Feb. 23,1924 Teosinte Maize Hybrids 577 
The two negative coefficients with length of best inflorescence, and 
the positive correlations with days pollen to silk, are all coherences 
expected on the hypothesis that the ramose variation is correlated 
with maize characters. None of the other characters listed have signi¬ 
ficant coefficients, but it is of interest to compare this series of relation¬ 
ships with those shown for the crinkly variation in Tables IV and V. 
The absence of correlation between the ramose type and the number 
of rows of alicoles on the best spike seems worthy of note. Thus it 
seems clear from this hybrid that the ramose parent introduced the 
necessary factors for the production of normal unbranched ears quite 
independently of the branched character of the inflorescence, and, as a 
consequence, unbranched inflorescences were obtained closely resembling 
normal maize ears. 
BRACHYTIC X TEOSINTE 
Brachytic is a dwarf variation of maize in which the length but not 
the number of intemodes is reduced, and all other organs remain full- 
sized. In crosses with plants of normal stature this type of dwarf 
behaves as a simple Mendelian character recessive to the normal form, 
reappearing in approximately 25 per cent of the F 2 plants, apparently 
unaltered as to stature (11). 
Only one F x plant was obtained of the hybrid with teosinte, and from 
this self-pollinated plant a second generation of 290 plants was grown. 
Of these plants only 35, or 12.1 ± 1.3 per cent were classed as brachytic, 
a departure from the expected 25 per cent, and too significant to be 
ascribed to chance. One of these plants is shown in Plate 6. While 
in many maize crosses the percentage of brachytic plants recovered in 
the F 2 , like that in the case of most recessive characters, is below the 
expected, the departures usually are within four times the error. The 
percentage obtained in the teosinte hybrid is not referable readily to the 
interaction of two independent factors, since it lies midway between 
the 6.25 per cent expected where the character only appears when two 
recessive factors are homozygous, and the 18.75 per cent expected in 
cases where the character appears on the interaction of a dominant and 
a recessive factor. 
Not only is the number of brachytic plants too low for a simple reces¬ 
sive character, but the distinction between normal and brachytic plants 
is not as clear as in crosses with normal maize. This difference between 
the maize and the teosinte hybrids is shown by a comparison of the 
graphs in figures 76 and 83. The failure to obtain definite segregation 
probably accounts in a large measure for the low number of brachytic 
plants, and perhaps is due to the introduction of modifying factors 
through the teosinte parent. 
Two plants *were found that had brachytic intemodes on the lower 
half and normal intemodes above, and one plant was found with a 
brachytic main stalk and normal suckers. Such instances are not 
unique nor confined to the hybrid with teosinte, similar plants having 
been found in maize hybrids after repeated back crossing of brachytic 
with normal plants. 
All varieties of maize exhibit what may be construed as an indication 
of brachysm, and it is possible to view the origin of the brachytic variation 
as an example of homeosis, or the transfer of the shortened intemodes 
of the ear stalk to those of the main culm. 
