﻿BOTANICAL GAZETTE 



In the examination of a considerable amount of herbarium 

 material, I have seen no intermediates between M. dilatatum and 

 M. canadense or M. bifolium, although even if such occur it by no 

 means diminishes the possibility that M. dilatatum originated from 

 M. bifolium or M. canadense through a single mutation, for when 

 O. gigas crosses with O. Latnarckiana, intermediate hybrids are 



Fig. 6.—Maianthemum canadense Desf. (on the left) and M. dilatatum (Wood) 

 Nelson and Macbride (on the right; a portion of one leaf was accidentally broken in 

 taking the photograph). 



produced, and these again when crossed back with either parent 

 species produce new intermediate stages. O. gigas is also very 

 variable in foliage, probably as a result of the tetraploid condition. 

 If these two classes of variants were found in a population of typical 

 wild 0. gigas and 0. Lamar ckiana, they would prove very con- 

 fusing from the systematic point of view, but from the genetic 



