332 PROCEEDINGS OF THE AMERICAN ACADEMY 



lectors that are in our herbariums show in most cases how far it was 

 done successfully. Of his R. nitida, R. lucicla, R. Woodsii, R. Caro- 

 lina, and R. hlanda in part, there can be no doubt. Some specimens 

 that I would now place under R. Sayi are referred to R. hlanda, and 

 others to R. stricta, var. The species of the Northwest Coast (/?. Nnt- 

 Jcana) is referred in part to R. fraxinifuUa, following Scringe in the 

 Prodromus, and in part to R. cmnamomea. Some specimens of 

 R. Woodsii are referred to R. majalis ; and it is probable that the 

 specimens from Lake Huron named R. IcBvigata ? wei'e R. setigera. 



The next revision of any moment is that by Torrey & Gray, in 

 the Flora of North America (1840). Here R. setigera is for the first 

 time identified with R. rubifolia, and R. Carolina, R. nitida, and R. 

 blanda (mainly) are clearly defined, while R. lucida is made to include 

 all other forms of the Atlantic States. R. Woodsii and R. Califor- 

 nica are adopted with no other knowledge of them than that derived 

 from the original descriptions ; while R. cinnamomea, R. fraxinifolia, 

 and R. stricta are taken from Borrer, with little variation, for the 

 then little known western and northern forms. In addition, two re- 

 markahl}' distinct species of Nuttall's are for the first time described ; 

 viz., R. foliolosa, from Arkansas and Texas, and R. gymnncarpa, from 

 Oregon. No reference is made to Schweinitz's R. Sayi, which is 

 first noticed by Eaton & Wright, in the eighth edition of Eaton's 

 Manual (1840), who otherwise follow Borrer, adding R. parvijlora, 

 R. gemella, and R. rubifolia. 



R. Maximiliani, collected by Prince Maximilian von "Wied in a 

 journey to the Upper Missouri, and published by Nees von Esenbeck 

 in 1841, is the same as Lindley's R. Woodsii, as is evident from 

 Nees's careful description. 



Torrey, in 1843, in the Flora of New York, described the three 

 species of that State as R. Carolina, R. lucida, and R. blanda. his 

 R. lucida being the common 7?. hwnilis, from which he failed to dis- 

 tinguish such specimens of the rarer R. lucida as he may have seen. 

 In New England, where R. lucida is frequent, the difficulty in uniting 

 the species was greater, and we accordingly find Emerson in the 

 Woody Plants of Massachusetts (184G) recognizing R. lucida (the 

 '•early Wild Rose," R. humilis), R. Carolina (the "Swamp Rose"), 

 and R. nitida (the "Shining Rose"), the last including the real 

 R. lucida. Wood in his Class-Dook (1846, and later editions) in- 

 cludes 7?. lucida under R. Carolina, his R. lucida also being 7?. humilis. 

 Finally, Dr. Gray, who in the first edition of his Manual (1848) had 

 retained R. nitida as distinct, in the second (1850) and later editions 



