340 



BOTANICAL GAZETTE 



[NOVEMBER 



but one criticism; the characters of the female chosen to be the 

 mother of all these families might dominate such different char- 

 acters as were possessed by the hermaphrodites, in which case 

 all families would show identical composition regardless of the 

 variations in the pollen parents. This suggested dominating in- 

 fluence of the female is rendered untenable, however, by the fact 

 that the same female was pollinated by 7 other hermaphrodites 

 having different histories from those considered under the present 

 case, and also by n different males of diverse origin, and in every 

 case the males among the progenies were of the same type as their 

 pollen parent. 



CASE V 



WHEN FEMALE OFFSPRING OF SELF-FERTILIZED HERMAPHRODITES ARE CROSSED 

 WITH AN UNRELATED MALE 



These families were produced by pollinating 14 different females, 

 taken consecutively in 08115, with pollen from a single normal 

 male, 0855(36), in an unrelated family. The essentially equal 

 results of all these crosses indicate that there are no differences 

 among these females which were not dominated by the sex char- 

 acter of the pollen parent. As this pollen parent was a male from a 

 normal male parentage, it may be appropriately assumed to have 

 been free from any hypothetically possible hermaphrodite modifier 

 H. If such a modifier had been possessed by any of these 14 

 females, a more striking evidence of that fact should be presented 

 than is found in the occurrence of less than i per cent of hermaph- 

 rodite individuals among the offspring. This is a smaller percent- 

 age of hermaphrodites than has been found in one or two cases 

 among the offspring of a female pollinated by a normal male, neither 



