WAGENINGEN AGRICULTURAL UNIVERSITY PAPERS 94-4 (1994) Arevisio n ofLeucomphalo s including Baphiastrum and Bowringia (Leguminosae - Papilionoideae) F.J. Breteler Herbarium Vadense Department ofPlant Taxonomy WageningenAgricultural University,The Netherlands Date ofpublicatio n 19-8-1994 Wageningen mm Agricultural University CÄVI =%<^c, Q is) CIP-GEGEVENSKONINKLIJK E BIBLIOTHEEK, DEN HAAG Breteler, F.J . Arevisio n ofLeucomphalo s including Baphiastrum and Bowringia (Leguminosac-Papilionoideae) / F. J. Breteler, - Wageningen : Agricultural University, -111 . - (Wageningen Agricultural University Papers, ISSN 0169-345X; 94-4(1994)) ISBN 90-6754-362-4 Trefw.: leguminosac : plantentaxonomie © Agricultural University Wageningen, TheNetherlands , 1994 No part of this publication, apart from abstract, bibliographic and brief quotations embodied in critical reviews, may be reproduced, recorded or published in any form including print, photocopy, microform, electronic or electromagnetic record without written permission from the publisher Agricultural University, P.O. Box 9101, 6700H BWageningen , the Netherlands. Printed inth eNetherland s byVeenma n Drukkers, Wageningen Contents Summary 1 Introduction 2 Generic delimitation inth eBaphia group 3 Relation ofLeucomphalos wit h other genera ofth e Baphia group 6 Geography ofth eBaphia group 7 Cytology 10 Description ofth egenu sLeucomphalos 10 Keyt oth e species 11 Alphabetical treatment ofth e species 13 Excluded species 35 Note concerning Fig. 1-3 35 Literature cited 38 Index to scientific names 40 V Summary The related genera Baphiastrum, Bowringia and Leucomphalos are taxonomically revised. They share a singular seed character and mainly on that account they are united under the name Leucomphalos. Six species are recognized of which one is new from Liberia. They are described and mostly amply illustrated. A key to the species is provided and their distribution is mapped. Five species occur in Africa, mainly in Central Africa, one in East Asia. The distribution of the genus is com­ pared with the distribution of the other genera of the Baphia group of the Sopho- reae. Wageningen Agric. Univ.Papers 94-4 (1994) Introduction The present revision is the result of an investigation that started quite some time ago, in fact as far back as the early sixties when I was collecting specimens in Cameroun. The first observations concerned Baphiastrum brachycarpum and Bowringia mildbraedii, both well represented in the eastern part of that country. The seeds of these two species are rather peculiar but so much alike that one won­ derswh y they belong totw o different genera, whilejudge d from outside they could have easily originated from a single specimen! The problem was laid aside till the late sixties when a student named J.W. Meijeraan did some additional research. He studied the genus Bowringia from Africa and Asia, and found that the Bowringia material from Liberia represented ane w species. More than two decades have past since then before the original question wastake n up again. In the meantime much more material became available and the position of the genera inth eSophoreae was studied andbecam e more clear (Polhill, 1981).A new genus related toBaphiastrum was created by Brummitt (1968):Airyantha, and So- ladoye (1985)revise d the large genus Baphia. This revision answers the question put forward in the early sixties, and gives some comment on the classification of the remaining genera that were considered tob e closely related. WageningenAgric. Univ. Papers 94-4 (1994) Generic delimitation inth eBaphia grou p The genera Baphiastrum, Bowringia, and Leucomphalos together with Airyan­ tha, Baphia and Dalhousiea constitute the Baphia group in PolhilFs (1981) treat­ ment of the Sophoreae. In Yakovlev's (1991) classification the first 3 genera form group B,th e other 3grou pA ,an dbot h groupstogethe rhi strib e Baphieae. The genus Leucomphalos was published by Planchon in 1848 and so far re­ mained monotypic. Bowringia was based by Bentham (1852) on a species from Hong Kong. Later on 3 African species were added, one from West Africa, one from Central Africa and one from Madagascar. Harms (1913) described Baphias­ trum and based it on material of a single species from Cameroun. Several species were added, some of them being transferred to it from Baphia. The boundaries be­ tween the latter genus and Baphiastrum were elucidated by Brummitt (1968) when he created the genus Airyantha. Baphiastrum in fact appeared to be monotypic again. Polhill's Baphia group is mainly characterized by unifoliolate leaves, free sta­ mens with basifixed anthers and dehiscent fruits. As regards the calyx the group is rather variable: cupular with short lobes in Bowringia and Dalhousiea, or invagi- nate with small apical lobes or not (Leucomphalos). When lobed the lobes may be more or less imbricate (Airyantha) or valvate (Baphia, Baphiastrum). The corolla also shows a wide variation. In Leucomphalos s.s. the petals are only slightly un­ equal and have no claw (fig. 10.4) and the flowers are ± regular as in the Cadia group of the same tribe, while in the other genera of the Baphia group the corollas are distinctly papilionoid. The greatest variation within this group, however, is found in the characters ofth epo d and seed. This can be classified in 3 distinct sub­ groups.Th e first comprises the generaBaphia andDalhousiea with an elongate flat pod with discoidal unicoloured brown orblackis h seeds.Th e second subgroup with elongate spiral pods which are constricted between the ± lenticular, red seeds is represented byAiryantha. The third subgroup has short, few-seeded pods and com­ prises the genera Baphiastrum, Bowringia and Leucomphalos. The seeds of this subgroup show a character that isuniqu e in thePapilionoideae, viz. the short thick cotyledons with their opposing flat faces perpendicular to thepo d axis,an d provid­ ed with a distinct cupular strophiole or aril (small and ± flatfish in Bowringia dis­ color). Polhill (I.e.) does not mention this orientation of the cotyledons explicitly andth e unique character isno t dealt with inth e classification ofthes e 3genera . But Yakovlev (I.e.) does mention it and groups the genera accordingly. It is doubtful if thistyp e ofseed sma yb e classified asovergrow n (Polhill,I.e.) . The question has now to be raised whether the genera of the third subgroup showing this unique character, are to be united or to be kept separate, and where they have to be placed: in a separate group or tribe of their own, or are they to re­ main inth eBaphia group. Wageningen Agric. Univ. Papers 94-4 (1994) 3 O .2 e _ + S S " 2 « S? U -fc! y ^ X) g * ê'fi & <: o < S £ O U < so a, 'S* ~° X 7 s S I & S v. 0 o w <L» 6 <D u e 1 e j^ "cd td S rt Is« »en 3 e 3 O 3 O ^ 0 m 3 1 ir1j u 1 'S. -^1 3IZ] 'EH t->fi 'S. ^ â 'E. ai — cd =3 "Ecd . "—I cd O (N cd g "S *o -^o. w cd — -a 3 eS r- C 'T ^_^ o ^ PE x eSi «£> «î , t= o .- 0 3 XI CO % "S ~ "« s 1 <U & "5 g u o -O BB S •3 OÙ S D- _o O. •3 o t: 5 S -fi Ö <u CJ >* CJ u J= •* X>J> J * f« . •s -S -o .g •a -s P '1 • Som ^ 1 "5.S c > *- cd CJ S xî c x> OÙ S cj S g. » 08S O a s » o I g.Jä 0 O .S ^ u c •pu- 0 XI T3 ^ C .—1 cd cd S o. o .c cd ;> J- 3 ^ . - • ~ C! 3 V >-! a. '0 cd c « -o -o -s c •O •0 c -g a, ° Ms S g •- h w ,0 +1 Ic/ î PH <-> s .2 g +1 'S. •S s I E ca 1 Ils | D- P. o a c u 'S. " È o. o 3 cd fi 3 GH O o 5 «1 S t» 6 fi Ö e3 > >> C ^ I Ü .'S o a> ."S u X> x> •§ U S .S •g & H ~- Cd M _2 ^ t/3 T3 .fi KJ <D >* "TJ 'S >.^> Cl 43 ~- •^ _^ cd w >> >> § •S 5- -^ Xi 3 c - c ^ c 2 u S £ S -u S 30 in u t: cd •S P 3 3 « I DH cd ^ « C cd c« S en cd -fwi 3C/3 'w "cd (N o a. cd o b •S o - I T <~ -o o u s p. I ^3 WageningenAgric. Univ.Papers 94-4 (1994) In a note on a Zenker specimen of Leucomphalos capparideus in the Paris Herbarium, Pellegrin made the following observation: 'Pour moi ce genre me sem­ ble très très près de Bowringia. Voir si ce n'est pas le même genre. Mêmes fruits, mais calice et étamines différent'. Brummitt (1968)whe n comparing hisne w genus Airyantha with related genera mentioned the close resemblance of the pods and seeds of Baphiastrum, Bowringia and Leucomphalos. However, Soladoye (1985: 306) found that Baphiastrum had seeds like Airyantha. The 3 genera with similar pods and seeds areno t placed together inhi s scheme (p.305 ) showing the phenetic relationships withinBaphia proper andwit hrelate d genera. The species of the 3 genera forming subgroup 3 have been analysed to see whether they belong to one genus or not. In table 1th e main evolutionary charac­ ters are shown. The differences in the calyx (calyx-corolla ratio, and opening at an- thesis), used for a long time to separate Bowringia from the other two genera are rather gradually evolving from perfectly 2-valved without lobes in Leucomphalos s.s. to the short cupular one inL. callicarpus, the type species ofBowringia. As re­ gards the corolla all species, except L. capparideus, are quite papilionoid, but this character has never been employed in generic delimitation.