/ 



h 

 1 



i6 BOTANICAL GAZETTE ■ [july 





in H. excellens the egg cells have different numbers of chromosomes^ 

 owing to irregularities in the reduction divisions. When this form is 

 crossed with another species, the hybrids of the F^ have different 

 numbers of chromosomes. Whether this is the cause of the different 

 types in the F, is not ascertained. 



It seems probable, or at least possible, that there is a connection 

 between the type of hybridization in Oenothera and the chromosome 

 numbers, though just what that connection may be is at present quite 



4 



unknown. Apparently whatever determines the twenty or twenty- 

 one chromosomes determines also that the parent shall have the O, 

 Lamarckiana characters. There is thus a possibility that this may be 

 found to have a bearing upon the cause of dominance and recessivity 

 in Mendelian hybrids, the characters of the individual being deter- 

 mined in this case by the presence or absence of the extra chromo- 

 somes. Cannon^s^° suggestion that the Mendelian segregation of 

 characters results from the segregation of maternal and paternal 

 chromosomes in the reduction mitoses is as yet a hypothesis without ■ 

 proof, though Rosenberg's^^ recent paper, in which he states that in 

 the pollen development of the Drosera hybrid occasionally tw^o pollen 

 grains of the tetrad have the morpholog}^ of one parent and tw^o that 

 of the other, shows that in some cases, at least, segregation of char- 

 acters takes place at this point in the life history. 



Practically the same theory of the segregation of the characters 

 and purity of the germ cdls in plant hybrids on the basis of the chromo- 

 somes (idants) was put forward by Weismann^^ in 1892, in connec- 

 tion with his theory of the germ-plasm. At this time the Mendelian 

 hybrids had not been rediscovered, and complete segregation of the 

 characters of the parental types w^as regarded as a rare occurrence in 

 hybrids. It may be worth while to quote Weismann's statement 

 (p. 299): 



The germ -mother cells of the hybrid contain a group of idants (chromosomes) 

 derived from the paternal, and another from the maternal ancestral species. If 

 therefore the " reducing-division " halves the germ-plasm of these mother-cells in 



20 Cannon, W. A., A cytological basis for the Mendelian laws. Bull. Terr. Bot. 

 Club 29:657-661. fig. I. 1902, 



aJ Rosenberg, O., Erblichkeitsgesetze und Chromosomen. Botaniska Studier, 

 Upsala, pp. 237-243. figs, 5. 1906. 



aa Weismann, ArcusT, The germ-plasm. Engliiih translation. New York. 1898. 



