AGRIMONIA PUMILA MUHL The smallest of our species, high, erect or more slender and assurgent, simple or with a few ascending branches from the lower part of the elongated terminal raceme. Stem clothed with loose often coarse pubescence and hirsute with spreading hairs which become subappressed above and obsolete in the pubescent racemes. Leaves often crowded low on the stem, often trifoliate, small, I long, I wide, the upper ones much re duced. Leaflets firm, 1-2 pairs, the lower pair always small, oval to obovate or the odd leaflet cuneate, sessile or subpetiolulate, ob tuse or rounded at the apex or sometimes acute, crenate or den tate or sometimes boldly crenate-dentate, commonly y,' long, dark-green above and loosely appressed-hairy to nearly smooth, the margins loosely appressed-ciliate, paler and softly pubescent below, pilose-pubescent or hairy along the veins, obscurely, if at all glandulose ; leafstalk villous ; subleaflets wanting or a minute pair, entire or few-dentate. Stipules small, the main pairs rounded and cordate-clasping, acutely dentate-lobed or incised, some times dentate on the inner margin. Racemes loosely-flowered, often remotely-flowered below. Flowers small, the buds subglobose. Fruit small, 2" long, t"--1" wide, turbinate or subhemisphcric, minutely glandulose, often canescent in the sinuses when young, the disk mostly obscurely margined and rising to the short and broad calycular process, the bristles loosely ascending and erect.
Roots developing short tuberous thickenings. (Plate 283, fig. 5.) This species is nearly related to A. moth's and occasional de pauperate plants of the latter are with difficulty distinguishable from it. I think there can be no doubt, however, that the two species are distinct and that a comparative study of growing plants would disclose more pronounced differences than dried specimens have revealed. The uniformly small size and simple or nearly simple habit of pumila and its slender and elongated loosely-flowered raceme are usually sufficient to distinguish it. The pubescence seems to be generally coarser and more hirsute than in mollis, and the much smaller, mostly trifoliate leaves more crowded on the lower part of the stem and apparently of thicker texture. I have not seen satisfactory material in mature fruit. The range of the plant is clearly more restricted northward than that of mollis, and more extended southward as far as at pres ent known. • I adopt Muhlenberg's name prinda for this species with some hesitation and only to avoid the alternative of conferring a new name. Muhlenberg characterized his plant by the one word -little" and gave its habitat as Mississippi. Applying the rule of exclu sion it would appear that only this species could have been in tended. If it should be found that A. molltis extends into Missis sippi the name would have only a dubious claim to availability.