328 RYDBERG: NOTES ON ROSACEAE 
amendment and well knowing the publication of A. pubescens 
Wallr., which he gives as a synonym. 
Agrimonia Bicknellii (Kearney) Rydb. This was well de- 
scribed by Mr. Bicknell* as a variety of A. mollis, but without a 
varietal name. This was supplied a year later by Mr. Kearney. 
It is true that the characters separating A. pubescens (= A. 
mollis) and this species are not absolute and that intermediate 
forms are not lacking. A. pubescens is at home in the central 
states west of the Alleghenies, but extends east thereof into Vir- 
ginia and Georgia. The home of A. Bicknellii is the Atlantic 
coast, but it is found as far west as Pennsylvania and Tennessee. 
In the Mississippi valley only A. pubescens is found and on the 
Atlantic border only typical A. Bicknellii. The intermediate 
forms are found in the Alleghenian region, where the ranges of 
the two overlap. These two species are the nearest American 
representatives of the European A. Eupatoria, but have much 
smaller fruit. Especially is A. Bicknellii sometimes hard to 
distinguish from A. Eupatoria without the fruit. Two western 
specimens in the United States National Herbarium I have de- 
termined doubtfully as A. Bicknellii, viz. one collected at Fort 
Snelling, Minnesota, by Mearns, and the other at Naperville, 
Illinois, by Umbach. They may belong to A. Eupatoria L. 
Agrimonia Eupatoria L. Britton and Bicknell believed that 
this species was not found at allin America. No specimen has been 
seen from the East, where the species is most likely to be found 
introduced. There are, however, two specimens, one in the 
National Herbarium and one in the herbarium of the Missouri 
Botanical Garden, which without any doubt belong to the species. 
Holzinger’s specimen, especially, has the fruit so well developed 
that there is no question of the identity. The specimens are: 
MINNESOTA: Winona, 1889, Holzinger. 
Wisconsin: Mirror Lake, 1903, Eggert. 
Agrimonia striata Michx. This species has been badly mis- 
understood. Probably the real cause of this is that in Michaux’s 
herbarium there are two specimens on the sheet of A. striata. 
The left-hand specimen represents the plant, here treated under 
that name, and the right-hand specimen is one of A. rostellata. 
* Bull. Torrey Club 23: 517. 1896. 
