great and actually declines again in the highest row num- 
ber class. These results are again explicable in terms of 
modern maize. Normally heterotic vigor tends to increase 
row numbers within certain limits. In the archaeological 
material, the highest row number class probably tends 
to be beyond the range of introgression. As might be 
expected, the lack of heterosis in length of Tonto cobs 
is repeated in regard to kernel-row number. That is, 
introgression has only a general effect in reducing row 
numbers. 
A third factor distinguishing these collections lies in 
the relationship between cob diameter and introgression. 
Although these characters are independent at Richards’ 
Caves, diameters increase significantly with introgression 
in Tonto cobs. Thus, the more Tripsacoid Tonto cobs 
have greater diameters, lower row numbers and shorter 
ears. The first two of these characters suggest introgres- 
sion from the indurated Northern Flints, but the third 
one seems to rule this out since these flints have long ears. 
Fasciation and Row Number. Introgression tends to 
eliminate fasciation, or flattening of the rachis, by caus- 
ing a slight reduction in row number. Nevertheless, in 
comparison to modern maize, fasciation is extremely fre- 
quent in the lower row numbers. In modern maize, fas- 
ciation is usually confined to row numbers well above the 
range of this prehistoric material. These facts suggest 
that teosinte introgression may so modify the expression 
of genes for fasciation that increases in row number are 
produced without the association of a flattened rachis. 
SUMMARY 
1. Among archaeological cobs from Tonto and Rich- 
ards’ Caves in Arizona, there are many specimens which 
resemble, in appearance, those which occur in the F2 and 
backcross generations of maize-teosinte hybrids. 
[121 ] 
