Notes on Malesian Fabaceae ( Leguminosae-Papilionoideae ) 15 . Notes on Indigofera

Notes on some morphological features of Indigofera as well as taxonomic notes on several species are given. The recently described I. erectifructa is reduced to a synonym of I. luzoniensis.


INTRODUCTION
Indigofera L. is a pantropical genus of c. 700 species (Lewis et al. 2005), of which 17 occur in the Flora Malesiana area.The genus comprises herbs, shrubs and small trees with simple or unifoliolate, trifoliolate or imparipinnate leaves and small flowers in axillary racemes.It is often easily recognised by its biramous hairs (T-shaped or balance hairs) and reddish flowers.Several species are or have been cultivated for the blue pigment indigo (Lemmens & Wessel-Riemen 1991).
For the Flora Malesiana area and adjacent Southeast Asia the genus was revised by De Kort & Thijsse (1984).Since that time only one new species for Malesia was described: Indigofera erectifructa Y. Endo, H.Ohashi & Madulid (2005).In the following sections a few notes on morphology and taxonomy for several species are given.Corrections to typifications proposed by De Kort & Thijsse will be discussed.

Emergences
Several types of emergences are briefly discussed by Schrire (1995) and Wilson & Rowe (2004).Apart from biramous and simple hairs, glandular hairs, colleters and pearl bodies are found.Schrire (1995) calls the latter two structures pearl bodies, Wilson & Rowe (2004) tend to call most structures emergences.In the species of Indigofera of the Flora Malesiana area hair-like, dark-coloured emergences are common on axial parts, in axils of stipules, leaves, stipellae, leaflets and bracts.In notes to the species I have called these structures colleters.Pearl bodies are present in several species, often they are club-shaped (clavate), on the lower surface of the leaflets of I. glandulosa and I. trifoliata they are rounded, disc-shaped and called glands.These 'glands' are according to Wilson & Rowe (2004) similar to the hollow hairs (cavitated trichomes) described by Prabhakar et al. (1985) and Vyay Kumar et al. (1986).The 'colleters' may be similar to the multiseriate, cylindrical hairs described by Vyay Kumar & Ramayya (1986).

Tissue remnants
Pods of several species break off at or just before dehiscence leaving a parenchymatous tissue remnant or base.Schrire (1995: 201-203, pl. 15, f. 6d) calls this a unique feature of the tribe Indigoferae (pl.15, f.6d: persistent cup-like base remaining with calyx once pod valves are caducous), however, he does not use the character (or its loss) in his cladistic analysis.(The character was left out of the analysis because its homology with similar features in other genera is in doubt (pers.comm.Shrire).)In the species of Indigofera of the Flora Malesiana area this tissue remnant is present in: I. arrecta Hochst.ex A.Rich., I. glandulosa J.C.Wendl., I. linifolia (L.f.) Retz., I. nigrescens Kurz ex Prain and I. trifoliata L.

Development of hairs on ovaries
De Kort & Thijsse (1984) described several species with glabrous ovaries and hairy fruits.Closer inspection shows that ovaries of these species, especially older ones, have at least some hairs.The young pods are rather densely hairy, older pods may have less hairs.Probably hairs on ovaries develop late and pods may lose hairs on aging.
Richard (1847: 184) described I. arrecta using as the base a Hochstetter name for a Schimper specimen.The protologue reads: "Indigofera arrecta Hochst., in pl.Schimp.Abyss., sect.III, no.1923".He furthermore cites specimens collected by Guartin Dillon and Ant.Petit.Schrire (1998: 666) has taken the protologue as a kind of typification and gives Schimper 1923 as the type with the holotype at P and isotypes at K, L, LE, MO, P.However, all the 'type specimens' seen by me bear a printed label with the number 1923b.Moreover, as Richard cites three collections, a lectotype should have been selected.As the Paris herbarium holds at least three duplicates of Schimper 1923b the holotype should be indicated more precisely.Here P 00569218 is selected from among the material labelled Schimper 1923b as lectotype of Indigofera arrecta.

Indigofera arrecta
Hochst.ex A.Rich.(1847)184.-Lectotype (Schrire, here designated): Schimper III 1923b (P 00569218; iso K, L, LE, MO, P), Ethiopia, Tigray Region, Adoa, Oct. 1842.The glandular hairs often differ in length within as well as amongst specimens.In the Flora Malesiana area they are fairly uniform, the longest hairs on twigs are 0.1-0.8mm long and on pods 0.1-0.3mm long.In Australia the longest hairs on twigs are similar to those of specimens of the Flora Malesiana area, but up to 1.1 mm long; in continental Asia they are 0.5-1.7 mm long; in Africa 0.6-2.5(-5)mm long.There the longest glandular hairs of specimens present in L are found in Stolz 694 (4 -5 mm) and Tanner 3142 (3.5 -4 mm).As these two specimens have pods with few endocarp spots they may belong to I. zenkeri Harms.The type specimen of I. zenkeri has glandular hairs up to 4 mm long.Although the important seed character could not be assessed, due to the few pods available, I am convinced that Nemoto et al. 10630, the type of I. erectifructa, belongs to I. luzoniensis.The synonymy of I. luzoniensis becomes: Rottler sent to Willdenow by Klein, annotated in Rottler's handwriting with Indigofera glandulosa Roxb.This specimen was the base of Willdenow's I. glandulosa as the sheet also contains comments in Willdenow's handwriting.As Wendland and Willdenow were describing the same species, indicating it with the same German name, I have selected the Rottler specimen from the Willdenow herbarium as the neotype of Wendland's I. glandulosa.The synonymy becomes: Indigofera angustifolia.This species was renamed Indigofera tinctoria by Naves (1880).Both names are illegitimate.Merrill (1918) identified this plant as Indigofera hirsuta L. De Kort & Thijsse (1984) named Merrill's representative specimen Spec.Blanc.403 as the type of the names of Blanco and Naves.However, as no Blanco material of his I. angustifolia is known, they should have selected a neotype.As such I designate Merrill Spec.Blanc 403.In the synonymy of I. hirsuta it becomes: Indigofera tinctoria Blanco, seemingly with no references to the earlier description of I. tinctoria by Linnaeus.According to Merrill (1918: 179) the Blanco species is identical with the Linnaean one.De Kort & Thijsse (1984: 136) followed Merrill (1918) and placed the Blanco name in the synonymy of I. tinctoria L. As the type of I. tinctoria Blanco they cited Merrill's representative specimen Spec.Blanc.826.However, this cannot be the type.As no Blanco material of his I. tinctoria is known they should have selected a neotype.Here I select as neotype of I. tinctoria Blanco Merrill Spec.Blanc 826.
Kort & Thijsse (1984)m the Lesser Sunda Islands (Cinatti 108, Iboet 123, Jaag 417, Pleyte 93, Spanoghe s.n.) have very few or no glandular hairs.Identification of such specimens by means of the key in De Kort & Thijsse (1984) may lead to I. suffruticosa.Indigofera erectifructa Y.Endo, H.Ohashi & Madulid In 2005 Endo et al. described a new species of Indigofera from Luzon: I. erectifructa.The fruits of this species are almost erect and therefore the authors made a comparison with erect-fruited species such as I. galegoides DC., I. doachengensis Y.Y.Fang & C.Z.Zheng and I. densifructa Y.Y.Fang & C.Z.Zheng(Endo et al. 2005).The unknown plants were clearly different from these species and a new species was described.Although the authors consulted the treatment by DeKort & Thijsse (1984)no comparison was made with species recorded for the Philippines.Amongst others De Kort & Thijsse's new species I. luzoniensis has ± erect fruits, called 'ascending' by DeKort & Thijsse (1984:  127).After receiving the type material of I. erectifructa on loan a careful analysis of the characters showed that the differences between I. erectifructa and I. luzoniensis are very small.In 1798 Wendland described I. glandulosa with the German name 'Drüsige Indigo'.I asked Göttingen to search the Wendland collection for a specimen of his I. glandulosa, but I have not received any answer to this request.According to Sanjappa (1995) Wendland's herbarium in Hanover was destroyed.In 1803 Willdenow described an I. glandulosa with the same German name, contributing the epithet to Roxburgh.In the Willdenow herbarium, sub nr.13891, there is a specimen collected byIn 1841 Spanoghe described a new species Indigofera albicans based on 'Hallia glauca Zipp.'.Zippelius wrote the name 'Hallia glauca' on labels of several specimens he collected in Timor.One of these specimens bears a Spanoghe label 'Indigofera albicans', but lacks the indication that Zippelius collected it.As this specimen was seen and annotated by Spanoghe this collection(L 908.116-1306)is selected as the lectotype of I. albicans Span.In the synonymy of I. linifolia it should read: Indigofera albicans Span.(1841) 190. -Lectotype (here designated): Anon (Zippelius?) s.n.(L, L 908.116-1306), Timor.In the synonymy of I. tinctoria L. this becomes: Indigofera tinctoria L. (1753) 751 Indigofera tinctoria Blanco (1837) 591; (1845) 413; (1879) 499; Merr.(1918) 179.-Neotype (here designated): Merrill Spec.Blanc 826 (L; iso BM, BO, K, US), Philippines, Luzon, Rizal Prov.Antipolo.See for more complete synonymy De Kort & Thijsse (1984).