EXrElMMi:N'I\S AT KOCHKSTIuK. 



27 



number of crosses I'roin tree to tree must have been iu the aggregate very 

 large. In all these eases, however, the insects brought only Bartlett 

 pollen and did not i)erform true crosses. Tliere is scarcely any doubt 

 that had there been other varieties Avliieh Idoom at the same time, sucli 

 as White Doyenne or Clapps Favorite, scattered throiigli the orchard, 

 a fair amount of fruit would have set. As it was, insects carried from 

 tree to tree and from flower to flower only Bartlett pollen, which was 

 not capable of inducing the fruits to form. The relativelj^ few tlowers 

 which were hand-pollinated with Bartlett pollen and which f\xiled to set 

 were an index to what actually occurred over the whole orchard. On 

 the other hand, the cross-pollinations, resulting in a high percentage of 

 fruits, show what might have been had otlier varieties been at hand so 

 that insects could have performed true crosses. 



Examinations of the stigmas were made during flowering time and 

 they were invariably found to be covered with germinated pollen grains. 

 Microscopic examination showed that the pollen tube penetrated the 

 tissues. Bartlett pollen was examined and tested, and Avas found to 

 be sound and capable of germination in water or in a sugar solution. 

 The question arose whether it was simply the crossing that was needed 

 or whether Bartlett pollen was of itself inferior on its own as well as 

 on other stigmas. Thefruitfuliiess of the single Anjou tree surrounded 

 by Bartletts suggests that Bartlett i)ollen is effective on the Anjou, since 

 this variety had been shown to be self- sterile in tlie Brockport series. 

 The Rochester experiments settled this i)oint, as Avill be observed later 

 on. 



SERIES IIT.— EXPERIMENTS AT ROCHESTER, N. Y. 



General statement. — These experiments were carried on in the large 

 variety orchard of Ellwanger & Barry, within the limits of the city of 

 Rochester. Every facility for making them was kindly aftorded by the 

 proprietors. Without this fine collection, many trees of which Avere 

 planted years ago, and the generosity of the owners in placing it at 

 our disposal, the number of results possible would have been very 

 much smaller. 



This pear collection, which was the pride of the late Patrick Barry, 

 probably contains more varieties than any other in America. On a 

 few acres of ground there are here to be found about all the pears com- 

 monly cultivated in the United States and many of the rarer ones. 

 It is scarcely necessary to say that the trees have received intelligent 

 and excellent cultural treatment, and were, with very few exceptions, 

 in flne condition. 



The orchard was reached May 7, before the flowering began. The 

 blossoms commenced to open May 14; on May 15, the earlier- blooming- 

 varieties Avere about one-fourth out and the others beginning to ojjen, 

 and on May 16 all but a few exceptionally late-blooming sorts were well 

 in flower. I^aper bags were put on a large number of varieties on Mav 



