Origin o/ CEnothera Lamarckiana De Vries. 24 x 



Thus Don' described in 1832 "CE. biennis'' from "sand banks on 

 the coast of Lancashire " and noted that it " is common in gardens 

 and often escapes from thence into rich waste ground," but the 

 statements " flowers large " and " delicately fragrant " are relative 

 matters and we cannot be certain from this or other parts of the 

 account whether the plants were nearest to Lattiarckiana De Vries 

 or to forms of biennis such as that from Holland (the biennis of 

 Linnaeus). Lindley^ in 1833 gave an account of "CEnothera 

 biennis var. grandiflora " and stated it to be " not uncommon in 

 gardens." The figure of this plant shows petals drawn about 

 3.5-4 cm. long, as large as those of Lamarchiana De Vries, but a 

 basal or perhaps a rosette leaf is too narrow for Lamarckiana, nor 

 will it do for the Dutch biennis or for grandiflora. The bracts are 

 figured broad at the base and sessile as in both Lamarckiana and 

 biennis, but the inflorescence with long internodes is not as in those 

 species but is more like that oi grandiflora ; the bracts, however, 

 are not those ot grandiflora. These contradictions are most puzzling, 

 but may there not be specimens preserved at this period which will 

 clear the obscurity. 



I am aware that the CEnothera floras in other parts of England 

 offer opportunities for investigation that should not be overlooked 

 in the study of the problems presented in this paper. I have, 

 however, confined my suggestions to the remarkable assemblage on 

 the sand hills of Lancashire for the reason that this region seems 

 to be best known and to have been under observation for the 

 longest period. 



' Don, George. A General System of Gardening and Botany, vol. 2, 

 p. 685, 1832. 



^ Lindley, John. CEnothera biennis:, var. grandiflora. Edwards's Botanical 

 Register, vol. 6, p. 1604, 1833. 



University of Pennsylvania. 



March, 1913. 



