108 Self- Incompatibility in Hermaphrodite Plants 



which were not fertilized. It may be, however, that in some of these 

 phenospermic achenes embryo abortion occurs as an expression of in- 

 compatibility after fertilization. This appears also to be the case for 

 feebly self-fertile plants of Eschscholtzia californica as noted above. 

 The ability of such plants to produce abundant seed to compatible cross- 

 pollinations at the same time that phenospermic seeds are produced 

 among heads selfed indicates that any embryo abortion which may occur 

 is associated with incompatibilities rather than with inability of parent to 

 nourish young 'embryos. It is to be recognized however that in chicory 

 the development of the head as a whole may in large measure depend 

 on the number of ovules fertilized properly, and that the development 

 of the individual seeds in a head may thus be influenced by local relations. 

 If such is the case, the effects would be most liable to appear in plants 

 partially or feebly self-compatible. 



Considerable cytological investigation has been directed to the study 

 of these conditions in chicory, but at the moment of writing the observa- 

 tions are not conclusive as to the extent to which true embryo abortion 

 may occur. The results, however, indicate that in plants fully self-incom- 

 patible there is complete failure in fertilization due to the poor growth 

 of pollen tubes. In plants feebly self-fertile there appear to be various 

 stages at which the processes of fertilization may fail. 



III. The heredity of self -compatibility in selffertilized lines of descent 

 in the variety red-leaved Treviso of Cichorium Intyhus. 



The immediate parents of this generation were highly self-fertile 

 plants which had descended from three generations of ancestry known 

 to have been self-sterile at least during the early part of their period of 

 bloom. The race had been kept in culture by rather close inbreeding. 

 This generation therefore aiforded material for the study of the heredity 

 of self- fertility after it has arisen sporadically in an inbred variety. 



The number of plants grown in the /j, and also the I^, was so large 

 that it was impossible to test all of these by controlled hand pollination 

 (described 1916, p. 362) as was done in previous studies with this species. 

 The plants were tested by autonomous self-pollination. Flower heads 

 were allowed to open and close under a bag. Such heads were appro- 

 priately tagged and allowed to continue development in the open. As 

 a rule flower heads of chicory are open but a few hours during the fore- 

 noon ; by midday they are closed never to open again, and the next day 

 the corollas and styles fall. The entire afternoon can be devoted to 



