460 
Journal of Agricultural Research 
Vol. XXIV, No. 6 
The crossed seed was sown in the fall of the same year and produced 
80 plants, 2 of which were winterkille^d and 5 of which were not crosses. 
The remaining plants were harvested individually and the seed sown in 
the spring of 1920. 
In the second generation a population of approximately 5,000 plants 
was grown. In order to facilitate observations the seeds were sown at 
intervals of 3 inches, in rows i foot apart. The date of emergence of the 
first head on each plant was noted. The hybrids fpimed a continuous 
series for date of head emergence, beginning with those which came out 
at the same time as the spring parent, to those which did not emerge at 
all. In this way they resembled the winter parent. (PI. i, B.) 
The time of heading was divided into weekly periods. One week from 
the day on which the first plant headed, tags were placed on plants on 
which one or more heads had emerged. These comprised the fost class. 
All of the Marquis control plants headed during the same period as did 
those plants included in the first class. One week later, tags were 
attached to all plants which had headed since those of class i. These 
constituted the second class. This process was continued for 8 weeks, 
after which period no more plants headed. The plants which did not 
head were classed as true winter types. The winter parent controls failed 
to head, thereby falling into the same class as the winter hybrids. 
From the first 7 heading classes in the Fg, 65 families were grown in the 
F3. The plants from the seventh class produced only a few seeds, while 
those of the eighth class headed so late that no seed was produced at all. 
In sowing, the seed was again spaced as in the second generation so that 
a study could be made of the individual plants. 
Several of the families were uniform for heading period in the Fg. 
In these cases the entire plot was given a general heading date, as in a 
varietal test. Others of the families were heterozygous for date of head 
emergence. In those plots a final count was made at harvest time of 
the number of plants which failed to head and the number which pro¬ 
duced heads. 
The rust studies were made on plants Rowing in the field under an 
artificial epidemic produced with several different biologic forms of stem 
rust and also on inoculated seedlings in the greenhouse. 
In the second generation all the plants were grown in the field under 
an artificial epidemic produced by spraying the plants with a suspension 
of urediniospores of several different biologic forms. Both parents were 
susceptible to some of the rust forms which were used. All of the hybrid 
material was as susceptible as either parent in this epidemic and there¬ 
fore a detailed genetic study of the inheritance of rust resistance under 
field conditions could not be made. Studies on the inheritance of 
resistance and susceptibility were made in the greenhouse by inoculat¬ 
ing the Fg seedlings with cultures of urediniospores of known biologic 
forms. 
The seedlings in the greenhouse were grown in 4-inch pots and inocu¬ 
lated when they were to 2 inches tall. After inoculation they were 
placed in a glass-topped chamber and incubated 48 hours. The notes 
on infection were taken 12 to 14 days after the date of inoculation. 
Some of the plants were completely susceptible and the others were im¬ 
mune since there were no intermediate types of infection. The uredinia 
were large, coalescing, and normal in every respect on the seedlings both 
of Marquis and of susceptible hybrids, while the seedlings of Kanred and 
