172 



Davis. 



on the sand dunes of Holland since pre-Linnaean days and no species 

 of Oenothera is more free from suspicion of gametic impurity. American 

 forms of Oenothera with points of strong resemblance to the Dutch 

 biennis have recently been found but an exact duplicate has not as yet, 

 so far as I am aware, been discovered. However, it seems not unrea- 

 sonable to expect that such a tyjie will eventually be isolated from 

 among the many allied species and races that grow in America. 



Fig. 3. Rosette of biennis, nearly mature, showing tlie close cluster of broadl}'- 

 elliptical. crinkled leaves. Contrast with Figs. 4 and 5, muricnta. 



Oenothera muricata in the accounts of DE Vrees refers to a plant 

 with taxonomic features quite as characteristic as those of O. biennis 

 Linnaeus. There is no possibility of confusing the two species. 

 Oenothera muricata has also been in Holland for a long period although 

 not so long as 0. biennis. "We have it in America represented by a 

 series of forms very closely related and apparently widely distributed. 

 I have grown plants from Woods Hole, Massachusetts, side by side with 

 plants of Dutch origin and liavc found only minor differences, and 



