94 Fertility in Cichorium iiitybus 



obtained in the offspring of parents whose self-fertility is higher 

 than 30%. 



Discussion. 



Physiological sexual compatibility in chicory is decidedly sporadic in 

 its heredity. That its expression in the individuals of self-fertilized 

 lines of descent is continually fluctuating is clearly in evidence from the 

 behaviour of the various self- fertilized lines of descent reported above. 

 The number of generations and the number of plants in the main sub- 

 families have been, it would seem, sufficiently large -to establish these 

 points. 



Starting with self-compatible plants that arose sporadically among 

 the Fy progeny of self-sterile parents, lines of descent have now been 

 grown through three further generations {F.., F^, and ^^4) and in every 

 generation, and in every series (excepting a single one of only two 

 plants) of each family, self-sterile plants have appeared, and usually 

 these have been in considerable numbers. A general summary of the 

 different generations and families shows that about half of the plants 

 have been self-sterile : that is, there has been considerable regression 

 in each generation, in each family, and in each series to the condition of 

 self-incompatibility which appeared to be the rule in the original stocks 

 from which these plants descended. Furthermore, the average per- 

 formance has been quite the same for the successive generations. 



The first self- fertile parents of these families and lines of descent 

 were offspring of parents whose self-sterility had been thoroughly tested 

 and found to be complete. As individuals they appeared to be com- 

 pletely self-sterile ; the races or strains to which they belonged, however, 

 are not to be considered as absolutely self-sterile. The self- fertile plants 

 used as the first parents for the cultures here reported had thus one 

 generation of parentage known to be self-sterile, but I have elsewhere 

 shown (1917) that self- fertile plants may arise after three generations 

 of ancestry self-sterile on both male and female sides. In the develop- 

 ment of self-compatibility, these plants then differed sharply from their 

 immediate parents, and from the gi'eater number of their sister plants. 

 The extremes, self-fertility and self-sterility, it would seem, are two 

 (juite decidedly contrasted characters. In the apparent suddenness of 

 the occurrence of self-fertile plants among the offspring of self-sterile 

 parents there is much that is suggestive of what is quite generally called 

 mutation. 



