revealed 

 little departure from the normal behavior of the tree as a whole. However, the 

 seed data revealed that the apples on the low spreading branches contained mere 

 seeds than the apples on the vigorously growing upright leaders. This indicated 

 a possible difference in the ability of limbs to hold apples. Hoffman (11) ob- 

 served that drop is heaviest in the lower half of a tree as the result of more shade. 



Influence of Length of Stem 



Stem length measurements were taken on some 6,000 drops at the time of 

 seed count in 1937. The stems varied from 5 to 26 millimeters in length. With 

 the crop of one of the three trees a low but significant correlation (+ . 136 ± .016) 

 between stem length and time of drop was obtaineo. Considering that very low 

 negative coefficients were obtained with the other trees, it appears that length 

 of stem in itself played a minor role in fruit drop. It has alreadx' been suggested 

 that a short stem might enhance the chances for early dropping for the reason 



