Studies of Teratological Phenomena. 



107 



families were 353 X 301, 301 X 396 and 301—1—5 X 373. The cause 

 of the high abnormal condition of 324 X 301 Fi flowers is easily- 

 accounted for by the character of the 324 parent, as the flowers of 

 this race are 5 — 7-petaled and almost constantly in possession of a 

 three-loculed ovary. In all cases except the one just mentioned and 

 301 X 396, one parent is normal. My Fi data on this subject would 

 be more satisfactory if I had used only a single fasciated plant as the 

 abnormal parent in these crosses. Personally, I believe the race to be 

 a homozygous pure line, and if the factor A cannot be modified by 



Fig. 19. Flowers of parents and F^ hybrid (303—1—13 X 327). 

 Three center flowers belong to hybrid. 



selection, any one of the pure strain individuals would be similar to 

 every other one, so that the flowers of an Fi family grown from 



303 X 373 should not be more nor less abnormal than those from a 

 301 X 373 culture, providing both were grown under the same environ- 

 ment. An experiment of this kind is in progress. Two or more crosses 

 of the same kind, so far as my experiments have gone have always 

 given similar data on dominance, leading one to infer that internal 

 environmental differences of the normal races are responsible for the 

 variability in the expression of A in each of these hybrid families (Fig. 19). 



Cytology. The maturation phenomena in the Fi anthers of 



304 X 402 was investigated, but as compared with the pure strain, the 



