Small island but great diversity: thirty six species of Parmotrema (Parmeliaceae, lichenized Ascomycota), including sixteen new species, on Réunion (Mascarenes), with additional data from the Western Indian Ocean Author Masson, Didier 0000-0002-3340-5472 Vertevoye, 2860 route d’Irieu, F- 40390 Saint-Martin-de-Seignanx, France di. masson @ wanadoo. fr; https: // orcid. org / 0000 - 0002 - 3340 - 5472 di.masson@wanadoo.fr Author Magain, Nicolas 0000-0001-5409-9518 Evolution and Conservation Biology, InBios research centre, Sart Tilman B 22, Quartier Vallée 1, Chemin de la vallée 4, B- 4000 Liège, Belgium nicolas. magain @ uliege. be; https: // orcid. org / 0000 - 0001 - 5409 - 9518 nicolas.magain@uliege.be Author Sérusiaux, Emmanuël 0000-0002-3340-5472 Vertevoye, 2860 route d’Irieu, F- 40390 Saint-Martin-de-Seignanx, France di. masson @ wanadoo. fr; https: // orcid. org / 0000 - 0002 - 3340 - 5472 & Evolution and Conservation Biology, InBios research centre, Sart Tilman B 22, Quartier Vallée 1, Chemin de la vallée 4, B- 4000 Liège, Belgium e. serusiaux @ uliege. be; https: // orcid. org / 0000 - 0002 - 0456 - 0131 * Corresponding author & Vertevoye, 2860 route d’Irieu, F- 40390 Saint-Martin-de-Seignanx, France di. masson @ wanadoo. fr; https: // orcid. org / 0000 - 0002 - 3340 - 5472 di.masson@wanadoo.fr text Phytotaxa 2024 2024-06-27 657 1 1 138 http://dx.doi.org/10.11646/phytotaxa.657.1.1 journal article 10.11646/phytotaxa.657.1.1 1179-3163 13217474 Parmotrema dilatatum (Vainio) Hale (1974a: 335) MycoBank no. 343038 Parmelia dilatata Vainio (1890: 32) MycoBank no. 122344 Type :— BRAZIL . Minas Gerais : Sitio [now Antônio Carlos, Ahti 1998 ], 1000 m , s.d., E.A. Vainio s.n., distributed as Lichenes Brasilienses Exsiccati 397 ( TUR-V 2548 [image!], holotype ; BM, FH, M [image!], UPS, isotypes ; fide Hale 1965a ). ( Fig. 19 ) Thallus foliose, loosely to moderately adnate, subcoriaceous, up to 12 × 7 cm . Lobes imbricate, irregularly branched, 5–10 mm wide, rarely plane, frequently ascending when sorediate, occasionally convolute; margins undulated, often crenate or shortly laciniate, occasionally lobulate, with black rim; apices rounded; eciliate or very sparsely ciliate ( Fig. 19C ). Cilia black, rare, simple or rarely once bifurcate, up to 1.6 mm long. Upper surface pale yellowish grey near lobe tips to pale yellowish centrally, emaculate or faintly punctiform white-maculate, smooth near lobe tips but reticulated cracked in older parts, sorediate, lacking schizidia, pustules, dactyls, phyllidia and isidia; marginal lobules occasionally present, up to 1.5 mm wide. Soralia marginal at first, linear interrupted to somewhat labriform, then more or less subcapitate on short ascending laciniae, finally ± coalescing on arbuscular rising structures (up to 3.5 mm high) ( Fig. 19D ). Soredia granulose, (40)– 54.3 –(70) µm in diameter (n = 30, SD = 8.4 µm). Medulla white throughout, rarely with a yellowish tinge near the lower cortex. Lower surface smooth to verruculose, black and more or less mat to the margin or with a shiny, chestnut brown naked marginal zone (ca. 2–6 mm wide). Rhizines black, in scattered groups, simple to furcate, short (up to 0.5 mm long). Apothecia not present. Pycnidia rare, submarginal towards apices, black. Conidia not found (7 pycnidia investigated). Chemistry :— Spot tests and fluorescence : upper cortex K+ yellow, UV−; medulla K± yellowish, C− , KC + pink, P+ orange, UV−. Secondary metabolites ( TLC ): upper cortex with atranorin, chloroatranorin and usnic acid; medulla with protocetraric acid (major), echinocarpic acid (major), conechinocarpic acid, subechinocarpic acid and 3 undetermined substances. Mauritian collection with the same chemistry, except two undetermined substances instead of three. FIGURE 19. Parmotrema dilatatum . A : Distribution on Réunion (UTM 2×2 km grid system); B : Bioclimatic characteristics of collection site (abbreviations and threshold values for thermotype and ombrotype horizons from Rivas-Martínez et al . 2011: 17–18 ); C : Gross morphology of thallus ( Masson 974.4112 ); D : Eciliate sorediate lobes, with young linear interrupted to somewhat labriform soralia, and older, ± subcapitate soralia on short ascending laciniae, finally ± coalescing on arbuscular rising structures ( Masson 974.4112 ). Scale bars: A = 10 km; C = 10 mm; D = 4 mm. Geographical distribution :—Considered pantropical by several authors ( Louwhoff 2001 , Nash & Elix 2002 , Kurokawa 2006 ) but, given the divergent interpretations in the circumscription of the species (see Notes below), it is premature to have an accurate view of its distribution. Specimens with atranorin and usnic acid in the upper cortex and protocetraric and echinocarpic acids as main extrolites in the medulla are cited from: Kenya , Tanzania and Sierra Leone ( Krog & Swinscow 1981 ), NE Australia ( Elix 1994 ), Papua New Guinea ( Louwhoff & Elix 1999 ), Taiwan ( Kurokawa & Lai 2001 ), New Caledonia (Louwoff & Elix 2002), Mexico ( Egan et al . 2016 ), SE USA ( Brodo et al . 2001 ) and South America, where the species seems to be widely distributed (see map in Michlig & Ferraro 2012 ) and locally frequent (e.g. Donha 2005 , Benatti & Marcelli 2010 a , Cunha-Dias 2012). This species appears to be rare in the countries west and north of the Indian Ocean, as it has only been found at two locations, one in Kenya and one in Tanzania ( Krog & Swinscow 1981 ). In India , the status of Parmotrema dilatatum is still very confused. In their monograph on parmelioid lichens in India , Divakar & Upreti (2005) reported this species, but with chemistry that mentions only atranorin and protocetraric acid. However, on the corresponding chromatographic profile on the plate-50A (No10), a yellow spot, with a Rf similar to that of salazinic acid, appears above the protocetraric acid spot; it could well be echinocarpic acid. The more recent reviews of Awasthi (2007) and Mishra & Upreti (2017) do not mention P. dilatatum in India . The only data for Madagascar are from des Abbayes (1961) , who also reported the species from one locality in Réunion . Initially, des Abbayes identified his specimens as Parmelia robusta (= Parmotrema robustum ), but he followed Hale’s suggestion ( in litt .) that Parmotrema dilatatum and P. robustum might be conspecific. These two taxa were later recognized as distinct by Hale (1977) . We were able to examine the specimen collected by des Abbayes from Réunion ( REN 000044). Its morphological and chemical characteristics (only protocetraric and several undetermined fatty acids in the medulla) show that it is indeed P. robustum . Malagasy specimens still need to be checked. On Réunion this taxon is apparently rare, known from only one locality (present work, Fig. 19A ). There is also only one recent collection from Mauritius (Diederich & Ertz 2020). The other old collections cited in that paper, from Dodge (1959) , refer to Parmelia sieberi Dodge , a species synonymized with Parmotrema dilatatum by Hale (1965a) . The type collection of Parmelia sieberi clearly refers to a different species as it contains salazinic acid instead of protocetraric and echinocarpic acids ( Krog & Swinscow 1981 , Spielmann & Marcelli 2020 ); therefore, these old collections from Mauritius need to be re-examined. Ecology :—The single thallus collected on Réunion grew on the bark of a well-lit branch of Pandanus montanus , in a Pandanus submontane wet thicket ( 725 m elevation). The bioclimatic features of the locality are: bioclimate: pluvial tropical, thermotype belt: upper thermotropical (It = 495), ombrotype belt: ultrahyperhumid (Io = 26.1) ( Fig. 19B ). According to the limited ecological data available in the literature, Parmotrema dilatatum is a mainly corticolous species, occasionally lignicolous ( Donha 2005 ), which occurs at fairly low elevations (e.g. Krog & Swinscow 1981 , Louwhoff & Elix 1999 , Lai 2001 , Donha 2005 ). Notes :—More than 130 years after its description, the delimitation of this species is still uncertain. It seems clear that phenotypic characters (morphology, secondary metabolites) alone are insufficient to clarify what Benatti & Marcelli (2010a) have called the ‘ Parmotrema dilatatum complex’. The chemistry of the two specimens from the Mascarene Islands examined corresponds to that of the type collection [ isotype in BM examined by Krog & Swinscow (1981) ; isotype in M examined by Aubel (label dated 1985)]: atranorin and usnic acid in the upper cortex, echinocarpic and protocetraric acids as main medullary substances with 4–6 others undetermined. Krog & Swinscow (1981) also pointed out traces of pigments in the isotype in BM and considered the medulla of P. dilatatum as ‘inherently pigmented’, but with variable concentration of pigments. Elix (1994) reported secalonic acid A in the specimens from northeast Australia , but no pigment was detected in the specimens from the nearby Papua New Guinea ( Louwhoff & Elix 1999 ). Specimens with medullary pigments are considered by some authors ( Hale 1974 b , Kukwa et al . 2012, Michlig & Ferraro 2012 , Flakus et al . 2014 ) to belong to P. affluens (Hale) Hale. According to the protologue ( Vainio 1890 ) and the type collection, P. dilatatum is eciliate. This is also the case for the Mauritian sample, whereas that from Réunion has a few short and sparse cilia. Some authors report American specimens densely ciliated at the margin of the lobes ( Donha 2005 , Egan et al . 2016 ), which led them to consider that the development of cilia was variable in P. dilatatum . An alternative hypothesis is that their material is heterogeneous and includes more than one species. ITS sequences of two specimens identified as P. dilatatum are available on GenBank. They were generated and studied by Stelate et al . (2022) . In their phylogenetic tree, both are associated with a Portuguese sequence identified as P. robustum in a well-supported clade. Similarly, these two P. dilatatum were recovered within the P. robustum clade in our phylogenetic tree based on ITS sequences ( Fig. 4 ). We were able to examine these two collections from La Gomera (Canary Islands) preserved in MAF (see Selected specimens studied for comparison in the section on P. robustum ). Because of the lack of echinocarpic acid in their medullae, we believe that these two specimens fit better with P. robustum than with P. dilatatum . Despite three attempts, we were unable to study the DNA of the Reunionese material, but we had more success with the P. dilatatum specimen collected on Mauritius . In our 3-locus tree ( Fig. 3 ), this specimen was part of a strongly supported clade comprising P. robustum . The two species are strongly supported as different by bPP and Stacey analyses. Specimens examined :— FRANCE . Réunion : Saint-Benoît, Saint-François les Hauts, Sainte-Marguerite trail, elev. 725 m , 21°07’11”S , 55°40’39”E , Pandanus submontane wet thicket, on branch of Pandanus montanus , 28 August 2012 , D. Masson 974.4112 (Hb. DM). MAURITIUS . Plaine Wilhems District : Curepipe , Curepipe Botanic Gardens , elev. 565 m , 20°19’28”S , 57°30’49”E , on bark of tree, 30 July 2016 , P. Diederich 18308 ( Hb. P. Diederich ) .