A Problematic Transatlantic Disjunction in Paepalanthus (Eriocautaceae)

Abstract

The monocotyledonous family Eriocaulaceae is extensively distributed, the most widespread genus, Eriocaulon, occurring in the Americas, Asia and Africa with one species in northwestern Europe. The genus Syngonanthus is also well represented in Africa and Central and South America. However, Paepalanthus, the largest genus of the family, is mainly restricted to the New World except for P. lamarckii Kunth, which is widespread in eastern and northern Brazil and sparingly so in tropical Africa, and P. pulvinatus N.E.Br., which has hitherto been considered to occur only in Africa. While studying Brazilian and African material of Paepalanthus in the Kew Herbarium (K), we noticed the striking similarity between P. pulvinatus (from Sierra Leone) and P. hispidissimus Herzog, a species from northeastern Brazil. Closer comparison of the two species has shown that they are synonymous. P. pulvinatus is well-defined by its habit, with densely-arranged glaucous, rosulate leaves (more than 200 per plant), which are pilose on the upper surface and glabrous below. Other distinctive characters of this species are the truncate sheath (spathe) with long terminal hairs and the inflorescence which, unusually for the genus, develops laterally rather than terminally. This species has hitherto only been known from the type specimen collected by Bockstatt, and in his monograph of the family, Ruhland (1903) mentions that though he did not examine the type, it appeared from the original description to be of doubtful affinity with, and very different from the other African members of the family. P. hispidissimus was characterized by Herzog by its very numerous, densely rosulate-caespitose leaves, and hitherto has only been known from the type material cited below. A careful comparison has shown that the two species are indistinguishable from one another in floral and vegetative characters.