386 
Journal of Agricultural Research 
Vol. IX, No. ii 
Unlike Z. tunicata , Z. ramosa is a recessive variation. The dominance 
of normal maize over this variation seems complete. We have never 
been able in any way to distinguish between plants heterozygous for the 
ramosa character and normal maize. So far as observed, the character 
behaves as a simple Mendelian unit. 
Krafft (1870) described and figured a branch pistillate inflorescence 
that appears to have been of the same character as Z. ramosa . The 
reference is of interest as indicating that Z. ramosa , as well as Z. tunicata , 
may be looked upon as having originated more than once. 
NATURE OF THE VARIATIONS 
In normal maize one of the characters differentiating the male and 
female inflorescence is that the glumes are greatly reduced in the pistil¬ 
late inflorescence, so that the kernels are naked on the cob. In Z. tuni¬ 
cata this differentiation is lost, and the kernels are inclosed in glumes as 
long as those of the staminate inflorescence. 
Another specialized character of normal varieties of maize is the sup¬ 
pression of the branches in the pistillate inflorescence, or ear. In Z. 
ramosa this specialization is lost, and the ear is as completely branched 
as the tassel. A further loss of specialization in Z. ramosa is that the 
differentiation between branches and pairs of spikelets is lost, the one 
grading into the other in both ear and tassel. 
In considering these losses of specialization it is desirable to keep in 
mind the fact that in Z. tunicata, as in ordinary maize, the ear is a pistil¬ 
late h’omologue of the central spike of the tassel, while in Z. ramosa the ear 
is a pistillate replica of the entire tassel. The loss of specialization in 
Z. tunicata affects the characters of the floral organs and spikelets, while 
in Z. ramosa the general form of the pistillate inflorescence is changed to 
conform to that of the tassel. 
Both Z. ramosa and Z. tunicata are variations from normal maize 
toward the general type of grasses, and as such may be looked upon as 
reversions, since both cases involve a loss of a specialization that dis¬ 
tinguishes maize from practically all other grasses. 
Recent investigations have shown that many reversions may be 
explained as recombinations, but neither Z. ramosa nor Z. tunicata 
results from the recombination of latent factors,* or cryptomeres. Had 
the first appearance of Z. ramosa and Z. tunicata been the result of the 
fortuitous combination of factors, or cryptomeres, it would seem to follow 
that when the new combination was crossed with the parent stock these 
factors should have been again redistributed and the combination neces¬ 
sary to bring the new characters again into expression should have 
occurred much less frequently than the observed one in four individuals. 
Both of these variations may also be considered either as examples of 
homoeosis (Eeavitt, 1909) or as metaphanic variations (Cook, 1915). 
