No. 538] 



581 



nating forms ( /. ,<., those having a short stvle so that the 

 anthers surround the stigma in the laid) shall he classed 



"European hicnnis" has flowers which are larger than 

 any known Ameriean races, hat it is conveniently classed 

 as 0. biennis on the basis of its short style. The ques- 

 tion as to the characters of the particular plant on which 

 the species name was originally based, also of course 

 enters here. In connection with the early records, her- 

 barium specimens and figures are the chief means of 

 determining approximately the race to which a plant 

 under a given name belonged. The present paper con- 

 tains the writer's further conclusions concerning these 

 various races and species, and the reader will constantly 

 be referred in this connection to the extensive data al- 

 ready brought tog-ether in the paper previously men- 

 tioned (Gates, 'lie). 



Early Records and IIehharitm Specimens 

 We may now consider the identity of these various 

 forms in succession. In the paper just mentioned I have 

 given the nine species of Onagra listed by Tournefort 

 in his "Institutiones Rei Herbaria?," 1700. Of these the 

 first five are Oenotheras. Species number five, Onagra 

 angustifolia, caule rubro, flore minori, I have already 



