24:4 Emerson. 



suggests another) only hypothesis III, vegetative segregation after ordi- 

 när}' double fertilization. 



Since neither of the two hypotheses based upon abnormal endosperm 

 "fertilization'" can be used to explain the two anomalous seeds here 

 described, it is worth while to consider somewhat carefully the only 

 remaining hypothesis that has so far been proposed to account for such 

 cases, namely, the vegetative-segregation hypothesis. Let us then in- 

 quire into the possible nature of such vegetative segregations. This 

 inquiry will lead naturally to a consideration of their relation to bud 

 sports. In fact if it were not for the probability that they bear some 

 relation to bud sports and that their study might throw some light 

 upon the larger problem of somatic variations, they would hardly be 

 worth extended discussion. 



East and Hayes (9) proposed the h}q)othesis in the following 

 form : "It may be possible then that the cause of these seeds is Mendelian 

 segregation in somatic tissue, such as often occurs in bud sports''. This 

 statement indicates clearly that the authors regarded somatic segre- 

 gations as Mendelian in nature. But East (7), at a somewhat earlier 

 date in a summary statement regarding bud sports, apparently employs 

 the term "Mendelian segi'egation" to denote a change that later is 

 inherited in a Mendelian way — a very different usage from that commonly 

 followed, in which the term is restricted to the separation of the oppo- 

 sing units of all allelomoq)hic pairs present in a heterozygous condition. 

 East's statement follows: 



Any discussion regarding the mechanism of the production 

 of variation is of course pure speculation, yet one cannot help 

 feeling that the production asexually of changes that mendelize 

 throws some light on the subject. It follows that segregation 

 may occur in the somatic cells, and that by division of one of 

 these cells a line of descendants originates lacking a Mendelian 

 character that is present in the parent cell. This fact does not 

 develop anything new in regard to the relative imiwrtance of 

 the chromosomes and the cytoplasm as bearers of hereditary 

 characters, but it certainly seems to show that Mendelian segre- 

 gation is not limited to tlir ivdiictiou division in tlir maturation 

 of sexual ceils. 

 Cytological evidence of vegetative segregation in endosperm deve- 

 lopment or of vegetative segregation in bud sports is hardly to be 

 expected. The cases of anomalous endosperm foiiiiation are extremely 



