496 
Journal of Agricultural Research 
Vol. XXVII, No 7 
Inspection of the frequency distributions of the four F 2 progenies 
(Table III) shows sharp segregation into a spotless and a spotted group. 
The percentages of spotless individuals in the F 2 progenies and the 
departures from the 25 per cent expected with a single factor difference 
are stated in Table V. It is evident that the departure is significant 
in none of the progenies and that while the population obtained by 
combining the four progenies as one array shows a slight excess of spot¬ 
less individuals, the departure from 25 per cent barely exceeds its prob¬ 
able error. 
Table V .—Percentages of spotless individuals and departures from the expectation, in 
the F 2 progenies grown in IQ22 
F2 progeny. 
Number of 
plants. 
Percentage 
spotless.® 
Departure 
from the 
expectation 
(25 percent). 
£1-3-12-14X12) 20. 
47 
34-0 
9 - 0 ± 4.66 
(1-3-12-14X12) 28. 
57 
26.3 
i-3±3-93 
( 3 - 2 - 4 -iXi 3 ) 21,... 
28 
25.0 
o± 5 -Si 
(3-2-4-1x13) 24. 
47 
23-4 
i.6±4. 16 
4 F 3 progenies as one array. 
179 
27-3 
2.3 i 2.24 
a The probable error of the percentage is omitted, being the same as that of the departure from the expec¬ 
tation. 
Reference to Table IV shows that the mean of the spotless segregates 
in each F 2 progeny does not differ significantly from the mean of the 
progeny representing the corresponding spotless grandparent, except in 
progeny (1-3-12-14 X 12) 20, in which the spotless segregates gave a 
significantly lower mean. On the other hand, the means of the spotted 
segregates in F 2 are in all cases significantly lower than the mean of the 
progeny representing the corresponding spotted grandparent and 
approach the means obtained in Fj (Table II). This, of course, is to be 
expected if dominance is incomplete, as was indicated by the results in 
F r The incompleteness of the dominance is confirmed by consideration 
of the variability in F 2 . The spotless group in each F 2 progeny is not 
much more variable than the corresponding spotless parental population, 
as is made evident by comparison of the frequency distributions (Table 
III) and of the standard deviations (Table IV). On the other hand, 
both comparisons show the spotted group in each F 2 progeny to have 
been much more variable than the corresponding spotted parental 
population and the spotted groups in F 2 were likewise more variable 
than F x (Table II). 
That the heterozygotes of the spotted group are partly distinguishable 
from the pure dominants is indicated by the marked bimodality of the 
distributions of the spotted plants in F 2 progenies (1-3-12-14X 12) 20 
and (3-2-4-1X13) 24 (Table III). Because of the relatively small 
numbers involved and the nearly complete dominance of the spot, as 
well as the sensitiveness of this character to environmental influence, it 
is impossible to resolve the phenotypic ratio of approximately 3:1 into 
a 1:2 :1 ratio. The point will be further discussed in considering the data 
from F 8 . 
