EARLY HISTORICO-BOTANICAL RECORDS OF THE 
OENOTHERAS. 
By R. R. GxYtes. 
The present paper is an attempt to trace, as far as possible from 
available data, the history of the Oenotheras, particularly the large- 
flowered formas, in cultivation. An effort is also made to recognize, 
as far as this can be done, the precise characters of the various forms 
which have been figured or described during the last three centuries. 
Such records of course vary greatly in accuracy and value; for they 
are contemporaneous with the development of the science of botany it- 
self. Judging from the number of polynomials applied to them by 
different authors, the Oenotheras would appear to have been as va- 
riable then as they are now. And I may say that my cultures of 
Oenotheras derived from various sources indicate that at present many 
of these forms are no less variable or mutable than the 0. Lamar ckiana 
of DeVries’ experiments. 
I have been able to examine a large number of references and plates 
of Oenotheras — many of them pre-Linnaean — from the valuable sets 
of Herbals and leones in the library of the Missouri Botanical Garden. 
I wish to express my thanks to the Director, Professor William Tre- 
lease, for valuable aid in connection with the study of these early 
records. I am also indebted to Miss Cora J. Hogan, who has aided in 
deciphering the Snippendale manuscript and has also translated most 
of the Latin descriptions. I have attempted to trace, as far as pos- 
sible, the history of 0. Lamarckiana Ser., 0. grandflora Ait. and (in 
part) 0. Mennis L. from these early citations and plates. See also 
the important historical data supplied by Miss Vail in MacDougal 
(1903). The degree of accuracy of the plates varies greatly, but in 
many cases at least, one’s conclusions concerning the plants figured 
can rest on a pretty certain basis, when they have a minute knowledge 
of the differentiating characters of these forms. 0. Lamarckiana and 
0. grandiflora have often been confused with each other, and there 
was frequent failure to recognize these two as independent forms. The 
same is true of 0. hiennis and 0. Lamarckiana. DeVries (1905) has 
