most serious exception to this sequence of dates is the 
Maiz de Ocho from Stalling’s Island Mound, Georgia, 
where the major occupation probably pre-dates the 
Christian era. It is not possible, however, to determine, 
without radiocarbon dating, whether this eight-rowed 
maize is from this earlier occupation or from a later 
historic occupation (A.D. 1600-1700) which followed 
at the same site. The later occupation seems more 
likely. 
If Maiz de Ocho did originate in the highlands of 
Colombia, then its poor adaptation to the lowlands in 
the southern part of the Southwest might be expected. 
But a flow of germplasm from local races such as Chapa- 
lote, Reventador and Tabloncillo has apparently tended 
to acclimatize it to this area. ‘Thus, Harinoso de Ocho, 
as it lingers on, is extremely variable and mixed. Each 
ear from two collections of Harinoso de Ocho recently 
received from the Rockefeller Foundation in Mexico is 
different in size and shape and has thin dented kernels 
rather than thick floury kernels, as described for this race 
by Wellhausen et al. (1952). Although the kernels of 
Harinoso de Ocho are thinner (4.4 mm.) than those of 
Cabuya (6.36 mm.), they are still thicker than those of 
all other Mexican races except one, Cacahuacintle, which 
is thought to be also from South America and to be re- 
lated to Cabuya (Wellhausen e¢ al., 1952 and Grobman 
et al., 1961). The mixed nature of Harinoso de Ocho is 
also apparent in archaeological collections from north- 
western Mexico (Mangelsdorf and Lister, 1956) and ad- 
joining areas of the Southwest, as in a collection from 
several sites in southwestern New Mexico reported by 
Cosgrove (1947) and represented in Plate X XIII. This 
collection shows the distinct elements which were 
blended during the evolution of maize in the Southwest 
as follows: Chapalote (figs. 1, 2,38), tripsacoid Chapalote 
[ 128: | 
