A Genetic Study of Plant Height in Phaseolus Vulgaris. 9 
(Emerson 1910), while height of plants has been put in the 
second class. In certain crosses between tall and dwarf maize 
one type of behavior results, and in other crosses the other type 
is exhibited. 
Intermediate development in Fi and a wide range of variation 
in F2, characteristic of one of these classes, is now interpreted 
rather generally, tho by no means universally, ^ by the multiple- 
factor hypothesis, which postulates that, external influences 
aside, quantitative differences depend upon two or more independ- 
ently inherited, nondominant,^ genetic factors — a strictly Men- 
delian interpretation. Quantitative differences that show domi- 
nance in Fi, followed by a 3-1 segregation in Fo, must depend upon 
a single dominant, genetic factor or upon factors that are coupled 
in inheritance. That some quantitative differences are due to a 
single factor while others are due to many factors need not, it 
might seem at first, occasion any wonder. In maize, for instance, 
seeds may differ much or little in breadth. When the parents 
of a cross differ much, many more individuals must be grown in 
F2 than when they differ little, if the parent sizes are to be re- 
covered. In other words large differences in breadth of seeds 
appear to be due to more factors than do small differences. Is 
it not to be expected, then, that some differences may be due to 
a single factor? 
The fact of prime importance in this connection is that both 
large and small quantitative differences in the same plant parts 
seem to depend upon a single factor. Thus, in crosses between 
pole beans and bush beans, the Fi plants are always pole beans 
and the F2 plants consist of approximately three pole beans to 
one bush bean. This is equally true whether the pole bean parent 
normally grows to a height of say 12 feet or only to a height of 
6 feet and whether the bush bean is 2 feet or only 9 inches high. 
What happens when a tall bean is crossed with a very tall one? 
What results are to be expected from a cross between a short 
bean plant and a very short one? How does a cross between a 
tall bean and a short one differ from a cross between a very tall 
bean and a very short one? To answer these and similar questions 
was the purpose of the investigations reported here. 
^ For an interpretation based upon the assumption that genetic factors 
are commonly modified in crosses see Castle 1912 and 1914a. 
2 Both Shull (1914) and Muller (1914) have shown that lack of dominance 
is not essential to the multiple-factor hypothesis (referred to by Shull as the 
hypothesis of plural genes). Shull's argument is based upon the assumption 
of the possible interaction of dominant factors of opposite effect, as an in- 
hibitor and a stimulator. Muller's contention is practically the same but is 
stated in terms of plus and minus dominant factors. A third form of statement 
for the same results is that part of the positive factors may be fully dominant 
and part wholly recessive. 
