372 NEW JERSEY STATE AGRICULTURAL 



The small percentage of fruits resulting- in this test is probably 

 due to the fact that it was not made until late in the season, 

 when flowers outside of the bags did not generally set fruits. 

 The test will be continued and, in season for best results with the 

 hope of learning the effect of such close-breeding upon the off- 

 spring oif crosses as to the fixation of qualities and also, among 

 commercial varieties, as a test of its weakening tendency. 



Tomato Exclusion Experiment. 



In order to' insure self-fertilization only one bud, before it 

 opened, was allowed to remain on a flower stalk, which was cov- 

 ered with a paper bag to^ exclude foreign pollen. One hundred 

 buds of nearly the same age were thus covered until the time for 

 fertilization had passed. 



In order tO' compare the number of seeds in the fruits from the 

 self-fertilized with those grown in the open, an adequate number 

 of the latter were harvested from the same plant with the former, 

 and the results are shown in the table below : 



Type of Fruit. 

 'Cherry," 



"Plum" (Small), ... 

 "Plum" (Medium), . 



"Plum" (Medium), . 



"Plum" (Cross), . . . 

 "Pear" (Medium), . 



