5°6 
Journal of Agricultural Research voi. xxvii, No. 7 
The variation in grade of petal spot on the individual plant appears 
not to be due solely to variation in factors of the external environment. 
This is indicated by comparing flowers graded on the same plant on the 
same day. Of such comparisons, in 1923, 75 showed a difference of 2 
grades, 31 of 2)4 grades, 24 of 3 grades, 6 of 3K grades, 2 of 4 grades and 
3 of 5 grades. It may be argued, however, that since it is a rare occurrence 
for more than one flower to open on the same fruiting branch on the same 
day, these differences in the petal spot of flowers opening simultaneously 
on the same plant may be attributed to variations in the environment, 
either external or internal, during the period when the successive fruiting 
branches were being developed. 
The Fj hybrid progenies grown in 1921 and the F 3 hybrid progenies 
grown in 1922 did not show greater mean variation in petal spot on the 
individual plant than did the corresponding parental progenies. These 
facts, together with the absence of significantly greater individual varia¬ 
tion in the heterozygous than in the recessive plants of the segregating 
F s progenies grown in 1923 (Table XIV) indicate that the type of “veg¬ 
etative segregation” recently described by Gates 8 ' 9 does not occur in 
this case. 
RELATION OF PETAL SPOT TO SIZE OF THE FLOWER 
It has been noted, in normal Pima populations, that the petal spot often 
is less well developed in the small and sometimes misshapen flowers borne 
by stunted plants than in the larger flowers borne by well-grown plants. 
The spotless families described in this paper showed no apparent inferior¬ 
ity to the spotted families in the vegetative vigor and fruitfulness of 
the plants, when grown under comparable conditions. It has been 
observed, however, in grading petal spot in the spotless populations that 
flowers showing no trace of the spot often have a small corolla and 
relatively few and small stamens, while flowers of normal size in the same 
populations often exhibit a faintly developed spot. 
Comparison of the mean corolla length of spotted plants and of spotless 
plants in the F 2 progenies of 1922 showed a difference of only 1.5 ±0.39 
mm. in favor of the former, the mean length having been 60.4 ±0.20 
mm. in the spotted class (276 flowers) and 58.9 ±0.33 mm * ^ the 
spotless class (123 flowers). In 1923 a dominant and a recessive F 3 
progeny were compared, with negative results. The two progenies, 
descendants of the same F x plant, were situated in adjacent and con¬ 
terminous rows, hence soil heterogeneity was not a factor. One corolla 
was measured on one plant in each progeny, 41 flowers having been 
measured in the dominant progeny and 42 flowers in the recessive progeny. 
The mean corolla length was 62.3 ±0.36 in the former and 63.0±0.39 in 
the latter, the difference amounting to only 1.3 times its probable error. 
Although these comparisons of the mean corolla length of the spotted 
and of the spotless populations showed little or no difference, a signifi¬ 
cant positive correlation between grade of petal spot and length of corolla 
was found to exist within each class. In the F 2 progenies of 1922, 276 
flowers of the spotted class on which both characters were measured 
8 Gates, R. Ruggles. vegetative segregation in a hybrid race. In Jour. Genetics, v. 6, p. 237- 
253. pl. 9* 1917. List of references, p. 253-253. „ , ^ _ 
9 - A PECULIAR TYPE OP VARIABILITY IN PLANTS. In JOUT. GenetlCS, V. 13 , p. I3~45. 24 tig. 1923 . 
Literature cited, p. 44~45- 
