Clerodendrum umbellatum Poiret in de Lamarck (1804: 166)

Meerts, Pierre, 2023, The genus Clerodendrum (Lamiaceae) in the flora of Central Africa (D. R. Congo, Rwanda, Burundi), Phytotaxa 594 (1), pp. 1-42 : 33-35

26. Clerodendrum umbellatum Poiret in de Lamarck (1804: 166) ; Huber (1963: 442); Adam (1975: 1470); Verdcourt (1992: 97); Lebrun & Stork (1997: 514); Akoegninou et al. (2006: 985); Hawthorne & Jongkind (2006: 424); Lisowski (2009: 364); Lejoly et al. (2010: 295); Vande weghe et al. (2016: 744); Pollard (2022: 38), nom. cons. prop.

Type:— NIGERIA, près d’Oware, Palisot de Beauvois (G00366283), typ. cons. prop. (see notes) .

= Clerodendrum scandens Palisot de Beauvois (1808: 6 , pl. 62); Baker (1900: 304); De Wildeman (1905: 310; 1910a: 402; 1910b: 255; 1914: 191; 1920b: 173); Durand & Durand (1909: 440); Thomas (1936: 58) ≡ Clerodendrum umbellatum f. scandens (P. Beauv.) Moldenke (1983: 330) . Type:— NIGERIA, près d’Oware, Palisot de Beauvois (holotype: G00366283).

= Clerodendrum scandens var. asperifolium Thomas (1936: 59) ; Moldenke (1980: 219). Type:—REP. CONGO. Nola, 15 September 1913, Tessmann 2030 (holotype: B, not seen), nom inval.

= Clerodendrum congense Engler (1886: 65) ; Baker (1900: 301); Durand & Durand (1909: 438); De Wildeman (1912b: 47) ≡ Clerodendrum umbellatum var. congense (Engl.) Moldenke ( Moldenke 1980: 219) . Type:—D.R. CONGO. Boma, 5 September 1974, Naumann s. n. (syntype: B, not seen); Ponta da lenha, s.d. Naumann s. n. (syntype: B, not seen).

= Clerodendrum subreniforme Gürke (1900: 291) ; Baker (1900: 518); Durand & Durand (1909: 441); De Wildeman (1912b: 317); Thomas (1936: 56); Moldenke (1980: 219). Type:—D.R. CONGO. Sine loco, Dewèvre 917 (holotype: B? (not seen); iso-: BR0000008979452! (fragm.)).

= Clerodendrum fuscum var. lanceolatum Moldenke (1953a: 176) . Type :—D.R. CONGO. Bas-Uele, savane, 10 March 1935, Dewulf 755 (holotype: BR0000008976147!), synon. nov.

= Clerodendrum cordifolium ( Hochstetter 1842: 227) Richard (1850: 170) ; Baker (1900: 304); De Wildeman (1909: 132); Thomas (1936: 56). ≡ Volkameria cordifolia Hochst. Type:— ETHIOPIA. Near R. Tacazze, 8 December 1839, Schimper 1132 (First-step lectotype designated by Cufodontis (1962); second-step lectotype inadvertently designated by Verdcourt (1992): TUB (3 sheets: TUB003785, TUB003786, TUB003787); iso-: P00466157; G00366287, S08-10966; S08-10965).

Suffrutex, or small liana, (0.2–) 0.3– 4 m. Stem quadrangular, subglabrous, puberulent or, more rarely, pubescent; persistent petiole bases present, ca. 3 mm. Leaf: petiole (0.5–)1–7(–8.5) cm, glabrous, papillate or shortly pubescent; lamina ovate, 4–16 × 2.5–11 cm, cordate to truncate at base, more rarely rounded to obtuse, apex acuminate, membranose or coriaceous, 3–4(–5) veins on either side, upper surface dull to shiny, glabrous, puberulent, scabridulous or, more rarely, pubescent, reticulum inconspicuous or, more rarely, prominulent, lower surface gland-dotted, glabrous, or papillate to pubescent on veins, more rarely pubescent throughout, margin entire. Inflorescence in axillary cymes, on the 2–5 upper nodes, 2–5 × 3–8 cm, 11– 19(–40) flowers, peduncle 1–9 cm, branches puberulent to shortly pubescent, lowermost cymes subtended by foliaceous bracts; flower: pedicel 5–13(–15) mm, papillate to puberulent; calyx funnel-shaped to narrowly campanulate, 6–12(–15) mm long, tube 2–4 mm, obconical, glabrous to shortly pubescent, green, lobes ovate-triangular to triangular, acute, or, more rarely, acuminate, (3–)4–9(–11) × 1.5–4(–5) mm, gland-dotted, whitish, occasionally rose-tinged after anthesis, glabrous, or, more rarely, puberulent; corolla pleasantly scented, tube 12–22 mm, with sessile to shortly stipitate glands, lobes 7–13 mm long, with sessile glands on outer surface, white, with a red spot near throat; stamens exserted 20–40 mm, anther 2–3 mm. Fruit 10–13 × 11–12 mm, 4-lobed, black, shining, subtended by cup-shaped reddish calyx; seeds sunk in a yellow pulp.

Distribution in Central Africa:—D.R. Congo, Burundi.

Distribution elsewhere:— Benin, Cameroon, Central African Republic, Congo, Ethiopia, Gabon, Gambia, Ghana, Guinea, Guinea-Bissau, Gulf of Guinea Is., Ivory Coast, Kenya, Liberia, Nigeria, Senegal, Sierra Leone, Sudan, Tanzania, Togo, Uganda.

Habitat:—Forests, woodlands, fallow fields, cultivated fields, secondary forest, savannah, montane forest; 0– 2000 m.

Selection of representative specimens:— D.R. CONGO. Kasaï: Muetshi 70 Km WNW Lusambo, 1982, Casier 267 (BR!) ; Panzi , 16 June 1955, Devred 2021 (BR!) ; Bord du Sankuru , 3 July 1903, van den Bossche 34 (BR!). Bas-Katanga: Kipaka-Lutshi, June 1952, Germain 7721 (BR!) ; Territ. Mweneditu, Bakwanda (au N. De Katabaie), 12 May 1957, Liben 2937 (BR!) ; E of Gandajika, 30 July 1945, Luxen 508 (BR!) ; Kamina, Grelco , July 1936, Quarré 2626 (BR!, WAG). Forestier Central: Ponthierville [Ubundu], February 1921, Claessens 171 (BR!) ; Befale, 10 January 1958, Evrard 3186 (BR!) ; Urega , July 1932, Lebrun 5769 (BR!) ; Eala , 22 November 1946, J. Léonard 1047 (BR!) ; 20 km W of Yangambi, 15 August 1938, Louis 10809 (BR!) ; Lula , 5 February 1926, Robyns 1470 (BR!). Ubangui-Uele: Gangala na Bodio, s.d., Cornet d’Elzius et al. 710 (BR!) ; Niangara , 25 March 1952, Gérard 159 (BR!). Lac Albert: Djegu, Plaine Iswa, 1 October 1953, Taton 1306 (BR!). Lacs Edouard et Kivu: Route Kavumu-Walikale, km 110, 18 March 1957, Troupin 3148 (BR!). Haut-Katanga: 33 km WSW Kolwezi, 14 March 1983, Schaijes 1892 (BR!) ; Environs de Kolwezi , 17 August 2008, Malaisse & Kisimba 30 (BR!).—BURUNDI. Lacs Edouard et Kivu: Territ. Matana, route Matan-Rutovu, km 15, 26 February 1966, Lewalle 478 (BR!, MO) ; Bujumbura, Buhonga, 13 October 1967, Lewalle 2057 (BR!, MO) ; Mutumba , 28 August 1971, Reekmans 919 (BR!). Rwanda-Burundi: Cankuzo, 3 October 1974, Auquier 4384 (WAG) ; Bururi, Kitaba (= Kwitaba?), September 1932, Becquet 95 (BR!) ; Kwitaba , 19 October 1977, Reekmans 6537 (BR!, WAG) ; Kigwena , 5 October 1978, Reekmans 7135 (BR!) .

Notes:—1. The type specimen of C. umbellatum Poir. actually represents C. splendens G.Don. ( Meerts & Sosef 2022) . Therefore, following the rules of the ICN, Clerodendrum splendens would have to be regarded as a synonym of Clerodendrum umbellatum , while the species commonly known under that name would have to be called Clerodendrum scandens . This would be a highly undesirable situation. According to Art. 57.1, the only way to preserve a name that has been widely and persistently used for a taxon not including its type is by a proposal to conserve the name. Therefore, the name Clerodendrum umbellatum Poir. has been proposed as a nomen conservandum, with a conserved type with the aim of preserving nomenclatural stability (Art. 14.2 of the ICN) ( Meerts & Sosef 2022).

2. C. umbellatum is variable in leaf texture and pubescence; specimens from D.R. Congo often have leaves more pubescent than in other regions and the most pubescent forms have been described as “ C. scandens var. asperifolium B.Thomas », nom. inval., but variation in pubescence is continuous. Density of golden yellow glands on lower leaf surface is variable; the most glandular forms have been named Clerodendrum umbellatum var. congense (Engl.) Moldenke , but variation is more or less continuous. Plants from Burundi tend to have lamina more coriaceous and shining than plants from other regions; more work is needed to decide if they deserve taxonomic recognition.

3. The isotype of Clerodendrum subreniforme G ü rke in BR lacks complete leaves, but flower traits are diagnostic; the particular shape of the lamina mentioned in the protologue is probably an individual variation, not worthy of recognition.

4. In the savannahs of NE Congo (notably in the Garamba), C. umbellatum is represented by a suffrutescent, dwarf morphotype (e.g. De Saeger 1035), sometimes not exceeding 30 cm in height, but all intermediates with lianescent forms exist. C. fuscum var. lanceolatum Moldenke belongs here.

Second-step lectotypification of Clerodendrum cordifolium (Hochst.) A. Rich. The protologue of Volkamaria cordifolia Hochst. ( Hochstetter 1842) is based on materials collected by Schimper, but no collection number is mentioned. Cufodontis (1962) identified Schimper 1132 as the type (first-step lectotypification). This collection exists in several herbaria. Verdcourt (1992) mentioned Schimper 1132 (TUB) as the holotype, in error, which can be considered as inadvertent second-step lectotypification.