134 



soon in six-inch pots which were placed on a bench in a green- 

 house where they remained undisturbed throughout the summer 

 and the following autumn and winter. The vines grew well ; 

 some to a length of about ten feet. The roots became much 

 crowded, filling the pots and extending down into the ashes of the 

 bench. 



Of the lOO plants thus grown only seven bloomed but these be- 

 longed to six different pedigreed clons from seedlings grown 

 from seed supplied from Santo Domingo b}" Dr. B. H. A. Groth. 



One plant bloomed from November 22, 1924, until -\pril 16, 

 1925, producing a total of twenty-five flowers. The lowe>t num- 

 ber of flowers on a plant was five. 



Self-pollinations were made on all the plants and for several as 

 many as ten flowers were selfed. Not a single capsule was set 

 to self-pollination. 



Fourteen dift'erent cross-combinations' were made involving 

 from one to twelve flowers each. Only four cross-relations were 

 successful but from these eight mature capsules were obtained 

 (see FiGURK i). In the most successful combination two of the 

 five flowers crossed yielded capsules. 



CONCLU.SION 



The results obtained at St. Croix indicate that seed is not, as 

 a rule, there produced by sweet potatoes during the first part of 

 the period of their blooming. This suggests that some type of 

 sterility may be operating at this time. Whether this imoKes a 

 direct environnK-ntal influence on fruit setting, or the inii)crfect 

 development of flowers for a time, or a cyclic change in the com- 

 patibilities of fertilization is not now apparent. 



It seems very clear from all the evidence now at hand that 

 during the [)eriod of maximum blooming the type of sterilit} oper- 

 ating in sweet potatoes is that of incompatibility in the processes 

 of fertilization. ISlost seedlings and most clonal varieties appear 

 to be completely self-incompatible. 



Various combinations in cross-i)ollinatioii likewise fail, lail cer- 

 tain compatible relations are to i)e touml which readilx' } ield 

 fruit and seeds. This type of sterility is rather frequent in plants 



3 Afost of tlicsc were- made liy Miss Hester M. Iviisk, 'I'lchnical Assist- 

 ant at 'I"hc New York JiOtanical Garden, w liosc ctlicient assistance in this 

 and other i)r<>jccts of research deserves s|>ecial nunlion. 



