Revision of the fern family Marattiaceae in the Seychelles with two new species and a discussion of the African Ptisana fraxinea complex.

Bruno Senterre; GERMINAL ROUHAN; ISABELLE FABRE; CHARLES MOREL; MAARTEN J.M. CHRISTENHUSZ

Background and Aims Throughout the history of fern classification, familial and generic concepts have been highly labile. Many classifications and evolutionary schemes have been proposed during the last two centuries, reflecting different interpretations of the available evidence. Knowledge of fern structure and life histories has increased through time, providing more evidence on which to base ideas of possible relationships, and classification has changed accordingly. This paper reviews previous classifications of ferns and presents ideas on how to achieve a more stable consensus. † Scope An historical overview is provided from the first to the most recent fern classifications, from which conclusions are drawn on past changes and future trends. The problematic concept of family in ferns is discussed, with a particular focus on how this has changed over time. The history of molecular studies and the most recent findings are also presented. † Key Results Fern classification generally shows a trend from highly artificial, based on an interpretation of a few extrinsic characters, via natural classifications derived from a multitude of intrinsic characters, towards more evolutionary circumscriptions of groups that do not in general align well with the distribution of these previously used characters. It also shows a progression from a few broad family concepts to systems that recognized many more narrowly and highly controversially circumscribed families; currently, the number of families recognized is stabilizing somewhere between these extremes. Placement of many genera was uncertain until the arrival of molecular phylogenetics, which has rapidly been improving our understanding of fern relationships. As a collective category, the so-called 'fern allies' (e.g. Lycopodiales, Psilotaceae, Equisetaceae) were unsurprisingly found to be polyphyletic, and the term should be abandoned. Lycopodiaceae, Selaginellaceae and Isoëtaceae form a clade (the lycopods) that is sister to all other vascular plants, whereas the whisk ferns (Psilotaceae), often included in the lycopods or believed to be associated with the first vascular plants, are sister to Ophioglossaceae and thus belong to the fern clade. The horsetails (Equisetaceae) are also members of the fern clade (sometimes inappropriately called 'monilophytes'), but, within that clade, their placement is still uncertain. Leptosporangiate ferns are better understood, although deep relationships within this group are still unresolved. Earlier, almost all leptosporangiate ferns were placed in a single family (Polypodiaceae or Dennstaedtiaceae), but these families have been redefined to narrower more natural entities. † Conclusions Concluding this paper, a classification is presented based on our current understanding of relationships of fern and lycopod clades. Major changes in our understanding of these families are highlighted, illustrating issues of classification in relation to convergent evolution and false homologies. Problems with the current classification and groups that still need study are pointed out. A summary phylogenetic tree is also presented. A new classification in which Aspleniaceae, Cyatheaceae, Polypodiaceae and Schizaeaceae are expanded in comparison with the most recent classifications is presented, which is a modification of those proposed by Smith et al. (2006, 2008) and Christenhusz et al. (2011). These classifications are now finding a wider acceptance and use, and even though a few amendments are made based on recently published results from molecular analyses, we have aimed for a stable family and generic classification of ferns.

A new fossil eusporangiate fern Marattiopsis vodrazkae J. Kvaček, sp. nov. has been recovered from the Campanian Hidden Lake Formation of James Ross Island, Antarctica. Its fertile and sterile pinnules are described and compared to the other species of the fossil genus Marattiopsis Schimper and the living genera Marattia, Ptisana and Eupodium (Marattiaceae). In contrast to the other species of the genus Marattiopsis, M. vodrazkae is characterised by stalked synangia, a smaller number of sporangia per synangium, generally small sized pinnules, and the absence of venuli recurrentes. It shows a mosaic of characters present in the living Marattiaceae: it shares stalked sporangia with the genus Eupodium and some species of Marattia and exhibits a suture (an abscission scar at pinnule bases), a character typical for the genus Ptisana. Additionally, M. vodrazkae provides important palaeoenvironmental signals for climate reconstructions, arguing for warm (paratropical to warm-temperate) and humid climatic conditions on the Antarctic Peninsula and adjacent volcanic islands during the Campanian.

Phytotaxa 158 (1): 057–075 www.mapress.com / phytotaxa / Copyright © 2014 Magnolia Press ISSN 1179-3155 (print edition) Article PHYTOTAXA ISSN 1179-3163 (online edition) http://dx.doi.org/10.11646/phytotaxa.158.1.4 Revision of the fern family Marattiaceae in the Seychelles with two new species and a discussion of the African Ptisana fraxinea complex. BRUNO SENTERRE1, 2, GERMINAL ROUHAN3, ISABELLE FABRE2, CHARLES MOREL2, 4 & MAARTEN J.M. CHRISTENHUSZ5 1 Evolutionary Biology & Ecology - CP 160/12, Université Libre de Bruxelles, 50 Av. F. Roosevelt, 1050 Bruxelles, Belgium. E-mail: bsenterre@gmail.com 2 Plant Conservation Action group (PCA), P.O. Box 392, Victoria, Mahé, Seychelles. 3 Muséum national d’Histoire Naturelle, UMR CNRS 7205 ‘Origine, Structure et Evolution de la Biodiversité’, Herbier national, 16 rue Buffon, CP 39, F-75231 Paris cedex, France, E-mail: rouhan@mnhn.fr 4 Natural History Museum, P.O. Box 720, Victoria, Mahé, Seychelles. 5 Jodrell Laboratory, Royal Botanic Gardens, Kew, Richmond, Surrey TW9 3DS, U.K, E-mail: m.christenhusz@kew.org Abstract The fern flora of the Seychelles archipelago remains relatively poorly known. Current projects involving the development of the Seychelles National Herbarium and study of key biodiversity areas have resulted in extensive new explorations on the granitic islands. Historical literature, newly collected specimens and older collections have been revised for the fern family Marattiaceae as a prelude for a fern checklist, resulting in the rediscovery of Ptisana a century after it was last reported (as Marattia). Two species of Angiopteris are recognised in the Seychelles, one of them new to science. In total, three species are recorded from the granitic Seychelles islands, Mahé and Silhouette, and none from the coral islands, which are too dry to harbour Marattiaceae. These three species are described in detail. The newly described Angiopteris chongsengiana is characterized by the elongated terminal pinnules and progressively reduced distal pairs of pinnules. Based on phylogenetic molecular analyses the Ptisana fraxinea complex is discussed and several new combinations are proposed. The Seychellois Ptisana is most closely related to Ascension Island endemic P. purpurascens and is described as a new species, Ptisana laboudalloniana. It differs from P. fraxinea in its winged secondary rachises and from P. salicifolia in its smaller habit, with the laminae shorter than broad and shorter than the petioles. Preliminary revisions of other fern families indicate that more discoveries are likely to be made in the fern flora of the Seychelles. Key words: Angiopteris, Baton monsenyer, island flora, long distance dispersal, Marattia, Mauritia, montane rain forest, phytogeography, pteridophytes, trnSGG, Western Indian Ocean Introduction The granitic Seychelles form a tropical archipelago of continental fragments that belong to a Precambrian microcontinent called Mauritia, which “was separated from Madagascar and fragmented into a ribbon-like configuration by a series of mid-ocean ridge jumps during the opening of the Mascarene ocean basin between 83.5 and 61 million years ago” (Torsvik et al. 2013). This could have contributed to some phytogeographical links observed between the Seychelles, Madagascar and Sri Lanka (e.g. Nesolindsaea Lehtonen & Christenh. in Lehtonen et al. 2010). Only few synthetic studies exist focusing on the phytogeographical affinities of the Seychelles flora, and although extremely valuable, they are already outdated (Summerhayes 1931, Rakotondrainibe et al. 1996). Summerhayes (1931) observed that the indigenous flora of the Seychelles has been influenced by both Asia and Africa, and additionally has a strong Western Indian Ocean element (Mascarenes and Madagascar). His study indicated that the African influence tends to decrease when considering affinities of the Seychelles endemic Accepted by Marcus Lehnert: 2 Dec. 2013; published: 30 Jan. 2014 57 elements (using closest relatives known at that time), and reaches even lower values when considering the affinities at the genus level. Rakotondrainibe et al. (1996) also observed a relatively small African influence in the Seychelles fern flora compared to the percentage of African elements present in the other islands of the region. The phytogeographical link between the Seychelles, Madagascar and the Indo-Malayan region stems from the plate tectonical history and therefore floristic affinities may be observed at higher taxonomic levels. This phytogeographical pattern is also observed in the fern family Marattiaceae, which at generic level appears to reflect the geological history rather than long distance dispersal events (Christenhusz & Chase 2013). Contrary to this, speciation within the genus Ptisana appears to be influenced by rare long distance dispersal events, as suggested by our observations. Marattiaceae form an ancient fern lineage with fossils dating back to the Lower Carboniferous (Christenhusz 2007, Hill & Camus 1986). The family has since long been isolated from other extant fern lineages (a split estimated to have occurred ca. 325 million years ago) and is characterized by the fusion (partial or complete) of the eusporangia into a so -called synangium (Murdock 2008b). Additionally, Marattiaceae are unique in the presence of vascularized, starchy auricles at the base of petioles (often called ‘stipules’ or ‘trophopods’; Lellinger 2002), and by the presence of pulvini (swollen nodes) at the base of petioles, petiolules and leaflets. As with most fern families, the taxonomic treatment at the generic level has been relatively unstable, until a recent revision based on molecular phylogeny resolved the family into two major lineages (Murdock 2008a, b), representing six genera and over 100 species. One lineage is represented by the exclusively neotropical Danaea Smith (1793: 420) (ca 50 spp., Christenhusz 2010a), the other by a clade comprising the five genera Angiopteris Hoffmann (1796: 29) (with ca 30–40 spp., mostly in Southeast Asia; He & Christenhusz 2013), Christensenia Maxon (1905: 239) (1 or 2 paleotropical species; Rolleri 1993, Murdock 2008b), Eupodium Smith (1841: 190) (3 spp.; Christenhusz 2010b), Ptisana Murdock (2008b:744) (ca. 20 paleotropical species, especially in New Guinea; Murdock 2008b), and Marattia Swartz (1788: 128) (8 species in the Neotropics and Hawaii; Lavalle 2003). In the Western Indian Ocean islands, Angiopteris and Ptisana are the only two marattioid genera recorded. Angiopteris reaches its western limit of distribution with one species, A. madagascariensis de Vriese in Vriese & Harting (1883: 23), recorded from Madagascar, La Réunion, Mauritius (Badré 2008), the Comoros and the Seychelles, where it was initially addressed as A. evecta (Forster 1786: 81) Hoffmann (1796: 12) by Baker (1877) and where it is still frequently encountered. A second species of Angiopteris is reported in this paper as new to science. Ptisana is present in the Madagascan region with several variable taxa recognized in the species complex of Ptisana fraxinea (Smith 1790: t. 48) Murdock (2008b: 746; Roux 2009, Tardieu-Blot 1951). In the Seychelles, Christensen (1912) recorded P. fraxinea (as Marattia fraxinea Sm.), but it had not been found since and it was therefore believed to be an erroneous record (Tardieu-Blot 1960). As a prelude for a fern checklist of the Seychelles, we here revise the family Marattiaceae for the area, recognizing three species, two of which being proposed as new. All three species are here morphologically described. Their comparative placement in the phylogeny is assessed through phylogenetic analyses using molecular data, which, for the first time, includes samples from the Western Indian Ocean. Preliminary revisions of other fern families indicate that more discoveries are likely to be made in the fern flora of the Seychelles. Material and methods Morphology and Taxonomy:—Based on historical studies on the fern flora of the Seychelles (Awmack 1997, Baker 1877, Christensen 1912, Tardieu-Blot 1960), only eight specimens of Marattiaceae were originally cited from this archipelago. More recently, extensive field explorations resulted in many new collections, with the aim to develop the Seychelles National Herbarium in Victoria (SEY) and to study key biodiversity areas on the main granitic islands (Senterre et al. 2013). As a result of these collecting efforts, 27 specimens of Marattiaceae from the Seychelles are now available for study at SEY, accompanied by ca. 70 observation records, all from the islands of Mahé and Silhouette. We examined additional herbarium specimens from H, K, L, MAU, MO and P (herbarium acronyms follow Thiers 2010), representing taxa recorded for all major islands and archipelagos of the Western Indian Ocean. The descriptions proposed here are partly based on living plants and therefore also include some important characters that are usually lost during the process of herbarium preservation, e.g. aerating areas (see Holttum 1978), habit, colour and texture of leaves. It also includes characters of the rhizomes and of the auricles at 58 • Phytotaxa 158 (1) © 2014 Magnolia Press SENTERRE ET AL. the base of petioles, which are rarely collected and preserved in herbarium specimens, but have useful diagnostic characters. Measurements were taken from living plants except for microscopic characters (e.g. scales, synangia, venation, venuloids) or unless marked otherwise. In addition, measured ranges were checked using available dried specimens. We present field photographs of the new species (Figs. 2–4), and additional illustrations of Seychelles species are available online at seychellesplantgallery.com. Taxon sampling:—Our molecular phylogenetic analysis includes 79 specimens of Marattiaceae representing 37 species from all regions of the distribution area of the family. This dataset was primarily composed from GenBank sequences (http://www.ncbi.nlm.nih.gov/genbank/), available from the phylogenetic study of the family by Murdock (2008a). To establish the closest extant relatives of the three taxa found in the Seychelles, we added two specimens identified as Ptisana fraxinea var. fraxinea from Madagascar and three specimens of Angiopteris madagascariensis from Madagascar and the Comoros to the dataset of Murdock (2008a). The neotropical genus Danaea was selected as outgroup, because in all earlier analyses this genus has been shown to form the sister lineage to the remainder of the family (Christenhusz et al. 2008, Murdock 2008a). Voucher information and GenBank accession numbers for the nine newly obtained sequences are listed in Table 1. The updated taxonomy follows Murdock (2008b) and Christenhusz et al. (2011). TABLE 1. Voucher information for the nine specimens newly sequenced in this study for the plastid DNA region trnSGCU to trnGUUC (trnSGG); all other taxa used were taken from GenBank with accessions numbers following the taxa names on the tree (Fig. 1). Species Locality, Collector, collection number (herbarium) GenBank accession number trnSGG Ptisana laboudalloniana Seychelles B. Senterre & I. Fabre 6161 (P, SEY) KF649836 Ptisana fraxinea Madagascar G. Rouhan et al. 1206 (P, TAN) KF649839 Ptisana fraxinea Madagascar G. Rouhan et al. 338 (P, UC) KF649840 Angiopteris chongsengiana Seychelles B. Senterre & I. Fabre 6151 (P, SEY) KF649834 Angiopteris chongsengiana Seychelles B. Senterre & I. Fabre 6152 (P, SEY) KF649835 Angiopteris madagascariensis Seychelles B. Senterre & N. Labiche-Barreau 6123 (P, SEY) KF649833 Angiopteris madagascariensis Madagascar G. Rouhan et al. 368 (BM, NBG, P, UC) KF649838 Angiopteris madagascariensis Madagascar F. Barthelat et al. 576 (MAO, MO, P) KF649837 Angiopteris madagascariensis Comoros M. Pignal & A.B. Sifari 1860 (BR, K, MO, P) KF649841 DNA isolation, amplification and sequencing:—Leaf tissue was preserved in the field in silica-gel as the optimal routine collection of DNA-friendly material in these field conditions (Chase & Hills 1991, Gaudeul & Rouhan 2013). Total genomic DNA was isolated from these samples using the Qiagen DNeasy Plant Mini Kit (Valencia, CA, U.S.A.) following the manufacturer’s protocol. Plastid DNA region trnSGCU to trnGUUC including the trnG intron was amplified (trnSGG) (~1700 bp), using primers 3’trnGUUC (Shaw et al. 2005) and trnS GCU-F1 (Murdock 2008a) using the polymerase chain reaction (PCR) program: 5 min 94°C / 35× (45s 94°C / 4 min 66°C) / 10 min 66°C. PCR products were sequenced in both directions (forward and reverse) using the amplification primers and additional internal primers 5'-trnG2G and 5'-trnG2S (Shaw et al. 2005). Alignment and phylogenetic analyses:—All sequences were edited and assembled using Sequencher v. 4.8 (Gene Codes Corporation, Ann Arbor, Michigan). Consensus sequences were automatically aligned using Muscle v. 3.6 with the default parameters (Edgar 2004) and the resulting alignments were manually checked and revised using the MUST package (Philippe 1993), in order to maximize the similarity between sequences. Gaps were treated as missing data. The obtained data matrix was analysed using equally weighted maximum parsimony (MP) and Bayesian inference (BI). Maximum parsimony analyses were conducted using PAUP* v. 4.0b10 (Swofford 2002), with heuristic searches performed with 10,000 random-sequence addition replicates, tree bisection-reconnection (TBR), branch MARATTIACEAE OF THE SEYCHELLES Phytotaxa 158 (1) © 2014 Magnolia Press • 59 swapping and with the Multrees option on. The robustness of each node was assessed by a non-parametric bootstrap analysis (BS; Felsenstein 1985), with 10,000 replicates of similar heuristic searches (but each replicate with 1 random-sequence addition replicates). Bayesian analyses (BI) were performed using MrBayes v. 3.2.1 (Huelsenbeck & Ronquist 2001, Ronquist & Huelsenbeck 2003). The trnSGG alignment was assigned the GTR+Γ model of nucleotide substitution, suggested as the best fit to the data under the Akaike information criterion, as implemented in MrModelTest v. 2.3 (Nylander 2004). Two independent but parallel analyses were carried out using flat priors, starting from random trees and consisting of four chains each. The analyses were run for 3 million generations, sampling every 100 generations. Following completion of the analyses, we explored the temporal evolution of the output parameter estimates in order to recognize the point of convergence to the stationary distribution, using Tracer v.1.5 (Rambaut & Drummond 2007) and the first 500,000 generations (5,000 trees) were conservatively excluded as ‘burn-in’. We pooled the post burn-in trees (25,000 total trees) and computed a majority consensus with posterior probability estimates (PP) for all nodes. Results Both BI and MP analyses resulted in similar phylogenetic tree topologies (2,508,556 MP trees of 685 steps long, CI = 0.79 and RI = 0.96). Although based on trnSGG only, the overall topology (Fig. 1) agrees with the phylogeny presented by Murdock (2008a). The monophyly of all genera of the family is retrieved, with Marattia redefined as done by Murdock (2008b). Ptisana and Angiopteris (the only two genera occurring in the region) are each strongly supported as monophyletic (BS=100; PP=1). The newly included taxa from the Western Indian Ocean are placed in their respective genera. Within Ptisana, P. fraxinea var. salicifolia (Schrader 1818: 920) Murdock (2008b: 746) from South Africa forms the sister lineage to all other species of the genus. The second branching lineage in Ptisana has a weak signal supporting a dichotomy (PP=0.53; node not supported in MP) between a clade including the species from Asia, Australasia and the Pacific Islands and a clade formed by specimens from the Seychelles, Madagascar and Ascension Island (BS=97; PP=1). Within this clade of only four specimens, Ptisana from the Seychelles is strongly supported as the sister lineage to P. purpurascens (de Vriese in Vriese & Harting (1853: 7)) Murdock (2008b: 747) from Ascension Island (Atlantic Ocean), and these two species are separated with good support from typical P. fraxinea var. fraxinea from Madagascar. Ptisana fraxinea sensu lato does not appear to be monophyletic, although no material from Mauritius or tropical Africa is included in our analyses to allow unravelling the complex further. As shown by Murdock (2008a), the relationships within Angiopteris are not well-resolved, but there is strong support for the inclusion of Archangiopteris Christ & Giesenhagen (1899: 77). Angiopteris evecta, a poorly defined taxon (see Christenhusz & Toivinen 2008), is unsurprisingly found to be polyphyletic. The six specimens from the Western Indian Ocean (Madagascar, the Comoros and the Seychelles) are nested among five taxa mostly from South East Asia (PP=0.56; node not supported in MP), most notably A. fokiensis Hieronymus (1919: 275), which is diagnosed by its conspicuously verrucose petioles. This clade is included in a wider polytomy of further species from Asia and Australasia. Standard sequence data show very little variability among Angiopteris species, and this might be explained by the close relationships between most species of Angiopteris perhaps resulting from a recent evolutionary radiation. More studies are needed to resolve the taxonomic intricacies of Angiopteris. Taxonomic Treatment Even though over 200 names of Angiopteris are available (e.g. Christenhusz 2007, He & Christenhusz 2013), most of them have been based on fragmentary herbarium material and the actual number of species in the genus is much lower, probably less than 50. Only one of the species in the Seychelles could be morphologically matched with a known species, Angiopteris madagascariensis. In spite of the numerous names available, the second species could not be matched with any known taxon. Therefore we hereby describe the second species of Angiopteris as new to science. Further molecular studies on the genus may allow us to tease out this species complex. 60 • Phytotaxa 158 (1) © 2014 Magnolia Press SENTERRE ET AL. FIGURE 1. Phylogenetic relationships within Marattiaceae resulting from the BI (50% majority-rule consensus tree) and MP analyses (strict consensus tree) based on the plastid region trnSGG. Values above and below branches indicate the support values for the BI and MP analyses (PP/BS, respectively), with thick branches for the most robustly supported (PP>0.95 and BS>80%). Dotted lines are not retrieved in MP tree. Names of the six marattioids genera are reported on the right. Bold taxa are from the Western Indian Ocean area and were sequenced for the purpose of this study. MARATTIACEAE OF THE SEYCHELLES Phytotaxa 158 (1) © 2014 Magnolia Press • 61 The taxonomy of Ptisana in the African region is even more contentious. Based on the position of synangia and width of pinnules, Tardieu-Blot (1951) distinguished four species of Ptisana in the Indian Ocean area (as Marattia boivinii Mett. ex Ettingshausen (1864: 246), M. microcarpa Mett. ex Ettingshausen (1864: 246), M. fraxinea and M. salicifolia Schrader (1818: 920)), and a few additional names have been applied to specimens from mainland Africa and neighbouring islands. Most other authors (e.g. Roux 2009) opted for merging most Ptisana taxa from Africa and the Mascarenes into the single species P. fraxinea, except P. purpurascens (endemic to Ascension Island) and P. boivinii (endemic to Madagascar). Murdock (2008b) conserved four taxa, three of them as varieties of P. fraxinea plus Ptisana purpurascens. The latter is undoubtedly of African origin, since there are no Ptisana species in the Neotropics. Our study reveals that it is most closely related to the Seychelles plants, together forming a clade distinct from the African and Madagascan clades. As detailed below, morphological evidence supports the description of the Seychelles Ptisana as a new species. Key to Marattiaceae on the Seychelles 1. -. 2. -. False veins (venuloids) absent, pinnae mostly (sub-)opposite, sporangia completely fused into a two-lipped synangium ... ...........................................................................................................................................................Ptisana laboudalloniana False veins present between true veins, pinnae mostly alternate, sporangia not completely fused into a synangium, fused only at base ............................................................................................................................................................................ 2 Secondary rachises distinctly winged; terminal pinnules up to 25 cm, much longer than ultimate pairs of distal pinnules (6–10 cm); petioles with conspicuous wart-like bases of old scales ........................................... Angiopteris chongsengiana Secondary rachises not winged; terminal pinnules similar in size to other pinnules (ca 20 cm); petioles with wart-like bases of old scales, but these small and inconspicuous .................................................................... Angiopteris madagascariensis Angiopteris chongsengiana Senterre & I.Fabre, sp.nov. Fig. 2, Fig. 3 TYPE: —SEYCHELLES. Mahé: Mont Cotton, vallée à l'Est du Mont Cotton, au pied de Pérard, dans les hauts de la rivière Grand St. Louis, 700 m, 17 September 2011, B. Senterre & I. Fabre 6151 (holotype P!, 2-sheets P-02432630, P-02432631, isotype SEY!). Diagnosis:—This species differs from other Angiopteris with verrucose petioles by its lateral pinnae with elongate terminal pinnules and reduced distal pinnule pairs. Its venuloids are submarginal. Petioles show pinkish flesh when cut and secondary rachises are broadly winged distally. Perennial ferns, 2–3 m tall. Rhizomes 30–65 cm tall, 30–45 cm in diameter, starchy, upright, globular to trunk-like, covered with remaining bases of old leaves, unbranched. Roots branched, stout, 7–8 mm in diameter (2 mm for secondary roots). Stipules 8–10 cm (5 cm when dry) × 7–9 cm (4 cm when dry), fleshy, apparently not proliferous, black, covered with brown scales, the scales smaller than those of the petioles, aerating areas 3–4 × 1 mm, mostly towards base of stipules, slightly immersed, elliptic, whitish, stipule margins irregularly laciniate, with narrow 1.0– 1.5 cm long lobes, radiate (not falcate), very thin, breaking off easily so that margin appears subentire. Fronds 350– 400 cm, spirally arranged, densely set, 15 cm apart, 8–12 functional fronds per plant, arching. Petioles 160–180 cm long, 5 cm in diameter, terete, rounded adaxially, with a basal pulvinus (ca. 10–15 cm long, 8 cm in diameter at base), not winged, green (when fresh), flesh pinkish inside, densely scaly especially towards base of young leaves, scales 6–9 × 0.5–0.8 mm, mixed with much smaller scales, golden brown, presence of conspicuous wart-like bases of old scales, aerating areas (15–)25–30 × 1 mm, abundant throughout up to rachis. Laminae 200–220 × 180 cm, bipinnate with undivided pinnules (basal pinnae never more than once pinnate), elliptic, longer than petiole, longer than broad. Rachises terete, rounded adaxially, not winged, green (when fresh) or yellow (when dried), sparsely pubescent or glabrescent. Pinnae petiolulate; petiolules 60 mm (up to 140 mm for basal pinnae, 10–35 mm in distal pinnae), inarticulate, rounded adaxially or with two small crests, pulvini up to 35 mm long (10–20 mm when dry), 15–18 mm in diameter at base (4–6 when dry). Pinnae alternate or rarely subopposite, 6–10 on each side of the frond, 16–24 cm apart (closer towards apex), not overlapping. Basal pinnae 85–100 × 30–38 cm, progressively and slightly reduced, perpendicular to rachis or ascendant towards apex (acrotropic), oblong or deltoid. Middle pinnae 90–94 × 35 cm, ascendant towards apex, oblong. Distal pinnae 50–54 × 24 cm, progressively and slightly reduced, ascendant towards apex, oblong or elliptic. Terminal pinnae absent, the rachis ending in a mucron. Secondary 62 • Phytotaxa 158 (1) © 2014 Magnolia Press SENTERRE ET AL. rachises terete, rounded adaxially, broadly winged distally, wings to 0.5–1.0 mm broad on either side, yellow (when dry), sparsely pubescent or glabrescent. Pinnules (observed on a middle pinna) sessile or shortly petiolulate. Secondary petiolule 0–1 mm, not articulate. Pinnules alternate to opposite, (15–)25–30 on either side, (1.6–)2.0– 2.5 cm apart, not overlapping. Basal pinnules (6–)13–18 × (1.6–)2.2–2.5 cm, similar to other pinnules, perpendicular to rachis, elongate-lanceolate. Middle pinnules (10–)17–19 × (1.7–)2.3–2.5 cm, perpendicular to rachis, lanceolate. Distal pinnules (6–)8–10 × 1.3–1.8 cm, typically progressively reduced, slightly ascendant towards apex (acrotropic), lanceolate. Terminal pinnules 14–25 × 2.5–3.5 cm, more developed than other pinnules, much longer and often much broader, lanceolate. Pinnule blades entire, bases slightly attenuate, symmetrical or slightly asymmetrical, margins slightly serrate with 8 teeth per cm, glabrous, slightly revolute, apices progressively acuminate (with teeth more distantly spaced, 2–3 per cm), acumen 2–3 cm long, dark green above and light green below (when fresh), membranaceous or somewhat coriaceous. Venation pinnate, midvein reaching the apex, glabrescent below, secondary veins distinct, undivided or furcate (more or less close to the midvein), 1.0–1.5 mm apart, straight, oblique, terminating in teeth, venuloids 3–4 mm, intramarginal, ending after the line of sori, evanescent, nearly up to mid-way between the margin and the costa if observed in transparency, straight, occasionally fragmented. Sori widely set, 120–130 per pinnule, 1 row on each side of the midvein, intramarginal, 1.0–1.5 mm from the margins, not immersed or weakly immersed, oblong, 0.6–1.2(–2.0) × 0.4 mm, not indusiate. Sporangia (8–)10–30(–40) per sorus, basally fused, mostly free above. FIGURE 2. Angiopteris chongsengiana: A. Aerating areas of the petiole, B. Details of the venation and shape of pinnules, C. Distal part of a pinna showing the elongated terminal pinnule and winged rachis, D. Petiole base and ‘stipules’, E. Pinnule apex, F. Base of a middle pinna, G. Ripe sporangia. (illustrated by Katy Beaver from living and dried specimens, B. Senterre & I. Fabre 6172) Representative specimens examined:—SEYCHELLES. Mahé: Congo Rouge, 540 m, 14 August 2011, B. Senterre & I. Fabre 6172 (SEY); Mare aux Cochons (Mare d'Antin), 615 m, 12 April 2012, B. Senterre & N. Labiche-Barreau 6259 (SEY); Mont Cotton, 580 m, 17 July 2011, B. Senterre & I. Fabre 6152 (P-02432634, P02432633, SEY); Montagne Planneau (Mont Harrison), 641 m, 27 July 2011, B. Senterre & I. Fabre 6158 (SEY), 624 m, 27 July 2011, 6160 (SEY), 628 m, 27 July 2011, 6162 (SEY); Varigault, 585 m, 10 August 2011, B. Senterre & I. Fabre 6171 (SEY). MARATTIACEAE OF THE SEYCHELLES Phytotaxa 158 (1) © 2014 Magnolia Press • 63 Distribution:—Only known from the granitic Seychelles; possibly endemic on the island of Mahé, where it is found in the ravines between Morne Seychellois, Pérard, Congo Rouge and Mont Cotton in the North of the island, as well as in the massif of Montagne Planneau to Varigault in the center. It has been searched for on neighbouring Silhouette Island, but was not found there, although it could be expected. Ecology:—Angiopteris chongsengiana occurs typically in ravines of the lower montane evergreen rain forest belt, i.e. mostly at an elevation between 500 and 750–780 m above sea level. That belt, due to telescoping effect (Senterre et al. 2009), corresponds to ca. 1500–1800(–2000) m above sea level in continental areas or larger islands at similar latitudes and is characterized by the dominance of tropical montane forest species. Conservation:—This species is rare as it is known from only a few small populations, isolated in ca. 12 ravines of two separate mountain massifs. We estimate the number of mature individuals per ravine to ca. 10–20. The total number of mature individuals is estimated to ca. 120–250. The area of occupancy (AOO) and extent of occupancy (EOO) are respectively ca. 0.1 km² and 17 km². Considering IUCN criterion D (very small population, IUCN 2011: 14, 58), this species can be classified as endangered (EN). Half of the population is in the Morne Seychellois National Park and the other half is in poorly accessible areas, recently proposed to be classified as a new National Park (Senterre et al. 2013). All populations seem healthy but with relatively limited regeneration. Invasive species are present but not abundant and we do not think that they are currently threatening these populations, although this should be assessed more carefully. If the reduced regeneration is confirmed and if it appears to be due to invasive species, then the IUCN threat category could be changed for critically endangered (CR). Etymology:—The specific epithet honours Seychellois botanist Mr. Lindsay Chong-Seng who contributed significantly to the botany, natural history and nature conservation of the Seychelles. For visiting and local naturalists, Lindsay has since long been one of the main sources for knowledge on the Seychelles flora, practical information and guidance. Vernacular name:—This species is relatively rare and difficult to find, and is also easily confused with the more common Angiopteris madagascariensis. Because it has only recently been recognised, it has no Creole name yet. We name it ‘Baton monsenyer-d-gran bwa’ as this species is found in more pristine forests and at higher elevations in the mountains. Morphological affinities:—This species is distinguished by its much longer terminal pinnules of lateral pinnae compared to the distal pinnules pair of the same pinna. This character is usually well-preserved in herbarium specimens, facilitating comparisons of material. The secondary rachises are distinctly winged, which makes it easily distinguishable from the more common Angiopteris madagascariensis. Angiopteris madagascariensis de Vriese in Vriese & Harting (1853: 23). Fig. 3 Type: —MADAGASCAR. Without locality, 1834, J. Goudot s.n. (holotype P!-00466551, isotypes P!-00696299, P!00696300). Synonym:—Angiopteris evecta auct. non (G.Forst.) Hoffm., e.g. sensu Baker (1877: 517). Perennial ferns, 3–4 m high. Rhizomes 50–90 cm tall × 50 cm in diameter, upright, globular or trunk-like, covered with remaining bases of old leaves. Roots branched, stout, 6 mm in diameter. Stipules 5–7 × 5–9 cm, fleshy, often proliferous, black, covered with brown scales, the scales smaller than those on the petioles, aerating areas 3–4 × 1 mm, mostly towards base, slightly immersed, elliptic, whitish, stipule margins regularly laciniate or digitate, with falcate linear lobes. Fronds 400–500 cm, spirally arranged, densely set, 10–12 cm apart, 5–7 functional fronds per plant, arching. Petioles 120–170 cm long, 5 cm in diameter, terete, rounded adaxially, with a basal pulvinus (ca. 10 cm long, 6–8 cm in diameter at base), not winged, yellow (when dry) or green (when fresh), flesh whitish inside, densely scaly (especially towards base on young leaves), scales 8–10 × 0.1–0.2 mm, golden brown, inconspicuous wart-like bases of old scales present, small, aerating areas 2.0–2.5 × 1.0–1.5 mm, abundant throughout up to rachises, only slightly discolorous. Laminae 280–330 × 200–220 cm, bipinnate, with undivided pinnules except for the 1–3 basal pinnae that are bipinnate towards base, elliptic, longer than the petioles. Rachises terete, rounded adaxially, not winged, yellow (when dry) or green (when fresh), sparsely pubescent. Pinnae petiolulate. Petiolules 30–50 mm (up to 200–350 mm for basal pinnae, 20 mm in distal pinnae), not articulate, rounded adaxially, pulvini 30 mm long, 20 mm in diameter at base (when fresh). Pinnae alternate, 10–12 on each side of frond, 23–26 cm 64 • Phytotaxa 158 (1) © 2014 Magnolia Press SENTERRE ET AL. apart (closer towards apex), not overlapping. Basal pinnae progressively and slightly reduced, perpendicular to rachises, oblong. Middle pinnae 100 × 35 cm, ascendant towards apices (acrotropic), oblong. Distal pinnae progressively reduced. Terminal pinna absent, the rachis terminated in a mucron. Secondary rachises terete, rounded adaxially, not winged, green (when fresh), sparsely pubescent. Pinnules (observed on a middle pinna) petiolulate. Secondary petiolules ca. 5 mm, not articulate, rounded adaxially. Pinnules alternate or subopposite or opposite, 34–36 on either side, 2.2–3.3 cm apart, not overlapping. Basal pinnules similar to other pinnules. Middle pinnules 20–22 × 2.5 cm, slightly ascendant towards apices (acrotropic), elliptic to nearly oblong. Distal and terminal pinnules similar to other pinnules. Pinnule blades entire, bases attenuate or obtuse, slightly asymmetrical, not decurrent on the petiolules, margins scarcely serrate, teeth 12–13 per cm, margins glabrous, slightly revolute, apices progressively acuminate, acumen 2–4(–5) cm, green or olive above, yellowish below, membranous, glabrous. Venation pinnate, midveins reaching the apex, secondary veins distinct, undivided or furcate (at bases and up to more than midway to the margins), 0.7–0.8 mm apart, straight, oblique, terminating in teeth apices, venuloids 2.0–3.5 mm long, intramarginal, ending after the line of sori, evanescent, nearly up to mid-way between the margin and the costa if observed in transparency, straight. Sori more or less densely placed, 300–350 per pinnule, in one row on each side of the midvein, intramarginal and from the base nearly to the acumen, at 1–2 mm from the margin, shortly immersed, oblong or elliptic, not indusiate. Sporangia 6–12 per sorus, nearly free, basally fused. FIGURE 3. A–E: Angiopteris chongsengiana. A. View of the ‘stipules’ and rhizome (photograph by B. Senterre), B. Winged secondary rachis, with sessile pinnules (photograph by I. Fabre), C. Mature sporangia and venuloids (photograph by B. Senterre), D. Wart-like bases of old scales on petiole (photograph by B. Senterre), E. Typical apex of a pinna (photograph by I. Fabre); F–J: Angiopteris madagascariensis. F. Massive rhizome ca. 60–70 cm high (photograph by B. Senterre), G. secondary rachis not winged (photograph by I. Fabre), H. Old sporangia and venuloids (photograph by B. Senterre), I. ‘Stipule’ (photograph by I. Fabre), J. Typical aspect of the basal pinna, bipinnate on mature individuals (photograph by I. Fabre). Representative specimens examined:—SEYCHELLES. Mahé: L.H. Boivin s.n. (P-01646348, P-01646344, P-01646338, P-01646346), 400 m, 27 February 1882, Th. Delacour 36 (P-01332779, P-01646336), J. Stanley Gardiner s.n. (K), J. Horne 199 (K), L. Humblot s.n. (P-01332835, P-01332836), A. Pervillé 207 (P-01332837, P- MARATTIACEAE OF THE SEYCHELLES Phytotaxa 158 (1) © 2014 Magnolia Press • 65 01646350, P-01646349), H.J. Schlieben 11726 (K), s.c. s.n. (P-01646337), Mériau s.n. (P-01646342, P-01646341); Casse Dent, 400 m, 30 May 2008, B. Senterre 5353 (SEY), 550 m, 6 July 2011, B. Senterre & N. Labiche-Barreau 6123 (P-02432632, SEY); Casse-Dent à Congo Rouge, 18 April 1972, H. Jacquemin 988 (P-01332834, P01646339); Congo Rouge, 300-800 m, July 1970, J. Procter 4076 (K, P-01646332, SEY); La Réserve, 400 m, May 1987, F. Friedmann 5485 (P-01646333); Mare aux Cochons, near Mt. Jasmin, M.J. Whitehead 56 (K); Mission Viewing Lodge, D. Lorence 1810 (K); Morne Blanc, près du sommet, November 1982, F. Friedmann 4367 (P01592104); Trois Frères, April 1981, F. Friedmann 3752 (P-01646334, P-01646335); Silhouette: 500 m, 14 September 1993, C.S. Awmack 407 (SEY); Sans Sentiment, J. Stanley Gardiner s.n. (K). Without locality: Barkly s.n. (BM), L.H. Boivin s.n. (P-01646345), A. Pervillé s.n. (P-01646343, P-01646340). Distribution:—Endemic to islands in the Western Indian Ocean. It is known from Madagascar, Mayotte, La Réunion, Mauritius and the Seychelles, where it is widely distributed on the islands Mahé and Silhouette. It has been expected and searched for in the ravines of nearby Praslin, but was not found on that island. Ecology:—Common in submontane and lower montane ravine evergreen rain forests between 300 and 750 m, progressively replaced by Angiopteris chongsengiana and Ptisana laboudalloniana at elevations above 750 m; penetrating the lowland belt in wet ravines down to ca. 150 m as isolated individuals. Conservation:—This species is locally relatively common and also occurs on other Mascarene Islands and in Madagascar. We estimate its IUCN red list category to be LC (Least Concern), although a formal assessment is yet to be made. Vernacular name:—‘Baton monsenyer’ (Creole), a name given in reference to the resemblance of the crosier, or developing leaves, with the cross of cardinals. Morphological affinities:—Originally, specimens from the Seychelles were identified as A. evecta. This was the first species of Angiopteris to be described and because herbarium material is fragmentary and therefore not diagnostic, the majority of Angiopteris were initially identified with that name before the genus became better known. As we understand now, Angiopteris evecta is a species from Australasia and the Pacific (holotype from Tahiti) and has occasionally naturalised elsewhere in the Tropics (see Christenhusz & Toivonen 2008). It is distinguished by the presence and length of its false veins, i.e. almost to the costa vs. intramarginal for A. madagascariensis. The Seychelles specimens are very much like those from Madagascar, although we have not found any Madagascan specimens with the basal 1–2 pinnae having pinnate basal pinnules (observed only in fully developed individuals). Nevertheless, this is a common feature in other Angiopteris species and may be environmentally induced. Ptisana laboudalloniana Senterre & I.Fabre, sp.nov. Fig. 4 TYPE:—SEYCHELLES. Mahé: Montagne Planneau (= Mont Harrison), crête au Sud du sommet, 628 m, 27 July 2011, B. Senterre & I. Fabre 6161 (holotype P!, 2 sheets P-02432635, P-02432636, isotype SEY!, 2 sheets). Diagnosis:—This species is characterized by its secondary rachises widely winged along most of their length, few pinnae, 2–4 pairs, petioles 1.0–1.6 m long, erect, verrucose, brownish to purplish on fresh, laminae shorter than the petioles and shorter than broad, pinnules not thick, not fleshy, (sub)sessile, with an apex progressively narrowed, and intramarginal synangia. Perennial ferns, 1.6–2.4 m. Rhizomes 18–20 × 15–20 cm in diameter, upright, globular, covered with remaining bases of old leaves. Roots branched, stout, 7 mm in diameter. Stipules 7.0 × 5.5 cm (when fresh), fleshy, rarely proliferous (observed once), glabrescent, with or without visible aerating areas, surface irregular, stipule margins entire, irregularly cracked. Fronds 190–260 cm, spirally arranged, densely set, 10 cm apart, ca. 4 functional fronds per plant, strongly ascending. Petioles 100–160 cm long, 3–4 cm in diameter, terete, rounded adaxially, with a basal pulvinus (ca. 5–7 cm long, 5 cm in diameter at base), not winged, yellow (when dry), purple or brown to black (when fresh), flesh pinkish inside, densely scaly (on young leaves), scales 6–7 × 1.5–2.0 mm, mixed with caducous narrower scales (1–3 cells wide), golden brown, with strong wart-like bases of old scales (almost prickly on old leaves), aerating areas abundant throughout up to rachises, 20–40 × 1–2 mm. Laminae 90–100 × 125–130 cm, bipinnate with undivided pinnules, rhomboidal, shorter than the petiole, shorter than broad. Rachises terete, rounded adaxially, not winged, yellow (when dry), green (when fresh), glabrous. Pinnae petiolulate. Petiolules (40– )50–70 mm (up to 100 mm in basal pinnae, 30 mm in distal pinnae), inarticulate, rounded adaxially or with two small crests, pulvini (1.5–)2.5–3.0 mm long, 3–5 mm in diameter at base (when dry, up to 12 mm when fresh). 66 • Phytotaxa 158 (1) © 2014 Magnolia Press SENTERRE ET AL. Pinnae opposite (rarely subopposite or alternate in a few leaves), 2–4(–5) on each side of the frond, 18–20 cm apart (closer towards apex), not overlapping. Basal pinnae similar to other pinnae, often curved backwards (basitropic) or perpendicular to the rachis. Middle pinnae 70 × 30 cm, oblong. Distal pinnae 52 × 25 cm, progressively reduced, ascendant towards apices (acrotropic), oblong. Terminal pinna absent (paripinnate) or similar to lateral pinnae (rarely). Secondary rachises terete, rounded adaxially, almost entirely winged, wings 1.0–1.5 mm broad (on each side), plane, green, glabrous. Pinnules (of middle pinnae) shortly petiolulate or sessile. Secondary petiolules (0–)2– 3 mm, articulate, rounded adaxially. Pinnules alternate, subopposite or opposite, (9–)14–16 on either side, 2.0–3.5 cm apart, not overlapping. Basal pinnules similar to other pinnules or progressively and slightly reduced, curved backwards (basitropic), perpendicular to rachis or rarely ascendant towards apex (acrotropic), oblong, 8–11(–14) × 1.8–2.5 cm. Middle pinnules slightly ascendant towards apices (acrotropic), oblong, (11–)15–22 × 2.1–2.9 cm. Distal pinnules similar to other pinnules, ascendant towards apices (acrotropic), oblong. Terminal pinnule absent (paripinnate), similar to other pinnules or more developed than other pinnules (and often irregularly pinnate distally). Pinnule blades entire, bases obtuse, rounded or subtruncate, slightly asymmetrical (more acute on the distal side), shortly decurrent on the petiolules, margins serrate, glabrous, plane or slightly revolute, teeth 6 per cm, more developed towards the acumen, apex progressively acuminate, acumen 1.7–3.0 cm, somewhat to strongly discolorous, dark green or olive above, light green or glaucous below, sometimes reddish on developing leaves, coriaceous, glabrous. Venation pinnate, midveins reaching the apex, secondary veins distinct, undivided or furcate (at base or before midway to the margin), 2.2–2.5 mm apart, straight, slightly oblique, terminating in teeth apices, venuloids absent. Sori (1.0–)1.4–2.2 × 1.0 mm, widely set, 40–130 per pinnule, intramarginal and basal (often restricted to the basal half of the pinnule), at 1–2 mm from the leaf margin, not immersed or slightly immersed, oblong, more rarely elliptic, not indusiate. Sporangia 12–16 per sorus, fully fused into a synangium. FIGURE 4. Ptisana laboudalloniana: A. A large plant with relatively small rhizome; Congo Rouge, Seychelles (photograph by B. Senterre), B. Leaves; Pérard, Seychelles (photograph by B. Senterre), C. Closer view to the rhizome and “stipules” (photograph by I. Fabre), D. common variation in the apex of pinnae (photograph by I. Fabre), E. Winged pinna rachis (photograph by I. Fabre), F. Unripe sporangia (photograph by I. Fabre), G. Ripe sporangia (photograph by B. Senterre). Representative specimens examined:—SEYCHELLES. Mahé: J. Stanley Gardiner s.n. (K); Congo Rouge, 730 m, 30 May 2008, B. Senterre 5348 (SEY); Congo Rouge Est, 820 m, 6 July 2011, B. Senterre & N. Labiche- MARATTIACEAE OF THE SEYCHELLES Phytotaxa 158 (1) © 2014 Magnolia Press • 67 Barreau 6126 (SEY); Mont Cotton, 560 m, 17 July 2011, B. Senterre & I. Fabre 6141 (SEY); Montagne Planneau (= Mont Harrison), 628 m, 18 August 2013, B. Senterre 6593, 6594 (SEY); Varigault, 576 m, 10 August 2011, B. Senterre & I. Fabre 6168 (SEY). Distribution:—This species appears to be endemic to the island of Mahé in the granitic Seychelles, on the highest summits of the Morne Seychellois National Park and in the massif of Montagne Planneau. It has been searched for on Silhouette Island (close to Mont Dauban), but was not found there. Ecology:—Usually as understory plant of the “tree fern lower montane forest belt” (Elzein 2011, Senterre 2011), of evergreen rain forests, mostly above 750 m elevation. Due to telescoping effect this elevational belt would correspond to 1800–2500 m above sea level in continental areas or larger islands at similar latitudes. This elevational belt is characterized by the dominance of tropical montane forest species and tree ferns (Cyatheaceae). Ptisana laboudalloniana is occasionally found at lower elevations (> 500 m), but these are isolated individuals in wet ravines. Conservation:—The geographical distribution pattern and population size of Ptisana laboudalloniana are very similar to those of Angiopteris chongsengiana, the two species being often found in association. Ptisana laboudalloniana is slightly more common (found in ca. 15 ravines, especially on the summits of Congo Rouge and Pérard) and each ravine contains slightly more mature individuals (ca. 20-30). Therefore, considering IUCN criterion D (very small population, IUCN 2011: 14, 58), this species can be classified as vulnerable (VU), i.e. less than 1000 mature individuals and area of occupancy less than 20 km². The total number of mature individuals is estimated to ca. 300–500. The area of occupancy (AOO) and extent of occupancy (EOO) are respectively ca. 0.1– 0.2 km² and 17 km². All populations seemed very healthy, with abundant regeneration and little impact from the small populations of invasive species. Etymology:—The specific epithet honours Seychellois botanist Mr. Victorin Laboudallon who contributed significantly to the botany, natural history and conservation of the Seychelles. Victorin Laboudallon is currently the chairman of a local NGO. Vernacular name:—This species is rare in the Seychelles and has not been observed for a long time. If found, it was confused by local people with Angiopteris madagascariensis, which is superficially similar. Therefore it lacks a Creole vernacular name as yet. Due to its smaller size compared to the two other species of Marattiaceae in the Seychelles, we propose to name it ‘Pti baton monsenyer’. Morphological affinities:—This species was first recorded on the Seychelles by Christensen (1912) as Marattia fraxinea, a species described from Mauritius and currently called Ptisana fraxinea var. fraxinea. Nevertheless, the Seychelles plants differ from typical P. fraxinea in being an exclusively montane species and having strongly winged secondary rachises. Another taxon with winged secondary rachises has been described from South Africa and is known as Ptisana fraxinea var. salicifolia. The latter taxon is larger than the Seychelles Ptisana (with leaves up to 3 m long), with laminae longer than the petioles and has at least 6 pairs of pinnae. Our molecular results show that Ptisana laboudalloniana is related to the rare species Ptisana purpurascens, endemic to Ascension Island, and that the South African Ptisana fraxinea var. salicifolia forms a distinct clade (Fig. 1). Ptisana purpurascens has slightly winged secondary rachises and leaves with few pairs of pinnae, but it differs from P. laboudalloniana especially in having thicker, fleshier and much shorter (4–10 cm) pinnule blades. In mainland Africa, we recognize another four species that all differ from the Seychelles species in having pinnules with abruptly caudate-acuminate apices combined with the secondary rachises not being winged: Ptisana odontosora (endemic to Guinea), P. robusta (endemic to São Tomé) and two undescribed species (one being commonly found from Cameroon to Gabon and the second found from Guinea to tropical East Africa). Synoptic revision of the genus Ptisana in Africa and neighbouring islands The revision of the Seychelles specimens required the study of the complex of species traditionally associated with Ptisana fraxinea. This complex has been discussed in detail by Pichi Sermolli (1969) and was further discussed by Murdock (2008a, 2008b) based on molecular results with samples from South Africa (identified as P. fraxinea var. fraxinea and P. fraxinea var. salicifolia) and from Ascension Island (identified as P. purpurascens). Here we further discuss this complex based on new evidence, but a monographic revision, including a study of more material from Africa, will be needed to confirm our observations. 68 • Phytotaxa 158 (1) © 2014 Magnolia Press SENTERRE ET AL. Much confusion originated due to repeated descriptions of plants from Mauritius and La Réunion (Marattia cuneiformis de Vriese in Vriese & Harting (1853: 7), Marattia macrophylla de Vriese in Vriese & Harting (1853: 3), Myriotheca fraxinifolia Bory (1804: 266), Myriotheca sorbifolia Bory (1804: 267)) where only one taxon is nowadays commonly recognized: Ptisana fraxinea var. fraxinea (Badré 2008, Roux 2009). Consequently, P. fraxinea has been interpreted as a variable taxon and most of the specimens from Madagascar to West Africa were considered to fall within the variability of this species (Roux 2009), although varieties were recognized for South African specimens (P. fraxinea var. salicifolia) and plants from São Tomé (P. fraxinea var. robusta). The endemic P. purpurascens from Ascension Island in the Atlantic Ocean has often been associated with, but never included in this complex (Pichi-Sermolli 1969, Murdock 2008b). The results from our molecular analyses (Fig. 1), including specimens from Indian Ocean islands (from which Ptisana fraxinea was originally described), suggest a more complex situation, because the specimens from the Seychelles are well supported as sister to P. purpurascens (Ascension Island) and are strongly supported as distinct from the specimens of Ptisana fraxinea from nearby Madagascar. Because of this geographical discrepancy and morphological differences, the Seychelles specimens must be considered an undescribed species and Ptisana fraxinea as traditionally circumscribed is not monophyletic. Below we discuss the taxa that require to be redefined following our observations, and a few new combinations in Ptisana are proposed. Key to the species of Ptisana in Africa and surrounding islands 1. -. 2. -. 3. -. 4. -. 5. -. 6. -. 7. -. 8. -. Pinnule veins curved. Synangia perfectly marginal, almost in the teeth of the margin. (swamps in Guinea)......................... .................................................................................................................................................................. Ptisana odontosora Pinnule veins straight. Synangia more than 1 mm from the margin ...................................................................................... 2 Very large plants (leaves > 3 m, petiole > 1.5 m on mature individuals), with trunk-like rhizomes (> 30 cm high), laminae with at least 6 pairs of pinnae ................................................................................................................................................ 3 Medium sized to relatively small plants (leaves < 3 m long, petiole < 1.5 m on fully mature individuals), with hemispherical rhizomes (< 30 cm high), laminae with less than 5 pairs of pinnae (rarely more and then just on a few leaves, see P. fraxinea) ................................................................................................................................................................................ 4 Broadly winged pinna rachises between the terminal pinnules and the last 5–8 juga. Distal pairs of pinnules mostly gradually and evidently smaller than the others. Terminal pinnules longer and often pinnatifid or pinnatipartite at the base. (South Africa, Zimbabwe, Malawi and Mozambique)................................................................................ Ptisana salicifolia Pinna rachises not winged or at most with narrow wings distally up to the 1–2 distal pairs of pinnules. Distal pairs of the same size as the others. Terminal pinnules similar to the lateral ones. (Guinea, Ethiopia, Kenya, Democratic Republic of Congo) ................................................................................................................................................................ Ptisana sp. A Synangia median, situated at midway between the margin and the costa. Pinnule apices rounded or progressively narrowed, not acuminate. (Madagascar incl. Île de Sainte-Marie)....................................................................... Ptisana boivinii Synangia marginal, situated at 1–2 mm from the margin. Pinnule apices acuminate............................................................ 5 Pinna rachises prickly (São Tomé)................................................................................................................. Ptisana robusta Pinna rachises not prickly ...................................................................................................................................................... 6 Pinnule blades thick, fleshy, leathery. Pinna rachises thick, robust. (Ascension Island) ...................... Ptisana purpurascens Pinnule blades more or less thick but never fleshy. Pinna rachises not robust ..................................................................... 7 Relatively small plants, petioles up to 60 cm. Pinnules at most 3 times longer than wide, with the apex abruptly caudateacuminate. (Cameroon, Equatorial Guinea, Gabon) ........................................................................................... Ptisana sp. B Medium sized plants, petioles generally more than 1 m. Pinnules more than 3 times longer than wide, with the apex progressively acuminate ............................................................................................................................................................. 8 Plants with a broad elevational range, from sea level upward. Secondary rachises at most slightly winged distally (except occasionally in juvenile plants). Terminal pinnules always similar to the other ones. (Madagascar, Mascarenes)................. ........................................................................................................................................................................ Ptisana fraxinea Plant exclusively in montane forests. Secondary rachises strongly winged on most of their length. Terminal pinnules often irregularly pinnate distally. (Seychelles) ...........................................................................................Ptisana laboudalloniana Ptisana boivinii (Mett. ex Kuhn) Senterre & Rouhan, comb. nov. Marattia boivinii Mett. ex Ettingshausen (1864: 246), as ‘Boivini’. Type:—MADAGASCAR. Without locality, without date, Boivin s.n. (possible holotype L, isotype P-00466552!). This species is poorly known but very characteristic. It is a relatively small plant, ca. 1 m high or less, with a petiole MARATTIACEAE OF THE SEYCHELLES Phytotaxa 158 (1) © 2014 Magnolia Press • 69 ca. 30–50 cm and few pairs of pinnae (1–4), the lamina is shorter or slightly longer than the petiole. It is easily recognized by the synangia placed nearly mid-way between the margin and the midvein. The pinnules are narrowly elliptic, not acuminate at apex and clearly petiolulate. We do not know other Ptisana species that are morphologically similar. Distribution:— Ptisana boivinii is known from a few specimens collected in southern Madagascar at Sainte Marie (Boivin 1605, P-01592077!, P-01592078!, P-01592079!). Ptisana fraxinea (Sm.) Murdock (2008b: 746). Maratttia fraxinea Smith (1790: t. 48). Type:—MAURITIUS. Without locality, without date, Thouin 91 (holotype LINN!, possible isotype G). Myriotheca fraxinifolia Bory (1804: 266). Type:—LA RÉUNION. Quartier de Ste. Rose, without date, J.B.G.M. Bory de St Vincent s.n. (holotype P-00466550!, isotype FI, possible isotype K!). Myriotheca sorbifolia Bory (1804: 267). Type:—LA RÉUNION. Habitation Fabien sur les hauteurs de St. Denis, without date, J.B.G.M. Bory de St Vincent s.n. (holotype: P-00466549!, isotypes B, FI). Marattia macrophylla de Vriese in Vriese & Harting (1853: 3). Lectotype (designated by Pichi Sermolli 1969):— MAURITIUS. Without locality, without date, Bojer s.n. (K!). Marattia cuneiformis de Vriese in Vriese & Harting (1853: 7). Type:—LA RÉUNION. Without locality, without date, s.c. no. 21 (possible type K!, 2 sheets). Marattia microcarpa Mett. ex Ettingshausen (1864: 246). Lectotype (designated by Pichi Sermolli 1969):—MADAGASCAR. Nosy-Be [Nossibe], 1846–1848, Boivin s.n. (W, fragment and photo BM!) Marattia salicifolia auct. non Schrader (1818: 920), sensu Tardieu-Blot (1951), see H. Humbert 3211 (P-01647489!). After reviewing specimens and literature, we conclude that Ptisana fraxinea does not reach mainland Africa but is restricted to the islands of the Western Indian Ocean. The presence of Ptisana fraxinea (as defined here) in Sri Lanka (as Marattia calliodous de Vriese in Vriese & Harting (1853: 6), Gardner s.n., holotype: K!) is possible but unlikely and should be confirmed on the basis of field work. Since most descriptions available for P. fraxinea are doubtful or fragmentary, we considered specimens from Madagascar, Mauritius, La Réunion and Mayotte, which were accompanied with good descriptions of key characters (see mostly specimens T. Cadet 3598, F. Badré 879, K.U. Kramer 9277, H. Tuyguy 1082, J.-N. Labat 2749, G. Rouhan 338, 1206). These plants have pinnae similar to those of P. fraxinea, as circumscribed above, and are characterized by leaves of up to 2.2(–3.0) m long, with petioles 0.8–1.8 m long (most often longer than the lamina to occasionally slightly shorter), and 2–4 pairs of pinnae, up to 6(–7) pairs for some of the largest leaves only. These characters indicate the morphological affinity with P. laboudalloniana and P. purpurascens, which is also confirmed by our molecular data. On the contrary, the plants from Ethiopia called M. fraxinea by Pichi-Sermolli (1969) are described as being always larger plants, with the petioles typically shorter than the laminae and with a more massive rhizome, so that “this fern resembles Angiopteris sp.” (see Pichi-Sermolli 1969: 348). Although the pinnae are very similar, their shape and apices of pinnules are slightly different. Therefore, we think that Pichi-Sermolli (1969), in his otherwise excellent revision, misapplied the name Marattia fraxinea in Ethiopia, because he focused on characters of pinnae and probably because he had not been able to observe living specimens from the Mascarenes. This idea is supported by the fact that Pichi-Sermolli (1969), who extensively studied specimens from Ethiopia to South Africa, considered his M. fraxinea as being a very close but distinct species to the South African plants, i.e. Marattia salicifolia, which indeed have the same leaf characteristics, but are not closely related to true Ptisana fraxinea, based on our molecular phylogenetic results. Therefore, we conclude that M. salicifolia is an independent species and that the Ethiopian plant cannot be associated with the name Ptisana fraxinea. The East African Ptisana is very similar to some specimens from Guinea, which we discuss below (see Ptisana sp. A). Selected specimens examined :— LA RÉUNION (F. Badré 879, P-01592037; T. Cadet 3598, P-01592042; 4526, P-01592038; K.U. Kramer 9277, P-01592049), MADAGASCAR (T. Janssen et al. 2529, P-00590721; Ms. Marie 55 ex Herb. Paris, K!; Razafitsalama 1151, P!; F. Rakotondrainibe 1281, P-00064683; 4823, P-00134961; Rouhan et al. 338, P-00749288; 1206, P-02432738; s. coll. s.n., labeled “oppositifolia”, MO-1854058), MAYOTTE (as M. microcarpa: Pichi Sermolli 1969; J.-N. Labat 2749, P-00052881; H. Tuiguy 1082, P-00144987, P-00144988, P-00144989). Distribution:— Mauritius, La Réunion, Madagascar, Mayotte (as Marattia microcarpa: Pichi Sermolli 1969) and the Comoros (as M. microcarpa: Pichi Sermolli 1969) 70 • Phytotaxa 158 (1) © 2014 Magnolia Press SENTERRE ET AL. Ptisana odontosora (Christ) Senterre & Rouhan, comb.nov. Marattia odontosora Christ (1909 : 19). Lectotype (designated by Pichi Sermolli 1969: 343):—GUINEA. Tolo, 23 December 1905, O. Caille 15629 (P-00507743!, isolectotypes P-00507741!, P-00507742!) This species is clearly morphologically distinct as demonstrated by Pichi Sermolli (1969), but the name has been mostly misapplied, due to the inclusion of two species in the cited material of the original description. Ptisana odontosora is characterized by ascending, curved veins (not oblique and straight as in most Ptisana species), distal pinnules with their base strongly decurrent on the secondary rachises, and synangia very close to the margin, almost within the teeth. Its ecology is also unusual, found in marshes outside the Guineo-Congolian region where Ptisana sp. A occurs and with which this species has often been confused. The apex of the pinnules is variable but sometimes abruptly caudate acuminate, which suggest a possible affinity with other taxa from mainland Africa. Distribution:—This species is known only from the type locality in Fouta Djallon, a highland region in the centre of Guinea. Ptisana robusta (Alston) Senterre & Rouhan, comb. nov. Marattia robusta Alston in Exell (1956: 8). Ptisana fraxinea var. robusta (Alston) Murdock (2008b: 746). Type:—SÃO TOMÉ. Monte Café, 1200 m, 17 January1949, J.V.G. Espírito Santo 191 (holotype BM!, isotypes COI, LIS-C000002!). Ptisana robusta is a large plant (about same size as Seychelles plants, i.e. P. laboudalloniana), with sessile, symmetrical pinnules and a remarkably prickly pinna rachis. The conclusions made by Pichi Sermolli (1969) regarding the affinities between Marattia robusta and M. fraxinea (as M. macrophylla), as well as confusions between M. fraxinea and M. salicifolia, can be explained by the importance given to pinnules width, which we think should only be considered as a secondary character. Distribution:—Endemic to São Tomé. Ptisana salicifolia (Schrad.) Senterre & Rouhan, comb. nov. Marattia salicifolia Schrader (1818: 920). Ptisana fraxinea var. salicifolia (Schrad.) Murdock (2008b: 746). Type:—SOUTH AFRICA. Cape of Good Hope, Hesse s.n. (holotype LE). This species has a relatively wide distribution and it forms a complex not with Ptisana fraxinea but with an undescribed species (here treated as Ptisana sp. A) found further north into tropical East Africa and westwards to Guinea. Diagnostic characters of Ptisana salicifolia are detailed by Pichi-Sermolli (1969: 336). It differs from most species by its large leaves that can grow to 3 m long, arching (not upright), with the laminae longer than the petioles, with at least 6 pairs of pinnae, and it has very stout rhizomes, more than 30 cm tall. It differs from Ptisana sp. A by its pinnae rachises that are manifestly winged between the terminal pinnule and the last 5–8 pairs, and the last upper lateral pinnules mostly gradually and evidently smaller than the others, except for the terminal pinnules which are longer and often pinnatifid or pinnatipartite at the base. Distribution:—Typical Ptisana salicifolia occurs in South Africa (e.g. A. Rehmann 384, P01647484!), Zimbabwe, Malawi and Mozambique (see online flora of Zimbabwe: http://plants.jstor.org/flora/fz7513). Ptisana sp. A Synonyms:—Marattia chevalieri Christ, nom. ined. (A. Chevalier 13445, P-00507739). Marattia fraxinea auct. non Sm., sensu Pichi-Sermolli (1969). This species is typically found in Kenya and Ethiopia and it most likely extends into Rwanda, Burundi, Tanzania, Malawi, Mozambique and the Democratic Republic of Congo. After reviewing specimens from Guinea, Côte d’Ivoire and Sierra Leone, we have concluded that these specimens belong to the same species as P. fraxinea of Pichi-Sermolli (1969) from Ethiopia. Pichi-Sermolli also confirmed this by placing a determinavit ‘Marattia fraxinea’ on the specimen A. Chevalier 13445 (P). Ptisana sp. A has also been confused with P. odontosora by Chevalier, before the description of the latter, and consequently Christ included the material misidentified by MARATTIACEAE OF THE SEYCHELLES Phytotaxa 158 (1) © 2014 Magnolia Press • 71 Chevalier as syntypes for M. odontosora (see detailed explanation in Pichi-Sermolli 1969: 341–343), which explains why the name M. odontosora has been commonly used for specimens from tropical East Africa. Nevertheless, P. odontosora is a totally different species from marshes in the Fouta Djalon (Guinea). Considering that Pichi-Sermolli (1969) carefully studied Ptisana sp. A and identified its closest relative as P. salicifolia, and that he recognized it as distinct at species level, we suspect we are dealing with an undescribed species rather than as a subspecies of P. salicifolia. It differs from P. salicifolia by the “pinnae with rachis not winged or with a very narrow and straight wing between the terminal pinnule and the last two or three juga”, and by the “terminal pinnules similar to the lateral ones” (Pichi-Sermolli 1969: 336). As for other taxa in the Guinean region, the apices of pinnules are “gradually tapering and usually abruptly terminated by a long caudate point or sometimes ending with an acuminate tip” (Pichi-Sermolli 1969: 336, 349). If only characters of pinnae are available, it is difficult to distinguish Ptisana sp. A from P. fraxinea, which explains why it has traditionally been confused with that species. Ptisana sp. A differs from P. fraxinea by its distal pairs of pinnules being similar in size and shape to the other ones (progressively larger distally for P. fraxinea), the pinnules with a more or less abruptly caudate acuminate apex (progressively narrowed or acuminate for P. fraxinea), a larger rhizome, up to 40 cm tall (up to 20 cm, hemispherical, for P. fraxinea), leaves of 2.5–3.0 m long, up to 4 m (Verdcourt 1999), petioles 0.9–1.2(–1.5) m (thus shorter than the lamina), with 6–9 pairs of pinnae. Representative specimens examined:—GUINEA. Chevalier 12329 (P-01647448!), 17052 (P-01647426!). IVORY COAST. Chevalier 19685 (P-01647424!). SIERRA LEONE. Jaeger 6873 (P-01647431!). ETHIOPIA. Mooney 8809 (K). KENYA. P. Kamau & M.J.M. Christenhusz 638 (EA, K). DEMOCRATIC REPUBLIC OF CONGO. de Nere 2184 (P-01647389!). Ptisana sp. B Synonym:—Marattia fraxinea auct. non Sm., sensu Tardieu-Blot (1964). This undescribed species is widespread in submontane evergreen rain forests of Atlantic Central Africa, but is unlikely to reach West Africa (e.g. Liberia and or Côte d’Ivoire as stated by Tardieu-Blot 1964), where we think that it has been confused with Ptisana sp. A. Ptisana sp. B is one of the smallest Ptisana of the region, with petioles up to 60 cm, and sessile pinnules with the apices abruptly caudate to acuminate. It differs from Ptisana robusta and Ptisana sp. A by its smaller habit and by its shortly oblong pinnules (less than three times longer than wide). Representative specimens examined:—EQUATORIAL GUINEA. R. Pérez Viso 2665 (MA-748717!). CAMEROON. F. Breteler 2500 (P-01365197!). GABON. E. Bidault 572 (MO!). Distribution:— This species is found in Equatorial Guinea (Velayos et al. 2008: 313), Gabon and Cameroon (Tardieu-Blot 1964, as Marattia fraxinea). Conclusions The fern family Marattiaceae contributes to the understanding of phytogeographical affinities of the Seychelles flora. The three present species belong to main chorological groups. Angiopteris madagascariensis corresponds to the Western Indian Ocean islands centre of endemism. It illustrates the strong affinity of the Seychelles flora with the other islands of that region. Angiopteris chongsengiana, without morphologically close relatives, belongs to the endemic element of the Seychelles flora, substantiating its long isolation. The presence of the genus Angiopteris underlines the affinities of the Western Indian Ocean islands with tropical Asia, as previously noted by Summerhayes (1931). Finally, the endemic Ptisana laboudalloniana shows affinities to species from continental Africa. Isolated islands represent good opportunities for migratory birds to have a rest on long ocean crossings, and the Seychelles are no exception to this (L. Chong-Seng, pers.comm.). Some shearwater birds (i.e. Puffinus bailloni Bonaparte 1857) have been observed in the montane ravines of Seychelles and may be the vector for such long distance dispersal of montane plants (G. Rocamora, pers.comm.). Wind, on the contrary, could eventually bring spores through long distance dispersal, but has a lower probability for establishment, especially in suitable habitats. 72 • Phytotaxa 158 (1) © 2014 Magnolia Press SENTERRE ET AL. The spores of Marattiaceae are also relatively large and short-lived, making wind-dispersal over long distances less likely. We therefore hypothesize that long-distance dispersal associated with birds could play an important role in speciation processes in the genus Ptisana, but not at higher taxonomic levels (see Christenhusz & Chase 2013). Birds could be carrying both the spores of the fern and the associated mycorrhizae, facilitating these rare colonization events. The absence of Ptisana laboudalloniana on Silhouette Island (although suitable habitat does exist there) appears to support this hypothesis. If our interpretation of Marattiaceae in the African-Madagascan region is correct, i.e. including the recognition of several local endemics (Ptisana odontosora in Guinea, P. robusta in São Tomé, P. purpurascens in Ascension Island, P. laboudalloniana in the Seychelles and P. boivinii in Madagascar), it will be necessary to study plants of this family more carefully, both in herbarium and in the field, especially in isolated mountains and remote islands, with more attention drawn to elevational gradients and micro-habitats. Our study suggests the presence of two additional species on the African mainland. In total, we recognize nine species of Ptisana in Africa and neighbouring islands: two from Atlantic Central Africa and São Tomé, one from Ascension Island, three from Guinea to Ethiopia (at the periphery of the Guineo-Congolian region) and to South Africa, and three from the Western Indian Ocean islands. Phylogenetic affinities between Ptisana fraxinea sensu stricto (confirmed to occur only on the Mascarenes and Madagascar), the southern African P. salicifolia and the mostly East African P. sp. A remain to be studied into more details. A broad geographical sampling using highly variable molecular markers (e.g. AFLPs) may be needed to more precisely investigate the biogeography of Marattiaceae and to fully resolve the Ptisana complex in Africa and surrounding islands. Acknowledgements The explorations that resulted in the discoveries published here were funded by UNDP-GEF, via a consultancy on the Seychelles Key Biodiversity Areas and a Small Grant Programme (SGP) for the development of the Seychelles National Herbarium. We are grateful to the Seychelles Ministry of Environment for supporting the necessary material exchange (material transfer agreement) between the Seychelles National Herbarium and the Muséum national d’Histoire naturelle (MNHN), Paris. We are also thankful to the following authorities for allowing us to work in the forests and for assistance in the field: Madagascar National Parks, the Ministère des Eaux et Forêts, and CNRE in Madagascar, CNDRS in Comoros, and the Seychelles National Parks Authority in Seychelles. Molecular work was conducted in the Botany-Entomology-Mycology (BoEM) lab of the Paris Museum, and funded through the ATM ‘Taxonomie moléculaire, DNA Barcode & gestion durable des collections.’ Additionally we thank curators of the herbaria BR, H, K, L, MAO, MO, P, SEY and TAN for their assistance in preserving and locating crucial specimens for this study. We are grateful to Marcus Lehnert and an anonymous reviewer for their valuable comments that greatly contributed to the improvement of the manuscript. Last but not least, we thank Katy Beaver for her drawing of the new Angiopteris species. References Awmack, C.S. (1997) The fern flora of Seychelles - report of visit to Seychelles, January–February 1997. IACR Rothamsted, Harpenden, Hertshire, UK, 16 pp. Badré, F. (2008) Marattiacées, pp. 59–62, in: Bosser, J., Badré, F. & Guého, J. (eds.), Flore des Mascareignes - La Réunion, Maurice, Rodrigues. Ptéridophytes. 1. Psilotacées à 26. Marsiléacées. Institut de Recherche pour le Développement, Paris; Mauritius Sugar Industry Research Institute, Ile Maurice; The Royal Botanic Gardens, Kew, 455 pp. Baker, J.G. (1877) Flora of Mauritius and the Seychelles - a description of the flowering plants and ferns of those islands. Reeve & Co., London, 557 pp. Bory de Saint-Vincent, J.-B. (1804) Voyage dans les quatre principales îles des mers d’Afrique vol. 1. F. Buisson, Paris, 430 pp. Chase, M.W. & Hills, H.H. (1991) Silica gel: an ideal material for field preservation of leaf samples for DNA studies. Taxon 40: 215–220. http://dx.doi.org/10.2307/1222975 Christ, H. (1909) Diagnoses plantarum Africae, plantes nouvelles de l’Afrique tropicale Française décrites d’après les collections de M. Auguste Chevalier. Journal de Botanique ser. 2, tom 2: 19–24. MARATTIACEAE OF THE SEYCHELLES Phytotaxa 158 (1) © 2014 Magnolia Press • 73 Christ, H. & Giesenhagen, K. (1899) Pteridographische Notizen. 1. Archangiopteris nov. gen. Marattiacearum. Flora 86(1): 72–79. Christenhusz, M.J.M. (2007) Evolutionary history and taxonomy of neotropical marattioid ferns: studies of an ancient lineage of plants. Annales Universitatis Turkuensis ser. AII 216: 1–134. Christenhusz, M.J.M. (2010a) Danaea (Marattiaceae) revisited: biodiversity, a new classification and ten new species of a neotropical fern genus. Botanical Journal of the Linnean Society 163: 360–385. http://dx.doi.org/10.1111/j.1095-8339.2010.01061.x Christenhusz, M.J.M. (2010b) Revision of the neotropical fern genus Eupodium (Marattiaceae). Kew Bulletin 65: 1–7. http://dx.doi.org/10.1007/s12225-010-9178-3 Christenhusz, M.J.M. & Chase, M.W. (2013) Biogeographical patterns of plants in the Neotropics – dispersal rather than plate tectonics is most explanatory. Botanical Journal of the Linnean Society 171: 277–286. http://dx.doi.org/10.1111/j.1095-8339.2012.01301.x Christenhusz, M.J.M. & Toivonen, T.K. (2008) Giants invading the tropics: the oriental vessel fern, Angiopteris evecta (Marattiaceae). Biological Invasions 10: 1215–1228. http://dx.doi.org/10.1007/s10530-007-9197-7 Christenhusz, M.J.M., Tuomisto, H., Metzgar, J. & Pryer, K.M. (2008) Evolutionary relationships within the Neotropical, eusporangiate fern genus Danaea (Marattiaceae). Molecular Phylogenetics and Evolution 46: 34–48. http://dx.doi.org/10.1016/j.ympev.2007.09.015 Christenhusz, M.J.M., Zhang X.-C. & Schneider, H. (2011) A linear sequence of extant families and genera of lycophytes and ferns. Phytotaxa 19: 7–54 Christensen, C. (1912) On the ferns of the Seychelles and the Aldabra group. Transactions of the Linnean Society of London, 2nd Series, Zoology 15: 407–422. http://dx.doi.org/10.1111/j.1096-3642.1912.tb00109.x Edgar, R.C. (2004) MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Research 32: 1792–1797. http://dx.doi.org/10.1093/nar/gkh340 Elzein, H. (2011) Caractérisation des types de forêts montagnardes sur l'île de Mahé (Seychelles). Master Thesis, Université Libre de Bruxelles, Service Eco-Ethologie Evolutive, Bruxelles, 93 pp. Ettingshausen, C. Ritter von (1864) Die Farnkräuter der Jetztwelt zur Untersuchung und Bestimmung, der in den Formationen der Erdrinde eingeschlossenen Überreste von vorweltichen Arten dieser Ordnung nach dem Flächen-skelett bearbeitet. Carl Gerold’s Sohn, Vienna, 298 pp. http://dx.doi.org/10.5962/bhl.title.69309 Exell, A.W. (1956) Supplement to the Catalogue of the Vascular Plants of S. Tomé, (with Principe and Annobon). British Museum (Natural History), Dept. of Botany, London. Felsenstein, J. (1985) Confidence limits on phylogenies: an approach using bootstrap. Evolution 39: 783–791. http://dx.doi.org/10.2307/2408678 Forster, G. (1786) Florulae Insularum Australium Prodromus. J.C. Dieterich, Göttingen, 103 pp. http://dx.doi.org/10.5962/bhl.title.10725 Gaudeul, M. & Rouhan, G. (2013) A plea for modern botanical collections to include DNA-friendly material. Trends in Plant Science 18: 184–185. http://dx.doi.org/10.1016/j.tplants.2012.12.006 He, Z. & Christenhusz, M.J.M. (2013) Marattiaceae, pp. 82–89, in: Wu, Z., Raven, P.H. & Hong, D. (eds.) Flora of China vol. 2–3, Lycopodiaceae through Polypodiaceae. Science Press, Beijing, Missouri Botanical Garden Press, St. Louis. Hieronymus, G. (1919) Bemerkungen zur Kenntnis des Gattung Angiopteris Hoffm., nebst Beschreibungen neuer Arten und Varietäten derselben. Hedwigia 61: 242–285. Hill, C.R. & Camus, J.M. (1986) Evolutionary cladistics of marattialean ferns. Bulletin of the British Museum (Natural History), Botany 14: 219–300. Hoffmann, G.F. (1796) Descriptiones et icones plantarum. Commentationes Societatis Regiae Scientiarum Gottingensis 12: 22– 37. Holttum, R.E. (1978) The morphology and taxonomy of Angiopteris (Marattiaceae) with description of a new species. Kew Bulletin 32: 587–594. http://dx.doi.org/10.2307/4109664 Huelsenbeck, J.P. & Ronquist, F. (2001) MrBayes: Bayesian inference of phylogeny. Biometrics 17: 754–755. http://dx.doi.org/10.1093/bioinformatics/17.8.754 IUCN Standards and Petitions Subcommittee (2011) Guidelines for Using the IUCN Red List Categories and Criteria. Version 9.0. Prepared by the Standards and Petitions Subcommittee, 87 pp. Lavalle, M. d. C. (2003) Taxonomía de las especies neotropicales de Marattia. Darwiniana 41: 61–86. Lehtonen, S., Tuomisto, H., Rouhan, G. & Christenhusz, M.J.M. (2010) Phylogenetics and classification of the pantropical fern family Lindsaeaceae. Botanical Journal of the Linnean Society 163: 305–359. http://dx.doi.org/10.1111/j.1095-8339.2010.01063.x Lellinger, D.B. (2002) A modern multilingual glossary for taxonomic pteridology. Pteridologia 3: 1–263. 74 • Phytotaxa 158 (1) © 2014 Magnolia Press SENTERRE ET AL. Maxon, W.R. (1905) A new name for Kaulfussia Blume, a genus of marattiaceous ferns. Proceedings of the Biological Society of Washington 18: 239–240. Murdock, A.G. (2008a) Phylogeny of marattioid ferns (Marattiaceae): inferring a root in the absence of a closely related outgroup. American Journal of Botany 95: 626–641. Murdock, A.G. (2008b) A taxonomic revision of the eusporangiate fern family Marattiaceae, with description of a new genus Ptisana. Taxon 57: 737–755. Nylander, J.A.A. (2004) MrModeltest, version 3. Distributed by the author, Uppsala. Pichi Sermolli, R.E.G. (1969) Adumbratio florae Aethiopicae 16. Marattiaceae. Webbia 23: 329–351. Philippe, H. (1993) MUST, a computer package of Management Utilities for Sequences and Trees. Nucleic Acids Research 21: 5264–5272. Rakotondrainibe, F., Badré, F. & Stefanović, S. (1996) Etude floristique et phytogéographique des ptéridophytes des îles du sud-ouest de l'Ocean Indien, pp. 109–120, in: Lourenço, W.R. (ed.) Biogéographie de Madagascar. ORSTOM, Paris. Rambaut, A. & Drummond, A.J. (2007) Tracer v1.5. http://beast.bio.ed.ac.uk/Tracer. Rolleri, C.H. (1993) Revision of genus Christensenia. American Fern Journal 83: 3–19. http://dx.doi.org/10.2307/1547357 Ronquist, F. & Huelsenbeck, J.P. (2003) MRBAYES 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 1572–1574. http://dx.doi.org/10.1093/bioinformatics/btg180 Roux, J.P. (2009) Synopsis of the Lycopodiophyta and Pteridophyta of Africa, Madagascar and neighbouring islands. Strelitzia 23: 1–296. Schrader, H.A. (1818) Analecta ad floram capensem. Sect. I. Filices, Lycopodineae. Göttingische gelehrte Anzeigen unter der Aufsicht der Königl. Gesellschaft der Wissenschaften 92: 913–920. Swartz, O. (1788) Nova genera & species plantarum seu Prodromus, descriptionum vegetabilium maximam partem incognitorum. M. Sweder, Stockholm, Uppsala, Turku. http://dx.doi.org/10.5962/bhl.title.4400 Senterre, B. (2011) On the forest types of the Seychelles. Kapisen 12: 5–7. Senterre, B., Gerlach, J., Mougal, J. & Matatiken, D.E. (2009) Old growth mature forest types and their floristic composition along the altitudinal gradient on Silhouette Island (Seychelles) - the telescoping effect on a continental mid-oceanic island. Phytocoenologia 39: 157–174. http://dx.doi.org/10.1127/0340-269x/2009/0039-0157 Senterre, B., Henriette, E., Chong-Seng, L., Gerlach, J., Mougal, J. & Rocamora, G. (2013) Seychelles Key Biodiversity Areas Output 6: Patterns of conservation value in the inner islands. Consultancy Report, Ministry of Environment-UNDP-GEF project, Victoria, Seychelles, 65 pp. Smith, J. (1841) An arrangement and definition of the genera of ferns, with observations on the affinities of each genus. Journal of Botany, being a second series of the Botanical Miscellany 4: 147–198. Smith, J.E. (1790) Plantarum icones hactenus ineditae, plerunque ad plantas in herbario Linnaeano conservatas delineatae, vol. 2. J. Davis, London, t. 26–50. http://dx.doi.org/10.5962/bhl.title.49832 Smith, J.E. (1793) Danaea. Mémoires de l'Academie Royale des Sciences 5(1790–1791): 420, t. 9, fig. 11. Shaw, J., Lickey, E.B., Beck, J.T., Farmer, S.B., Liu, W., Miller, J., Siripun, K.C., Winder, C.T., Schilling, E.E. & Small, R.L. (2005) The tortoise and the hare II: Relative utility of 21 noncoding chloroplast DNA sequences for phylogenetic analysis. American Journal of Botany 92: 142–166. http://dx.doi.org/10.3732/ajb.92.1.142 Summerhayes, V.S. (1931) An enumeration of the angiosperms of the Seychelles Archipelago. Transactions of the Linnean Society 19: 261–299. http://dx.doi.org/10.1111/j.1096-3642.1931.tb00129.x Swofford, D.L. (2002) PAUP*. Phylogenetic analysis using parsimony (*and other methods), version 4.0 Beta. Sinauer Associates, Sunderland. Tardieu-Blot, M.-L. (1951) Marattiacées, Ophioglossacées, Hymenophyllacées, Cyathéacées (1–4), pp. 1–7, in: Humbert, H. (ed.) Flore de Madagascar et des Comores (plantes vasculaires). Muséum National d’Histoire Naturelle, Paris. Tardieu-Blot, M.-L. (1960) Les fougères des Mascareignes et des Seychelles. Notulae Systematicae 16: 151–201. Tardieu-Blot, M.-L. (1964) Ptéridophytes. In: Aubréville, A. (ed.) Flore du Cameroun. Muséum National d'Histoire Naturelle, Laboratoire de Phanérogamie, Paris, 371 pp. Thiers, B. (2010) Index herbariorum: A global directory of public herbaria and associated staff. New York Botanical Garden. Available at http://sweetgum.nybg.org/ih/, accessed 1 September 2013. Torsvik, T.H., Amundsen, H., Hartz, E.H., Corfu, F., Kusznir, N., Gaina, C., Doubrovine, P.V., Steinberger, B., Ashwal, L.D. & Jamtveit, B. (2013) A Precambrian microcontinent in the Indian Ocean. Nature Geoscience 6: 223–227. Velayos, M., Aedo, C., Cabezas, F. & De la Estrella, M. (2008) Flora de Guinea Ecuatorial, vol. 1. Psilotaceae-Vittariaceae. Consejo Superior de Investigaciones Científicas. Madrid, 381 pp. Verdcourt, B. (1999) Flora of Tropical East Africa. Royal Botanic Gardens, Kew. Vriese, W.H. de & Harting, P. (1853) Monographie des Marattiacées. Arnz, Leiden & Düsseldorf. MARATTIACEAE OF THE SEYCHELLES Phytotaxa 158 (1) © 2014 Magnolia Press • 75

RELATED PAPERS

RELATED TOPICS

Log In

<p class="HeadingRunIn"><strong>Revision of the fern family Marattiaceae in the Seychelles with two new species and a discussion of the African <em>Ptisana fraxinea</em> complex.</strong></p>

<p class="HeadingRunIn"><strong>Revision of the fern family Marattiaceae in the Seychelles with two new species and a discussion of the African <em>Ptisana fraxinea</em> complex.</strong></p>

<p class="HeadingRunIn"><strong>Revision of the fern family Marattiaceae in the Seychelles with two new species and a discussion of the African <em>Ptisana fraxinea</em> complex.</strong></p>

<p class="HeadingRunIn"><strong>Revision of the fern family Marattiaceae in the Seychelles with two new species and a discussion of the African <em>Ptisana fraxinea</em> complex.</strong></p>

<p class="HeadingRunIn"><strong>Revision of the fern family Marattiaceae in the Seychelles with two new species and a discussion of the African <em>Ptisana fraxinea</em> complex.</strong></p>

Related Papers

RELATED PAPERS

RELATED TOPICS