A Genetic Study of Plant Height in Phaseolus Vulgaris. 63 
axis. The difference in variation between the pole and bush 
segregates in F2 is no greater than that between pole and bush 
races. 
In the hypothetical formulae employed here, the factor A 
for habit of growth is assumed to differ in no way from the 
factors B and C, except in regard to dominance and to the mag- 
nitude of its effect upon number of internodes. As a matter of 
fact, the factor A must differ in another respect from B and C, 
for the axis of a pole bean with relatively few internodes is not 
terminated by an inflorescence and can be forced to elongate 
almost indefinitely, while the axis of the tallest bush bean is 
terminated just as abruptly as that of the shortest bush bean. 
If A were a factor for habit of growth alone without other relation 
to internode numbers and B and C and their like mere modifying 
factors, some of them, perhaps, concerned in general vigor of 
growth, time of flowering and of seed production, etc., the results 
would be the same. It is probable that size factors in general 
interact one upon another during development, a single factor 
sometimes being concerned in the development of several charac- 
ters commonly regarded as quite distinct and several factors at 
times being concerned in the development of a single character. 
Several indications of this sort have been observed in maize 
(Emerson and East 1913). In the present paper numerous 
illustrations of the interrelations of growth habit, internode num- 
ber, internode length, time of flowering and seed production, and 
the like have been pointed out. 
By way of conclusion, it may be said that the known facts 
in regard to the inheritance of the quantitative character height 
of plant in beans yield readily to an analysis based upon a modi- 
fied multiple-factor hypothesis. The proposed modification con- 
sists merely of the assumption of inequality in dominance and 
inequality in potency of some of the factors concerned. It is 
probable that no geneticist would maintain that all the genetic 
factors concerned in the development of any quantitative charac- 
ter are necessarily equal in dominance or potency^ but, since the 
actual potency of distinct quantitative factors has as yet been 
determined in no case and since the assumption of equal potency 
of two or more factors affords a simple and sufficiently accurate 
interpretation of quantitative inheritance, no other assumption 
seems necessary except in cases like the one now under considera- 
tion. Whether this modified multiple-factor hypothesis will 
account for the facts of inheritance of plant height in case of 
such types as tomatoes, peas, dwarf maize, etc., is a question for 
future determination. 
^ For a discussion of this matter, see Shull 1914, and Muller 1914. 
