258 Emerson. 



useful informatiou iu the more fundamental studies of cell organization 

 as pointed out in an earlier paper (Emerson (12)). 



Summary. 



Two anomalous seeds of maize, resulting from a cross between 

 colorless parents one of which was homozygous starchy and the other 

 sugary, are described. One seed was starchy throughout but about half 

 colorless and half colored (purple). The other seed was purple through- 

 out, but part of its endosperm was starchy and part sugary. A third 

 anomalous seed, the result of a cross between colorless-seeded and 

 heterozygous, colored-seeded parents, is described. The seed was wholly 

 colored but half purple and half red. 



These three seeds are shown to be inexplicable either on the 

 hypothesis of Correns and of Webber, that the second male nucleus and 

 the fused polar nuclei may each independently develop a part of the 

 endosperm, or on the alternative hypothesis of Webber, that the second 

 male nucleus may unite with one polar nucleus to produce a part of 

 the endosperm, the other polar nucleus developing independently to 

 produce the other part of the endosperm. These seeds are explainable 

 on the basis of the hypothesis of East and Hayes, namely, that sub- 

 sequent to normal endosperm fertilization there occurs a vegetative 

 segregation of genetic factors. It is shown, however, that, if such a 

 segi'egation occurs, it is not a typical MeudeHan segregation, because, 

 in neither of the three cases reported, could all the independently in- 

 herited genetic factors present in a heterozygous condition have been 

 involved. It is suggested that such seeds may also be interpreted as 

 somatic mutations: that is that they may be due to a change in genetic 

 constitution rather than to a segregation of genetic factors. 



A fourth anomalous maize seed described by Collins, a seed 

 with a small part colored and waxy and the remainder white and horny, 

 is shown to be explainable on the basis of either of the four hypo- 

 theses. It has special interest, however, in connection with the segre- 

 gation and mutation hypotheses because at least two factors, instead 

 of a single one, are here concerned. 



The relation of such somatic segregations or mutations in the 

 endosperm of maize to bud sports is considered. Reasons are given 

 for the belief that in certain cases the production of self-color as a bud 

 sport in variegated plants (Antirrhinum flowers, de Vries; Mirahilis 

 leaves, Correns; maize ears, Emerson) is to be regarded as a somatic 



