384 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN 



made to test the inheritance of red coloration. The plant {A X C) 

 no. I, as were its parents A and C, was without any apparent 

 trace of red coloration in leaves and stems, while in the leaves and 

 stems of the "improved striped-leaf" plants there was a very 

 strong red coloration. Of the Fi generation a total of 30 plants 

 tested proved to be completely self-sterile as is shown in table 8. 

 These plants were all blue-flowered, but they showed almost 

 every conceivable variation in respect to the amount and distri- 

 bution of red coloration. As there were no self-fertile plants, 

 further generations were not grown. The complete self-sterility 

 of the plants of the generation grown is, however, an indication 

 that crossing between widely separated strains does not neces- 

 sarily give the development of self-fertility. 



In the various self-pollinations that have been made, a total 

 of 631 plants have been tested for self-sterility; the number of 

 flower heads manipulated is about 4, 100, and the number of flowers 

 concerned is hence about 74,000. The studies were at first quite 

 general for wild plants and for plants of eleven different cultivated 

 varieties whose pedigrees were not known. The later studies, 

 which embrace the greater number of plants, were made on plants 

 of pedigreed stock, as will appear from the foregoing data. 



Phenomena of physiological inter-incompatibility 



AND compatibility 



During the experimental work of 1912 it became evident that 

 cross-fertility such as has been reported in table 2 does not always 

 occur in Cichorium, and that not only is there self-sterility, but 

 also cross-sterility, quite as Correns ('12) afterward reported for 

 Cardamine. 



Results of crosses involving A and various of its progeny. — Several 

 crosses attempted in 191 2 between wild white-flowered plants 

 grown from seed of the plant A completely failed. On this 

 account opportunity was taken during the summer of 1913 to 

 test the behavior of certain sister plants when crossed among 

 themselves and with a parent plant. For this purpose 8 plants, 

 all white-flowered and derived from open-fertilized seed of the 

 plant A, were tested as fully as the conditions allowed. These 

 plants were in their second year of growth and had all proven to 



