three-cleft. Petals rather more yellowish, scarcely so long as the stamens. They are represented too narrow and acute in M. Redouté's figure. Capsules combined almost all the way up, making together a roundish-elliptical three-lobed figure, larger than a coriander seed, of a light brown, transversely corrugated, their points widely spreading, crowned with the short thick styles, and capitate stigmas. The seeds are extremely numerous. The wooden cut of the old authors above cited, is quite as expressive as the copper-plate of Seguier, commended by Villars, and copied by Lamarck.
Gathered by Kalm in North America, but in what part we are unable to determine. Three of his specimens are preserved in the Linnæan herbarium.
This, which Linnæus did not distinguish from his Anthericum calyculatum, is most akin to our Tofieldia alpina, with which it accords in size and habit, as well as in bearing two, sometimes three, alternate distant leaves on the stem. The inflorescence is a dense obtuse cluster, one inch and a half long, interrupted in the lower part. The bracteas however afford a clear specific difference, being lanceolate, and extending not only to the summit of the short and thick partial flower-stalks, but often reaching much beyond the calyx, which latter is very broad and shallow, unequally three-cleft, sometimes with a few supernumerary teeth or notches. The petals are very different in shape from both the preceding species, being lanceolate and acute. Their colour seems a greenish white. Anthers pointed. Germens ovato-lanceolate. Styles twice as long as the last.
Tab. VIII. Fig. 1 represents a dried specimen of T. stenopetala, of the natural size, with the separate parts of fructification mag-