2 BOTANICAL GAZETTE 
tation of American roses. Watson did not always seize upon th 
natural affinities which hold together certain forms, being misled 
one taking the name RX. Sayi Schwein.; then later into a thir 
species, R. Engelmanni These three species are actually bu 
three varieties of the same specific type. Also he has recogniz 
as a distinct species R. Arkansana Porter, which is perhaps but 
variety of R. blanda Ait. He has preserved asa separate specie 
R. lucida Ehrh., which is indeed but a variety of R. ha a 
Marsh. Finally, he has organized two specific groups under th 
names RX. Fendleri Crépin and R. Woodsii Lindl., each of whic 
seems to me to be composed of heterogeneous elements. A 
for his R. Mexicana, the few specimens which have been collected 
scarcely permit me to know whether he has well separated it as 
a distinct species. In any case it seems to belong to the section 
CAROLINA. 
In the preface of his monograph Watson says that if the 
roses were reduced to their primary types North America woul 
contain but nine species. This proposed condensation indicate: 
Mr. G. N. Best, took u 
which he has publish 
deserve to be consult 
p the study of American roses, concerni 
ed a number of interesting notes,’ whi 
ed by all who study the genus. 7 
Thanks to researches and multiplied observations which have — 
_ been made, the acquaintance with a certain number of species is 
3 Garden and Forest, 1887, 
46N 
88 
i 
ouvelles remarques sur les roses américaines,” iid. 26: —. 1887, and 
5 Remarks on 
the group CAROLIN# of the genus Rosa, in Bull Torr. Bot. C 
1887. Remarks on the group CINNAMOME of the North American Roses, id 
Ameri Toses; remarks on the characters with classification, in /é -“ 
Trenton Nat. Hist. Soc. 1889, : 
