Novelties in Psiadia Jacq. ex Willd. (Compositae) from Madagascar and Comoros

2025  47  22 adansonia Adansonia est une revue en flux continu publiée par les Publications scientifiques du Muséum, Paris Adansonia is a fast track journal published by the Museum Science Press, Paris Les Publications scientifiques du Muséum publient aussi / The Museum Science Press also publish: Geodiversitas, Zoosystema, Anthropozoologica, European Journal of Taxonomy, Naturae, Cryptogamie sous-sections Algologie, Bryologie, Mycologie, Comptes Rendus Palevol Diffusion – Publications scientifiques Muséum national d’Histoire naturelle CP 41 – 57 rue Cuvier F-75231 Paris cedex 05 (France) Tél. : 33 (0)1 40 79 48 05 / Fax : 33 (0)1 40 79 38 40 [email protected] / http://sciencepress.mnhn.fr © Publications scientifiques du Muséum national d’Histoire naturelle, Paris, 2025 ISSN (imprimé / print) : 1280-8571/ ISSN (électronique / electronic) : 1639-4798 Directeur De la publication / Publication director: Gilles Bloch Président du Muséum national d’Histoire naturelle réDacteur en chef / editor-in-chief : Thierry Deroin réDacteurs / editors : Porter P. Lowry II ; Zachary S. Rogers ; Mathieu Gardère assistant De réDaction / assistant editor : Emmanuel Côtez ([email protected]) Mise en page / Page layout : Emmanuel Côtez coMité scientifique / scientific board : F. Blasco (CNRS, Toulouse) M. W. Callmander (Conservatoire et Jardin botaniques de Genève) J. A. Doyle (University of California, Davis) P. K. Endress (Institute of Systematic Botany, Zürich) P. Feldmann (Cirad, Montpellier) L. Gautier (Conservatoire et Jardin botaniques de Genève) F. Ghahremaninejad (Kharazmi University, Téhéran) K. Iwatsuki (Museum of Nature and Human Activities, Hyogo) A. A. Khapugin (Tyumen State University, Russia) J.-Y. Lesouef (Conservatoire botanique de Brest) P. Morat (Muséum national d’Histoire naturelle, Paris) J. Munzinger (Institut de Recherche pour le Développement, Montpellier) S. E. Rakotoarisoa (Millenium Seed Bank, Royal Botanic Gardens Kew, Madagascar Conservation Centre, Antananarivo) P. H. Raven (Missouri Botanical Garden, St. Louis) G. Tohmé (Conseil national de la Recherche scientifique Liban, Beyrouth) J. G. West (Australian National Herbarium, Canberra) J. R. Wood (Oxford) couverture / cover : Réalisée à partir des Figures de l’article/Made from the Figures of the article. Adansonia est indexé dans / Adansonia is indexed in: – Science Citation Index Expanded (SciSearch®) – ISI Alerting Services® – Current Contents® / Agriculture, Biology, and Environmental Sciences® – Scopus® Adansonia est distribué en version électronique par / Adansonia is distributed electronically by: – BioOne® (http://www.bioone.org) 353 ADANSONIA, sér. 3 • 2025 • 47 (22) © Publications scientifiques du Muséum national d’Histoire naturelle, Paris. www.adansonia.com MOTS CLÉS Asteraceae, Astereae, Comores, Madagascar, synonyme nouveau, lectotypifications, espèce nouvelle. KEY WORDS Asteraceae, Astereae, Comoros, Madagascar, new synonym, lectotypifications, new species. Calvo J. & Callmander M. W. 2025. — Novelties in Psiadia Jacq. ex Willd. (Compositae) from Madagascar and Comoros. Adansonia, sér. 3, 47 (22): 353-362. https://doi.org/10.5252/adansonia2025v47a22. http://adansonia.com/47/22 ABSTRACT Madagascar and the other islands of the western Indian Ocean are the foremost center of species diversity of the genus Psiadia Jacq. ex Willd. (Compositae, Astereae). Revising the material housed in the historical herbaria P-JU and P-LA, we realized that the name P.madagascariensis (Lam.) DC., based on Lamarck’s name Conyza madagascariensis Lam., was misinterpreted and should be the priority name for the widely used Psiadia altissima (DC.) Benth.& Hook.f. ex B.D. Jacks. To avoid further misinterpretation, the types of the names Conyza glutinosa Lam. [accepted as Psiadia glutinosa (Lam.) Willd.] and Glyphia lucida Cass. [accepted as Psiadia lucida (Cass.) Drake] are discussed and clarified. In this frame, lectotypes are designated for six names. Moreover, the species P.lucida is recorded for the first time in the Comoros (Grande Comore). Finally, a new disciform species of Psiadia from Madagascar, i.e., Psiadia salicifolia J. Calvo& Callm., sp. nov., is described and illustrated. RÉSUMÉ Nouveautés dans le genre Psiadia Jacq. ex Willd. (Compositae) de Madagascar et des Comores. Madagascar et les autres îles de l’océan Indien occidental sont le principal centre de diversité d’espèces du genre Psiadia Jacq. ex Willd (Compositae, Astereae). En révisant le matériel conservé dans les herbiers historiques P-JU et P-LA, nous avons réalisé que le nom P.madagascariensis (Lam.) DC, basé sur le nom de Lamarck Conyza madagascariensis Lam., a été mal interprété et doit être le nom prioritaire pour Psiadia altissima (DC.) Benth.& Hook.f. ex B.D. Jacks, qui est largement utilisé. Pour éviter d’autres erreurs d’interprétation, les types des noms Conyza glutinosa Lam. [accepté comme Psiadia glutinosa (Lam.) Willd.] et Glyphia lucida Cass. [accepté comme Psiadia lucida (Cass.) Drake] sont discutés et clarifiés. Dans ce cadre, des lectotypes sont désignés pour six noms. De plus, l’espèce P.lucida est signalée pour la première fois aux Comores (Grande Comore). Enfin, une nouvelle espèce disciforme de Psiadia de Madagascar, Psiadia salicifolia J. Calvo& Callm., sp. nov., est décrite et illustrée. Joel CALVO Martin W. CALLMANDER Conservatoire et Jardin botaniques de Genève, ch. de l’Impératrice 1, C.P. 71, 1292 Chambésy (Switzerland) [email protected] [email protected] Submitted on 19 March 2025 | accepted on 14 May 2025 | published on 15 December 2025 Novelties in Psiadia Jacq. ex Willd. (Compositae) from Madagascar and Comoros 354 ADANSONIA, sér. 3 • 2025 • 47 (22) Calvo J.& Callmander M. W. INTRODUCTION The genus Psiadia Jacq. ex Willd. (Compositae, Astereae) comprises c.60 species distributed in Southern and East Africa, Madagascar, the Mascarene Archipelago, and Arabia, plus one species, P.ceylanica (Arn.) Grierson, in India and Sri Lanka (Humbert 1960; Scott 1991, 1993; Beentje 2002; POWO 2025). The genus is represented by 30 species in Madagascar (28 endemic; Humbert 1960; Nesom 2020) and at least 30 species on the surrounding islands (27 endemic to the Mascarene Archipelago; Strijk etal. 2012), which makes this region the foremost center of species diversity of Psiadia. The species of Psiadia are shrubs or erect suffrutescent plants (rarely scandent) characterized by having the leaves alternate, often viscid; synflorescences corymbiform; capitula radiate or disciform; involucres 3-4-seriate with imbricate bracts; receptacle naked or somewhat fimbrillate; ray florets (when present) pistillate with small limbs and disc florets functionally staminate, 5-lobed, corollas white or yellow; anther bases obtuse to shortly auriculate; style branches of the disc florets short, papillose; achenes subprismatic, 5-10 ribbed; and pappus composed of barbellate, uniseriate bristles. In Madagascar, Psiadia was comprehensively revised in the frame of the Flore de Madagascar et des Comores by Humbert (1960), who described most of the species included in the treatment. A narrow endemic from Mayotte was described afterwards (Labat& Beentje 2003). More recently, the species Conyza ageratoides DC. and Psiadiella humilis Humbert were transferred to Psiadia (Nesom 2020) based on the phylogenetic results previously presented by Strijk etal. (2012). Thanks to the study of the specimens housed at the historical herbaria P-JU and P-LA, we realized that the name Conyza madagascariensis Lam. has been misinterpreted so far and that it has priority over the widely used Psiadia altissima (DC.) Benth.& Hook.f. ex B.D. Jacks.; the required nomenclatural adjustments are made below. Likewise, the types of the names Conyza glutinosa Lam. and Glyphia lucida Cass. are also discussed in order to avoid further misinterpretation. Including the synonyms, six names are here lectotypified. Lastly, the new species Psiadia salicifolia J. Calvo& Callm. is described from central-eastern Madagascar. MATERIAL AND METHODS This study is based on the revision of the available literature and the examination of herbarium specimens kept at G and P. Additionally, digital herbarium specimens from B, BM, GH, JE, K, M, MO, MPU, TAN, and W were revised. The preliminary conservation status of the new species was assessed following IUCN Red List Categories and Criteria (IUCN 2012). We calculated the extent of occurrence and the area of occupancy (with a 2× 2 km grid) using the online “GeoCAT” software (http://geocat.kew.org; Bachman etal. 2011). NOMENCLATURE AND TAXONOMY Family Compositae Giseke Genus Psiadia Jacq. ex Willd. Species Plantarum, Ed. 4, 3: 2384 (Willdenow 1803). type speCies. — Psiadia glutinosa (Lam.) Willd. 1. Psiadia glutinosa (Lam.) Willd. (Fig. 1) Species Plantarum, ed. 4, 3: 2384 (Willdenow 1803). — Conyza glutinosa Lam., Encyclopédie méthodique. Botanique 2(1): 89 (Lamarck 1786). — Typus: Madagascar • “de l’Isle de France”; s.d.; Anon. s.n.; lectotype (designated here): P-LA[P00356056]!; isolectotypes: MPU[MPU014968] image!, P-LA[P00356055]!) (Fig. 1). Solidago viscosa Schrad., Sertum hannoveranum 1 (fasc. 1): 12, t.6 (Schrader 1795), nom. illeg., non Lamarck (1779). — Psiadia glutinosa Jacq., Plantarum rariorum horti caesarei schoenbrunnensis 2: tab. 152 (Jacquin 1797), nom. inval. — Typus: Iconography; lectotype (designated here): Schrader (1795: tab. VI, sub Solidago viscosa). Psiadia altissima var. latifolia Humbert, Mémoires de la Société linnéenne de Normandie 25: 50, 175, 284 (Humbert 1923).— Typus: Madagascar • Anosy Region [Prov. Toliara], Fort-Dauphin; 1897; G. Paroisse 3; lectotype (designated here): P[P02488135] image!; isolectotypes: P[P02488129] image!) • Atsimo-Andrefana [Prov. Toliara], Andrahomana; XI.1900; Ch. Alluaud 113; remaining syntype: P[P02488090] image! iConography. — Schrader (1795: tab. VI, sub Solidago viscosa); Jacquin (1797: tab. 152); Humbert (1960: 305, fig. LXI: 25-31). Distribution. — Endemic to southeastern Madagascar. notes Some authors attributed to Nikolaus Joseph von Jacquin (1727-1817) the authority of Psiadia glutinosa (e.g. Humbert 1960), but this is an error because Jacquin’s name was not validly published under ICN Art. 38.5 (Turland etal. 2018). Since he cited the name Solidago viscosa Schrad. [1795] as a synonym, an illegitimate name under Art. 53.1 but validly published, both names become homotypic, and therefore, the requirement (b) for accepting the descriptio generico-specifica provided by Jacquin as validly published is not fulfilled. Two specimens of Conyza glutinosa Lam. are kept at P-LA. One of them (P00356056), which is selected here to serve as lectotype, bears a label handwritten by André Thouin (1747-1824), the Chief Gardener of the Jardin du Roi in Paris at that time, with the following information: “C. glutinosa L. / nouvelle en 1770 / de l’Isle de France”. The second specimen at P-LA (P00356055) bears a short description in Lamarck’s handwriting but is clearly part of the same gathering and considered here as an isolectotype. Another duplicate of this collection is extant in Thouin’s herbarium now at MPU (MPU014968) (see Callmander etal. 2019). 355 Novelties in Psiadia Jacq. ex Willd. (Compositae) from Madagascar and Comoros ADANSONIA, sér. 3 • 2025 • 47 (22) fig. 1. — Lectotype of Conyza glutinosa Lam. (Psiadia glutinosa (Lam.) Willd.), P-LA[P00356056]. 356 ADANSONIA, sér. 3 • 2025 • 47 (22) Calvo J.& Callmander M. W. Psiadia glutinosa is unknown on l’Île de France, now Mauritius (Scott 1993). Indeed, it is considered as endemic to Madagascar (Humbert 1960), which suggests that the provenance of the seeds planted in the Jardin du Roi was mistakenly annotated. Confusion between Madagascar, Réunion, and Mauritius was not rare at that time. Those islands were often visited during the same voyage, leading to errors in the origin of the plants cultivated in France. Another case exemplifying this is the specimen of Psiadia glutinosa kept at B-W (B-W16745 -010), where the label on its blue cover indicates “Habitat in insula Burbon [Bourbon]”, now Réunion, where the species is unknown. This uncertainty probably also explains the fact that this species was omitted by Humbert in his first treatment of the Malagasy Compositae (Humbert 1923). Later, he treated P.glutinosa as an accepted species (Humbert 1960) and placed his P.altissima var. latifolia Humbert in synonymy. Two syntypes were cited in the protologue of Psiadia altissima var. latifolia: Alluaud 113 and Paroisse 3. The second consists of two specimens kept at P, P02488135 being the most complete and informative. This is therefore selected as lectotype. Psiadia glutinosa belongs to a taxonomically complex assembly of taxa centered around P.altissima (synonym of P.madagascariensis, see below) along with P.coarctata (Humbert) Humbert, P.serrata (Humbert) Humbert, and P.angustifolia (Humbert) Humbert. These taxa are currently treated as five distinct species (Humbert 1960; POWO 2025), but Humbert (1923) initially treated them as infraspecific taxa of P. altissima, except for P.glutinosa that was omitted as already explained above. 2. Psiadia lucida (Cass.) Drake (Fig. 2) Histoire physique, naturelle et politique de Madagascar 36: t. 488. (Drake del Castillo 1903). — Glyphia lucida Cass., Bulletin des sciences, par la Société philomatique de Paris 1818: 141 (Cassini 1818).— Glycyderas lucida (Cass.) Cass., Dictionnaire des sciences naturelles 59: 63, 74 (Cassini 1829).— Typus: Madagascar • s.l.; s.d.; P.Commersons.n.; holotype: P-JU[P00667271]!; isotypes: G[G00618076]!, P[P02475737]!; probable isotype: P[P02475738] image! (Fig. 2). Microglossa sessilifolia DC., Prodromus systematis naturalis regni vegetabilis 5: 320 (Candolle 1836). — Typus: Madagascar • s.l.; 1835; W.Bojer 6; lectotype (designated here): G-DC[G00455785]!; isolectotype: K[K000273809] image!) • s.l.; s.d.; J. Forbes s.n.; remaining syntypes: BM[BM000903879] image!, G-DC[G00455808]! Conyza flexilis DC., Prodromus systematis naturalis regni vegetabilis 5: 385 (Candolle 1836). — Typus: Madagascar • Alaotra-Mangoro Region [Prov. Toamasina], “province Bé tani-mena, forêts de Bé-fouroun”; 1835; W. Bojer 3; holotype: G-DC[G00453951]! Microglossa mikanioides Baker, Journal of the Linnean Society. Botany 20: 182 (Baker 1883). — Typus: Madagascar • Central Madagascar; s.l.; s.d.; R. Baron 1496; lectotype (designated here): K[K000273806] image! Psiadia tortuosa Klatt, Botanische Jahrbücher für Systematik, Pflanzengeschichte und Pflanzengeographie 12 (Beibl. 27): 23 (Klatt 1890).— Typus: Madagascar • Vakinankaratra Region [Prov. Antananarivo], “Nord-Betsiléo, Sirabé”; VIII.1880; J.M. Hildebrandt 3537; holotype: GH[GH00303964] image!; isotypes: BM[BM000903880] image!, G[G00018143, G00018144]!, JE[JE00019268, JE00019269] images!, M[M0147048] image!, P[P00435708, P00435709] images!, W[W1889-0074243] image! iConography. — Drake del Castillo (1903: pl. 488); Humbert (1960: 299, fig. LIX: 13-25). D istribution . — Madagascar (widespread except for the West) and Comoros (Grande Comore). The species is here recorded for the first time in Comoros based on the specimen Bojer s.n. (G01094005) and confirmed by recent field observations (Y. Ibrahim, pers. comm.). This represents the third species of the genus for the country. notes Humbert (1960) treated Microglossa sessilifolia as a heterotypic synonym of Psiadia lucida. The type of the former name, kept at G-DC, has small capitula, with external bracts shorter than inner ones. In contrast, the type of Psiadia lucida shows remarkably larger capitula with external involucral bracts almost as long as the inner ones. The French botanist considered the latter as an atypical form with large capitula and suggested that it might be the result of a teratology (see isotype P02475737). He also pointed out the existence of intermediate forms such as Scott Elliot 2621 (P02475730) and Baron 4362 (P02475713: “forme à gros capitules, faisant passage vers la forme à capitules encore plus gros et à bractées élargies appelée Glycideras lucida par Cass. [form with large capitula, leading to the form with even larger capitula and enlarged involucral bracts called Glycideras lucida by Cass.]”). Scott Elliot 2621 originating from Fort-Dauphin has involucres slightly smaller than those of the type of P.lucida but very similar in their overall shape, whereas the capitula of Baron 4362 from Central Madagascar perfectly match the typical form with smaller capitula. Humbert (1923) also indicated that both the typical form and that with large capitula might be found in the same individual as in Viguier& Humbert 973 (P02475681, P04308799). These specimens show the typical small capitula and one or two abnormal capitula that are actually galls, rather than a plant developing two types of capitula. One specimen that really possesses large capitula was identified by Humbert himself as “Psiadia madagascariensis forma macrocephala” (P02475635) but the name was never published. All the specimens with large capitula come from the vicinity of Fort-Dauphin in southeastern Madagascar, whereas the typical form is distributed through the eastern part of the Island from North to South, including a few localities nearby Fort-Dauphin. Since the two forms lack a differentiated geographic distribution and intermediate forms exist, we follow Humbert’s treatment and do not recognize them as distinct. Future phylogenetic data may shed more light on this taxonomic decision. 357 Novelties in Psiadia Jacq. ex Willd. (Compositae) from Madagascar and Comoros ADANSONIA, sér. 3 • 2025 • 47 (22) fig. 2. — Holotype of Glyphia lucida Cass. (Psiadia lucida (Cass.) Drake), P-JU[P00667271]. 358 ADANSONIA, sér. 3 • 2025 • 47 (22) Calvo J.& Callmander M. W. 3. Psiadia madagascariensis (Lam.) DC. (Fig. 3) Prodromus systematis naturalis regni vegetabilis 5: 320 (Candolle 1836).— Conyza madagascariensis Lam., Encyclopédie méthodique. Botanique 2: 85 (Lamarck 1786). — Baccharis madagascariensis (Lam.) Pers., Synopsis plantarum 2: 423 (Persoon 1807). — Typus: Madagascar• s.l.; s.d.; P. Commerson s.n.; holotype: P-JU[P06986451]! (Fig. 3). Microglossa altissima DC., Prodromus systematis naturalis regni vegetabilis 5: 320 (Candolle 1836). — Nidorella altissima (DC.) Benth.& Hook.f. ex Klatt, Annalen des Kaiserlich-Königlichen Naturhistorischen Hofmuseums 7: 296 (Klatt 1892). — Psiadia altissima (DC.) Benth.& Hook.f. ex B.D. Jacks., Index Kewensis 2: 640 (Jackson 1894), syn. nov. — Typus: Madagascar • “Province d’Emirne”; 1835; W. Bojer 5; holotype: G-DC[G00455755]! Psiadia stenophylla Baker, Journal of the Linnean Society, Botany 22: 490 (Baker 1887).— Typus: Madagascar • Central Madagascar; s.l.; s.d.; R. Baron 3549; lectotype (designated here): K[K000273973] image!; isolectotypes: BM[BM000903966] image!, P[P00435635, P00435636] images! Psiadia decurrens Klatt, Botanische Jahrbücher für Systematik, Pflanzengeschichte und Pflanzengeographie 12 (Beibl. 27): 23 (Klatt 1890).— Typus: Madagascar • Vakinankaratra Region [Prov. Antananarivo], “Nord-Betsiléo, Sirabé”; VIII.1880; J.M. Hildebrandt 3541; holotype: GH[GH02290259] image!; isotypes: BM[BM000903967] image!, G[G00002313, G00393789]!, JE[JE00006214] image!, P[P00435616, P00435617] images! iConography. — Drake del Castillo (1903: pl. 483 sub. Psiadia altissima; pl. 484 sub. P.decurrens); Humbert (1960: 305, fig. LXI: 1-24, sub. P.altissima). Distribution. — Madagascar (widespread) and îles Éparses (Europa and Juan de Nova). notes The type of Conyza madagascariensis, kept at P-JU, shows a flowering branch of an erect shrub that perfectly matches the concept hitherto known under the name P.altissima. The specimen bears a modern determination label by unknown hand with the identification “Psiadia altissima”. The name Psiadia madagascariensis has priority over the widely used P.altissima (Humbert 1960; Strijk etal. 2012; POWO 2025). Humbert (1960) misinterpreted Candolle’s name P.madagascariensis as being heterotypic to Lamarck’s Conyza madagascariensis, which is its basionym. In turn, it seems that Candolle misinterpreted the concept of P.madagascariensis since the single specimen in G-DC labelled as such (G00455808) corresponds to P.sessilifolia (see below) and the taxonomic entity named Conyza madagascariensis by Lamarck was described by himself as Microglossa altissima DC. The intended combinations by Bentham& Hooker f. (1873) of Microglossa altissima in both Psiadia and Nidorella are to be treated as not validly published (see ICN Art. 35.2 Ex. 6.). 4. Psiadia salicifolia J. Calvo& Callm., sp. nov. (Figs 4; 5) Diagnosis. — Among the disciform species of the genus, Psiadia salicifolia is unique for its linear leaves whitish-lanate on the abaxial surface. t ypus . — Madagascar • Alaotra-Mangoro Region [Prov. Toamasina], Ambatondrazaka, Réserve de Zahamena, lit de la riv. Manambato; 17°30’49”S, 48°45’13”E; 950 m alt.; 23.VII.1997; S.Wohlhauser, M.O.Laivao& M. W. Callmander s.n.; holotype: G[G00618075]!; isotypes: MO!, P! phenology. — Collected in flower in June and July. Distribution anD eCology. — Endemic to Madagascar, known from the central-eastern medium altitude moist evergreen forests (sensu Gautier etal. 2018). This species grows on rocky riverbeds. e tymology . — The epithet salicifolia has been chosen because of the resemblance of the leaves of the new species to some species of the genus Salix L., e.g., S.eleagnos Scop. Conservation status. — Psiadia salicifolia sp. nov., is known from two locations encompassed within the Corridor Ankeniheny-Zahamena and Zahamena protected areas. The new species has an extent of occurrence (EOO) and an area of occupancy (AOO) of 8km². Deforestation due to fires for slash-and-burn agriculture is considerable in the extreme south of the Corridor Ankeniheny Zahamena where one of the two known locations occurs (Goodman etal. 2018). Due to these plausible threats, P. salicifolia, sp. nov., is assigned to a preliminary risk of extinction status of “Endangered” [EN B1ab(ii, iii, iv, v) + 2ab(ii, iii, iv, v)] using the IUCN Red List Categories and Criteria (IUCN 2012). aDDitional speCimens examineD. — Madagascar • Atsinanana Region [Prov. Toamasina], Vatomandry, Ambalabe, Sahanionaka, forêt de Vohibe, au bord de la riv. Talaviana, de Tsitondroina vers l'ouest; 19°10’44”S, 48°32’29”E; 728 m alt.; 1.VI.2010; A. Razanatsima 848; MO[MO-3028060] image!, P[P00982567]!, TAN image! DesCription Subshrubs 0.3 m high, stems leafy on upper part, lower parts naked with conspicuous leaf-scars. Leaves alternate, simple, sessile, densely arranged; laminas 35-55× 2-3.6 mm, linear to slightly broadened in distal third, base attenuate, apex rather acute, margins entire except for distal third somewhat crenulate, glabrescent on adaxial surface, whitish-lanate on abaxial surface, venation pinnate with secondary veins conspicuous, not covered by indumentum. Synflorescences terminal, corymbiform, composed of usually 10-15 capitula; bracts linear to subulate, decreasing in size from base to capitula, basal ones similar to upper leaves but smaller. Capitula on pedicels 7-32 mm long, with remnants of arachnoid trichomes; receptacle flat to slightly convex, epaleate, fimbrillate. Involucres 4-5× 5-7 mm (in dried specimens), hemispherical, glabrous, straw-colored; involucral bracts 31-33, 3-4-seriate, imbricate, subequal, outer ones linear-subulate, 2.9-3.1× 0.5-0.6 mm, acute, entire, middle ones linear-oblong, 3.8-4.1× 0.9-1 mm, acuminate, scarious marginally, inner ones as middle ones but narrower. Peripheral florets c.23, 1-seriate, pistillate, tubulose; corolla 2.4-2.8 mm long, with trichomes in middle part, irregularly 3-5-lobed (lobes somewhat atrophied), whitish to pale yellow; style c.3.6 mm long, branches c.1 mm long, not broadened apically, rather smooth, pale yellow. Disc florets c.19, functionally staminate, tubulose; corolla 2.9-3.2 mm long, with trichomes in middle part, 359 Novelties in Psiadia Jacq. ex Willd. (Compositae) from Madagascar and Comoros ADANSONIA, sér. 3 • 2025 • 47 (22) fig. 3. — Holotype of Conyza madagascariensis Lam. (Psiadia madagascariensis (Lam.) DC.), P-JU[P06986451].