lO BOTANICAL GAZETTE [july 



fourteen so far as observed. There has been no indication whatever 

 that the number is ever higher. In the Lamarckiana hybrid it has 

 been found from a number of counts in the floral tissues that the 

 number of chromosomes in the somatic tissues is tv^enty or twenty-one, 

 and I have found no reason to think it varies. In the pollen mother 

 cells the number of chromosomes in the metaphase and anaphase of 

 the heterotypic mitosis is usually twenty or twenty-one. Occasionally 

 the number in early anaphase appears to be less, but this may be due 

 to a tardy separation of some of the paired chromosomes in the 

 equatorial plate; further study is required to determine this point- A 

 large number of telophases of this mitosis were counted which showed 

 ten chromosomes each. In one case there appeared to be eleven 

 chromosomes in a telophase, though the greater majority at least 

 show ten in each daughter nucleus. Occasionally a chromosome lags 

 behind in the equator of the spindle until late anaphase, and this may 



the number twenty-one. Such a case is figured in my former 

 (pL 4^ fig. jg). This is not always the case, however, as shown 



make 



by fig 



nucleus, but none left behind in the equator of the spindle. It is 

 possible that a single unpaired chromosome might be present in meta- 

 phase and be lost in the cytoplasm as the paired chromosomes scpa- 



Mm 



rated and went to the poles. It is certain that in many pollen mother 

 cells ten chromosomes go into each daughter nucleus in the hetero- 

 typic mitosis. Whether the number in the somatic tissues is. twenty 

 or twenty-one has not been determined with certainty, as it is a matter 

 of ver}^ considerable difficulty. The further discussion will reveal 

 the reason for suggesting that the sporophyte number of chromo- 

 somes in the Lamarckiana hybrid is twenty-one rather than twenty. 

 O. Lamarckiana^ the parent, has fourteen chromosomes as the 

 sporophyte number, so far as the plants have been examined. This 

 count was made constantly in the floral tissues. The homotypic mito- 

 sis in the pollen mother cell shows seven chromosomes constantly so 

 far as observed, a large number of counts being made in various 

 "*""-s of this mitosis. It is doubtless true that chromosome counts as 

 ordinarily made come from a very few individuals. In the case of the 

 Lamarckiana hybrid I have examined material from two individuals 

 and from different flowers in each; these plants were grown in 



o 



