936 
Journal of Agricultural Research 
Vol. XXXI, No. 10 
carefully planned breeding experiments in which the environmental 
factors are under the control of the operator. Published results of 
such experiments are conspicuous by their absence, and the pres¬ 
entation of data on this subject is the purpose of this paper. 
PRELIMINARY OBSERVATIONS 
Among plats of several varieties of hemp grown at the Bussey 
Institution of Harvard University during the summer and winter of 
1921-22, the writer observed that in addition to the appearance of 
the usual intersex types in the hothouse during the winter there 
appeared a number of monoecious plants in the summer plat of the 
Simple Leaf variety. Almost without exception these individuals 
had the carpellate vegetative development and had been first classed 
as pure carpellate plants. The staminate buds were first observed 
when the seeds were about half mature. As a rule they were not 
restricted to any particular part of the plant, but were fairly well 
distributed among the seed-producing branches. Practically all 
staminate flowers thus produced were identical in appearance with 
those produced by pure staminate plants. Inasmuch as this plat 
of hemp was grown during the summer season, when the occurrence 
of “sex reversal 77 is rare, the production of functional flowers of both 
sexes on single individuals furnishes a starting point for breeding 
experiments to determine the extent to which such a phenomenon is 
hereditary. At first sight it may seem that such a case of “gex 
reversal 77 is different from that which occurs so commonly in hot¬ 
houses during the winter. The two cases are fundamentally the 
same, however, and differ only in degree. The only difference lies 
in the fact that the abnormal flowers so common during the winter 
were absent from the monoecious Simple Leaf plants. The real 
question to be answered in either case is, what is the difference 
between the pollen produced by a carpellate type and that produced 
by a pure staminate type? 
The sex ratios observed in the preliminary experiments (Table I) . 
show a few peculiarities of minor importance. The most common 
condition seems to be about equal numbers of carpellate and stam¬ 
inate types. The minor fluctuations from this may be due to differ¬ 
ential germination, death, or sampling of the seed. The sex ratio 
in the Simple Leaf variety, however, may be due to other causes 
not yet known. Since this variety arose as a mutation, it is suggested 
that its sexual state may be genetically different from that of the 
other varieties tested. 
Table I. — Sex ratios in different varieties of hemp 
Variety 
Environment 
Number 
Sex ratio 
Notes • 
Ohington 
Field plat... 
472 
136 
341 
401 
382 
124 
103 
105 :100 
106.: 100 
130 :100 
212 :100 
104 : 100 
100 :100 
93 :100 
No intersex types. 
Do. 
1 intersex. 
56 monoecious. b 
1 intersex. 
44 intersexes. 
17 intersexes. 
Xftjn .. _ 
.do__ 
Tochymington_ 
_do____ 
Simple Leaf _ _ _ 
.do__ 
C5hinfft.nn 
Hothouse in summer. .. 
t 
1 
1 
1 
1 
1 
1 
• 
1 
1 
1 
1 
1 
J 
o c 
jne 
Hothouse in December-January__ . _ _ 
_do.... 
a All plants which produced abnormal flowers are, for convenience, classes as intersexes. 
& These plants were classed as carpellate in the sex ratio. 
