SMITHSONIAN CONTRIBUTIONS TO BOTANY NUMBER 8 F. R. Fosberg Three Indo-Pacific Thelypteris Species and M.-H. Sachet Reinterpreted and a New African Species Described SMITHSONIAN INSTITUTION PRESS CITY OF WASHINGTON 1972 ABSTRACT Fosberg, F. R., and M.-H. Sachet. Three Indo-Pacific Thelypteris Species Reinter- preted and a New African Species Described. Smithsonian Contributions to Botany, number 8, 10 pages, 3 figures. 1972.-Two species of Thelypteris which have been confused and to which a number of names have been applied were found to be separable by a number of characters. They must now be called Thelypteris opulenta and T. wagneri. Their distribution is Indo-Pacific, extending to Africa, and T. opu- lenta introduced into tropical America. T. wagneri and an African species, T. fadenii, are here described as new, The name Thelypteris interrupta, heretofore applied to this complex, is shown to be the correct name for what has commonly been called Dryopteris gongylodes and Thelypteris totta. Library of Congress Cataloging in Publication Data Fosberg, Francis Raymond, 1908- Three Indo-Pacific Thelypteris species reinterpreted and a new African species described. BiblioaraDhv : D. (Smithsonian contributions to botany, no 8) -, . 1. ThGlypteris. I. Sachet, Marie-HClhe, joint author. 11. Title. 111. Series: Smithsonian Institution. Smithsonian contributions to botany, no. 8. QKlS2747 no. 8 [QK524.P7] 581?.08s [587?.31] 77-39682 Official publication date ic handstamped in a limited number of initial copies and is recorded in the Institution?s annual report, Smithsonian Year. For sale by the Superintendent of Documents, U.S. Government Printing Office Washington, D.C. 20402 - Price 26 cents (paper cover) F. R. Fosberg Three Indo-Pacific Th elyp teris S p ec ies and M.-H. Sachet Reinterpreted and a New African Species Described Among the common Indo-Pacific ferns are a pair of Thelypteris (Cyclosorus) species with lobed pin- nae, nearly marginal sori, a sterile zone on both sides of the midrib of the pinna, and with at least a few, usually many, yellow particles (glands?) on the lower surface of the blade, especially on the lobes. They may be found in herbaria under many names, some properly applied to ferns of this relationship, others not. The present attempt to unravel the spe- cific taxonomy and nomenclature of this group was stimulated by our realization that a plant of the Marquesas Islands, described by Dr. Elizabeth Brown as Dryopteris sulfurea, was apparently identical with what we had previously called Dryopteris extensa (Blume) Kuntze, or Dryopteris interrupta (Willd.) Ching, in Micronesia. Since the latter two species were originally de- scribed from Java and India, respectively, it seemed appropriate to examine past treatments to determine what plants were originally so named. This survey showed considerable variation in opinions as to whether one or several species are involved and as to the names to be applied to them. The criteria by which they were separated also varied considerably, from author to author. Most of the purported differ- ences are rather intangible-depth and shape of lobing, ranges of thickness and habit of rhizome, F. R. Fosberg, Department of Botany, Smithsonian Znstitu- tion, Washington, D. C. 20560. M.-11. Sachet, Department of Botany, Smithsonian Institution, Washington, D. C. 20560. abundance of yellow glands, and details of venation. An attempt to sort the available material from the western Pacific in the United States National Her- barium according to these criteria was not very successful, except to suggest a geographical separation of two entities. A further attempt on these lines, separating the Malesian from the Micronesian and Polynesian material, showed a reasonably consistent difference in the shapes of the leaf apices of the plants from these two regions. The Pacific specimens have the apical segment broad at base, pinnatifid or at least deeply lobed in the basal half, gradually attenuate distally into a caudate tip. Most of the specimens from the Malesian region have the apical segment essentially similar to the pinnae, linear- oblong, with one or two similar but reduced segments at its base, either separate from it or fused with it. These segments are abruptly much smaller than the next lower pinnae. It is surprising that such a clear-cut difference apparently has not been used before to distinguish these plants. Those with the pinnatifid tip include all material seen from Polynesia and Micronesia and a few specimens from the Philippines, Malacca, Pulu Penang, Java, Rawak, Thailand, Ceylon, and west- ward to East Africa, and those from tropical America (introduced) . They have most frequently been called Nephrodium extensum or Dryopteris extensa, but this epithet has also on occasion been applied to the other plant. The plant with the linear-oblong terminal segment 1 2 SMITHSONIAN CONTRIBUTIONS TO BOTANY is found from New Guinea, the Malesian and Philip- pine Islands, westward to India and Ceylon, Central Africa and Fernando Po Island. It has commonly been called Nephrodium or Dryopteris pteroides, or more recently Dryopteris interrupta or Cyclosorus interruptus. Whether we are dealing with two species or only with varieties of one is a difficult question. Most pteridologists, including Copeland and Holttum, have maintained the two as separate species, though not always on the same grounds. Wagner, in a manu- script on file in the U. S. National Herbarium Divi- sion of Ferns but apparently never published, re- garded them as varieties of one. They are big, unwieldy ferns and not always fairly represented by herbarium material. Unfortunately, the rhizome is lacking on most specimens, making it difficult to apply the criterion that one has a short, thick erect or ascending rhizome with congested fronds, whereas the other has a more slender, creeping rhizome with scattered fronds. Our present impression is that this is not a very reliable difference, but that it does rep- resent a real tendency. Since the difference in the apical segment seems reasonably clear, and since that is supported by a number of less easy-to-use characters, and since the extensive geographical ranges have only partial over- lap, we are for now regarding the two plants as separate species. Future studies on a more represent- ative range of material may possibly show this to be wrong. In any event, the two form a closely related group, defined by yellow glands, lobing only part way to the costae of the pinnae, and marginal or submar- ginal positions of the sori, which are usually lacking in the united portions of the pinnae. The margins tend to be a bit revolute around the sori, making the lobes appear narrower and the sori more marginal than they really are. The indusia are well developed. The spores are bilateral and very deeply rugose with short curved or sigmoid ridges. The group seems related to the plant that is currently called Thelypteris totta (Thunb.) Schelpe or T. goggilodus (Schkuhr) Iwat- suki, which is discussed under Thelypteris interrupta. This may be only a superficial resemblance, due to the submarginal sori. The group with yellow glands has been confused with a number of other species, notably T. pennigera (Forst. f.) Allan and its close relatives. The rest of this paper will attempt to sort out the synonymy and establish the correct names for the two species under discussion, and to discuss two other species that enter into the problem. Unfortunately, several of the synonyms are hard to place without authentic material. These are indicated by a query. All that are definitely included in the two synonymies, without a query, and not as ?sensu? particular authors, are plants with the above combination of features that characterize the group. The names that seem to have been based on plants of this affinity were all studied with reference to their original descriptions and types (when available), as well as to subsequent discussions of their typification, and in the light of the combinations of characters of the two species as indicated above. It was fairly easy to assign most of the synonyms to one or other of the two species, thus compiling the synonymies listed FIGURE 1 .-Thelypteris opulenta, portion of frond showing terminal segment and lobing of pinnae. NUMBER 8 3 FIGURES 2 AND 3.-Thelypteris wagneri, portions of fronds showing terminal segment and lobing of pinnae. under the species. Names that have been misapplied to these species are indicated as ?sensu? the misapply- ing author, ?non? the correct author of the name. Herbaria where the specimens cited were examined are indicated by the standard Index Herbariorum abbreviations except those still in our possession and not yet deposited, which are marked (Fo) . Thelypteris opulenta (Kaulf.) Fosb., new combination Aspidium opulentum Kaulf., Enum. Fil. 238, 1824. Type: Polystichum pennigerum sensu Gaud., Bot. Voy. Uran. 328, Nephrodium impressum Desv., Prodr. 259, 1827. Type: Aspidium extensum Blume, Enum. PI. Jav. 156, 1828. Guam, Chamisso. 1826, non Polypodium pennigerum Forst. f. 1786. Timor, Hb. Desv. (P) . Type: Pulu Pinang, Blume. Aspidium multijugum Wall., Cat. 348, 1829, nomen nudum. Aspidium prionophyllum Wall., Cat. 355, 1829, nomen Aspidium terminans Wall., Cat. 386, 1829, nomen nudum. Nephrodium cumingii J. Smith in Hook., J. Bot. 3:411, Nephrodium terminans J. Smith, Bot. Mag. 72 (Comp.) : 32, Lastraea malaccensis Presl, Epim. Bot. 35, 1849 [ 185 11. Nephrodium conioneuron Fee, Gen. Fil. 308, 1852? Nephrodium extensum (Blume) Moore, Ind. Fil. 91, 1858. Aspidium pteroides vars. a?, b (pro parte), Mett., Ann. Mus. Lugd. Bat. 1:231, 1864. Aspidium pteroides sensu Luerssen, J. Mus. Godeffr. 1 :55, 1873, non Polypodium pteroides Retz. 1791. Nephrodium parasiticum var. multijugum C. B. Clarke, Trans. Linn. SOC. Bot. 1:533, 1880. Type: Penang, Wallich 348. Dryopterir extensa (Illume) Kuntze, Rev. Gen. P1. 2:812, 1891. nudum. 1841, nomen nudum. 1846, nomen nudum. Type: Malacca, Cuming 391. 4 SMITHSONIAN CONTRIBUTIONS TO BOTANY Nephrodium wakefieldii Baker, Ann. Bot. 5:326, 1891. Type: Mombasa, Kenya, Wakefield. Dryopteris sulfurea E. Brown, Bish. Mus. Bull. 89:23, 1931. Type: Marquesas, Brown 293. Phegopteris pennigera sensu Hosokawa, Trans. Nat. Hist. SOC. Formosa 26:233, 1936, non Polypodium penni- gerum Forst. f., 1786. Dryopteris impressa (Dew.) Posth., Verh. K. Akad. Wet- tensch. Amst. 36(5) : 14, 1937. Cyclosorus extensus (Blume) Ching, Bull. Fan Mem. Inst. Biol. Bot. 8:182, 1938. Cyclosorus interruptus sensu Wagner and Grether, Occ. Pap. Bish. Mus. 19:50, 1948, non Pteris interrupta Willd., 1794. Thelypteris extensa (Blume) Morton, Amer. Fern J. 49: 113, 1959. Thelypteris interrupta sensu Stone, Micronesica 2 : 3, 1966, non Pteris interrupta Willd., 1794. Thelypteris sulfurea (Brown) Reed, Phytologia 17 : 3 18, 1938. Rhizome tending to be rather thick and short; fronds broad, .oblong, pinnate; pinnae deeply lobed usually one-half to three-fourths the way to the mid- rib, the lobes tapering somewhat, the apex of fronds deeply pinnately lobed or pinnatifid in lower half strongly tapering to a caudate tip; sori submarginal around the lobes, appearing marginal because of rev- olute margins of lobes; yellow glands on lower surface of pinnae especially on veins and near tips of lobes. SPECIMENS EXAMINED.-~NSULAE OCEANI PACIFICI : For- ster (BM) (a wilted specimen but probably this). TUAMOTU Is.: Bow Island [Hao Atoll], Barclay 165(?) MARQUESAS Is.: Dupetit-Thouars in 1843 (P, 2 sheets). Uapou I.: alt. 200 m, Jones 1105 (BISH, BM); Haka- hetau valley above village, 25-250 m, common name: ?Autapa?amei,?? Decker 2283 (Fo, US), Uahuka I.: Han- nay, 150 m, Jones 2662 (BISH); Jones 2665 (BISH); Upper Ha?avei valley, 200-250 m, Decker 2469 (Fo); in- tenor Vaipae?e Valley, 200 m, Decker 1537 (Fo), 2549 (Fo, US) ; Upper Vaipae?e, 170 m, Decker 1982 (Fo, US) ; west side of Vaipae?e, 100 m, Decker 2793 (Fo, US). Hiva?oa I.: 300 m?, Brown 793 (BISH, type of Dryopteris sulfurea); central Puamau, 85 m, Decker 802 (Fo, US); Puamau village, 25 m, Decker 883 (Fo, US); SW Puamau, 200 m, Decker 2012 (Fo); central Puamau valley, 2 km inland, 225 m, Decker 1080 (Fo) ; E. Puamau valley, 175 m, Decker 2233 (Fo) ; Puamau, along Puamau-Atuona trail, 500-650 m, Decker 1200 (Fo). Nukuhiva I.: Hakaui, Brown 790 (BISH) ; Taiohae, Brown 31 (BISH, 2 sheets); Taiohae-Hatiheu trail, 450 m, Decker 21 70 (Fo, US). Fatu- hiva I.: Omoa Valley, 500 m, Brown 987 (BISH); Hana- vave, 100 m, Jones 2808 (BISH); s.l., Cheesman in 1925 (BM). Motane I.: 100 m, Jones 2725 (US, BISH). Eiau I.: 700 m, Jones 1525 (US, BISH). Tahuata I.: Vaitahu, 150 m, ]ones 2745 (BISH). (BM). SOCIETY IS.: Meetia I.: Fatia-Po to Fareura, 250 m, St. John 24289 (BISH). AUSTRAL Is.: Rimatara I.: Stream near Mutuaura, 20 m, Fosberg 22053 (US, BISH). Rurutu.: Moerai, 10 m, St. John 26636 (BISH); Puputa Valley, above Moerai, 25 m, Fosberg 12937 (BISH). SAMOAN Is.: Tau: 1 mi. S Siufaga, 60 m, Yuncker 9020 (BISH) . BONIN Is.: Nakata-ioto, Tuyama (reported by Ito, 1937: 714). MARIANAS Is.: Gaudichaud (P, 2 sheets) ; Gaudichaud in 1824 (P). Agrigan I.: canyon on E side, Fosberg 31633 (US, BISH, Fo, MICH). Alamagan I.: s.1. Falanruw 1923 (US) ; 60 m, Anderson 412 (US, BISH, Fo, MICH, IJ, NY). Saipan I.: s.1. Kanehira 2229 (NY); Hosokawa 6659 (A) ; Raishingan, Hosokawa 8032 (BISH, A). Rota I.: Savana, Hosokawa 7593 (BISH, A), 7646 (A); Necker R 88 (US); 1300 ft, Kondo in 1952 (BISH); 800 ft, Grether 4470 (US) ; slopes above As Malote, S side of island, 250 m, Fosberg 32875 (US, BISH, Fo, MICH) ; NW coast of W end of main part of island, between Rota and Tataacho Point, Fosberg 25137 (US, BISH, Fo, MICH, IJ). Guam 1.: Mertens (BM); Wagner 3292 (US); Connover 534 (US); Herb. Richard in 1824 (P); Marche 167 (P), 47 (P), 73 (P) ; Haputo Point, Johnson and Necker 2 (US) ; Mt. Santa Rosa, Glassman 272 (Fo) ; Hosaka 3251 (US, BISH, Fo) ; 1 mi. S of Taguac, on north plateau, Fosberg 35479 (US, BISH, Fo, MICH, IJ, NY) ; Agaiia, Safford and Seale 2086 (US), 2088 (US), 1089 (US) ; Agania Swamp, E of Agniia, 1 m, Fosberg 31226 (US, BISH, Fo, MICH, IJ, NY); 5 m, Bryan 2084 (US, NY, BISH) ; Agania Marshes, Grether 3823 (US, BISH) ; lower Fonte River, 15 m, Fosberg and Evans 46297 (US, BISH) ; 2 km S of Piti, Necker 201 (US) ; shores of Piti Harbor, Bartsch 202 (US), 110 (US); Yigo, 180 m, Bryan 1239 (BISH) ; Lonfit River, Grether 3720-A (US) ; Sigua River Valley, Grether 3728 (US, BISH) ; near junction of Lonfit and Sigua Rivers, Grether 3817 (US, BISH) ; W of Mapao, 3 mi WNW of Talofofo Bay, 100 m, Fosberg 35455 (US, BISH, Fo, MICH, IJ); Talofofo, Mt. Almzgosa, Hosaka 3184 (US, BISH, Fo); Umata, Marche 243 (P). CAROLINE Is.: Sonsorol 1.: Berry 2 (US). Palau Is.: E end Koror I., Fosberg 50632 (US, BISH); Tor Uil I., be- tween Koror and Malakal I., 15 m, Fosberg 47598 (US, BISH) ; Aulupse?el I., 10-15 m, Fosberg 47556 (US, BISH, Fo, MICH, IJ, NY, L); E end of Urukthapel I., road to lighthouse, 1-5 m, Fosberg 32036 (US, BISH, Fo, MICH). NEW GUINEA : Kaiser-Wilhelmsland, Walder von Woble, 180 m, Schlechter 26349 (BISH). Rawak I.: Gaudichaud in 1819 (P); in 1845 (?) (P); Gaudichaud (P). TIMOR I.: Herb. Desvaux (P, prob. type of Nephrodium impressum). PHILIPPINES: Luzon I.: Prov. Laguna, Los Banos,Baiios, Mount Maquiling, Elmer 27905 (BISH) ; Prov. Ilocos Norte, Ramos 7772 (BISH); Prov. Ilocos, Burgos, Ramos 27212 (US). Balabac I.: Mangubat 415 (NY, US). Min- danao I.: Cuming 293 (BM); District of Zamboanga, Santa Maria, Reillo 26502 (US). MALAY PENINSULA: Ayer Hitam: Matthew in 1908 (K, NUMBER 8 5 2 sheetsj; G. Pardok, Matthew in 1911 (KO). Negri- Sembilan: Lukut Fort, sea level, Hose 272 (K). Malacca: Cuming 391 (K, 3 sheets, one young, doubtful, US, P, isotypes of Lastrea malaccensis Presl, photo of Paris sheet in US, Morton Neg. No. 17450) [cited by J. Smith (1841 : 411) as his Nephrodium cumingii]. Penang: Wallich 1829 (K), 348 (K, P, photos in US, basis of Aspidium multi- jugum Wall., Morton Neg. Nos. 17451, 17452), 355 (P, photos in US, basis of Aspidium prionophyllum Wall., Morton Neg. Nos. 17453, 17454) ; Curtis 3367 (K). Pulu Penang: Blume (photo in US of specimen in L, type of Aspidium extensum Blume) . Kedah: Pulau Langkawi, Holt- turn 17414 (US); West Coast, Pulau Dayang Bunting (Pulau Langkawi) Henderson 29132 (US). Johore: Ves- terdal 295 (US). SINGAPORE I.: Blakang Mati I.: Sea level, Molesworth- Allen 2523 (US). JAVA: Goenung Parang, Blume (BM). THAILAND (SIAM): Murok Lek, Marcan 806 (BM); Bandon, Surat, near sea level, Smith 3004 (BM), 3007 (BM, US); Setun Puket P. Siam, Smith 2469 (BM, US), 2459 (US). CEYLON: G. Wall in 1884 (PDA); Thwaites (BM); ?Gandola,? Herb. 1. Smith 2206 (BM) ; Kandy, Chevalier [?] (BM), [Thwaites] C. P. 975 (PDA); s.1. ex Herb. William Ferguson 238 (US, 2 sheets). CHAGOS Is.: Diego Garcia: Minni Minni, Stoddart 832 AFRICA: Kenya: Mombasa, Wakefield (K, type of Nephrodium wakefieldii, 3 sheets) ; Tanzania (as Tangan- yika Terr.) : Pangani, Tanga Prov., Tanner 2836 (K) ; Ulugurus, Kimbosa, 1000 ft, Bruce 133 (US, BM). Pemba I.: Vaughan 456 (K). Mozambique: Manica e Sofala: Cheringoma, Inhaminga, Mendoca 4377 (BM) ; Zambezia, Mendoca 1448 (BM). COSTA RICA: Prov. Puntarenas: 6 km S of airstrip, Osa Peninsula, 500 ft, Mickel 2770 (ISC, LP, US). MARTINIQUE : Stehle 4986 (US). BRITISH GUIANA: Maguire and Fanshawe 22983 (US). COLOMBIA: Putumayo, 15 km NW of Puerto Asis, 1000 ECUADOR : Santiago-Zamora, Taisha, 1500 ft, Caralet (K). ft. King & Gueuara 6222 (US). & Pennington 7752 (US). We have not been able to examine Chamisso?s specimen, described by Kaulfuss as Aspidium opu- lentum, but the description is very clear and fits the species under discussion very well. This is the only species in Guam that corresponds to the Kaulfuss de- scription, and there seems no reasonable doubt that the epithet opulenta is correctly applied here. The type of Nephrodium impressum Desv., from Timor, is probably a specimen from the Desvaux herbarium in Paris which is this species. Blume?s description of Aspidium extensum gen- erally fits; a specimen in Leiden, named in Blume?s hand, photographed by C. V. Morton (Neg. No. 1128), and considered to be the holotype by Morton, has the caudate pinnatifid apex of T. opulentum. Nephrodium conioneuron Fee has usually been referred to Dryopteris pteroides (Retz.) 0. Kuntze or D. interrupta (sensu auct., non Polypodium inter- ruptum Willd.) . The type has not been located by us. A comparison of the description of this and Nephro- dium oreopteris Fee, published together, shows that N. conioneuron should have more deeply lobed pin- nae than N. oreopteris. Isotype material of the latter represents one of the few intermediates between T. opulenta and T. wagneri. Its pinnae are more deeply lobed than those of T. wagneri. Since the pinnae of N. conioneuron are described as still more deeply lobed, it seems safe to refer N. conioneuron tenta- tively here. Aspidium pteroides var. a Mett. was based on Nephrodium conioneuron Fee, and var. b Mett. was based on three synonyms, two of which, Aspidium opulentum Kaulf. and A. extensum Blume, belong here. No descriptions were provided for Aspidium termi- nuns, Nephrodium terminans, and Nephrodium cu- mingii. One of the two specimens cited with the latter, Cuming 391 from Malacca, is T. opulenta. The other, Cuming 293, from Mindanao, is probably T. wagneri, but the US specimen of this number is frag- mentary and the determination doubtful. We have not seen the Wallich specimen of Aspidium termi- nuns. Smith merely cites A. terminans Wall. and N. cumingii J. Smith as the protologue of N. terminans, hence it is also a nomen nudum, unless provided with a description in some publication missed by the in- dexers. Ching refers Nephrodium parasiticum var. multi- jugum C. B. Clarke, Wall. Cat. 348, from Penang, to Dryopteris extensa (= opulenta), and our exami- nation of Wallich?s material confirms this. The type of Nephrodium wakefieldii, from East Africa, certainly belongs here, but the material seen from there is very luxuriant and the pinnae are very deeply lobed. N. wakefieldii was reduced by Schelpe ( 1965 : 266-267) to Thelypteris extensa (Blume) Morton, which belongs here. We have seen ample material of Dryopteris sul- furea from the Marquesas, including that studied by Dr. Brown, and find no distinction from T. opulenta. 6 SMITHSONIAN CONTRIBUTIONS TO BOTANY References of Guam material of this species to de- rivatives of Polypodium pennigerum Forst. f., of P. pteroides Retz., and of Pteris interrupta Willd. are simply misidentifications. The two latter are discussed elsewhere in this. article. Polypodium pennigerum is a Thelypteris that superficially looks a bit like T. opulenta, but lacks yellow glands, has the sori differ- ently disposed, and does not actually seem very closely related. Thelypteris opulenta var. hirsuta (Ching) Fosb., new combination Cyclosorus interruptus var. hirsutus Ching, Bull. Fan Mem. Inst. Biol. 8: 186, 1938. Lectotype: Thailand, Rock 1716. This differs from var. opulenta in the short pale pu- bescence of the underside of the frond. This pubes- cence is most marked in the Rock 1716 specimen cited, here designated as lectotype. The Smith 473 specimen is intermediate between T. opulenta and T. wagneri (see discussion below under T. wagneri) . Thailand, Prov. Chiengmai, Chieng Rai, lower slopes of Doi Chang Mt. near Hue San (420-1, 260 m, Rock 1716 (US, lectotype, annotated by Ching). Thelypteris wagneri Fosberg and Sachet, new species Polypodium pteroidcs Retz., Obs. Bot. 6:39, 1791, type: Ceylon, Konig; non Thelypteris pteroidea (Klotzsch) Tryon, Rhodora 69:8, 1967, which was based on Poly- podium ptcroidcum Klotzsch, 1847 ; non P. ptcroides Lam., 1778. Aspidium ptcroidcs (Retz.) Swartz, Schrad. J. Bot. 1800 (2) :33, 1802. Ncphrodium orcopteris Fee, Gen. Fil. 304, 1852, type: Philippines, Cuming 48; non Thelypteris orcopteris (Ehrh.) Rydb., 1917. Aspidium pteroidcs var. b Mett., Ann. Mus. Lugd. Bat. 1:231, 1864, pro parte. Dryoptcris ptcroidcs (Retz.) Kuntze, Rev. en. P1. 2:813, 1891. TYPE.-Java, Raciborski in 1897. Rhizoma repens, glabrum; frons oblonga, ad api- cem abrupte contracta, segment0 terminali pinnam simulante, pinnis lobatis lobis brevibus vix ad dimi- dium costae fissis, infra flavo-granulatis, soris submar- ginalibus. Rhizome slender, 3-4 mm thick, creeping, shortly but rather densely hirtellous, covered by brown linear- lanceolate caducous scales ; fronds widely scattered, long-stipitate, up to 1 m or more long, the stipes dark brown, minutely retrorse-hirtellous, clothed on basal several cm with brown linear-lanceolate scales, the blades pinnate, oblong, up to 50 cm long, abruptly contracted toward apex, the rachis lighter brown than stipe, hirtellous, the pinnae lance-linear to linear ob- long, tapering in distal third to an acuminate tip, 13-28 cm long, 1-2.5 cm wide, the width and length varying independently, well separated, alternate, the lower sometimes subopposite or even opposite, the lowermost somewhat to rather strongly reduced on some specimens, these apparently early fronds in the dcvelopment of the plant (Smith 190, 299, Moles- worth-Allen 2844, Merrill 594, Williams 2218, Otanes, 18728, Topping 659), pinna lobed to about half or less the distance to the midrib, lobes obliquely rounded or obtuse to apparently acute but this be- cause of the margins curving round sori, lobes ap- pearing slightly falcate, pinna sparsely pilose above and beneath, especially on the veins, and densely so on upper side of costa, underside with yellow bulbous or globose particles or ?glands? scattered on veins and especially near tips of lobes, 3 veins from each lobe uniting in a vein running to the sinus between 2 lobes, sori submarginal but appearing marginal be- cause of the somewhat revolute margins, sometimes 1-3 pairs below sinuses but a broad sorus-free band along pinna-midrib, this rarely with a few scattered sori, distal two or three pinnae somewhat reduced, terminal one essentially identical with lateral ones, little dilated basally; indusium orbicular, somewhat hirtellous; annulus with 15 or 16 cells; spores bilat- eral, very deeply rugose with short curved or sigmoid ridges. DISTRIBUTION.-New Guinea to Ceylon, west- central Africa (Oubangi), and Fernando Po Island in the Gulf of Guinea; most abundant in the Philip- pines, Malesia, and Indo-China. SPECIMENS EXAMINED.-NEW GUINEA : Papua: Central Distr., near Douramoko Village, Vanapa River, 50 ft, Shoddc 2637 (US). HAINAN: Wang 33474 (US; very young wilted specimen, determination doubtful). PHILIPPINES : Cuming 48 (US, isotype of Ncphrodium orcopteris Fee?) . Luzon I. : Rizal Prov. : Montalban, Lohcr 944 (US), Ramos 5203 (US); Antipolo, Ramos and Edoiio 29542 (US). Tayabas Prov.: Edaiio 26947 (US); NUMBER 8 7 Atimorian, Topping 1310 (US), 1315 (US). Pangasinan Prov. : Umingan, Otanes 18278 (US). Laguna Prov. : McCregor 2311 I (US) ; Los Baiios, Topping 659 (US), 680 (US); Mount Maquiling, Elmer 17905 (US). Ilocos Norte Prov.: Ramos 7756 (US). Mindoro I.: Puerto Galera, Ramos 46426; Calapan, Merrill 984 (US) (in most respects this species, but the terminal segment slightly broader and more lobed at base than usual, thus tending in this character toward T. opulenta). Balabac I.: Mangubat 420 (US). Culian I.: Merrill 484 (US), 594 (US). Min- danao I.: Gulf of Leyte, Cuming 293 (US, doubtful, speci- men fragmentary) ; Davao Distr. : Copeland 636 (US) : Todaya, Mount Apo, Elmer 11, 111 (US, BISH). Zam- boanga Prov.: Binalanay, Copeland in 1917 (US); San Ramon: Copeland 1604 (US), Williams 2436 (US); Sax River, Williams 2218 (US). Palawan I.: San Antonio Bay, Merrill 5254 (US). Sulu Is.: 8 mi SE of Cagayan, Santos 4677 (US). MALAY PENINSULA: Kedah: Alorstan (Alor Sta.) , Rid- ley 14764 (K); West Coast, Dayang Bunting, Henderson 29144 (K). Malacca: Crifith (K). Perlis: Kalu Bukit, Kiah 35250 (K); Bukit Ketri, Matthew in 1914 (K). Pahang: Ayer Besar, P. Tioman, Henderson 18767 (K, US). Penang: Molesworth-Allen 2844 (US). SINOAPORE I. : Lobb 24 (K). INDONESIA: Borneo I: Banguey Is: Boden Kloss 19234 (US, BM). Java 1.: Prov. Tegal: Between Pogal and Gerbo, 800-900 m, Mousset 15 (US, BISH); Tengger Moun- tains, Pogal-Gerbo, Moussct [Ros. Fil. Exs. 831 (US); Lebaksioe, Raciborski in 1897 (US, 2 sheets, type and iso- type), in 1898 (US). Prov. Batavia: Buysman 124 (US); Dessa Dramaga near Buitenzorg, Schifner in 1893 (BISH). Celebes I.: Near Menado, Nielsen 754 (US, doubtful, very fragmentary). INDO-CHINA: Pierre (US, 4 sheets) ; (poor material, de- termination doubtful). CAMBODIA: Angkor, H. M. Smith 299 (US). THAILAND: Singora: H. M. Smith 190 (US). Chanta- bun: Vesterdal 477 (fragment, US). Gulf of Siam: Koh Tao Island, H. M. Smith 263 (US). Phitsanolok: Tung Salaeng Luang National Park, 500 m, Tugawa 11795 (US). CEYLON: Ferguson 137 (US). AFRICA : Republique Centre-Africaine: Oubangui-Chari, region de M?Baiki et Boukoko, Tisserant 1170 (P). Fer- nando Po I., Vogel (K) [possibly introduced]. Names based on Polypodium pteroides Retz. ( 1791) have usually, until recently, been applied to this species, and correctly. Then Ching ( 1933 : 566- 568) confused the application of the epithet pteroides of Retzius by saying he had examined the type of Polypodium pteroides Retz., a specimen from Madras collected by Konig, which he says ?really represents a species of the group of Aspidium sagenioides Mett. (Dryopteris sagenioides 0. Ktze.) , now considered as a free-veined tectarid fern, nearest to, or perhaps identical with Aspidium fuscipes Wall. or Nephro- dium membranifolium Presl, both common ferns in Tropical Asia.? This, however, ignores the statement in Retzius? description that the sori are marginal, and the fact that Retzius cited a Konig collection from Ceylon, not the one from Madras. Therefore, except for the possibility of regarding Polypodium pteroides as a later homonym of P. pterioides Lam. ( 1778) , the epithet pteroides would have been available until 1967, when Rolla Tryon transferred P. pteroideum Klotzsch to Thelypteris. Perhaps it would still be the correct epithet if the plant were regarded as belong- ing to another genus. However, it is unclear from Art. 73 of the Code whether Polypodium pterioides Lam. and P. pteroides Retz. (and P. pteroideum Kl.) are or are not homonyms. If they had not been described in the same genus it might be possible to regard the epithets as different. However, in one genus they will certainly be confusing, and we choose to regard the three epithets as orthographic variants or at least as too close to be maintained under Art. 75. If the Cuming 48 duplicate in the U.S. National Herbarium may be taken as an isotype of Neplzro- dium oreopteris Fee this name can be referred here with some hesitation. It seems rather intermediate between T. ulagneri and T. opulenta, having the abrupt terminal contraction of the frond of T. wag- neri, the pinnae more deeply cut but rather nearer T. wagneri in the shape of the lobes, and the terminal segment somewhat more deeply lobed at base than is usual for this species. In any event, the use of this epithet in Thelypteris is precluded by the combina- tion T. oreopteris (Ehrh.) Slosson ex Rydb. (1917). Mettenius based his Aspidium pteroides var. a on A. conioneuron Fee, his var. b partly on P. pteroides Retz. and partly on plants here referred to T. opu- lenta, and var. c on A. obtusatum Swartz, so it is just as well he did not name the varieties. Since the names applied to this species are either unavailable or incorrectly applied to it, it seems nec- essary to propose a new name. We take great pleasure in dedicating it to Prof. W. H. Wagner, Jr., outstand- ing and profound student of ferns and teacher of botany. To avoid confusion, we have proposed it as a new species, rather than as a renaming of one of the old ones, thus attaching it to an extant type of which we are certain. The boundaries between T. wagneri and T. opu- lenta are unclear in Thailand. The rhizomes of the plants cited as T. opulenta in Thailand tend to be 8 SMITHSONIAN CONTRIBUTIONS TO BOTANY more slender and creeping than is usual in that spe- cies. The specimens cited below are variously inter- mediate in the shape of lobes and depth of lobing of the pinnae, the abruptness of contraction of the ter- minal portion of the frond, and the shape and degree of cutting of the terminal segment. THAILAND: Wang-Djao, near Koh-Yai, Hosseus 89 (BM); Raheng, Nakawn Sawan, 150 m, E. Smith 3018 (US) ; Tap Sai, Chantabun, 200 m, Kerr 9604 (US) ; Prov. Udon, H. M. Smith 424 (US); Gulf of Siam, Koh Tao Island, H. M. Smith 256 (US). Khun Tai Mts., H. M. Smith 473 (US; annotated by R. C. Ching as Cyclosorus interruptus var. hirsutus) ; Sriracha, near sea level, E. Smith 3011 (BM); Sriracha, Nong Kaw, Marcan 1218 Thelypteris interrupta ( Willd.) Iwatsuki Thelypteris interrupta (Willd.) Iwatsuki, Jap. J. Bot. 38: 314, 1963. Pteris interrupta Willd., Phytogr. 13, Tab. X, 1794. Type: India, Klein. Polypodium tottum Thunb., Prodr. F1. Cap. 172, 1800. Type: Cape, Thunberg. Aspidium obtusatum Swartz, Schrad. J. 1800(2) :33, 1802. Type: no specimen is mentioned, but probably the pinna in BM thus labeled by Swartz and discussed below is a fragment of what he described. Aspidium goggilodus Schkuhr, 24 K1. Krypt. Gew. 1:193, t. 33c, 1809 [spelled goggylodus on plate, usually cor- rected to gongylodes]. Type : Essequebo, from Gartner. Aspidium resiniferum Kaulf., Enum. Fil. 237, 1824. Type OWahu [Hawaiian Is.], Chamisso. Cyclosorus gongylodes (Schkuhr) Link, Hort. Berol. 2 : 128, 1833. Dryopteris gongylodes (Schkuhr) 0. Ktze, Rev. Gen. P1. 2:811, 1891. Thelypteris gongylodes (Schkuhr) Small, Ferns SE States 248, 475, 1938. Thelypteris totta (Thunb.) Schelpe, J. So. Afr. Bot. 29:91, 1963. Names based on Pteris interrupta Willd. have com- monly been applied to one or the other of the two species described above, distinguished most readily from other similar Thelypteris (Cyclosorus) species by the tiny yellow particles or globules (glands?) on the lower sides of the lobes of the pinnae. However, while trying to separate and determine the correct names for these two plants, our suspicions were aroused by the similarity of Willdenow?s plate to the common pantropical marsh fern usually called Dryop- teris (or Cyclosorus) gongylodes (Schkuhr) Kuntze or more recently Thelypteris totta (Thunb.) Schelpe. This possibility was dismissed as unlikely, since no one had ever suggested it in print, so far as we could de- termine, and the epithet interrupta was in common use for the other plants mentioned above. While looking atplants of this affinity in the British Museum in 1968, we encountered a sheet on which were mounted three separate pinnae from the Swartz herbarium. One of these is labeled, in Swartz? hand ?pinna ex Pteridis interrupta Willd. (ab ipso!)- mihi aspid. obtusatum.? Another is labeled, also by Swartz, ?Aspidium obtusatum Sw. prodr.? Both of these are labeled in pencil, probably by Gepp, ?D. gongylodes var. proliferum.? Very careful examina- tion shows no globular yellow particles, the venation of the lobes is of the peculiar ascending type and the undersurface has the sessile glistening resinous glands characteristic of what has been called Dryopteris gon- gylodes. These two pinnae belong to the hirsute form that is common over most of the range of the species. It seems clear that the pinna from Willdenow must be a fragment of the specimen on which he based his Pteris interrupta, and which he sent to show Swartz what this species was. This specimen was, presumably, one collected by Klein in India. Dr. D. E. Meyer has very kindly sent a photograph of the Klein specimen in the Willdenow Herbarium in Berlin. This photo shows a plant with exactly the appearance of ?Dryopteris gongylodes? except that the large pale scales sometimes found on the rachis do not show, unless some rather shadowy grayish blotches are the scales. The sheet bears an annotation, possibly in Willdenow?s hand ?Pteris sp. n.? Aug. 1794. Auf den Reise nach Madras. (Klein)? and another ?Klein Ind. 1797 W,? as well as annotations by Ching and Hieronymus. This sheet and another without a label but probably a duplicate are both numbered 19770 in the Willdenow Herbarium. They both bear the annotation in the upper right corner ?Asp. obtusatum? ?la? and ?lb,? presumably added by Willdenow when informed by Swartz of their iden- tity with his species in the genus Aspidium. Interest- ingly enough, the sheet marked ?la? has one pinna missing, probably that sent to Swartz. Dr. Meyer also kindly sent a photo of the type sheet of Aspidium goggylodus Schkuhr, from the Schkuhr Herbarium. This consists of the drawing reproduced as t. 33c of Schkuhr (1809) and a pocket containing a single pinna, with the note ?Durch Gartn. aus Essequebo. Eine pinne habe an Mohr und Sw. ge- NUMBER 8 9 sandt.? Although Dr. Meyer says that ?Aspidium goggylodus Schkuhr und Willdenow No. 19770 sind nicht identisch,? they appear to belong to the same species. Schkuhr apparently later regarded his A. goggilodus as the same as A. obtusatum Swartz, since on a small ticket attached to the sheet bearing the drawing for his Tab. 33b he has crossed out ?gon- gylod? and written ?obtusat.? Why this is on the sheet of this plate, which is his Aspidium serra is not clear. In any event, there seems little doubt that the original material of Pteris interrupta Willd., Aspi- dium obtusatum Swartz, and Aspidium goggilodus Schkuhr belong to the same species, the common pan- tropical marsh fern. Willdenow?s epithet was earlier and must replace the several later ones that have been applied to it, e.g., totta, obtusata, and goggilodus (goggylodus, gongylodes). The necessary combination was made by Iwatsuki and later by Stone (1967: 135), both under the impression that Willdenow?s Pteris interrupta was the widespread species now to be known as Thelypteris opulenta. For a pantropical species, Thelypteris interrupta shows remarkably little variability over most of its range. However, East African specimens exhibit con- siderable variation in depth of lobing of pinnae, scali- ness, and the hairiness of the underside of the fronds. Material from Mafia Island is unusually hirsute; glabrous plants are found in Tanganyika, Zanzibar, and Pemba, as well as in South Africa, New Cale- donia, and Tahiti. Other African Species with Yellow Glands The presence of yellow particles or glands on the underside of the fronds is apparently a rather un- common feature in Thelypteris, so much so that their presence may cause people to refer quite unrelated plants to the species complex discussed above. Thelyp- teris microbasis (Baker) Tard. has abundant yellow glands, especially on the indusia. It is a plant with much more delicately cut fronds and does not seem especially related to any of the species treated in this paper. The only specimen we have examined is from Liberia, in the vicinity of Mount Coffee, Cook 22 Specimens from Uganda, at first referred to Cyclo- sows quadrangularis, were, when found to have such yellow glands, tentatively placed by the collectors in (US) ? Cyclosorus extensus. The glands on this material, however, are not identical with those of the T. opu- lenta-T. wagneri complex. Mr. Robert Faden and Dr. K. E. Lye, of the East African Herbarium and Makerere University, respectively, have kindly placed the available material of this species at our disposal for study. This has enabled us to distinguish it clearly from the species discussed above and, hopefully, to indicate its relationship. Since it does not seem to have been described it is proposed as a new species. Thelypteris fadenii Fosberg and Sachet, new species TYPE.-Uganda, Lye 51 86. Planta terrestris, rhizomatibus repentibus, frondi- bus hirsutis ovatis ad basim non contractis, pinnis lobatis, loborum costulis infra glandulis clavatis vel fusiformibus aureis instructis, indusiis valde hirsutis. Terrestrial plant with a creeping rhizome about 3 mm thick, this shortly hirsute and very scaly with lance-linear caudate caducous brown scales; fronds up to 100 cm long, irregularly disposed along rhizome from congested to 2-3 cm apart, the stipe pale dull brown, shorter than, to rarely as long as, blade, the longest stipe seen 50 cm, the bases of stipes hirsute and scaly with large glossy brown caudate caducous scales, the scales at first abundant, on old fronds scarce; blades up to 48 cm long, narrowly ovate in outline, broadest at base or the basal pinnae slightly shorter and spreading to more or less deflexed, without reduced pinnae, narrowed apically to a caudate ba- sally pinnatifid terminal segment, fertile almost to tip in large fronds; pinnae 16-18 well-developed pairs, subopposite to alternate, linear-lanceolate, acuminate, up to 13.5 cm long, up to 1.8 cm wide, lobed to 7 mm or even deeper, the lobes very slightly falcate, rounded to almost obliquely truncate at apex, hirsute on both surfaces, beset beneath, especially along costae, with abundant or rather sparse clavate or fusi- form to rarely bulbous or globular golden brown to rarely pale yellowish glands, these glossy and trans- lucent, appearing resinous; one pair of veins united below the sinus into a single vein ending in the sinus; sori about one-third in from margins of lobes but sometimes appearing submarginal because of the rev- olute lobe margins, rather closely set along the sides of the lobes, absent from the tips, sparsely scattered to abundant on united part of pinna, on most fronds reaching to near midrib, on some reduced to one or 10 SMITHSONIAN CONTRIBUTIONS TO BOTANY two pairs per lobe, these then near midrib; indusium well developed, orbicular-reniform, prominently long white hirsute. Found in Uganda, on forest floor in West and East Mengo districts ; one specimen, incomplete but prob- ably this, from St. Helena Island. AFRICA: Uganda, West Mengo District, Kyadondo County, Kampala, Makerere University Hill, 1200 m, K. E. Lye 5186 (MHU, holotype, isotypes, US, EA, NY, UC, GH, BM, IS, P, 0, B, MO) ; Cooper 2 (MHU, sterile, yellow glands very pale); Mawakota County, Mpanga Forest Reserve and Research Station, 1150 m, Faden, Evans ti3 Lye 69/998A (EA), Lye et al. 3822 (MHU; yellow glands pale and not abundant). East Mengo Dis- trict: Kyagwe County, Ssezzibwa Falls, 10 km W of Lugazi, 1200 m, Faden &, Evans 69/985 (EA, US). ST. HELENA I.: Quelle, Rosemary, lust in 1901 (US). The St. Helena specimen consists of only two blades and agrees in all respects, including the clavate golden glands, with T. fadenii, although the distri- bution seems unlikely, This species seems without doubt closest to Thelyp- teris quadrangularis (Fee) Schelpe (Dryopteris con- tigua Rosenst.), in fact, so close that it might be re- garded as a variety of it. It differs in its creeping rather than erect rhizome, more scattered fronds, shorter and broader lobes of the pinnae, and espe- cially in the clavate to fusiform golden glands. As species go in tropical Thelypteris, this seems sufficient to justify specific rank for T. fadenii. A considerable series of Brazilian T. quadrangularis, determined by Alan Smith, and a series from Africa, not so uniform but agreeing in most essential characters, were exam- ined, especially for glands. Some of the Brazilian specimens, especially three from Vicosa, Minas Gerais (Mexia 4487-a, 4760, and 5498-a, all US) and one from Sao Leopoldo, Rio Grande do Sul (Leite 4282, US), have scattered to abundant tiny white particles on the undersurface of the lobe tips. A sheet from Sao Bento, Bahia (Luettelberg 271a, US), has such particles and a few tiny globose pale to brownish glands. These could be regarded as approaching T. fadenii, but the ?glands? are much smaller and not the same shape, and the other characters are those of T. quadrangularis. These two species do not seem particularly related to the T. opulenta-T. wagneri group, nor to T. intar- rupta (T. goggilodus, T. totta), but perhaps more so to T. dentata (Forsk.) E. St. John and T. parasitica (L.) Tard. Literature Cited Ching, R. C. 1933. Notes on the Herbarium Willdenow. Lingnan Journal of Science, 12: 565-570. Ito, H. 1937. Filices Japonenses. VI. Botanical Magazine (Tokyo), 51 :709-714. Schelpe, E. A. C. L. E. 1965. A Review of the Southern African Species of Thelypteris. Journal of South African Botany, 3 1 : 259-269. Schkuhr, C. 1809. Vier und zwanzigste Klasse des Linneischen Pfianzensystems odes Kryptogamische Gewachse. 1 : 1-212, t. 1-173, lb-56bJ Wittenberg. Enmeratio filicum Philippinarum. Hooker?s lour- nal of Botany, 3:392-422. Smith, J. 1841. Stone, B. C. 1967. Further Additions to the Flora of Guam, 111. Micronesica, 2: 133-141. U.0. QOVWNMMT PRlNTlNQ OlClCLr lB7P 0-484-DI6/Sl