40 



THE PLUM IN KANSAS. 



The range of varieties in this series of experiments is barely large 

 enough to be suggestive, although the large number of blossoms cov- 

 ered and the relatively small number of fruits set give a satistactorily 

 clear-cut indication of a generally prevalent self-sterility. It seems 

 clear that one could expect little fruit from De Soto, the Original 

 Minnesota, Pottawatomie, RoUingstone, and the Wolf Seedling No. 6, 

 unless the trees were favorably situated for cross-pollination. In fact, 

 Robinson was the only variety in the experiment with which the fruit 

 set by self-pollination seemed to be normal and vigorous. Our judg- 

 ments, made in the orchard — and such judgments may properly go 

 beyond the numbers in the tabulations — were that self-sterility was 

 extremely doubtful in all cases except that of Robinson. 



The question which naturally comes next in order is this : If a cer- 

 tain variety must have foreign pollen in order to set fruit, what other 

 varieties are the most efficacious pollen bearers ? This is a question 

 requiring very many experiments in artificial pollination. Our own 

 work for 1896 has been limited to twenty-one exj^eriments among 

 eleven varieties, and consisted of 319 artificial crosses. The record of 

 these pollinations is seen in table 5. 



Table 



-Ihcord of Crosses. 



* No record. t Tree died. 



Beyond an indication that the typical Western forms of P. ameri- 

 cana can be pollinated by the Eastern forms (var. nigra), these ex- 

 periments show nothing. No one who has had experience in hand 



