June n, 1917 Hybrids of Zea ramosa and Zea tunicata 387 
The transference of long glumes and the branched conditions from 
tassel to ear might be looked upon as excellent examples of homoeosis. 
There is, furthermore, a large series of abnormalities that occur in the 
terminal inflorescence of full-tunicate plants that lend themselves to this 
interpretation. Spikelets develop into branches, glumes develop into 
pistils, lodicules develop into glumes, and so on. 
On the other hand, it is difficult to think of the nicely graded length of 
the branches of the tassel of Z. ramosa as a translocation of characters. 
This character is better explained as a metaphanic variation or loss of 
differentiation, an explanation that would also apply to the branched 
ear of Z. ramosa and the lengthened glumes on the ears of Z. tunicata . 
But here the variation rather overshoots the mark; for, instead of the 
glumes of the ear being intermediate in length between glumes normal to 
the ear and the tassel, the glumes of the tunicate ear much exceed normal 
tassel glumes, and the tassel glumes are themselves much elongated. 
Neither do the other abnormalities of Z. tunicata appear to be typical 
metaphanic variations, since many of them are repetitions and accentua¬ 
tions rather than intermediate expressions. 
There is a sense in which homceotic as well as metaphanic variations 
may be viewed as a loss or partial loss of differentiation. In typical 
metaphanic variations the normal specialization of parts is replaced by 
organs of intermediate form that may be taken to represent an unspecial¬ 
ized ancestral condition. In homoeosis instead of the normal specializa¬ 
tion certain cells exhibit the characters and potentialities of cells belong¬ 
ing to an entirely different part of the individual. In either instance this 
loss of specialization or the more or less complete return of cells to an 
earlier and less specialized condition may be viewed as a reversion. 
After concluding that Z. ramosa and Z. tunicata may be classed as rever¬ 
sions, it is still possible to look upon them as mutations. Both are wide 
departures from the parent type, and the evidence is that both attained 
this departure at one step, so far as visible variations are concerned. 
With Z. ramosa the case is simple. It is a recessive mutation and the 
dominance of normal maize is complete. The inheritance of Z. tunicata 
is somewhat more complicated. If the interpretation advanced above 
is correct, and the homozygous form is practically sterile and the common 
forms heterozygous, Z. tunicata would constitute a dominant mutation 
in which the dominance is imperfect. 
The characters separating the two mutations are sharply contrasted. 
In Z. ramosa there is no tendency for the glumes to be elongated more 
than in normal maize. In tunicate maize the branches of the tassel are 
no more numerous than in normal maize; and, although the spikelets 
are much enlarged, the transition from branches to pairs of spikelets is 
as abrupt as in normal maize. When ears develop on full-tunicate plants, 
there may be branches near the base of the ear, but these are not at all 
94351°—17-2 
