of teosinte introgression. Of these three, the last, a phe- 
nomenon only recently recognized (Mangelsdorf, 1953), 
may have been the most important. In any case, the 
maize of northwestern Mexico, in a short period of time, 
not more than several centuries at the most, was almost 
completely transformed. 
APPROXIMATE DatTE oF EvoLuTIONARY CHANGES 
None of the maize specimens from these caves has yet 
been dated by radiocarbon determinations. Correlations 
of the cultural manifestations mentioned earlier in this 
paper suggest that the most recent archaeological re- 
mains should be assigned to about 1000 A.D. The sud- 
den changes begin in Level 2 which must represent a 
date slightly earlier, perhaps about 900 A.D. It is in- 
teresting to note that Martin et al (1952) in studying the 
maize from Tularosa Cave, found a gradual decrease in 
average row number which was attributed, in part, to 
the introduction of varieties with low kernel-row num- 
bers from outside the area. It was assumed that some of 
these varieties were 'T'ripsacoid. The change in row num- 
ber was most sudden between the Georgetown and San 
Francisco phases (ca. 700 A.D.). Mangelsdorf and Smith 
(1949) found the earlier maize from Bat Cave to be non- 
Tripsacoid, while the more recent maize, especially that 
from Levels V and VI, included a high percentage of 
Tripsacoid specimens. Cobs from Level VI are dated by 
radiocarbon determinations at 1752250 years (Arnold 
and Libby, 1950) or, at the latest, about 450 A.D. But 
estimates based on pottery put this level at between 500 
and 1000 A.D. More precise dating will undoubtedly 
become possible as the data from radiocarbon determi- 
nations and archaeological manifestations, from various 
sites, are correlated. In the meantime, it will suffice for 
our purposes to conclude that a very marked change in 
[ 178 ] 
