112 



White. 



conditions; while in aa segregates, normal conditions prevailed as in the 

 normal grandparental race. The proportion of contabescent anthers and 

 the cytological irregularities which they contain fluctuates as the 

 grosser alterations produced by the factor A vary. These facts are of 

 theoretical interest, because of their possible bearing on the chromosome 

 theory of inheritance (see White, 1913). 



N. tabacum fasdata (301 — 1 — 5) X N. tabacum havanensis (373). 



Two F2 families were 

 grown (Fig. 22). Tables 20 a 

 and b give the ratio of abnormal 

 to normal segregates and the 

 proportion of fasciate to non- 

 fasciate- stemmed plants. It 

 was impossible to accurately 

 classify the heterozygotes, 

 owing to the numerous modi- 

 fications the character fas- 

 ciation had undergone, as 

 compared with its appearance 

 in the pure strain and the 

 304 X 402 AA and Aa segre- 

 gates. The small proportion 

 of 1 fasciated stem to 5 normal 

 one was perplexing, inasmuch 

 as theory demanded three of 

 the former to one of the 

 latter. The individuals of this 

 population did not vary much 

 in height, but segregation of 

 factors governing color, habit 

 and leaf character had occurred. 

 The leaf count per plant varied 

 between 16 and 111; for those classified as normal, it ranged between 16 

 and 28 and for "possible heterozygotes" as low as 19 and commonly 20 

 to 25. Table 21 shows the character of an F2 population, in terms of 

 its flower variability and leaf n\imber per plant. 



iV. tabacum fruticosa (353) X N. tabacum fasciata (301). 

 Only two F2 families of this cross were grown and the ratio of 

 abnormal to normal plants may be found in Tables 20 a and b. The 



r» 



/- 



Fig. 22. 



Two F, segregate plants (AA) from 

 301—1—5 X 373. 



