way of determining their genotype, However, since we 
can duplicate the principal characteristics of such speci- 
mens by combining the components of the Zu locus on 
chromosome 4 with the major modifying factor 7% on 
chromosome 6 and with other modifying genes from 
various popcorn varieties and since neither we nor our 
critics have yet found any other way to duplicate these 
characteristics, we shall continue to regard these early 
Fic. 2. Diagrammatic cross section of the early Bat Cave cob 
od 
KA FT illustrated in Plate II, C and D. The slender rachis (inner 
ORNS No circle), the long rachillae (distance between rachis and base of 
, a ie Ly kernels), the long glumes (solid line), and the paleas (broken 
" . /--\ NY ) line) which almost enclose the kernels, all indicate that this is 
SE a form of pod corn. 
cobs as representing a form of pod corn and we shall ad- 
here to the conclusions of Sturtevant and of Mangelsdorf 
and Reeves that primitive corn was both a popcorn and 
a pod corn. 
Changes in the Rachis Diameter 
We shall, however, modify the conclusion reached in 
the description of the cobs from the 1948 collection in 
which we attributed changes in the cob/rachis ratio 
largely to replacement of higher alleles of the T'u-tu 
locus by lower alleles including one designated as tu”. 
We have not in our experimental cultures been able 
definitely to establish the existence of such an allele or 
if it exists to distinguish its effects from those of various 
modifying factors whose role, in causing variation in the 
diameter of the rachis and the length of the glumes, was 
postulated in the 1949 paper. Changes in the cob/rachis 
ratio in this material now appear to be more a matter 
of increase in the diameter of the rachis than in the re- 
duction in the length of the glumes. The data in Table 
II, based on 430 cobs in which the rachis diameter could 
be measured, show that there has been an increase in the 
[8] 
