574 
Journal of Agricultural Research voi. xxvn. No. s 
Height is not a feature of the ramose variation, as it is of crinkly, and 
no bimodality in the various measures of this character was to be expected. 
The figures will facilitate a comparison of the distributions of the plants, 
with respect to these characters in this hybrid, with the distributions of 
the crinkly and brachytic hybrid. 
A comparison of the distribution of the first six characters in the 
crinkly (fig. i to 6) and ramose hybrids (fig. 45 to 50) shows a striking 
similarity, and emphasizes the failure of the semidwarf stature of crinkly 
to behave as a simple character. 
As in the case of the hybrid with crinkly, many F 2 plants of the ramose 
hybrid were proterogynous, though the mean number of days from 
pollen to silk for the whole population and for each of the two groups 
comprising it was a positive value. 
Bimodality is evident in the distributions for the several characters 
indicating the type of staminate inflorescence (fig. 51 to 55), and a very 
high correlation with the ramose type also is apparent. It must not 
be overlooked, however, that in each case there are intermediate plants. 
The frequency distributions that measure the branching of the pistil¬ 
late inflorescence (fig. 61 and 66) also are bimodal, as well as the alicole 
index (fig. 71), which in this hybrid presents the best evidence as yet 
obtained for the discontinuous inheritance of this character. The F, 
plants have a high percentage of single female alicoles, showing a partial 
dominance of this teosinte characteristic over the maize parent, and 
bimodality of the F 2 is therefore not unexpected. The majority of the 
plants would be expected to have a high percentage of single female 
alicoles, which proves to be the case; and in this respect the ramose 
hybrid resembles the crinkly hybrid and, like it, differs from the hybrid 
with Tom Thumb (5). 
It will be observed (fig. 67) that none of the plants classed as ramose 
had double female alicoles. This condition may be related to the almost 
complete sterility of these plants, especially as they approach teosinte 
in the character of the number of rows of alicoles. Of the very few 
seeds borne by these ramose plants, all were on inflorescences that 
approximated maize in the number of rows of alicoles, and none was 
borne by the plants which resembled teosinte. Several plants with 
branched pistillate inflorescences were found that had but two rows of 
alicoles; but these plants were so completely sterile that there was no 
evidence of even the formation of spikelets; and the feathery branches, 
though not at all succulent, suggest the sterile “cauliflower” inflorescence 
of the pod-ramose hybrid (3). (See PI. 7.) In none of these two- 
rowed inflorescences was there found more than one branch from an 
alicole, though these inflorescences sometimes were ramified further. 
The almost complete sterility of the ramose segregates of the teosinte 
hybrid would account partially for the high percentage of plants with 
single female alicoles; and it is not without significance from the stand¬ 
point of the origin of the ear, suggesting as it does that teosinte is even 
further specialized with respect to branched inflorescences than is maize. 
This sterility indicates that the change of normal maize to the ramose 
condition is not in the nature of a reversion to teosinte. While it is 
possible to obtain a complete series of fertile forms connecting the pis¬ 
tillate teosinte inflorescence with the inflorescence of maize, and a similar 
series connecting the normal maize inflorescence with that of the ramose 
type, it is not possible so to connect the ramose inflorescence with that 
