FACTORS FOE YIELD AND QUALITY OF SPRING WHEAT 
67 
Table 66. — Correlation coefficients and probable errors of crude-protein content in 
F 2 and F 3 generations of Marquis-Hard Federation wheat crosses grown at Boze- 
man and Havre, Mont., in one or both of the two years 1923 and 1924 
Crude protein 
Correlated with— 
Number 
of F 2 
plants or 
F3 strains 
Correlation 
coefficient 
and probable 
error 
r, A . . fFs, Bozeman, 1924 . . . 
221 
287 
0. 097±0. 045 
Crude protein | p J Hayre> ^ 
. 093± .039 
508 
.095 
The coefficients obtained are positive but not significant. Under 
the conditions under which the crop was grown at Bozeman in both 
1923 and 1924 and at Bozeman in 1923 and Havre in 1924 no evidence 
was obtained to show that the segregation for crude-protein content 
that occurred was inherited in an important amount. 
SUMMARY 
This inheritance and environmental study was made for the pur- 
pose of determining the factors for yield and quality of spring wheat 
on the dry lands of Montana. 
Hard Federation was selected for crossing with Marquis, the most 
important commercial variety, because of several contrasting char- 
acters, principally the longer fruiting period of Hard Federation and 
the greater height of Marquis, which affect differently their adaptation 
in certain localities and seasons. 
The material was studied at three points — Bozeman, "under fairly 
humid conditions in the Gallatin Valley; Moccasin, under semiarid 
or average dry-land conditions in the Judith Basin; and Havre, 
under marginal dry farming conditions in the so-called triangle. 
Three distinct environmental conditions are thus represented. 
Twelve ¥ t families were studied in the F 2 generation. Six were 
grown at Bozeman and six at Moccasin. These were divided equally 
between reciprocal crosses at each station. No important maternal 
or paternal influence was found, although a few statistically signifi- 
cant differences were obtained. 
Dwarf plants appeared in the F 2 generation in all F t families in a 
ratio close to 13 normals to 3 dwarfs. In F 3 the normals bred true 
or segregated in ratios of 13 normals to 3 dwarfs or 3 normals to 1 
dwarf. The dwarfs segregated in the ratio of 3 dwarfs to 1 normal 
or bred true for dwarfs. The Bozeman F 2 data corrected on the basis 
of F 3 results gave a close fit to a 7 : 4 : 2 : 2 : 1 genotypic ratio for the 
.above breeding groups. The results were satisfactorily explained on 
a 2-factor genetic basis by assuming DD a factor for dwarfs and NN 
^n inhibitor or factor for normals. 
The inheritance of awnedness was studied by grouping the material 
into the three following classes: (1) Awnless, (2) apically awnletted, 
(3) awnletted. In the F 2 generation the combined classes 1 and 2 
compared with 3 were close to a 3:1 ratio. Class 1, in comparison 
with classes 2 and 3 combined, was close to a 1 : 15 ratio. In the 
F 3 generation, classes 1 and 3 bred true. The Moccasin F 2 data for 
the three classes were close to a 1 :11 :4 ratio, and the Bozeman F 2 
