﻿Rydberg: Notes on Rosaceae 151 



baccus. It is a less robust plant than that species and suggests 

 somewhat a slender depauperate woodland form of R. nigrohaccus, 

 but the stem is perfectly terete, the sepals narrowly lanceolate 

 and the fruit is elongate, tapering above, and of many small, rather 

 dry drupelets. The plant also suggests a hybrid between R. 

 nigrohaccus and R. Randii, but the latter species is not found in the 

 region of R. allegheniensis. It is not at all the same plant as the 

 hybrid collected by me in the Adirondacks. 



Rubus argutus Link. The most distinct of the species of Rubus 

 described by Blanchard is R. Andrewsianus . It is one of the 

 most common and most widely distributed of our blackberries. 

 It was not a "new" species, however, and was called R. fruticosus 

 by Marshall. The trouble has been that it has been confused 

 with R. villosus of American authors, that is with R. nigrohaccus. 

 Blanchard has done us the great service of pointing out the 

 distinctiveness of the two. His name is, however, not the oldest 

 available name, for it is the same as R. argutus Link. Link's 

 description is incomplete, for he evidently did not describe 

 the new shoots in his diagnosis. Mr. Blanchard is not willing 

 to admit that R. Andrewsianus is the same as R. argutus and claims 

 that at best the latter is an abnormal form of the former. Pro- 

 fessor Bailey has kindly sent me a photograph, taken from the 

 type at Berlin. This photograph shows plainly the strongly 

 angled, almost grooved stem and stout prickles, characteristic 

 of R. Andrewsianus, and to my mind there is no doubt concerning 

 the identity of the two. Even if R. argutus should not be the 

 same, R. Andrewsianus is not the oldest available name, for R. 

 floricomus, described by Blanchard himself a year earlier, can- 

 not be distinguished from R. Andrewsianus. 



Rubus fioridus Tratt. is closely related to R. argutus and repre- 

 sents it in the south. Mr. Blanchard does not distinguish the two. 

 R. fioridus as a rule, however, is rather distinct, has thinner 

 leaflets with finer pubescence, those of the new shoots usually 

 narrower and more finely toothed; the prickles are also more 

 curved and more flattened and the branches terete. The species 

 may hybridize, which may explain the intermediate form from 

 regions where their ranges overlap. 



Rubus hetulifolius Small is closely related to R. fioridus, but 



