70 The Journal 
and pointed as in normal tassels, while, 
in case of the pistillate flowers, these 
parts are shorter, broader, and more 
rounded. 
The terminal inflorescence in tassel 
ear, on the other hand, is always com- 
pact and distinctly ear-like (Fig. 16). 
The glumes and palae are short, broad, 
rounded, and in all respects much like 
those of true ears. This can be seen 
not only in immature tassel ears (Fig. 
14), but in mature ones as well, particu- 
larly when poorly pollinated (Fig. 16). 
The terminal inflorescences of both 
tassel seed and tassel ear are very sub- 
ject to attacks of smut, much more so 
than normal tassels. When attempts 
are made to guard these pistillate flow- 
ered tassels against foreign pollen in 
artificial pollination experiments, the 
smut fungus develops under the paper 
bags used in such work even more than 
when the inflorescences are exposed. 
Moreover, since the silks protrude from 
the sheaths while the upper leaves are 
still closely crowded together owing to 
the short upper internodes at this stage 
(Figs. 9 and 11), it is very difficult to 
protect the silks against accidental pol- 
lination. Either the upper leaves must 
be removed or enclosed with the silks 
in large paper bags. Again, the weight 
of the tassels when the seed has begun 
to develop often causes the tassels to 
break off in storms. But, fortunately, 
it is not necessary to make use of the 
terminal inflorescence in artificial pol- 
linations. If the tassels are removed 
as soon as the silks appear, the true 
ears develop with little delay and can 
be pollinated just as in case of normal 
maize. — Xt 
INHERITANCE OF TASSEL SEED. AND 
SS, A, 
pe TASSEL EAR + m5 
Mention has been made above df the 
fact that these abnormalities are feces- 
sive in inheritance. The original open- 
pollinated tassel-seed specimen  pro- 
duced 28 normal plants. Several tas- 
sel-seed plants occurring as segregates 
in later generations were crossed with 
normals, resulting in 64 normal plants. 
of Heredity 
Various I’, progenies were grown and 
gave a total of 238 normal to 67 tassel 
seed. This is a deviation of only 9.3 
-+- 5.1 seeds from the 3 : 1 relation 
expected when parents differ in a sin- 
gle pair of factors. When F, plants of 
some of these same crosses were back- 
crossed with the recessive tassel seed, 
there resulted 368 normals and 381 tas- 
sel seed, a deviation of only 6.5 + 9.2 
seeds from the expected equality. Four 
self-pollinated F, normals bred true in 
F,, giving a total of 128 normal plants, 
while 10 other F, normals broke up 
again, throwing both normal and tassel 
seed in F,. Evidently, tassel seed is 
differentiated from normal by the single 
factor pair Ts ts. It is assumed that 
the recessive tassel-seed plants would 
breed true if it were possible to test 
them. But, owing to the lack of stami- 
nate flowers, they can neither be self- 
pollinated nor crossed with other plants 
of the same type. 
Tassel-ear plants crossed with nor- 
mals gave 24 normal plants in F, and 
total F, progenies of 260 normal to 36 
tassel ear. This is too great a deviation 
from a 3 : 1 relation to be due to 
chance. The expected numbers on a 
3 : 1 basis with a total of 296 are 222 
and 74, and the deviation is a 38 + 5 
plants. Such a deviation could not be 
expected to occur by chance even once 
in some millions of trials. The possi- 
bility is at once suggested that normal 
and tassel ear differ by two factor 
pairs, and that the F, progenies ap- 
proach a 15 : 1 instead of 3 : 1 ratio. 
But the numbers calculated on this ex- 
pectation are 277.5 and 18.5, a devia- 
tion of 17.5 + 2.8. Even such a devia- 
tion as this would not occur by chance 
more than once in perhaps one hundred 
thousand trials. It is, of course, pos- 
sible that in some crosses the parents 
differ by one factor pair and in others 
by two pairs. But no F, family with 
large numbers approached closely either 
a 3:-1 ora 15 : 1 ratio. 
If two factor pairs. are concerned, 
-about half of the normal F, plants, 
taken at random, should bréed true nor- 
mal, while, in case a single factor pair 
