thetical ancestor of modern corn—in which the silks are 
exposed to pollen for their entire length. Also from 
Weatherwax’s laboratory has come what is perhaps the 
most significant evidence on this point. Farquharson 
(14) demonstrated that certain varieties of Tripsacum, 
when used as female parents, crossed readily with corn 
without the use of special techniques. She states: ‘‘In 
fact, it seems highly probable that this cross has occurred 
occasionally in nature. ’”’ 
Randolph’s objection to this part of our theory is based 
on his own experiments in crossing Mexican and Guate- 
malan varieties of corn with Mexican and Guatemalan 
Tripsacum. Although these experiments may seem to 
have the virtue of directness, they are far from critical 
in several respects. The number of ears pollinated, 612, 
seems impressive at first glance, but a critical analysis 
suggests that successful crossing on a scale proportional 
to the numbers involved could hardly have been ex- 
pected, especially because of the high frequency in these 
areas of the Ga gene which produces cross sterility. 
These experiments likewise fail to take into account two 
important facts: (a) that the corn and Tripsacum of to- 
day is not the same as was that of ancient times; (b) 
that even the modern types of corn and 'Tripsacum of 
the region have not been completely sampled. 
Cutler and Anderson (11), in a survey of the genus 
Tripsacum, did not find 7°. dactyloides south of the United 
States-Mexican border. Fourteen years later, Randolph 
(82) stated twice that 7. dactyloides has still not been 
reported south of the Mexican border and that this spe- 
cies therefore could not have been involved in the as- 
sumed recent hybridization with corn in Guatemala or 
neighboring areas. Apparently, he was not aware that 
Melhus (29) two years earlier had reported the occur- 
rence of this species in Guatemala. Mangelsdorf (unpub- 
[ 359 | 
