146 University of California Puhlications in Botany [Vol. 5 



to determine the extent to which the experimental results agree. 

 In the summary of results given on page 144 the statement is 

 made that the average hybrid corolla diameter is intermediate 

 in size between the parental diameters (see East, 1912, p. 247; 

 also Hayes, 1912). To this extent our results seem to coincide 

 wdth those reported by Darbishire. The difficulty is that, as 

 shown in the figure on page 125 (Darbishire, 1908), there are 

 no hybrid starch grains as round or as compound as the parental 

 c-grains, nor are there any hybrid grains as potato-shaped as the 

 parental p-grains. The hybrid grains are half-way between being 

 potato-shaped and being round and half-way between compound 

 and not at all compound — that is, they are true intermediates. 

 The corolla diameters of hybrid flowers are, however, both as 

 large as the diameter of the large-flowered parent and as small as 

 the diameter of the small-flowered parent ; both extremes occur 

 among the flowers of a single plant ; and, finally, only a numerical 

 average establishes an intermediateness of corolla diameter for 

 the flowers of a hybrid plant. The important point also is that, 

 breeding from a hybrid pea the starch grains of which exhibit 

 one-half the D influence and one-half the R influence, we might 

 be prepared to have the hybrids in succeeding generations exhibit 

 some definite degree of segregation. Similarly, if the hybrid 

 corolla diameter was truly intermediate, in that it exhibited as 

 above one-half the D influence and one-half the B influence, we 

 might predict segregation in Fo at least to the extent to which 

 it took place in the starch grains of the pea hybrids. But now, 

 when on a single plant of the Nicotiana acuminata hybrids flowers 

 appear the corolla diameters of which show the unimpaired D 

 influence and the unaffected B influence and every degree of 

 intermediateness between the two, what prediction can be made 

 as to the appearance of the F, individuals with respect to the 

 corolla diameters of their flowers? In other words, is there 

 evidence to show that largen&ss of flower vs. smallness of flower, 

 intermediate flower size vs. small flower size, large flower size 

 vs. intermediate flower size, etc., in each case, constitute the two 

 members of a Mendelian pair? 



It is interesting in this connection to note the partial report 

 of an experiment in which peas differing widely in size were 



