Oct., I92i] EMERSON — ^ ABERRANT CHROMOSOME BEHAVIOR 
421 
Pr Sti, while the female parents of all had sugary endosperm, su. In 
addition, the female parents of all crosses shown in groups i and 2 of the 
table had colorless aleurone, A c R for group I and A C r for group 2, and 
those of the crosses presented in group 3 had red aleurone, A C R pr. 
In all, 47 aberrant seeds are recorded. Of these, 34 had aberrant aleurone 
color, 20 involving C c Su su (fig. i,'C), 13 R r Su su (fig. i. A), and i 
Pr pr Su su; and 13 had aberrant endosperm texture, 5 involving C c Su su 
(fig. I, D), 6 R r Su su (fig. i, B), and 2 Pr pr Su su. Every one of the 
34 seeds that had aberrant aleurone (colored-colorless) were starchy through- 
out, and all of the 13 with aberrant endosperm (starchy-sugary) had colored 
aleurone throughout. 
In short, there have been observed (tables 2 and 3) a total of 85 aberrant 
seeds involving C c with Su su, A a with Wx wx, and R r and Pr pr with 
both Su su and Wx wx. In every case in which aleurone color was con- 
cerned, 72 in all, the aberrant spot showed the recessive color of the female 
parent but was invariably underlaid with the dominant corneous or starchy 
endosperm of the male parent; and in all of the 13 cases in which endosperm 
composition was concerned the aberrant spot exhibited the recessive sugary 
condition of the female parent but was invariably overlaid by the dominant 
aleurone color of the male parent. It is obvious, therefore, that for none 
of the 85 seeds could the aberrant spots, though they displayed in every case 
one or other (never both) of the two recessive maternal characters whose 
genes were carried in the polar nuclei, have been produced by the inde- 
pendent division either of one polar nucleus alone or of the two after fusion. 
To explain these cases on the basis of independent division of one or both 
polar nuclei would require the unwarranted additional assumption that in 
some cases — ^aberrant sugary spots — the polar nucleus or nuclei alone give 
rise to a part of the underlying endosperm but to none of the aleurone layer, 
while in other cases — -aberrant aleurone-color spots — they give rise to a 
part of the aleurone but to none of the underlying endosperm. Moreover, 
if such behavior of independently dividing polar nuclei were so common 
in cases involving the endosperm factors Wx ivx with the aleurone factors 
A a, R r, and Pr pr (table 2), and the endosperm factors Su su with the 
aleurone factors C c, R r, and Pr pr (table 3), why should not the same 
behavior of the polar nuclei be found where there is involved Wx wx with 
Cc (table i) or with li (Collins, 1913) or C c with Sh sh (Hutchison's 
data)? But the facts are that in the great majority of aberrant seeds 
involving C c and Wx wx (55 to 3) where the aleurone layer is colorless 
(maternal), the underlying endosperm is waxy (also maternal), and the 
correspondence in outline between colorless aleurone and waxy endosperm 
is strikingly exact. Certainly no single hypothesis that assumes inde- 
pendent development of either one or both of the polar nuclei can be made 
to fit all the data now available. 
That the somatic-mutation hypothesis suggested by the writer (1915) 
does not agree with the great majority of the observed facts when C c and 
