sinte and Tripsacum according to species and strains, 
but neither average numbers nor exact ranges in num- 
bers for the various groups have yet been adequately de- 
termined. The most recent chromosome-knob count 
which has come to our attention for 7°. dactyloides (n= 
18) is 22 to 26 (82). This is greater than any number yet 
reported for corn or teosinte, and even if this form of 
Tripsacum be regarded as tetraploid, its number of knobs 
per genom of nine chromosomes is greater than the aver- 
age of either corn or teosinte. As to corn and teosinte, 
perhaps the statement will stand without contradiction 
that, broadly speaking, corn has the lowest number and 
annual teosinte an intermediate number. 
As to position of knobs, those of 7. dactyloides have 
a strong tendency to be terminal, but Ting (49) has 
shown that in another form of Tripsacum some of them 
are obviously intercalary. Most of the knobs of corn are 
intercalary, but a few are terminal; the most common 
position is sub-terminal, or at least closer to the end than 
to the centromere. Annual teosinte is again intermedi- 
ate; more of its knobs are terminal than in corn, but 
more are intercalary than in 7Tvripsacum dactyloides. 
Many varieties show pronounced differences in this 
character; according to Longley (17), the knobs of the 
Guatemalan varieties from Progresso, Moyuta, Nojoyo 
and San Antonio Huixta are mostly terminal, but those 
of the Mexican varieties from Durango and Chapingo are 
more often intercalary, like those of corn. The interme- 
diate position occupied by teosinte, in both number and 
position of chromosome knobs, is one of the several char- 
acters which may be explained by the view that teosinte 
is a hybrid combination of knobless, pure corn and a form 
of Tripsacum similar to 7°. dactyloides with many knobs. 
No objection to the view that teosinte occupies an in- 
termediate position in these respects has come to our 
[ 369 ] 
