No. 538] 



MUTATION IN (ENOTHERA 



587 



lata, latifolia, Lusitanica was a large-flowered form be- 

 longing in the grandiftora-Lamarckiana series, and there 

 is no reason to doubt that his citation of Tournefort's 

 Onagra latifolia, ftoribus amplis as a synonym is cor- 

 rect. It is therefore highly probable that Tourncl'ort V 

 species (3) represents 0. grandifiora, or perchance 0. 

 Lamarckiana or some race between these two as we now 



The type specimen of Linmcus's 0. biennis in the 

 4 'Species Plantarum" (to be found in the Linmean Her- 

 barium) is, however, not the " European biennis," but 

 a smaller-flowered form representing one of the Ameri- 

 can races of 0. biennis, having rather narrowly lanceo- 

 late stem leaves. I have already pointed out ('11a, p. 

 104) that Linnaeus does not cite figures of this form in 

 his synonymy, although a good figure by Barrelier was 

 in existence. Instead he cites Mori son's figure which 

 now appears most probably to have been the " European 

 biennis;' and in the "Hort. Cliff." he cites Barretter's 

 figure of Lysimachia lutea corniculata latifolia lusitanica, 

 which was undoubtedly a large-flowered form. 



To summarize briefly the conclusions which seem 

 justified from all the available data, it appears that the 

 earliest introduction, as represented by the plants of 

 Bauhin, Parkinson, Morison and Ray, belonged to a 

 race of what we now know as the « ' European biennis," 

 having flowers larger than the present American races 

 but self-pollinated, although an open-pollinated form 

 with long style appears also to have occurred (speci- 

 men 3). 0. muricata was recognized by Tournefort in 

 1700, and both 0. muricata and the "American hioniis" 

 were figured by Barrelier in 1714, in addition to a large- 

 flowered species which must have been related to 0. 

 grandiflora or 0. Lamarckiana. This large-flowered 

 form appears to have been first recognized and briefly 

 described by Ray in 1686 (see Gates, '11 a, p. 100). Dif- 

 ferent hypotheses as to the relation between 0. grandi- 

 flora and 0. Lamarckiana will be considered later in this 

 paper. 



