370 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN 



' TABLE 3 — Continued 



Plant no. 



Record for heads pollinated 



^ „, , With no With 



Dates lotal gged seed 



Seeds per head 



Fertility 



{E22 X -4) 

 No. I . . 



3- 



4- 



5- 



10. 



14- 





 II 



5 



15 



5 



5 



10 



5 

 10 



5 



15 



5 



o 



9 



6, 6, 9, 10, 13. 



10 other plants of this series were 

 self-sterile 



0.03 



0.51 

 0.04 



(C X E22) 



No. 15. 



" 17. 



4 

 II 



15 

 24 



13 

 21 



2 i I, 12 I 0.05 



3 9, 14. 14 .■■••■. 0.07 



I 17 other plants of this series were 1 



i self-sterile 



The intensity of the character of self-fertility may be estimated 

 for the self-fertile plants of this generation and also of the following 

 generations and expressed in terms of the percentage of flowers 

 which produced seed. Statistical studies have been in progress on 

 the variation and the heredity of flower number and from these 

 data the average number of flowers per head is quite accurately 

 known for all but three of the self-fertile plants that have thus 

 far appeared. Multiplying this average by the number of heads 

 concerned in the controlled self-pollinations gives a -very accurate 

 estimate of the number of flowers upon which the percentage of 

 seed can be computed. The average number of flowers per head 

 for the self-fertile plants of the Fi generation ranged from 17.1 

 to 20.8, and the average fertility computed, as just stated, for 

 these plants varied from i per cent to 51 per cent. 



Some of the data, especially that when only one head produced 

 seed, may invoh-e an experimental error. In the case of {A X 

 E22) no. 5, thirty-one heads produced no seeds, while one head gave 

 five seeds, a result that might ha\e been due to experimental error 

 as well as to a low degree of self-fertility. In plant {A X E3) 110. S, 

 one head out of eight set seed, in (E22 X A) iw. 2 one head out 



