52 



Bulletin 393 



on a nuniber of different branches from one or more trees of each variety. 

 Normal branches from five to ten years old were chosen. They \'aried in 

 length between one meter and one and a half meters. The total nimiber of 

 flowers was determined by counting the number of flower-stem scars on the 

 ends of the spurs and the number of fruits remaining on the spurs (fig. i). 



TABLE I. Percentage of Flowers Developing into Fruits after 



THE First Drop 



From two-fifths to four-fifths of the flowers are lost during the early 

 drop. The varieties given in the table fall into two groups, depending 

 on the number of flowers lost. One group, represented by Maiden Blush 

 and Westfield, lost only half as many as did the other group. All varie- 

 ties were growing in the same orchard and they bloomed during the same 

 time; consequently they had equal chances of being cross-pollinated. Can 

 the variations noted be due to a tendency toward self-fertility in the first 

 group? Subsequent data may throw some light on this question. 



Spurs setting fruit after the first drop 



From a practical standpoint, it is interesting to note the percentage of 

 flower-bearing spurs that set fruit. Such data are presented in table 2 : 



TABLE 2. Percentage of Flower-bearing Spurs Retaining Fruit 



after the First Drop 



166 



