PONOBa7) 151 
Proposition 4. There is a resistance to current hybridization 
between the races of a species. 
Observations of a similar character have also been made within 
the agricultural species upon the effects of current cross-fertilization 
between the races, and of cross-fertilization upon the succeeding 
crop. This study, however, is more uncertain on account of the 
difficulty of recognizing the crossed types. Had we varieties so dis- 
tinct in color and kernel as to be at once recognized, such varieties 
would serve us a touch-stone, and enable us readily to classify. The 
nearest approach to such a variety that we have is the Black Mexi- 
can, among sweet corns. Yet, as I wrote in my classification of 
sweet corns, published in the L’wral New Yorker in 1884: ‘* Prob- 
ably there are two distinct varieties, which are always sold in a 
mixed crop, through not being recognized, the one earlier than the 
other,” and hence if two races of the Black Mexican are present in 
_ our seed, we must expect some confusion in interpreting its occur- 
rence upon ears of other corn through the current influence of its 
olien. We, however, will undertake the illustration, using the 
notes of 1883 and 1884, with the exception of the 1883 notes which 
were made upon incorrectly named samples. 
In studying examples of ears from many sources, and representing 
hundreds of local names and collections, | think I can recognize 
three races of corn within each agricultural species, and each 
race holding a parallel relation in the different species. These races 
I designate at present by the letters of the alphabet, A, B,C. The 
diagnostic characters seem to be quite distinct. Race A can be 
ealled the 8-rowed type; race B the 12-rowed type; race C the 
many-rowed type, although the number of rows need not, in any 
one variety be of constant count. In season relations race A is 
early, race B medium, race C late, and yet the relation is but a gen- 
eral one. Race A has rounded kernels, race B kernels with straight 
sides and shallow, race C kernels with straight sides and deep, yet 
these relations are not absolute in themselves. Race A has small to 
medium ear-stalks, race B medium to large ear-stalks, race C me- 
dium to small ear-stalks, and yet the relations not always constant. 
Race A has kernels which never round over the butt; race B ker- 
nels which scarcely, if ever, round over butt of ear; race C kernels 
which round strongly over the butt of the ear. Race A ears which 
are of cylindrical type; race B ears which tend to taper; race C 
which tend to taper strongly. The union of «these characters seems 
available to separate the races with distinctness when the ears are 
examined in numbers, although when single ears are examined some 
of the characters given may be absent or divergent. 
These races do not seem to cross with each other through current 
hybridization. A and B seem usually to be resistant, if not always, - 
while C never crosses with A or B, so far as I have observed. 
Using now these conclusions, we will attempt to class our sweet 
corns through the character of the current hybridization, placing in 
race © those which receive race C dent kernels, and in race A those 
- 
