THE PROBABLE ORIGIN OF OENOTHERA LAMARCKIANA SER. 591 



for sale seeds with the announcement that the same form may he 

 found as a wild flower in his own country, 



0. Lamarckiana has been, for many years at least, a component 

 of the flora of England, growing in many localities, especially on 

 the sand dunes along the coast. The most universally known 

 station is that of St, Anne's on the Sea, near Liverpool, which has 

 been studied by Bailey, Gates, and other botanists, and where 

 the species occurs in thousands of specin^ens. Davis received 

 seeds from different English stations and recognized the plant in 

 the cultures derived from them {op. cit. p, 237), In Lancashire 

 the species locally grows together with 0. biennis L,, exactly as 

 it does in the sand dunes of Holland. In such cases it produces 

 hybrids such as I have described under the names of laeta and 

 velutina and as Davis has isolated as small-flowered races from 

 those English localities (p, 237), 



Now, if we agree with Davis that the seeds of Carter and Co, 

 were derived from some English station, the probability at once 

 arises that these English stations themselves owe their origin to 

 the introduction of seeds from America, either by Michaux him- 

 self or by some other botanist of the same period. The history of 

 the species would then become a very simple and clear one. In 

 this respect it becomes of interest to look at the figure published 

 in 1807 in Smith's English Botany (vol, VI, pi 1534).^) According 

 to the description accompanying this plate, the "specimen was 

 gathered on the extensive and dreary sand banks on the coast a 

 few miles north of Liverpool, where millions of the same species 

 have been observed by Dr, Bostock and Mr. John Shepherd growing 

 perfectly wild and covering large tracts between the first and 

 second range of sand hills," In this same locality 0. biennis L. and 

 0. Lamarckiana are now growing in the same abundance of indi- 

 viduals, partly separated and pure in different valleys and partly 

 in mixtures which are known to contain also their hybrids. The 

 specimen of 1807 is designated 0. biennis, but both the flowers 

 have the lobes of their stigma above the anthers, which is a dif- 

 ferentiating mark of 0, Lamarckiana. Moreover, it is the only deci- 

 sive detail, all other characters of the figures applying equally to 

 both species. If it is allowable to trust to this detail, we should 

 be entitled to conclude that the station of Liverpool contained 

 both forms as early as 1807, even as it is known to do at the present 



i) Cf, Davis, op. cit. p. 532. 



