Wood Anatomy of Caesalpinia S.S., Coulteria, Erythrostemon, Guilandina, Libidibia, Mezoneuron, Poincianella, Pomaria and Tara (Leguminosae, Caesalpinioideae, Caesalpinieae)

Abstract

Caesalpinia s.l. traditionally comprised c. 140 species in the New and Old World tropics, and contained a maximum of 25 generic synonyms. The genus in its broadest sense has been shown to be polyphyletic in molecular studies, and most species have now been assigned to reinstated segregate genera: Caesalpinia s.s. (c. 25 spp.), Coulteria (10 spp.), Erythrostemon (13 spp.), Guilandina (c. 7 spp.), Libidibia (8 spp.), Mezoneuron (c. 26 spp.), Poincianella (c. 35 spp.), Pomaria (16 spp.) and Tara (3 spp.). About 15 Asian taxa remain unassigned pending more data, especially DNA sequences. In this paper we describe the wood anatomy of these nine segregate genera, outlining the features that consistently help define some of them. We have examined the wood of 27 species representing all the woody segregate genera and found wood descriptions of three more species in the literature. Most species lack well defined growth rings, vessels are solitary and in radial multiples, intervessel pitting is alternate and vestured, fibres are mainly non-septate, axial parenchyma is aliform to confluent and irregularly storied, and the rays are mainly 1–2-seriate, mostly non-storied, and of varying height. Prismatic crystals are in chambered axial parenchyma cells in all except Erythrostemon gilliesii (Hook.) Link, and in ray cells in many species. Libidibia is well defined, with storied axial parenchyma, narrow short storied homocellular rays and lacking crystals in ray cells. Tara is also well defined with non-storied heterocellular rays and some ray cells containing crystals. The other genera are less consistent in wood characters. In Caesalpinia s.s. the rays are not storied, and most species lack crystals in ray cells. Coulteria has some species with storied rays and all have homocellular rays and crystals in ray cells. Poincianella is particularly poorly defined from a wood anatomical point of view, perhaps indicating that it can be further segregated. A few Poincianella species have septate fibres, which are otherwise seen only in Libidibia corymbosa. Mezoneuron has non-storied, heterocellular rays. The two species of Guilandina we examined have wide vessels and heterocellular rays containing crystals. Only two species of Erythrostemon were examined and E. gilliesii was unusual in having ring porous wood and very wide rays (but the sample was cultivated at Kew, and we do not know its porosity in its native range). Caesalpinia decapetala (Roth) Alston (originally described as Reichardia decapetala Roth) and Caesalpinia sappan L. from the Old World have not been reassigned to a segregate genus. Pomaria is mainly herbaceous and we have included some information on it.