298 
Journal of Agricultural Research 
Vol. XXV, No. 7 
having tested these variations by interhybridization, it seems certain that 
at least the smaller of our dwarfs is identical with that known as dwarf 
by Emerson. 
This latter form has been used in the hybrids described in this paper, 
and to obviate circumlocution is referred to throughout as dwarf. The 
reader should bear in mind, however, that close somatic resemblance is 
no certain indication of genetic identity, though the very close similarity 
of all the teratological characters of these andromonoecious forms raises 
the question as to whether these variations are not due to some common 
causes. The andromonoecious dwarfs and semidwarfs seem always to 
appear as simple Mendelian segregates formed of a complex of characters 
whose component parts, or at least characters very similar in appearance, 
are known to occur separately and in different combinations in other 
nondwarf strains. The most striking characteristic of dwarf is its 
greatly reduced stature, which often is less than one-fifth that of normal 
sister plants, although there seems to be no compensating increase in the 
diameter of the culm. The leaves are reduced in length and increased 
in width, entirely altering the normal proportions and giving the plants 
a peculiar tobacco-leaved appearance. Associated with the reduction 
in stature, there is also a proportional reduction in the size of the tassel, 
and the number of branches seldom exceeds three. Perfect flowers or 
staminate spikelets with large anthers are found throughout the ear, 
which usually terminates in a staminate spike, but, notwithstanding this 
excess of staminate development, pollen is shed sparingly and the anthers 
are rarely fully exserted, and even more rarely dehisce. The plants 
seem exceptionally vigorous and sturdy and the leaves are a very dark 
green. The reduction in stature is accomplished entirely through a 
shortening of the intemodes—not through a reduction in their number, 
which remains the same as in normal sister plants. In this respect the 
plants of dwarf resemble the plants of brachytic, a type of dwarf in which 
the stature only is reduced (PI. 2, 3). 
The brachytic type also has appeared in unrelated stocks. The chief 
characteristic of this variation is a shortening of the internodes, which 
includes also the homologous parts of the tassel, resulting in a reduction 
of the branching space. This reduction seems unaccompanied by other 
changes in the tassel, with the possible exception of a slight increase in 
the number of tassel branches. There are a few other minor changes, 
such as an increased diameter of the stalk, but nothing of a striking 
nature comparable to the reduction in stature. The leaves are of the 
same size and proportion as in normal plants and there is no tendency 
to produce perfect flowered ears. There is evidence, however, that the 
brachytic, like the andromonoecious dwarf, is associated with the develop¬ 
ment of staminate spikes on the ears (<?). 
FIRST GENERATION 
In view of their common characteristic of shortened internodes, it 
might be supposed that crosses between dwarf and brachytic plants 
would produce nothing but plants of short stature. However, such is 
not the case, for the first generation of such hybrids consists of normal 
plants fully as tall as the normal plants from which the immediate 
brachytic parent is derived, the observed height being 21.0 ±1.03 dcm. 
These plants are also normal with respect to all other teratological 
characters of their parents. While it is not uncommon to find that 
