352 RYDBERG: NOTES ON ROSACEAE 
In the Gray’s New Manual, published September 18, 1908, the 
authors, Robinson and Fernald, stated under Potentilla Anserina 
as follows: ‘‘Var. grandis T. & G. is merely a luxuriant state in 
rich meadows.” It is strange that so shortly after, in January, 
1909, Professor Fernald should admit this variety as a good species. 
The difference in the achenes, pointed out by me, evidently per- 
suaded him that it could not be kept in Potentilla Anserina L. 
(Argentina Anserina Rydb.). It would have been asking too 
much if one expected him to admit all at once the eight species 
recognized in my monograph. My intention is not to defend them 
as species. The limitation between species and variety will al- 
ways be arbitrary, so also between variety and form. If Professor 
Fernald admitted Argentina argentea as a variety under the name 
Potentilla Anserina var. sericea, he should have admitted A. 
occidentalis, A. litoralis, and A. subarctica also as varieties. There 
are certain statements, also, in Fernald’s article, to which I must 
take exception. As Professor Fernald has admitted Argentina 
argentea Rydb. as.a variety I should feel content so far as this 
species is concerned, but I can not pass it altogether. The speci- 
men in the Gray Herbarium from St. John Valley, referred to in 
Professor Fernald’s paper is ‘‘in spite of its leaves being silvery on 
both sides,”’ not A.argentea. I remember the specimen very well. 
Neither is A. argeniea exactly the same as Potentilla Anserina 
sericea Hayne. Both have leaves silvery on both sides, but in the 
Rocky Mountain plant the leaflets as a rule are decidedly obovate, 
with comparatively few ovate teeth, while in the Eurpoean plant 
the leaflets are usually much longer, elliptic, and with numerous 
lanceolate teeth directed forward. I have not seen any specimens 
of Argentina argentea east of South Dakota. It needs therefore 
not to be considered in connection with the flora of eastern 
North America. 
In discussing the species with nongrooved achenes, Professor 
Fernald criticizes my key, in which I separate P. pacifica and P. 
occidentalis from the rest by the petals being “usually over 1 cm. 
long rounded-obovate,” while the others have petals ‘“‘6-8 mm., 
rarely I cm. long, usually elliptic-obovate.”” It may be true that I 
have “never known the full beauty of its large flowers” [A. litoralis] 
and that the given characters do not hold. I did not see the speci- 
