1912] Goodspeed: Nicotiana Hyhrids 151 



the parental corolla diameter characters are "latent" (Shiill,. 

 1908, p. 7 and the literature there cited) in F^ but appear both 

 fully activated and in various degrees of blending and combina- 

 tion. Finally, there is no evidence which goes to show that there 

 is a "variability .... in the potency of determiners" (Daven- 

 port, 1910, p. 135) in our experimental material or that we are 

 here dealing with the case of a "stronger determiner meeting a 

 weaker determiner in the germ" (Davenport, 1908). It must 

 at times be somewhat a matter of speculation just "what does 

 meet what" in the germ and, in the case with which we are 

 dealing, it seems more than commonly a subject for speculation. 



We can, in conclusion, simply suggest that the occvirrence 

 of large flowers, small flowers and flowers the corolla diameters 

 of which show every degree of gradation between large and 

 small in F^, of a cross between a large-flowered and small-flowered 

 variety of N. acuminata, represents the maximum degree of 

 "segregation" which occurs in such a hybridization ( see, in 

 this connection: Locke, 1906, and Emerson, 1910). The F. 

 plants grown from the seed produced by self-fertilizing the 

 flowers of this hybrid may give us a few plants bearing flowers, 

 the corolla diameters of which show small variation in size. Suc- 

 ceeding generations may reduce the variation still further until 

 we have regained either one or both of the parental varieties 

 which entered into the production of the F^ generation, or have 

 established a new "flower-size" variety which will thereafter 

 come true to a certain corolla diameter of flower which varies 

 only slightly in size. 



The two sets of experimental results reported in this com- 

 bined paper are, in a broad sense, identical. In one case it has 

 been possible to carry through the evidences of segregation which 

 appear in F^ plants and in the seed which they produce, into 

 the F2 generation ; in the other it has not as yet been possible 

 to do so. In both cases evidence of segregation appears after 

 the seed produced by cross-fertilization had been grown and 

 before the seed that was to produce the Fo individuals had been 

 germinated. In l)oth cases we have been forced to suggest ex- 

 planations for the phenomena we recognize which are not strictly 

 IMendclian. In one case the non-appearance of any ^Mendelian 



