INHERITANCE IN THE BARLEY SPIKE. 
23 
These results show that, although density is a very stable size 
character, in some crosses numerous factors are involved which, by 
recombination, produce homozygous forms showing an almost con- 
tinuous range of density from the very lax to the dense types. It is 
only reasonable to conclude that if a greater number of varieties had 
been studied, together with crosses between them, a continuous range 
for the average length of internode of homozygous forms could be 
ob tamed which would show only small differences in average density 
between types. These results are of considerable interest in barley 
classification. While dependable in the isolation and description of 
- 
9. 
\30 
\ 
/ 
Jz/rc 
?C/?/7 
3V 
/ 
/ 
1 * 2 
htf 
&/e 
/ 
""-"—« 
-^ 
r 
r 
36 
/ 
\ 
/ 
-°— -o, 
/ 
i n 
/ 
\ 
/ 
/ 
\ 
/ 
/ 
/ 
> 
/ 
) 
/ 
/ 
V 
\ 
\ 
> 
4S-/ 
/■* 
k 
\.''-?<?&-//-3 
f??0- 
"\ 
\ 
\ 
/ 
/ 
i 
/ 
\ 
/ 
' s 
// 
\ 
.._ 
y^tt 
p^^ 
^ 
^r-44 
9-J&.- 
'&aa : 
^^ 
<^V 
^55; 
/.*? /.& /.& e.O ?.2 S<? 2:& Z>.<ff 3.o 3.2 3.*? 3.6- 3.0 <0 -4.2 <•?.;? -?.6" *?.& &o &2 &.*? &.& 
Fig. 2.— Diagrams showing the densities of parental forms and of the F 2 generation in a cross between 
the Zeocrlton and Hanna barleys (upper), of four pure lines (middle), and of several heterozygous 
lines (lower). 
strains, groups founded on this character are likely to overlap and 
hence to be of limited value for taxonomic purposes. 
While the general genetic results of these crosses are explained on 
a broad factor basis of differences of one to three factors, the fact 
remahis that the homozygous segregates corresponding to the parents 
do not always have the exact density of the parents. Likewise, the 
forms homozygous for intermediate densities do not all fall together 
but in groups, which, in the Hanna X Zeocriton cross become almost 
continuous, even where limited numbers are concerned, and might 
become wholly continuous if it were possible to carry the full number 
to the fourth generation. Obviously, there are modifying factors, 
and so far as they affect density they may be considered as minor 
density factors. Several explanations are possible. These varia- 
