A Genetic Study of Plant Height in Phaseolus Vulgaris. 49 
that the F2 pole-bean segregates of the July-Triumph cross, as 
well as of the July-Red Marrow cross, showed greater variation 
than did July with respect to mean lengths of the first 15 inter- 
nodes. 
From Table 4 it is seen that the variation in mean lengths of 
the first five internodes of the Fo pole-bean segregates of the 
Snowflake-Triumph cross is less than that of either parent race. 
The Fo bush-bean segregates of this cross, on the other hand, had 
a greater variation than either parent race. When the individual 
internodes are considered separately, it is found that the standard 
deviations are, as a rule, greater for Fo than for the parent races. 
The very small mean internode lengths of Triumph, however, 
make the coefficients of variation for that race larger in many 
cases than those for the F2 plants. It is possible that the mean 
internode lengths of Triumph are relatively smaller than they 
should have been owing to the fact that the conditions under 
which the bush-bean families were grown were somewhat less 
favorable than those surrounding the pole-bean families. That 
this may be true is suggested by the fact that the mean internode 
length of the Fo pole-bean segregates is considerably nearer that 
of Snowflake than that of Triumph — but this difference may 
in part be due to a somewhat increased vigor arising from partial 
heterozygosis in Fo. If normal internode lengths in Triumph and 
Snowflake are more nearly alike than indicated by the values 
found here from the first five internodes, less variation would be 
expected in F2 than if the difference between internode lengths 
of the parents were greater, but this variation should certainly 
not, even then, be less than that of the parents. From the data 
at hand, it cannot be said that any segregation occurs in factors 
for mean length of the first five internodes. 
The only cross between pole and bush beans yet to be con- 
sidered with respect to mean lengths of the first five internodes 
(Table 4) is that between Red Marrow and Snowflake. The mean 
lengths of the first five internodes are nearly the same for both parent 
races, but the mean lengths for both generations of the crosses 
are greater than for either parent. This, like the results of the 
Triumph-Snowflake cross, may be due in part to the somewhat 
unfavorable conditions under which the bush-bean families were 
grown, whereby the internode length of Red Marrow was less 
than it would otherwise have been. If this is true, the same 
condition, which may have made the internode lengths of Snow- 
flake and Triumph differ more than they should, would tend to 
make the internode lengths of Snowflake and Red Marrow more 
nearly alike than they otherwise would have been. If, on further 
investigation, these conjectures are substantiated, that is, if 
