64 
PACIFIC SCIENCE, Vol. IX, January, 1955 
age of less than three seeds per capsule was 
obtained. 
The parental plants were selfed to see 
whether they were true breeding. Upon segre- 
gation of the offspring the ratios were re- 
corded. Several crosses among the parental 
plants were made, and the resulting Fi types 
noted. The Fi plants were subsequently selfed 
or, where convenient, backcrossed to get the 
segregating ratios. 
RESULTS AND DISCUSSION 
Flavone Colors 
white. With the white flowers, the limb of 
the corolla is self-white, but the tube is pale 
yellow, as mentioned earlier. The whites were 
found to breed true for white. Also, when 
two whites were crossed, only whites were 
obtained. When these white flowers were ex- 
posed to ammonia vapor, the corolla turned 
yellow as it does in many other flowers (Bux- 
ton, 1932; Lawrence, 1931; Mehlquist, 1939; 
Scott-Moncrief, 1936). When such a reaction 
takes place, it is believed that an ivory an- 
thoxanthin, probably apigenin, is present, 
which upon treatment with ammonia is con- 
verted into a yellow anthoxanthin, probably 
luteolin. 
yellow. The yellow pigment in this species 
is a water soluble flavone, which turns orange 
when placed in ammonia vapor, similar to 
that found in Dahlias (Lawrence, 1931). The 
coloration occurs in the inner epidermal layer 
of the limb as well as the hypodermal layers. 
The yellow parent plants bred true for yellow. 
When yellow flowers were crossed with white 
flowers, all of the Fi flowers were yellow. 
The subsequent F 2 generation segregated into 
a 3:1 ratio and the backcross to the white into 
a 1:1 ratio. The results are shown in Table 1. 
Yellow, then, is a simple dominant to white 
and Y has been designated as the gene neces- 
sary for the production of yellow pigment. 
Gene Y when homozygous recessive results 
in white flowers in the absence of other flower 
colors. 
Anthocyanin Colors 
A close examination of the purple colored 
flowers revealed that the anthocyanin sap pig- 
ment occurs in two definite patterns, one in 
which the pigment is found in the inner 
epidermal layer of the limb and the other in 
which the coloration occurs in the outer epi- 
dermal layer of the limb and extends to the 
base of the tube. In order to make a distinc- 
tion between the two patterns of coloration, 
the former will be designated hereafter as 
purple, and the latter as purpleback, due to 
the fact that the back of the tube is colored 
in addition to the outer epidermis of the limb. 
The anthocyanin pigment is basically purple, 
with its variations ranging from bluish purple 
to purplish red. It appears that where the back- 
ground is white, bluish purple seems to exist, 
whereas when the background is yellow, the 
color tends to be reddish purple or bronze. 
The purple pigment also seems to be affected 
by the light intensity, and the age of the 
blossom. With aging, the color tends to turn 
bluish. 
TABLE 1 
The Ratio of Yellow to White Flowers in the F 2 and Backcross Progenies 
of White X Yellow in Asystasia gangetica 
OBSERVED 
CALCULATED 
PROGENY 
CHI-SQUARE 
Yellow 
White 
Yellow 
White 
f 2 
33 
9 
31.5 
10.5 
.285 
Backcross 
25 
26 
25.5 
25.5 
.020 
Chi-square == 3.841 at 5 % level. 
