may be asuperficial one. This is probably the case. We 
expressed the opinion some years ago (36) that the flint- 
flour corns of the Plains Indians, some of which are in- 
cluded in Brown’s Northern Flints, show little admixture 
with Tripsacum in spite of their straight rows. Addi- 
tional experience has confirmed this earlier impression. 
The origin of the Northern Flints is still obscure, but the 
corn in Latin America which most resembles them occurs 
at high altitudes in Guatemala, is non-tripsacoid, and has 
low chromosome knob numbers (55). 
The question whether or not corn is improved by teo- 
sinte introgression is answered in part by circumstantial 
and in part by direct evidence. Vachhani (56) found no 
correlation between chromosome knob number and vari- 
ous morphological and agronomic characteristics, includ- 
ing yield, in 20 inbred strains, but added that these results 
are not necessarily in conflict with those of Mangelsdorf 
and Cameron since only relatively low knob numbers 
were involved. Three additional papers (overlooked by 
Randolph) contain evidence which suggests or shows that 
teosinte introgression, in some instances at least, results 
in improvement. 
Wellhausen et a/ (58) concluded that the more produc- 
tive races of maize in Mexico had undergone introgres- 
sion from teosinte and that some of the most valuable 
races, such as ‘Tuxpeno and Celaya, are the product of 
several independent introductions of teosinte germplasm 
into maize. More recently, Wellhausen and Prywer (54) 
showed that among inbred lines developed from Mexican 
varieties adapted to elevations from 4500 to 6000 feet, 
those with the higher knob numbers tended to be the 
parents of the more productive hybrids. The reverse was 
true at higher altitudes. These results are consistent with 
those of Mangelsdorf and Cameron which showed that, 
in Guatemala, teosinte introgression is slight at altitudes 
[ 894 ] 
