May 12, X923 
Resistance to Stem Rust m a Cross of Common Wheat 467 
In view of the preceding facts and the data presented in this paper, 
it is not at all surprising that various workers have drawn different con¬ 
clusions regarding the dominance of spring and winter characters in wheat 
and other cereals. The large number of F3 selections, homozygous for 
heading period and varying all the way from the pure spring character 
to the pure winter character, demonstrates how finely the differences for 
growth habit may be divided. Each F3 family (or selection) with a dif¬ 
ferent heading period is of a different genetic nature. And as soon as 
these types become fixed, one ought to be able by intercrossing the types, 
to produce varying degrees of dominance of spring and winter characters. 
Probably one of the most important principles that this study has 
demonstrated is the necessity for using known biologic forms of rust in 
the determination of the resistance or susceptibility of any given host. 
Attention was called to this principle by Stakman, Levine, and Leach (15) 
in 1919; by Hayes, Parker, and Kmtzweil (7) in 1920; and again by Put- 
tick (jj) in 1921. It has already been pointed out that previous to the 
discovery of the existence of biologic forms of stem rust of wheat, the 
breeding of resistant varieties met with failure time and again. Simply 
because a variety of wheat is resistant in a given locality for a period of 
years, there is no assurance that it will always be resistant in that locality 
or in any other locality in which it may be grown. And for the same rea¬ 
son one can not expect to prove conclusively, in an experimental plot, 
the resistance of any given variety, unless it is tested for resistance against 
all of the rust forms which exist in the area in which it is expected to be 
grown later. One can readily see the difl&culty of producing an artificial 
epidemic in the field with certainty that a number of forms are present in 
sufficient quantity to have equal opportunity to attack all plants. Even 
if this were possible in a practical way, it would bring about difficulties 
in the synthetic production of a resistant variety. This point was clearly 
demonstrated in the work from which the data presented in this paper was 
taken. 
The Fj population was grown in the field under an artificial epidemic 
produced with several different biologic forms of rust. One of the parents 
and many of the hybrid plants were resistant to some of these biologic 
forms, as was proved when they were inoculated with the pure culture 
in the greenhouse. But in the field, all of the plants were equally sus¬ 
ceptible, for all practical purposes. There was no method by which one 
could differentiate the plants one from another under this general epi¬ 
demic. If a genetic study was to be made of a given number of biologic 
forms in their reaction on certain host plants it was evident that the 
work must be carried on under controlled conditions. 
After the susceptible individuals have been eliminated in this manner, 
the general selection for desirable agronomic characters can be carried 
on in the field. As soon as this desired agronomic type is obtained, it 
will be crossed again with other varieties or selections which are resistant 
to other biologic forms and this process will be continued until a desirable 
agronomic type has been obtained which is resistant to all of the biologic 
forms of stem rust. 
The fact that resistance and susceptibility of the host to several different 
forms of rust are inherited as a single genetic factor makes this cross very 
valuable. Kanred is known to be immune from at least 11 different biologic 
forms of stem rust, and, as far as tested, the immune progeny possesses 
all the immunity of the Kanred parent. This has further demonstrated 
that a variety of common wheat may be produced synthetically which 
will be resistant to a large number of the biologic forms of stem rust. 
