A evergreen shrub or small tree of bushy habit; young shoots, buds and undersurface of the leaves clothed with a whitish felt. Leaves opposite, oval or ovate, toothless, blunt at the apex, tapered or rounded at the base, 1 to 3 in. long, 3⁄4 to 11⁄2 in. wide, dark lustrous green and glabrous above except when quite young, whitish, felted and conspicuously veined beneath; stalk 1⁄4 in., or less, long, felted. Flowers, solitary, produced in July from the lowermost leaf-axils of the current year’s shoots, usually two or four of them on each shoot; each flower is 11⁄4 to 13⁄4 in. wide and borne on a felted stalk 1 to 11⁄2 in. long. Sepals four, roundish-oblong, 3⁄8 in. long, felted outside, reflexed. Petals four, broadly oval, concave, finally reflexed, red in the centre, whitish at the margin. Stamens very numerous, erect, 3⁄4 to 1 in. long, rich crimson. Fruit an egg-shaped berry up to 2 in. long, with the remains of the calyx at the top, with a pleasant, rich, aromatic flavour. Bot. Mag., t. 7620.
Native of S. Brazil and Uruguay; discovered in 1819 by a German collector named Sellow, after whom it was named. This shrub is not hardy in the open at Kew, but grows and flowers well on a south wall there. With this protection it has been grown for many years in the R.H.S. Garden at Wisley and succeeded even in Norfolk, at Little Hadden Hall. At Wisley it seldom flowers, however. The flowers are beautiful and richly coloured, especially the brush of long, erect, crimson stamens which constitute the conspicuous feature of the blossom. It grows well in light loamy soil and should have full sunlight. It can be increased by cuttings placed in heat in July or August. Considering the length of time it has been known, this fine shrub has been rather surprisingly neglected in British gardens. It is 18 ft high at Trewidden in Cornwall.
This species very rarely sets fruits in this country even in the milder parts, but the reason for this may be that it is self-sterile. Messrs Duncan and Davies, the New Zealand nurserymen, recommend that to ensure cross-pollination at least two plants should be grown; these would, of course, have to be seedlings, or belong to different clones, for the cross-pollination to be effective.