1921] SCHAFFNER—HEMP 213 
dwarfed ordinary carpellate plants are also defective and do not 
sprout so readily as those grown under normal conditions. 
Seeds collected from carpellate plants which later developed 
stamens were planted in plot 15. ‘There was apparently no differ- 
ence in the behavior of the plants from these seeds from those raised 
from seeds obtained from pure parents out of doors, but the number 
of plants was too small to draw any definite conclusions. One 
would expect that the embryos of seeds developed under abnormal 
conditions would show less fixity of the sexual state than those 
developed under normal conditions. Since practically all the seeds 
from the out of doors “pure” parents will show reversal of sex 
with proper abnormal conditions, however, probably the only way 
to get any definite data indicating a difference would be to test such 
seeds out of doors in comparison with ordinary seeds. In such 
an experiment, however, the plants should be grown in an environ- 
ment which will just keep the plants from normal seeds pure. 
Such limits can probably be discovered. 
Recent work 
Various investigators have made observations on the sex ratio 
of hemp. It is not necessary to refer to the older papers here 
except to state that a considerable diversity in the proportion of 
the staminate and carpellate plants has been found. Roughly 
speaking, the ratio of carpellate to staminate plants is 1:1, with 
a deviation in either direction of at least 50 per cent in extreme 
cases, even when large numbers are counted. 
In 1916 PricHarpD (2) published his results on changing the 
sex in hemp by mutilation. By removal of leaves and flowers, and 
by certain other treatments, he was able to obtain 17.8 per cent of 
reversals in the established sexual state. In all, 25 carpellate 
and 4 staminate plants showed reversal of sex. 
YAMPOLSKY (7), working with Mercurialis annua L., found 
that some carpellate plants produced staminate flowers and some 
staminate plants produced carpellate flowers, and that both 
staminate and carpellate plants showed gradations in degree of 
maleness and femaleness. He also found that the offspring of 
