1896 | ROSA! AMERICANA 15 
tinguish it from R.d/anda. Its slender and straight prickles, 
and its inflorescence with small and usually numerous flowers, 
separate it from A. Nutkana. Lastly, the form of its prickles, 
which are straight and not curved or hooked at the tip, permits 
no confusion with R. Californica. 
Despite these differences, there is more or less confusion. 
Thus, specimens of R. pisocarpa received from the Siskiyou 
mountains, Washington, from Mr. Pringle and Mr. Suksdorf, 
have been reported by Watson as R. Californica. This error came 
from the appearance of the specimens and a too narrow concep- 
tion of R. pisocarpa. They do not always show the small rounded 
fructiferous receptacles as large as a pea, such as were described 
by Asa Gray. These receptacles can become notably larger, 
either strongly constricted at the neck, or ellipsoidal. On the 
other hand, the leaflets, which are glabrous or pubescent, some- 
times show small glands upon the lower surface. In the last 
case, the teeth either remain simple or become glandular-com- 
pound. Among the rich material from Washington sent to me by 
Mr. Suksdorf, and which I have placed in my herbarium under 
R. pisocarpa, there are forms whose leaflets are more or less 
attenuate at base, as in RX. d/anda, instead of rounded as in the 
type. Will the contraction of the lower part of the leaflets 
necessitate the making of a species distinct from R. pisocarpa ? 
This is impossible, for this difference is not to be regarded. 
Such contraction will always weaken the amount of distinctive 
characters which separate R. pisocarpa and R. blanda. 
R. pisocarpa seems to be subject to the same variations as are 
R. Nutkana and R. blanda from dwarfing and enlarging, and to 
show very marked differences between specimens from dwarf 
bushes and those from more vigorous and taller bushes. 
The geographical range given to R. pisocarpa by Watson 
seems to me to need extension, and that, too, probably at the 
expense of the two groups of forms which he has included under 
_ the names R. Fendleri and R. Woodsii. Towards the north Wat- 
son does not extend the limit beyond the southern part of 
British Columbia. Does the species occur farther north? I am 
