3^4 
Journal of Agricultural Research 
Vol. IX, No. ix 
special inquiry for this type of maize among well-informed people 
familiar with Indian agriculture, there is no mention of tunicate maize. 
In hybrids with nontunicate varieties the tunicate.character behaves 
as a dominant, but in our experiments we have never been able to secure 
a homozygous tunicate strain. Progenies resulting from the selfing of 
tunicate plants have, with us, always shown segregation into approxi¬ 
mately three tunicate plants to one normal. 1 
The tunicate plants in self-pollinated progenies are separable into 
two classes, one producing typical tunicate ears and thickened tassels 
like the parent plant (PI. 13, B; 14, B ), the other with greatly enlarged tas¬ 
sels containing both staminate and pistillate flowers, and with the ear 
either aborted or bearing greatly enlarged and usually sterile spikelets 
(PI. 13, A; 14, B; 15). This last class represents approximately one- 
third of the tunicate plants. Although these plants produce what 
appears to be normal pollen in the terminal inflorescence, the long glumes 
never open and the pollen is not shed; and we have not been successful 
in securing selfed seed of this form. 
The ratios in which the different classes occur would indicate that the 
class with the bisexual terminal inflorescence is the homozygous form 
and that the ordinary tunicate plants represent the heterozygous form, a 
cross between the form with the bisexual inflorescence and the normal 
nontunicate maize. If the ordinary tunicate type can occur in a homo¬ 
zygous form, we should expect one in four of the plants grown from a 
self-pollinated tunicate plant to be homozygous, and the progeny of 
such homozygous plants should be all podded, whether cross or self- 
pollinated. This has not been the case in our experiments. If only a 
few progenies were grown, the failure to secure an all-tunicate progeny 
might, of course, be ascribed to the accidental selection of heterozygous 
instead of homozygous parents. 
In the course of our experiments the progenies of 43 different tunicate 
plants have been grown and in all of these progenies, except one, non¬ 
tunicate plants appeared. The one exception produced only eight plants. 
And, since only two or three normal plants were expected, it is not sur¬ 
prising that none appeared. Of the remaining 42 parent plants, 14 
might have been expected to prove homozygous. That none proved 
to be homozygous can hardly be accidental, since the chances against 
it are over 400,000 to 1. It is therefore concluded that, in the material 
that has come under our observation, the ordinary type of tunicate 
plants represents a case of imperfect dominance, and that it is unfixable, 
1 A comparison of the ratios of tunicate to nontunicate plants shows the following: 
Tunicate. 
Self and tunicate Xtunicate. 288 
Expected. 286. 5 ±5. 7 
Tunicate Xnontunicate. 99 
Expected. 108 ±5.0 
Non¬ 
tuni¬ 
cate. 
94 
95 - 5 
117 
108 
