[Vor. 4 
2 ANNALS OF THE MISSOURI BOTANICAL GARDEN 
Nymphaea ‘‘Stella Gurney." This undoubtedly proves the 
parentage of the original ‘‘Stella Gurney,’’ which, according 
to Mr. James Gurney, was a spontaneous seedling through 
insect agency. The seeds of the Brachyceras group are сат- 
ried over the winter in the ponds outside, and readily germi- 
nate the next season during May, while those of Euryale ferox 
germinate even if the ponds have been drained. Nymphaea 
** William Stone’’ produced the same breaking up into blues 
and pinks as * Mrs. С. W. Ward,” and there was no indica- 
tion florally of the pistillate parent, N. flavo-virens, in either 
eross, but it was evident in the tubers and in the extremely 
long petioles. 
Reciprocal crosses were also made between (FV 9? ХИ) х 
(КҮ? Хх 2 roseas)s and (ЕУ? XZ rosea2) ? X (FV9 X 
Z23)6. The only result attained was the intensifying of the 
color of the flowers, whereas the same gradation of the blues 
and pinks appeared. А final cross ([FV9 x Za]9 X 
[FV $ x 2 rosea |) ? X ([FV ? X Z rosea? |? х [FV? х Z 
81) 8 produced a great variation from light pink to violet. 
The violet variety having the small floral character was 
nearest to the parent, N. flavo-virens, to date. As this factor 
was unusual, the flowers were carefully emasculated and self- 
pollinated. During 1915 ten plants were raised and finally 
planted out in 1916 for the summer development. The result 
was six plants of М. capensis var. zanzibariensis, the blue- 
flowered form, and four of N. capensis var. zanzibariensis f. 
rosea, the original pink form. There was no indication in 
either tuber or flowers of N. flavo-virens. The appearance of 
zanzibariensis types is interesting, especially as the type ma- 
terial had not been in the collection for two seasons. 
The impossibility of suggesting the law of heredity can 
readily be seen when some characters are entirely absent and 
others are intensified in the offspring. Another factor to bear 
in mind is that nymphaeas contain several hundred seeds in 
one earpel, while one ovary may contain twenty to forty 
carpels, thus aggregating thousands of seeds in one pod. То 
1FV = М. flavo-virens; Z = №. capensis var. zanzibariensis; Z rosea == 
N. capensis var. zanzibariensis f. rosea. 
