528 Davis: Lamarck's evening primrose 



2. Inflorescence. The bracts of the inflorescence are not 

 broad at the base and sessile as in the Lamarckiana of De Vries. 

 They are narrow-elliptical and short-petioled. 



3. Buds. The size and form of the buds present perhaps the 

 most important characters on the sheet. They are short and 

 stout, and these characters alone make it impossible that the 

 plant could have been 0. grandiflora Solander. (Compare plate 38 

 with PLATE 37.) Forms of 0. biennis frequently show these 

 peculiarities. The sepals have a greater pubescence than those of 

 grandiflora. 



4. Flowers. The flowers are medium-sized, petals probably 

 between 2 and 2.5 cm. long. They are not large enough for 

 grandiflora or for the large-flowered forms of De Vries's Lamarck- 

 iana. The stigma (5, plate 38) appears to be at about the level 

 of the anthers, the style not extending well beyond as in the types 

 mentioned above. The flowers, in size and in the relation of the 

 stigma to the anthers, agree with forms of biennis. 



5. Capsules. The capsules appear to be of the biennis type, 

 which is similar to that of De Vries's Lamarckiana. 



In conclusion, the forms of the leaves and bracts distinguish 

 this plant of Abbe Pourret from the Lamarckiana of De Vries's 

 cultures. The size and form of the buds, the size of the flowers, 

 and the position of the stigma distinguish it from O. grandiflora 

 Solander as well as from the larger-flowered forms of De Vries's 

 Lamarckiana. All of the characters- described above are repre- 

 sented in the assemblage of forms included under the name Oeno- 

 thera biennis. Since we know nothing of the rosette, general habit, 

 and lower foliage of this plant, it is quite impossible to follow its 

 determination further. 



Sheet 3. Specimens referred by De Vries to Oenothera 



grandiflora 



The specimens on this sheet (plate 39) are so imperfect and 

 their form so abnormal that a satisfactory determination of their 

 identity is probably impossible. De Vries (1901, footnote to p. 

 316) considered them to be Oenothera grandiflora Aiton = 0. 

 grandiflora Solander (0. suaveolens Desfontaines) . 



