RYDBERG: NOTES ON ROSACEAE 3857 
It is usually much taller and the leaflets are linear-oblong, 5—8 cm. 
long and only 1-2 cm. wide, obtuse or rounded at the apex and 
with lanceolate teeth. It was described as Comarum angusti- 
folium by Rafinesque. In the European form the leaflets are 
elliptic or oval, mostly acute at both ends, with broad and ovate 
teeth. This is also the prevailing form in arctic and subarctic 
America, but the two forms grade in numerous ways into each 
other, and C. angustifolium can be regarded scarcely more than a 
variety. So also C. tomentosum Raf. (Potentilla palustris villosa 
Lehm.). All grades of pubescence occur from a dense and sub- 
velutinous one to practically none. 
DUCHESNEA 
Dr. Wolf included Duchesnea indica (Andrews) Focke in 
Potentilla and associated it with P. reptans, P. canadensis, etc. 
With Dr. Wolf’s broad view of genera this was very natural, but 
I can not see how he could keep out Sibbaldia and even Fragaria 
from his Potentilla. All three have lateral styles, and Sibbaldia 
is even habitually as close to the typical Potentillae as Duchesnea is. 
FRAGARIA 
It is exceedingly hard to draw specific lines in this genus. It 
is easy to distinguish F. vesca or F. americana from F. virginiana 
or F. grandiflora, for in the former two the achenes are wholly 
superficial, while in the latter two they are set in deep pits in the 
fleshy receptacle, but such species as F. californica and F. mexicana 
connect the two groups, the achenes being in very shallow pits. 
A twelve years’ study of the genus since the publication of my 
Monograph of the North American Potentilleae, had not changed 
my views, and the recognized species remained practically the 
same in my new revision in the North American Flora, only that 
three new species were described, two varieties given specific 
rank, and F. multicipita Fernald and the introduced F. grandiflora 
Ehrh. added. The treatment in my monograph was not followed 
either by the authors of the Gray’s New Manual or by those of 
the New Manual of the Central Rocky Mountains; and I scarcely 
expected that it would. The former admits only two species and 
two varieties as growing in the northeastern United States, while 
