: THE FERN GAZETTE is a journal of the British Pteridological Society and

F A TeDED 08 all aspects of pteridology.

ee M eis may be abetted: and books etc. sent t for review, to: Prof. M. Gibby, — Royal Botanic Garden Edinburgh, 20A Inverleith Row, Edinburgh, EH3 SLR, UK poe oe 0131-248-2973 os E-mail: ‘FernGazette@eBPS. org.uk

ae - aie ee ‘Seis are on page 244 of this volume and also available aie a: Bo cvidetiets ae

Ce Sk ee eee ae ee oes ee ae ee vee — + bests I rved Aico oe

j aterial tc ] opying or

British

5 = ANN

Published ‘by THE BRITISH PTERIDOLOGICAL SOC IETY c/o Department of Botany, “The Natural History Museum, London SW7 S5BD, UK

ee oe Printed by Bishops Printers Limited | es “Fitzherbert Road, Farlington, Portsmouth, FOS IRU, rt a gate caress co.uk

Cover design by Hazel Sims

181 FERN GAZ. 17(4). 2005

British Pteridological Society with

Royal Botanic Garden Edinburgh Linnean Society of London

Ferns for the 21st Century

Proceedings of the International Pteridophyte Symposium at the Royal Botanic Garden Edinburgh, Scotland, UK 12-16 July 2004 Part 2

Editors: M. Gibby, A. Leonard, H. Schneider & J.C. Vogel MISSOURI BOTANICAL

MAR 0 8 2006 GARDEN LIBRARY

ROY AL BOTANIC GARDEN EDINBURGH

182 FERN GAZ. 17(4). 2005

THE DELEGATES

1)Marina Romanova 2) John Mickel 3) Eric Schuettpelz 4) Alan Smith 5) Judy Jernstedt 6) Marcus Lehnert 7) Klaus Mehltreter 8) Germinal Rouhan 9) Kunio Iwatsuki 10) Michael Kessler 11) Jens Pahnke 12) T. Boonkerd 13) Leon Perrie 14) Louis Chinnery 15) Stephen Blackmore 16) Niklas Wikstrom 17) Eka Iskandar 18) Jennifer Winther 19) Julie Barcelona 20) Barbara Parris 21) Paul Kenrick 22) Elena Estrelles 23) Michael Barker 24) Jonathan Krieger 25) Adrian Dyer 26) Ana Ibars 27) Peter Crane 28) Ruth Stockey 29) Elizbieta Zenkteler 30) Heather McHaffie 3 1) Gar Rothwell 32) Ruth Kirkpatrick 33) Joanne Sharpe 34) Peter Hovenkamp 35) Graham Ackers 36) Harald Schneider 37) Paulo Windisch 38) Daniel Ballesteros 39) Thomas Janssen 40) Carol Gibb 41) Alison Paul 42) Mary Gibby 43) Istvan Pintér 44) Irina Gureyeva 45) Tony Braithwaite 46) Royoko Imaichi 47) Mitsuyasu Hasebe 48) Johannes Vogel 49) Alasdair Wardlaw 50) Carl Taylor 51) Chris Page 52) Chie Tsutsumi 53) Frank Katzer 54) Wen-Liang Chiou 55) Patrick Acock 56) Yves Krippel 57) Masahiro Kato 58) Gregor Kozlowski 59) Maarten Christenhusz 60) Kathleen Pryer 61) Sunniva Aagaard 62) Jurgen Kluge 63) Hanna Tuomisto 64) Harriet Hunt 65) Atsushi Ebihara 66) Fred Rumsey 67) Virag Krizsik 68) Luis Quintanilla 69) Ruth Aguraiuja 70) Kai Runk 71) Layne Huiet 72) Nathalie Nagalingum 73) Mirkka Jones 74) Chan-Ho Park 75) Susan Klimas 76) Raymond Cranfill 77) Petra Korall 78) Tom Ranker 79) Sabine Hennequin 80) Michel Boudrie

FERN GAZ. 17(4): 183-203. 2005 183

GRAMMITIDACEAE (PTERIDOPHYTA) OF MOUNT JAYA, NEW GUINEA AND OTHER MONTANE MALESIAN LOCALITIES

B.S. PARRIS

Fern Research Foundation, 21 James Kemp Place, Kerikeri, Bay of Islands, New Zealand 0470 (Email: bsparris@igrin.co.nz)

Key words: Grammitidaceae, Pteridophyta, Mt Jaya, Mt Kinabalu, Seram, Gunung ulu, Malesia

ABSTRACT

Mount Jaya (5000 m) in Papua (Indonesian New Guinea) is the highest mountain in Malesia and 56 species of Grammitidaceae are known from it. They are listed, together with their altitude ranges, habitat types and geographic distribution. Comparisons are made with Grammitidaceae of Murkele Ridge in Seram, Moluccas, Indonesia (3000 m, 55 species), Mount Kinabalu in Sabah, East Malaysia, Borneo (4100 m, 76 species) and Gunung Mulu, Sarawak, East Malaysia, Borneo (2400 m, 43 species). A Malesia-wide core of 23 grammitid species is identified.

INTRODUCTION

The limestone massif of Mount Jaya (c. 5000 m) in Papua (Indonesian New Guinea) ts the highest peak in Southeast Asia. It is the only peak in the region high enough to support glaciers and has long been a magnet for biologists. Previous names for the mountain are Mount Carstensz and Punjak Sukarno. The Royal Botanic Gardens, Kew, has organised five botanical expeditions to the mountain during 1998 to 2000, during which many collections were made along an altitudinal transect on the southern side. A detailed description of the alpine flora of Mount Jaya (above 3000 m) is in preparation at the Royal Botanic Gardens, Kew (Johns et al. in prep.). While writing up the treatment of Grammitidaceae for this account, all relevant material collected on Mount Jaya was identified, not just that from high altitudes, and included specimens gathered by Kloss on the Wollaston expedition on the mountain during 1912-1913 and those collected by Miller in 1991 and 1992. Although nearly all of the available material has been collected from only the southern side of the mountain, it is of sufficient interest to document here in the following terms: |) how many species?. 2) what are they?, 3) what are their altitude ranges?, 4) what vegetation types are they found in?, 5) where else do they occur?

Comprehensive collections of pteridophytes from specific localities are rare in Malesia, because many botanical collectors have specialised in gathering only flowering plants and have tended to overlook the smaller and more obscure species of pteridophytes, including most Grammitidaceae. There are only three other localities in Malesia where the fern flora, including Grammitidaceae, has been relatively well-collected over a wide altitudinal range: Mount Kinabalu Park in Sabah, East Malaysia, Borneo, Gunung Mulu National Park in Sarawak, East Malaysia and Murkele Ridge in Seram, Moluccas, Indonesia. More than 40 species are known trom each of these localities and they provide a useful comparison with Mount Jaya. The following questions are relevant: 1) how much do the floras of the three areas have in common

184 FERN GAZ. 17(4): 183-203. 2005

€

with Mount Jaya?, 2) is there a distinct East Malesian element present on Mount Jaya and Murkele Ridge?, 3) is there a distinct West Malesian element on Mount Kinabalu and Gunung Mulu?, 4) is there a Malesia-wide group of species of predictable occurrence?, 5) do other pteridophytes show the same geographical distribution patterns and do they also have Malesia-wide species of predictable occurrence?

Map | shows the location of Mount Jaya, Murkele Ridge, Mount Kinabalu and Gunung Mulu. Map 1. Malesia.

Philippines ie PACIFIC OCEAN

10°N

Peni Mt Kinehety

Malaysia

‘ Gunung Mulu re : M O’N New Guinea Borneo Murkele Ridge ? Seram Mt Jaya Lesser Sunda Is. 10°S 100°E 110°E 120°E 130°E 140°C 150°

Most species of Grammitidaceae are epiphytes on trunks and branches of a variety of trees and shrubs, or on the trunks of Cyathea in alpine communities, often associated with bryophytes, and a few species, particularly those of subalpine habitats, are lithophytes.

Mount Jaya 56 species of Grammitidaceae are known from Mount Jaya. They are listed, together with their altitude range, vegetation types and distribution, in Table 1.

Tropical lowland rainforests are generally poor in members of Grammitidaceae, Which are far more numerous above the average daily lower limits of cloud. On Mount Jaya only five (9%) species occur in lowland rainforest, which contains the tree genera Chisocheton, Ficus, Mvristica, Octomeles, Pometia and Vatica, with undergrowth of ferns including tree ferns, Arthropteris and Marattia, and Gesneriaceae. All five species are also found in heath forest. Heath forest, the other lowland forest habitat on Mount Jaya, supports a surprisingly large number of species, 16 (29%). It occupies old Pleistocene outwash terraces at the foot of the mountain (R.J. Johns pers. comm.) and includes the trees Alstonia spectabilis R. Br. Calophyllum, Casuarina, Dacrydium, Metrosideros and Vatica, together with Nepenthes, Davallia, Lecanopteris, Hymenophyllaceae and Polypodiaceae. Two of the four Mount Jaya endemics, Ctenopteris hymenophylloides (also in lowland rainforest) and Grammitis velutina, occur here, together with two other species (Crenopteris nutans, Grammitis scabristipes) disjunct from much higher altitudes in midmontane forest, upper montane

PARRIS: GRAMMITIDACEAE OF MOUNT JAYA, NEW GUINEA

SIWApUd RAUIND MON DIWUAPUd BAUIND MON BIUOPa|B.) MON O} BALL BOUIND) MON) O} ISAMRING BOUIND) MON 0} BISARRIA “Ud BOUIND MON 0} BIVRUINS s1uapud BAEr JUNO

SIWapUs BOUIND) MON

BOUIND MON 0} BUTYSOPU]

‘S] AJOID0G 0} BIPU] “Ss dIWapUa BOUIND MON njyenuevA 0} BISa|R ‘VISY 4S DIWUAPUS BOUIND) MIN

ROUIND) MOK ‘sRdoNO|| DILUDPUd BOUINH) MAN,

"S] UOLUOJOS 0} BISAR] RI, “Ud BOUIND) MON OF BRUINS

DIWIApPUd BIUINDH MIN

uoHNg!ysiG

JW AWW VS

AJWW

4H

VS AWW 4H AIWW ala

VS

JWW AWT a1 dH

VS AWN AWN AWW VS

AIWW

VS

VS ANW AWW AWT AW’

UO BIIBIA,

W OO8T-OO8 I WOOTE

W OYOPC-OOTC W Y09-00T

W (O9C-O0S W 009T-O00C W QOS-O00T W OQOSE-OOEE W OOET-ONS | W QO8-00T

W 009E-009T W QOLT-OOST W OO09E

W O08 I

W (YORE

W 00St-00EC W QOET-OOTI Ww QOL!

SPNNOTV

‘yadoy (AyD) Uppoqyoie sHiuWRIH ‘g]

‘jadoy (CMiapyy) euurdnepnqns "Dy “Z|

‘jado,) (*|]JO7Z) BIDAaSSIPOpUNdasgns “+ ‘g|

‘Paul SLIP RIBAS "D °¢

‘us f (QuIN|g) suRINU "+ “PT

‘yado,y (dung) BlosaypIuU "DE |

sued soproyAydouswiAy “) “Z|

‘adoD eosnj “TI

NIIPIVE WAY’ D (oumnyg) eyeypnonuap “yD ‘ol JOYIAIDH) 32 AUTEM HA (‘AdID) Saprlouysayq “Dd ‘jado,y (Aayeq) epyneuuidig suajdouayy SIV (loyeg) BUBLARISSNW SUUeIdOSAIY)

6

8

a ‘jadoy saylurel “9 ‘jadoy sapoiuw ay *¢ v

Jado) sysey *D

[Salg’d (AwNn]g 2 SaaN) smiey[NoN "Dy *¢ MOO | (YOOH) JayIAvjO uopowuuAyeD 7 ISLYD'H GSUYyO'H) Waiyoayyos sniosoiy "| saad

‘eAer JUNO JO IVIOEPHIWIWUPID "| Iquy

FERN GAZ. 17(4): 183-203. 2005

186

"s] sesanbie|y] 0} eIpuy S dIWApUd ROUIND MAN BOUIND) MON 0} OdUIOG stuapus vARP JUNO BOUIND MOXY O} BIVBUING njenuevA 0} BQeUINS

‘S] UOLIO]OS 0} sedONTOYW o1wapua eAer JUNO BOUIND MON 0} BIVRUWING BOUINDH MOXY O} RAVENS SIWWOPUD BOUIND MON SIWIAPUd BOUIND MAN BOUINH) MANY O} BRUINS fly 0} BISAR]eYy Ae;NsUIUdg BOUIND) MON 01 BISY FS SIWApUd BOUIN MIN dIWApUS BOUIND) MON DIWOPUd BOUND MON BOUIND MAN 0} OAUIOg SIWUApUd BOUIN MAN SIWUIapUd BARS UNO

d1WdpUd RaUINDH MAN

VS AWW 4H dH

dWW AW VS AWN JINN AWW

VS

dINW AW AH VS

dNW AWT VS AWW dWN

AWW

VS INW

a1 4H

VS

AWW

VS

W OQOET-O08 | W OOPE-OO0T W QOT

W OOb-O00C W 0QO9CT-O0LI W OO8E-O0F sec WOOET-OO8 W 006E-009T W O06!

W OQ06E

W OQOST-O0S W OQO8E-OOSE W OQOET-OO8 W OOS E-008 | W QO9T

W YOST

W OOTH-O0ET WW Q09-00S

W O09E

W OOPC WOOSE

[S%dg*D (1S1O4H) ensyuod ‘g * SLUB_ (“JSUISOY) BIL][LALUNDIIO ‘g ° SLUeg (Joyeg) 1aspiqing endesodg ° SLUeg PUINIOA

‘jadoD (sayeg) euRsyeWNs

‘Jado (isussoy) seprowuessoinayd

ms yy ‘[adoy (4ayeg) sadusiuqeros ‘+ ° % a

sled (Map y) sisuanded

‘Jado. (4ayeg) sisuasueped ‘fH -

‘Jadoy (‘ISUdsOY) BUIISSHBUIO ‘T° 5D

‘Jado usddip-safiaur *

‘jedoy (Map y) BsopNdo] ‘DH * ‘jado,y (4ayRg) eIL]JAd0] °D ° ‘JadoD (4ayeg) eurIpsoysinuy ‘DH ° SIURq (SLY) eBSsiwuonUT tr ° ‘Jadoy (ayeg) eidnuaqut ‘pH - ‘Jedo.D epidsty + °

‘Jedod CIP) eprsuy +H ‘[ado,y (Jado) erOsoYsOp ‘H ‘JedoD vryoytigap "Dp

SLUR BIRARID DH

‘jado,y edurs0je199 FH °

OF 6t

187

nA

A, NEW GUIN

, 4

PARRIS: GRAMMITIDACEAE OF MOUNT JA)

‘pasuvs) Supe ulvsd) JO aunqyeu snoyidioaid JO asnevsaq sysiuKjOg MAY AQ Pd}da] [09 JOU LE QOS E-OOXT Wues apne ION

‘(W QOTP-OOEE) SOUIUNWIOD dUIdjeqns “WS “(LU QOLT-QO9Z) ISMO} SuRJUOLE Jaddn “4PYA “(LUIQORZ-OO8 1) JSOIOJ] QURJUOWIPIL “IV “(WU QOL | -OOR) ISAOJ DULJUOL JAMO] “YIV] “(LU QOS-OOT) ISAOJUIeL PUR] MO] “477 “(LU YO9-QOT) ISO} YIBIY “4H

BOUIND) MON O} BISARTRIY “Ud BOUINE) MAN O} BISARTRIY “Udd SIWIApUd ROUIND MON

BOUIND) MAN, O} RISAR]RI, “Ud ‘S] UOLUOJOS 07 sauIddiyiyg SIWApUd ROUIND MON

BOUIND) MON O} RAVENS SIWAPUd BIUIND MAN sesanbiey 0} ByuRy] US

"S| UOWIOJOS OF puRyIeYL

‘S] UOLUOJOS “BaUIND MAN BOUIND MON O} BARS

DIWApUd BAUINH MIN DIWApPUd BOUND) MON

BOUIND) MON OF BIPUT S

BOUIND) MON O} BISARTRI] “Ud

W 009E-008 |

WQQ9-O0S

W QO9T-OOFT

W (09-00S

W YOPC-O0TI

W (QO9IT

WOOT

W OOPE-OOET

Wi QO9

W (09

W O0OLC

W OOPC

W OOS I

W 008

W OOS

W QOLE-O0FE

‘yado,y (Quinjg) epyneuurdgns “xX “9¢

Sled (MIOP]Y) Brayl[MIas “KX “SS

sued (MIaply) stpeaidsopnasd -yx “ps

ado.) (asneig) sisuaevofpiaos suaidoydiyx “¢¢ stued (‘Jado.)) Wapod “| “7S

sled (jado.) wintpoyluruess “y 1S

sled (Suinjg) winqelosey “pL “OS

SLUR (MIOP]Y) WNOSO]]|R WNTPAWdY_ | “6P “MIOPLY (UyNy) WNyRoINs *S “Rp

“MIOP]Y (Quinyg) wunyisnd *S “Lp

‘jado,y wintoplounf winssojso1a]9g “Op IUD W (azuNy xd aun] g) BSO]NUDA ‘gq “Sp SLued (Pry) stisuad “qd “pp

‘paul siueg eieyjided -g ‘¢p

WOW (owinyg) eienbijgo ‘gq ‘7p

Jodo) (oye FY TENA) Paoery|eaep ‘d “1p

188 FERN GAZ. 17(4): 183-203. 2005

forest and subalpine communities. Crenopteris serrata, Grammitis pleurogrammoides, Prosaptia burbidgei, Scleroglossum pusillum, S. sulcatum and oa ies en setulifera are known only from heath forest on Mount Jaya, but occur in other communities elsewhere in New Guinea and further afield in Malesia. The aie: may exist elsewhere in New Guinea, but apparently no members of Grammitidaceae have been collected from it. Heath forest, also called kerangas, is known from Peninsular Malaysia, Borneo, and elsewhere in Malesia. It develops at low altitudes on very poor or thin soils over rock or in areas of impeded drainage and is characterised by straight, pole-like small to medium sized trees and an understorey including pitcher plants (Nepenthes). Usually it contains very few, if any, species of Grammitidaceae.

Apart from heath forest, the most species-rich habitats on Mount Jaya are midmontane forest (28, 50%), and subalpine communities, with 19 species (21%). Lower montane and upper montane forests are less rich, each with 9 species (16%). Lower montane forest contains the fern Se//iguea (no other information concerning the vegetation of lower montane forest is available on the specimen labels). The midmontane forest is dominated by Nothofagus; Casuarina, Dacrydium, Fagraea, Loranthaceae, Macaranga, Papuacedrus, Rhododendron and Timonius are also present together with the sedge Mapania, the ferns Asplenium, Belvisia, Cyathea, Elaphoglossum and Loxogramme, and shrubby Urticaceae. Grammitis clavata is the only Mount Jaya endemic in midmontane forest. Only eight species are restricted to midmontane forest, with another eight extending downward to lower montane forest and sometimes further, to heath forest and lowland rainforest, and 13 extending upwards to upper montane forest and sometimes to the subalpine communities. Upper montane forest is dominated by Papuacedrus papuana (F. Muell.) Li and Podocarpus brassii Pilg., with Dicksonia, Drimys, Elaeocarpus, Nepenthes, Rapanea, Rhododendron, Sloanea and Utricularia. Grammitis interrupta, Scleroglossum Juncifolium and Themelium graminifolium are restricted to upper montane forest on Mount Jaya, but occur in other vegetation types elsewhere. Another five species occur in midmontane forest as well as upper montane forest, and two, Crenopteris bipinnatifida and Grammitis padangensis, extend to the subalpine zone. The subalpine grasslands are dominated by Cyathea tree ferns, while the subalpine shrubland contains Cyathea, Coprosma, Drimys, Rhododendron and Vaccinium, and the subalpine forest contains Dacrycarpus compactus (Wassch.) de Laub. and Podocar- pus. Nine species are restricted to subalpine habitats: Calymmodon fragilis, C. ramifer, Ctenopteris fusca, C. subulatipinna, Grammitis ceratocarpa, G. debilifolia, G. locellata, G. meijer-dreesii and Prosaptia davalliacea. With the exception of G Jocellata and P. davalliacea, all are New Guinea endemics. While the nine species all occur on Mount Jaya at altitudes higher than that of the nearest subalpine site to New Guinea, Murkele Ridge (3000 m) in Seram, the two species that are widespread through Malesia, G /ocellata and P. davalliacea, occur at lower altitudes on Murkele Ridge.

Several species found on Mount Jaya are infrequently collected and are known from very few localities. They include medium to high altitude Papua endemics known elsewhere only from Lake Habbema - Grammitis hispida, Themelium allocotum, Mount Trikora - Ctenopteris fusca, G. meijer-dreesii, and/or Mount Doorman, C. subulatipinna, T. allocotum, and Mount Goliath and Hellwig Mounts - Xiphopteris pseudospiralis. Xiphopteris govidjoaensis (syn. X. exilis Parris) is known from two na localities, Gunung Belumut in Peninsular Malaysia and Govidjoa in Papua New

uinea.

PARRIS: GRAMMITIDACEAE OF MOUNT JAYA, NEW GUINEA 189

Four species (7%) are endemic to Mount Jaya; Crenopteris hymenophylloides, Grammitis clavata, G. papuensis and G. velutina. Surprisingly, none occurs in the subalpine communities, as might be expected when considering that several other mountains in New Guinea have high altitude endemics, e. g. Grammitis dictvmioides Copel, G habbemensis Copel., Xiphopteris antipodalis Copel. (all Lake Habbema), G. reptans Parris (Mount Trikora), G montana Parris, G. trogophylla Copel. (both Mount Sarawaket), G murrayana (C.Chr.) Copel. (Wharton Range) and G. salticola Parris (Mount Giluwe) (Parris, 1983).

A number of other high altitude New Guinea species found on Mount Trikora and/or Lake Habbema, and elsewhere in Papua New Guinea, may be expected on Mount Jaya, including Ctenopteris pendens (Rosenst.) Copel., C. whartoniana (C.Chr.) Copel. and Grammitis tomaculosa Parris.

New Guinea endemic species are important on Mount Jaya, comprising 36% (20 species) of the total. While they occur in all of the vegetation types they are best represented in the high altitude subalpine communities, where they form 20% (11 species) of the total number of species. The largest number of species, 24 (43%) are Malesia-wide in the sense that they occur in both West Malesia and East Malesia, on both sides of Zollinger’s line (see later) between Borneo and Sulawesi, sometimes extending into Thailand, Indochina and Taiwan to the north-west and into Melanesia (Solomon Islands, Vanuatu, New Caledonia and Fiji) to the south-east. A small group of species (4, 7%) is very wide-ranging, from South India to the Society Islands (Ctenopteris blechnoides), South India to the Marquesas Islands (Prosaptia contigua), Sri Lanka to the Marquesas Islands (Sc/eroglossum sulcatum) and from South India to New Guinea (P. obliquata). These four species are amongst the most common and widespread members of Grammitidaceae in the Asia-Malesia-Pacific region. Another four (7%) species are restricted to East Malesia; Crenopteris serrata ranges from Sulawesi to New Guinea, Calymmodon mniodes is in the Moluccas and New Guinea, Grammitis pleurogrammoides extends from the Moluccas to the Solomon Islands and Scleroglossum juncifolium occurs in New Guinea and the Solomon Islands.

The grammitid flora of Mount Jaya inevitably invites comparison with that of Mount Kinabalu (4100 m) in Sabah, Borneo, East Malaysia, the highest peak between the Himalayas and Papua, and the two other areas sufficiently well-known to permit comparison of their grammitid floras: the Murkele Ridge, a limestone massif including Gunung Binaiya (3000 m) in Seram, Moluccas, Indonesia, and Gunung Mulu National Park (including Gunung Mulu, sandstone, 2400 m and Gunung Api, limestone, 1700 m) in Sarawak, Borneo, East Malaysia. It is useful to examine the composition of these montane grammitid floras with respect to Zollinger’s line, which divides West Malesia, comprising Peninsular Malaysia, Sumatra, Java, Borneo, the Philippines and the Lesser Sunda Islands, from East Malesia, consisting of Sulawesi, the Moluccas and New Guinea, and separated by the Makassar Strait between Borneo and Sulawesi. Wallace’s line is similar, but includes Bali, the westernmost of the Lesser Sunda Islands, in West Malesia, and the remaining Lesser Sunda Islands in East Malesia. As the Lesser Sunda Islands are considered a discrete botanical region, Zollinger’s line is preferred to Wallace’s line. Mount Kinabalu and Gunung Mulu National Park (subsequently referred to as Gunung Mulu) are in West Malesia, while Murkele Ridge and Mount Jaya are in East Malesia. Both Mount Kinabalu and Murkele Ridge have subalpine areas above the treeline, but Gunung Mulu is forested to its summit.

The Moluccas have been of biogeographical interest ever since Wallace (1869)

190 FERN GAZ. 17(4): 183-203. 2005

suggested that the region has been constantly receiving immigrants from New Guinea. Kato (1989a) also points out that the affinity of the fern flora of Seram at species level is with New Guinea to the east rather than with Sulawesi to the west. Parris (1993) notes that the endemic pteridophytes of Seram are derived on the whole from wide-ranging taxa or those from East Malesia, and that Seram is the end of the line for numerous species moving east from West Malesia and moving west from New Guinea. Grammitidaceae of Seram collected by the Japanese-Indonesian expeditions of 1983, 1985 and 1986 (but not those collected by Parris in 1987) have been listed (Kato & Parris, 1992), the pteridophyte flora of Mount Kinabalu has been documented (Parris ef al., 1992), and a checklist of pteridophytes of Gunung Mulu National Park has been compiled (Parris, unpublished). Numerous changes have been made to the taxonomy of Grammitidaceae in Seram and on Mount Kinabalu since the publications of Kato & Parris (1992) and Parris et al. (1992).

Murkele Ridge 55 species of Grammitidaceae are known from Murkele Ridge. They are listed, together with their altitude range, vegetation types and distribution, in Table 2.

It is useful to compare the grammitid floras of Mount Jaya and Murkele Ridge, the two East Malesian localities, which have very similar numbers of species, 56 in the former and 55 in the latter. 25 species are shared between the two, and nine of these are restricted to East Malesia. The remaining 16 are more widespread, occurring in East and West Malesia and sometimes beyond. Only Ctenopteris serrata extends throughout East Malesia from Sulawesi to New Guinea, so a well-defined widespread East Malesian endemic element is lacking in Grammitidaceae. Only one species, Acrosorus Jriderici-et-pauli, reaches its eastern limit of distribution on Murkele Ridge in Seram, but eight others, Calymmodon mniodes, Ctenopteris whartoniana, Grammitis ahenobarba, G. collina, G. parva, G. pseudolocellata, G. subfasciata and Prosaptia engleriana, reach their western limit in Seram, either on Murkele Ridge or slightly further to the north-west on Gunung Kobipoto (G. collina) or Gunung Roihelu (G parva). They amply demonstrate Wallace’s point (1869) that the Moluccas receive immigrants from New Guinea. In addition, some of the Murkele Ridge endemic species have obvious vicariants in New Guinea, e. g. Ctenopteris themelioides with C. bipinnatifida, and Themelium pseudallocotum with T. allocotum. The similarity in species numbers between Mount Jaya and the much lower Murkele Ridge is surprising, and the reasons for it are not obvious.

Comparisons between Murkele Ridge and the two West Malesian localities, Mount Kinabalu and Gunung Mulu, are interesting. 33 species are shared with Mount Kinabalu, more than are shared with Mount Jaya. They include seven widespread species collected on Gunung Mulu and known from New Guinea that are absent from Mount Jaya: Calymmodon gracilis, C. pectinatus, C. reconditus, C tenopteris brevivenosa, Grammitis adspersa, G. impressa, G. reinwardtii. All are expected on Mount Jaya. 19 species are shared with Gunung Mulu and, with the exception of Calymmodon conduplicatus, all are also known from Mount Kinabalu.

31 species (56%) on Murkele Ridge are found in both East and West Malesia and another 5 (9%) extend well beyond Malesia, while 9 species (16%) occur in East Malesia and another 9 (16%) are endemic to Murkele Ridge. The amount of local endemism is much higher than on Mount Jaya, which has only four endemics. One might expect local endemism at high altitudes on Murkele Ridge, the highest part of the

19]

PARRIS: GRAMMITIDACEAE OF MOUNT JAYA, NEW GUINEA

BOUIND) MIN O} ISAMP[NS DWAPUd ISPly IOxINJA| BOUIND) MOK O} BISARIRIAY “Ud BOUIND MON O} BRUINS BOUIND MAN, O} BULYDOPU| njenued 0} purypliey

“S] AJOIDOG 0} BIPUT

njyenurA 0} BIsayey “VISY JS BOUIND MIN 0} OOUIOg BOUIND) MON O} BALL

BOUINL) MON “SBdONO|| BOUIND MON O} RISYV AS

"S| UOWO]OS 0} BISARTRI “Ud BOUND) MAN 0} OdUIOG BOUIND MAN O} BVBWINS dWAapuUds dIspry sayin dIWapuds dIspry aon

seoonjoyy 0} BISY 4S

uOHNgLySIC

AW

AW AWN ‘SAWT AT dW) AWN AW]

AW

AW

VS JINN AW AWA AW AW

VS

AW

JWT

U0IRIIGIA

We QOOT-009 WwW OQO9T

W 006C-008 I W YOO0T-O08 | WOOT!

W OOPC

W OOET-OOS W QOOT-O00TI W (YOST WQO81-0091 W O08 1

W 0081

W QO8C-008 1 W QOSC-O0CI W QOOC-OOLI W OQO8T WOOL

W QOOT-O0ET

apmynty

‘Poul SLUR IBIS */) S|

"POUL SLLIBY SISUDWURIAS *) “L |

"US f (QUIN; g) suBINU *) “QT

‘jadoy (auinyg) eyosaqpu "DS |

NdIpPIeL WAY D (oun | UNVJOH CMIOply) BSOUDSATAAIG “DE ]

—

Bye|NOQUap "DO ‘pI

JoyIasH 2 AUseA\’ HM (AIID) Saplouysayq sluaidoualD ‘Z| SLUR (IDV) BPURIARIDSNU SILULURIBOSAIY) *| |

‘paul sued SNIpUuodal “Dd ‘0

‘pour sued snyeunsad *y

‘jado.y sapotuuwt “>

‘JadoD (a94) stplovss * >

[SXq’) (OuIN|g WZ SdaN) sniBT[NINd * +)

‘jadod (asneig) snyeotjdnpuos *)

DOO [| (YOO) JOJIAB]S “+

‘pour sLueg sisuddreulg UOpoUWAye)

aa tu 6K wo oO

‘Poul SLUR SNIRI[PA “VY

‘yadod (syd H) yned-ja-19apLy snsoso1ay

_

saiadg

‘ISpPly apayn JO avaowpniwweIyH “7 9qUL

FERN GAZ. 17(4): 183-203. 2005

192

BOUIND MON O} BQBUINS BOUIND) MAN “SBOON[O|y njenue, Oo} BQeUINS

‘S] UOWIOJOS 0} BYU] LIS BOUIND) MAN “SsBdoNjO|W BOUIND) MON ‘sedoNjop] BOUIND MON O} BAVRUINS SRIIN|O|] 0} OAUIOG d1WapUd dSpry spayany BOUINDH MAN OF BVBUINS if1.J 0} BISARTRIY Ae[NSUIUDY dIWapUd dSpry sjayINy BOUIND) MON 0} UBMIB] BOUIND MANY O} BRUINS BOUIND) MON O} ODIO BOUIND MAN ‘sBdoN[O| BOUIND MON O} BIZRUINS BOUIND MON ‘seoonjoyy fly 01 BIS JS

BOUIND) MON ‘sKdoNjoYy dIWapud sspry sjayIN YY BIUOPs[BD MAN O1 BALLS

dW)

AW

VS JWA AW

VS

VS AWN AWT AW

VS

VS ANN AW

AW

WQ00C-00E I W 0061-00F1 W 006C-00CT W QOOC-O0L W006C

W Q06C-008 I tu QOOC-O08 | WOOL! W009 W (06T-008T W Q00C WW 09

Ww 000C-009 | WwW oo0e!

WU (OOT-008 W OOF!

W OOTC

WwW O09

Ww 009]

W 0067

UW QO8T-O00T Ww O0E!

‘jedoD (4ayeg) BUREUINS DH “Op

* ado (suasoy) eIVIOSRIQNS TD “GE ‘yadoy (4ayeg) sadnsuqess ‘ry “g¢ aun] MpaweMuUlal FH “LE ¢

SLUR BIL]]AOJOpNasd ‘FH ‘g¢ ‘jadoD (asneig) eared tp “¢¢ ‘Jado. (4ayeg) sisuasdueped "Fp “pe ‘Paul sLUeY STOW "DH “EE

‘poul SLY SIsuajasnueul DH “7E ‘jado,y (Aayeg) BIE]]A00] T “TE ‘Jado (4ayeg) vuvipsoysinuy D “o¢ ‘POUl SLLIBG SISUBYOYIULY 1) “67 SLUR (ISUYD'H) BSSIWONUI TD “gz ‘jado.y essaidun 1 “47

‘Jado (‘jado,)) vsosoysijop "DH ‘gz SLUR BULL[OS ‘Ty ‘SZ

auinyg vsoudsaes ty “pz

SLUR BqIeqoudye T “FZ

auwinyg (owinyg) esusdspe sniutuPIin *7z ‘Jado, (4YyD*D) BuriuoEYM * “1 Zz ‘Poul SLUBY SAPlOlaWay) “J ‘QZ

‘}adod (]]07Z) vdasstIpopunsasqns ‘61

193

PARRIS: GRAMMITIDACEAE OF MOUNT JAYA, NEW GUINEA

‘(WE QOOE-O0PT) SAH

-luNLUWOD auIdyeqgns “WS “(W QOSZ-001Z) Ise} ouRJUOW Jaddn “yi “(LU QOL Z-O09) ISAL0J DUBIUOW JAMO] “AIT “(LU QOY) ISesOJ puR[MO] “47]

BOUIND MON O} BISARTRI “Ud BOUIND) MON O} BISARIRI] “Udd ‘S] UOWO]OS 0} sourddiyiyd BOUIND MON 0} OOUIOG dWIapUda dBpry san] sesanbiey 0} ByURT] LIS

“S| UOLUOJOS 0} pueypeyL BOUIND MON O} BALL

dIWIApUd sBdON]O||

d1WdpUd ddply sa4ANJA| BOUIND MON O} BIPU] S BOUIND MAN ‘SBdoN|O/ BOUIND MON 0} BISARTRIY “Ud "s] sesonbieyy 0} eIpUuy S

otwapud aspry spoyINy|

dW AW AW dW dW AW AW)

AW] AW

W008 I

W 0001-006 WwW YOOT-008 I W (OCC

We OOE|

WwW OOET-0001 WwW QO 1-O001 W OOTZ

WwW Q091-00E1 W 0Q00C-008 | W Q00C-009 W 008-009

W 0Q06C-O08C W QO00C-00L WO00EI

‘jadoy (aun) epyyneuurdgns "x “¢¢

Sued CAOpPlY) elayipnias suaydoydry “ps sled (Jadod) MApoAd “| ES

SIIB (DAVE) SOPIOIPOXR] “| ZS

‘pour siueg WNnjosojjepnasd wintyawayy |S “MIOPTY (UyNy) WUNRO[NS “S "OS

“MIOppy (aunyg) tunypisnd winssojSo19[9g “Gp IUD’ W (azuny xo awn] g) BSO[NUDIA ‘g ‘Rp sluvg (sLuRg 2 OY) BIqe{sqns ‘gq “Lp ‘POU SILUIBY SISUSWIRIAS ‘gf ‘OP

NOW (ounpg) eienbijgo “g “Sp

‘jadoy (asneig) euBLio|sud ‘g “pr

‘Jodo (oyeg 7 ‘[ONW4) B9ORTT[BARP ‘g “EP |SUq*D (1810,,'H) ensyuos endesolg ‘Zp SLU, ZW OY" sisuans0ya 1 ‘|p

194 FERN GAZ. 17(4): 183-203. 2005

Moluccas and the only part of the region to be above the tree line, but only Calymmodon binaiyensis is restricted to subalpine habitats. Crenopteris themelioides ranges from lower montane forest to subalpine habitats while the other local endemics are all found in lower montane forest. Only one species, Prosaptia subglabra, is endemic to the Moluccas.

Mount Kinabalu

76 species of Grammitidaceae are known from Mount Kinabalu. They are listed, together with their altitude range, vegetation types and distribution, in Table 3.

Mount Kinabalu and Mount Jaya make an interesting contrast. The former has far more species of Grammitidaceae, 76, than the latter, with 56. Mount Kinabalu has a long history of botanical collecting, however, over more than a century, which may explain its species richness in part (Parris ef al., 1992). It should be noted, however, that nine species have not been collected on the mountain in the last 60 years (Parris, 2001) and may well be locally extinct. 23 species occur on both mountains; all of them, except Themelium fasciatum, are also found on Murkele Ridge, and 10 of them are also found on Gunung Mulu. Mount Kinabalu has nine endemic species of Grammitidaceae (16%), more than double the number found on Mount Jaya. As with Murkele Ridge, one might expect the endemic species to be associated with the high altitude vegetation communities above the tree line, but only Grammitis kinabaluensis extends above the tree line, and it also occurs in lower montane and mid-montane forest. Grammitis graniticola, G. muscicola and G nubicola are restricted to upper montane forest, while Calymmodon innominatus and Grammitis ramicola are found only in lower montane forest and Calymmodon kinabaluensis, Grammitis havilandii and G. ultramaficola occur in both lower montane and upper montane forest. 44 species (58%) are Malesian-wide, a higher number and percentage than for either Mount Jaya or Murkele Ridge. Another eight species (11%) extend well beyond Malesia. 10 species are restricted to West Malesia, six of which are also found on Gunung Mulu. The West Malesian Crenopteris fuscata occurs at high altitudes on Mount Kinabalu (3000-3800 m) and is not found on Gunung Mulu (2400 m), but Grammitis congener (1100-2500 m) and Themelium curtisii (1400-2400 m) are absent from Gunung Mulu, but may be expected. Xiphopteris nudicarpa occurs on ultramafic rocks on Mount Kinabalu and on Palawan Island in the Philippines. Ultramafic rocks, and_X nudicarpa, are absent from Gunung Mulu. Five species are endemic to Borneo and all are shared with Gunung Mulu.

Gunung Mulu 43 species of Grammitidaceae are known from Gunung Mulu. They are listed, together with their altitude range, vegetation types and distribution, in Table 4.

Gunung Mulu is lower than Mount Jaya, Murkele Ridge and Mount Kinabalu, and lacks vegetation communities above the tree line. It has no endemic species, and 36 of its 43 species are also found on Mount Kinabalu, 22 species (51%) are Malesian-wide and five (12%) range well beyond Malesia. The nine West Malesian species (21%) all occur on Mount Kinabalu. Seven species are endemic to Borneo: five also occur on Mount Kinabalu, but Scleroglossum crassifolium and Themelium sarawakense are not known there.

In Table 5 the geographic distribution of Grammitidaceae on Mount Jaya, Murkele Ridge, Mount Kinabalu and Gunung Mulu is summarised.

199

GRAMMITIDACEAE OF MOUNT JAYA, NEW GUINEA

PARRIS

BOUIND) MON 0} BISARTRIY “Ud BOUIND MON 0} BISY FS soulddiyiyg 0} eiskeyeyy ‘Udd ROUIND MAN O} B.BUINS soulddiyjiyg 0} Biskeyeyy ‘udd sourddiyiyg 0} purprey,| BOUIND MON 0} BUIYyDOpUT nyenue, 0} puryieyy,

“s] AJOI90g 0} BIPUT “Ss nyenurd 0} BISy 4S sourddipiyg 0) eyurT] Ls BOUINL) MANY 0} OSUIOG ROUIND) MON 0} BARS soulddiiyg 0} eave

"S] UOLUOTOS 0} BISY 4S o1luapus nyjeqeury JUNO dUdpUs Njeqeury JUNO BOUIND) MAN O} BALL

BOUIND) MAN O} BISY FS

"S] UOWO]OS 0} BISARRIY ‘Ud BOUIND MAN O} BBUINS dIWINpUsa OAUIOg

BOUIND MON 0} OOWIOgG soutddiyiyg 0} pueyrey.| SRdONJO|| 0} BISY FS

woHnqLysig

Uu0ne}I99A4

WU OOS E-O09T W ()06C-00S | WwW OOF]

W 0081 WOOL

W OO8E-O00E W 0O08T-00TI W Q00E-00F | W 0001-008 WwW OOTC-O0FI Ww OOS WOOT €-008 | W QOST-00S I W OOS |

W OOPC-O0S I W OO8C-O0S | W OOP

W 001C-00FI W OOOE-00TI W QO00E-00TT W 009C-00S | WwW 000-008 I W OOFE-008 I W 0061-008 W 0061-001

apnynty

“WS 'f (QuIN}g) suBINU ) “SZ AZUNY (AWIN|g A SAIN) BWIOTIJOUL "-y “PTZ wiINyjOH (awinyg) enue > °

‘yadoy (auinyjg) eloyaypru “>

‘jado,y (duinjg) ensidsuosut °°

azuny (auin|g) BIBdIsNy ° ‘OZ

naipsep Wu’ D (auinjg) erepnonuap *>y 6] wINOH] (Maply) BSOUDATADIG "5 ‘81

JOYINID w 1dUseAA HA ('AdID) saprouysayg siuaidouany */ | sued (Jayeg) eueiaridsnu “> “9|

sued ((Wg'f) esojnpurys sntuuUeIBOsAIY) “S| ‘poul sLueg snyipuodsal “> “P|

‘paul stueg smeunoad 9 “¢]

‘jadoy (Jado) sapioosnu “5 “Z|

‘jado,y (ultu0g) snuelUassiany “> “||

‘Poul Sie” Sisuon[eqeuly *) ‘Ol

‘poul sLueg smeuluouul y “6

‘jadon snaidossoisAy 8

‘Jado> (994) sifowid yy *L

[SUd'D (Aunjg 2 SdenN) smepnond yg AOOW'L (CYOOH) JAPAR]I“ *¢

‘poul sue sIsuddu0g "Dy “p

‘jadoy snyosuye uopowmmAjey “¢

‘jadoy (aayeg) snyAydoidans “y °*z

‘Jadog (suyD'H) yned-ja-19Uapuy snruosoloy ‘|

-—- ANS

oral

sapods “nyeqeury juno) Jo avacepHiUUeID “¢ afquy

FERN GAZ. 17(4): 183-203. 2005

196

dIlUAapUd OdUOg

‘S| UOWOJOS 0} ByURT LIS dlWdpud Nyeqeury JUNO soutddiiyg 0) enewng BOUIND MAN 0} BIeUINS ‘S] UOLUOJOS 0} BsyRUINS dWdpUd Njeqeury JUNO dIWIapUd Nyeqeury iUNO| SROON|O|Y 0} OOWIOG ROUIND MON 0} BRUINS ify 01 eIsAeyeyy sepnsurusg dIWAapUs Njeqeuly JUNOPy ‘S] UOWO[OS 0} BISY 4S BOUIND MON O} URATE], BOUIND) MAN 0} B.YBUUINS ISIMEING O} BISARTRIA| "Ud d1WapUd Nyeqeury JUNO dWapus Njeqeury JUNO] ISAMBINS ‘OOUIOg

BOUIND MANY 0} OAUIOg sourddiiyd 01 vIsy 4S ROUIND MON 01 OOWIOg ROUINDH MON 0} BRUINS d1uIapus OsUI0g

ify 01 BISy AS

BIUOpa[eD MON 0} BARS OaWOg 0} BAyEUINS

‘S| UOWO]OS 0} BYU] US

AW

vs AWN AWN AW VS AWN AW JW dW dW

AW

AWA AW dW AW JW AWN AWT AWN AWT AW'T

Vs AWN

W OOET-OOTI W OQO81-00T1 W 0097-005 | WwW YOST-0091

W OObE

W007! W 00TE-O00E

W OOPE W OOS 1-00t1 W OOLE-00TE W QOLT-00S | UW QOOb-008 I W QOQIC-OOFI W OOPC-00F I

WwW O0TI W 0067-0021 W OOTE-O0S I W OOPE-OOEE W (Q0L7-0007 W OOOE-O0TT W OOSC-OOTI

WW 0O81-00S I Ww OOS I QOPT-O0LI QO0C-00F I UW QOLT

‘jadod saplonpieMulal ty “¢¢

SUN] | HIPIBMUTDI DT) *7C

‘Poul SLUBY BjOOTWIBI ‘FH “1 ¢

aunt g eBjnosnisoyid ‘fH *Q¢

‘jadoy (s9yRq) stsuasueped ‘F “Gp ‘jadoy (s94eg) BIL[ODOULIGO "DH ‘gp

‘pout sled Bjoorqnu DH “Lp

‘pour SLUeg BJOSIOSNU ‘DH ‘Op

“POUT SIUB_ SI[JOU fT) “Sp

‘Jodo (4axeg) BIEI[P00] ‘+H “pp

‘jadoy (4oyeg) vuLIpsoysinUy "DH “Ep ‘Jedo9 (edo) sisuonjeqeury 'D “Zp BME] (UYNY Xd “WoOY)) PUBLIOTEL ty “|p SLU (SY D"H) BSSIWONUI T “Op ‘jadoz essoidut tp ‘6¢

‘Jado tuinyjoy '*D “g¢

‘jodog (sayeq) HpuryiAey 'D “L¢

‘Poul SLUB_ BjoontUvIs ‘DH ‘O¢

‘}edoD (suy9"H) yned-19-19L19pLy °D “SE ‘Jodo (‘jedoD) erosoysijop 'D “PE auIn|g 19udTUOd ‘FD ‘EE

siivdg (‘[adod) ovtsuouiajo 1p “zZE

auinjg vsoydsavs ry “| ¢

‘jadoy (‘jedop) stsuv08u0g Dy “Y¢ owing (ounjg) essodspe siptuUBsn “67 ‘edo ([9Z) Bessipopunsasqns *> *gz wN}OH (Maply) BynuILUgns *> °2Z

naipsel, FAY" (HE) enpuedas *> “gz

197

PARRIS: GRAMMITIDACEAE OF MOUNT JAYA, NEW GUINEA

‘(WIQORE-OOSE) Son

-IUNLULUOD dUIdyEgns “WS “(QOS E-009Z) IS2404 SURIUOUL Joddn “4A (LUQO9T-OOT 1) ISAIOJ BULIVOW J9MO] ALT “(LUT 1-008) 805 |]!Y “AH

BOUIND) MIN O} BISAR[RIY ‘Ud soutddijiyg ‘oow0g

OoWOg 0} puRyey]

ROUIND MAN 0} OOWIOg souiddijiyg ‘oauw0g

BOUIND) MAN O} BRUINS ooUOg ‘eyRUINS

sesonbieyy 0) Bye] US

‘S] UOWIOJOS 0} puRpeYyy d1Wapus OdUIOg

BOUIND) MAN O} BISALRY “Ud BOUIND) MON O} BARS

BOUIND) MAN 0} BIPUT S BIUIND) MIN, 0} CAWOg BOUIND) MOXj 0} BISALIRI “Ud "S| sesanbseyy 0} eIpuy § BOUIND MAX O} puRyeY d1LUapUd OdUIOg

sourddijiyd 01 BIsy 4S

BOWS 0} BIPUT S

o1luapus nyeqeury jNop BOUIND MAN O} BILLINGS nyenur, 0} eayeung

AW)

AWN AWT AWN AWT

Wt (OTC-OOS | W (06T-OOLT W OOLT-O0TI W QOOE-OOET W OOIT

W OOFE-OOLT WwW OOPCT-O0F | W QOOE-O0S I W YOTE-OONS | WwW OOO!

W 009T-008 I W (06T-00SC Ut OORT-O0S | WOO1C-O0S I W OO8E-006C W 006C-00T1 W QOTC-OOS I W OOIT

W OOS 1-00T1 W008

Ww OOOE-O0FT W QO0O0E-00S I W O00E-008 I

‘jodo,y (owing) epyneuurdgns "yx *97 SLUvd (OD 3 BOW’) edivoipnu "x ‘G/ wnyyoH (ayo’a) usnuAuoaty "Xx “PZ SLueg (oyeg) SOploiIpoxey "| EL

stueg (‘jadod) asusuosjey “fp *7Z

sued (Suing) Wingelosey pT L

sluvg (Joyeg) USNS WUINTPOWY Lf ‘OL “MAIOPLY (UYNy) LUINIwOINS “S "69

“MIOpLY (atuN}E) wuNyjisnd “g “g9

“pour slueg wUNUUTTLOBIs “"S “79

‘MIAPLY (UYNS) a]IGop LUNssO[TOIIIIS “99 dd DW (azuny Xd dwn] g) BSO]NUDA ‘gq ‘S9 HOW (owing) eyenbiygo “q “p9

sued (jadod) eyepneonjnw “g “¢9

Jado) (1oxeE 7 “TONA) BOdBI|[BARP “qd *79 [SAld’ (Iss94'yH) ensyuod ‘g “19 ‘SIRM|'Y ZW BMBBKY (QWIN]E]) BIIGI]99 ‘gq ‘09 ‘POU SLUIBY SISUDDUIOG ‘qd *6S IUD W (aye) vypAydosyyeseg “g 8 IsLYyD'H (ouinyg) viepe endesosg *L¢ ‘poul SLUR BfOOTRUTRYIN "Fy ‘OCG

‘jadod (s94eq) BURBUINS "FD ‘SS

‘jadog (sayvg) sadnsiuqeos tH “pS

a

FERN GAZ. 17(4): 183-203. 2005

198

"S| UOWO]OS 0} vISy AS ROUIND MON 0} BBN ISOMB[NG ‘ooulOg o1WIapud CoUIOg

ift,y 0} eISY JS

OoulOg 0} ByBUUINS BOUIND MON 0} BISAReI “Ud BOUIND MAN 0} BISY FS ooulog ‘eyeunsg nyenue, 0} pueypreyy

“S| AJOI90§ 0} vIPU] nyenue, 0} vIsy AS ROUIND MON 0} OdUIOg BOUND) MON O} BALL

‘S] UOWO]OS 0} BISY 4S BOUIND MIN 0} BISY AS ‘S] UOWIOJOS 0} BISAR] RIA ‘Udg BOUIND MON 0} OoUIOg dIWIApUs OSUIOg saurddiyiyg 07 pueprey,y sedON]OP| 0} BISY FS

uoOnngLsig

AWN

qT

JANN AWT ‘SAWT AT

u0oNeI980A,

Ww 0091

W 0071 WOOLI

W 00S 1-006 W 0081-0001 WwW QOLI

W OOET

W OO0ET-00IZ W 00r

W (QOQOT-00EI W OQOTI-00Z WW QOS 1-007 1 W 00€T-0007 W 0081-0071 UW (081-00L1 W 0007-006 WwW 0061

tu O08 I

WW QOL1-00F1 W QO81T-00S1 W OO81-00T1

apmnty

BMBSEL (UYNY XO “Nal) BULLIOSel ‘FD “1 Z

‘jodod essoidutt 5 “QZ

‘JedoD GsuyD"H) Yned-y-1ovapLy 'D “6]

‘Jado (Jado) sisus0Buo0g ‘FH *g]

auinyg (ownjg) esiadspe sniutuesn */ |

wiN}OH CAJop]y) BINUIUIGNs ‘> ‘9]

‘WS 'f (euN]g) suBINU “Dp “S|

aZUNyY (OWIN|g 7 S9dN) BUIOSITJOW “> “p]

Sled (MIOp| VY) BISYOOIG *D “E]

WIN}JOH (MJaply) BSOUDAIAAIG *D *Z]

Japan 2 laude’ H’A\ (Ad) saprouydsaq sisajdouayy *| | sled (Joyeg) vuvlavsssnur siuuessosAsyy ‘| ‘pour Suey SNIpuosal“g *6

‘poul sLuvg snyeunoad “5

Jodo, (ulwog) snuviuasssany ‘5

‘Jado9 (994) silos “9

[SA1q*D (Suin|g 2 SdeN) snre]]Nond “> ‘jadoy (asnesg) smeordnpuos +5

Poul sie sisusauog UOpOWUATeD

‘jadoD (4axeq) snyAydojdans “y

Jodo, (asiyd"H) tned-j9-1oLepiy sniosolny

m™ o6

—Narwtune

sapods

‘NN, Bunun jo avaoepyiuwesy “p aquy,

199

PARRIS: GRAMMITIDACEAE OF MOUNT JAYA, NEW GUINEA

‘(WOOET-009|) ISd0} suRUOW J9ddn “AIA ‘(WOD6 | -OO) ISAO} OULIVOL JOMO] “YT “(LUQOR-OO]) Sa1OJUTeA puLpMO] “477

oauJog 0} puryieyy OdUOg 0} BISAR]R “Ud d1lUapus OaWOg soutddipiyg ‘oauiog sesanbieyy 0} ByUe] US "S] UOWO]OS 0} puLyeYyL d1Wapua OaUlOg d1lapua CaO g

‘S| Sesonbsey] 0} BIpUy § BOUIND MON O} pueyey L d1apua OdUOg soulddijiyg 01 BISy 4S BOLURS O} PIPU] S ODUIOG 0} BISARR “Ud BOUIND) MON O} BIRWING ISOMBINS 0} BIBUING d1luapua OaUIOg

‘S] UOWO]OS 0} ByUR] US

souiddiyiyg 0) eryeuins BOUIND MON 0} BAJBUINS ‘S] UOLUOJOS 0} B.BLUNS

ifi.J 01 BISAR]eYY JB]NsUIUDY

AW AT cig

AW

AW

AW’ AT JANN AWT dW

dW

AW dT dW AW AWN AW1 AT AWT 47 qT

AWN JNO AW

W OO8 1-00F W009

w 00S |

W O08 1

WW OO8 1-007 W 00-009 | W 0071-0001 W 0091-006 W O08 1-00 W Q0CTT-008 | W OOET

WW QO8-009

WwW OQOTI-O01 W QO

WwW QO0T

W 0007-008 I W Q00C-00F | WOOL

W 0007

W OOET

W 0071-0001 W Q00C

winnporH (ay gD) usnuiAuosaty "Xx “Ep ‘jadoy (‘sa,j) suaprusaye stuajdoydiy “Zp SUB (SLUR) DSUdYBMBILS “Lp

sled (jadoy) asuauosjey tuniaway | ‘Op “MOply (UyNY) WiNwd]Ns “S “BE

‘Map| Yy (auinyg) tunjjisnd “g “gE

“Poul SLUBY WIN] |IOBIT “S “LE

IYO’ (4oyeg) wintpojisses9 wiNsso[soI19]9g ‘O¢

[Sd (IS1O4'D) ENB UOd ‘g “SE “SIBM[ SY WZ BMBSL] (DUNE) BdIGQI]99 ‘g “PE ‘Poul SLUR SISUIDUIOY ‘g “EE

dH DW (oye) eyAydosyyeseg ‘g ‘7 ISLIYD' H (auinyg) eyeye endesolg “[¢ SLU (AY DD) BlOP BWIA Ty “OE ‘jadog (4ayegq) BuRABUINS TF “67

OUIN|E] BSOIDS Ty ‘QZ

‘jedod soplonpsemulal ‘D *LZ

suIN| | MMpseMulsl LT) “97

auinyg Bjnosnisoyid ‘fy “¢Z

‘Jado (4ayeg) sisuadueped ‘Fy ‘pZ ‘jadod (sayeg) BIR}odoURIGO ‘DH '€Z ‘JadoD (19yeq) BUBIPIOJsINUY *D “ZZ

FERN GAZ. 17(4): 183-203. 2005

198

"S] UOLUO]OS 0} BISY 4S BOUIND MON 0} BIVBUINS ISAMP[NS “OIUIOG SIWWUOpUd OdUIOg

fly 01 BISY 4S

OsUIOg 0} BIVBUINS BOUIND MON 0} BISARTRIA, “Ud BOUIND MON 0} BISY FS ooulOg ‘PBVeUUNS nyenuva O} puryiey

*s] AJOIN0S 0} BIPUT njenued 0} vISyY 4S BOUIND MAN 0} CdUIOg BOUIND) MAN O} BARS

‘S] UOLUOJOS 0} BISY FS BOUIND MON 0} BISY 4S "S] UOLUO]OS 0} BISAR]RIY “Ud BOUIND MAN 0} COUIOg SIWApUd OSUIOG sourddiyiyg 0} pueypreyy seoonjopy O01 BISV FS

uonNqLysiq

AW

dW

dW]

dW AWN

ATT

JINN AW ‘SANT AT

AWT AW] AWN AWT AWA SW AW AW AW

UORIIBIA,

WwW O09T

W Q0cI WOOL!

We OOS 1-006 W (081-0001 WOOL!

1 O0EC

W QOET-OOIC W OOP

W Q00C-00E! WwW Q0CT1-00C W QO81-O0TI W OQOET-O00T W 0081-001 W 0081-0021 W QO0C-006 WwW 0061 woos!

W QOL1-O0FI W O08 1-008 1 W 008 T-00CI

apmanty

BMeSeL (UYNY X9 “Wap\) BURLIOSel DH “|Z

‘Jado essaidut "pb “(QZ

‘Jado (suyD'H) yNed-1-19UapLy ‘DH “|

‘jadog (‘jadod) sisuso3u0g FH ‘g]

sun} g (ouwin]g) esiodspe situwelD */ |

wunyjoOH CMJaply) Binurwiqns “> *9|

‘wS'f (oun; g) suRInuU “yD “S|

azuny (SWIN|G 2 SddN) BUIODTTJOU “D “pI

SLB (MIOP[Y) eBISYOOIg “Dd “¢]

Wn}JOH (MIOPly) BSOUDAIADIG “D ‘Z|

JOYIOIN 2 19UsSeA\’ HA (ADIN) saprlouysayq siajdouay *| | SLUeg (JOyeg) VURIARISSNW SIIWULUPISOSAIYD “(| ‘poul sLUeg SNIPUOdaAI "Dy “6

‘poul slueg smeunsad yg

Jado (uluiog) snueiuassian] “QL

‘Jodo (994) sypoeis “9

[SAudg’) (OuIN]g 2 S2AaN) smeR]]NOND "Dg ‘Jado (asneig) smeoydnpuos yp

Poul SLB SISUIDUIOG UOPOLIWIATR “¢

‘JadoD (s9y4eq) snyjAydoidans y °z

Jado (suy)'H) Wned-ja-1ouaplLy sniosoly

—

saodsg

‘NYY Sununy jo asvaoepniwutuey “p apqey

199

GRAMMITIDACEAE OF MOUNT JAYA, NEW GUINEA

PARRIS

‘(WUQOEZ-009 |) ISes0} SURJUOUW! Jaddn “4A “(WD 1-008) ISAUOJ OURJUOL IOMO] “IAT “(LUQOS-OO |) asofures puL] MO] “477

oOdUOg 0} puRyey | OIUIOG 0} BISAR[RIA, “Ud dIWOpUd OdUIOg soulddijiyg ‘ooulog sesonbivyy 0] ByURT] LS “S| UOLUOJOS O} puLpeYy SIWIAPUd OdUIOG dIWIapUd OdUIOG

“s] sesonbieyy 0} eIpuy § BOUIND MON OF puR[eyL dIWIDpUd OdUIOG soulddijiyd 01 eIsy 4S BOIS 0} BIPU] S oOdUIOg 0} BISAR[RI “Ud BOUIND MON O} BVRUUNG ISAMPING 0} BIVRUUNG dIWIAPUS OSUIOG

‘S] UOWO]OS 0} ByURT US

soutddiyiyg 0} Byewing BOUINDH MAN O} BATBUINS “S] UOLUOJOS 0} BIVBUINS

iff 0) BISkee Je[NsuIUdg

AW dT al

AW)

JWT

AW 47 JW) AW AW

AW

AW 47 AW) AW) AWN AW’ AT AAT dT AT

JWA JW) AW JW) AWT AW] JIA JW AW AWN

W QO81-00P WwW Q09 woos!

WwW 0081

W QO8 1-007 W QOPC-0091 W Q0T1-0001 W 0091-006 W QO81-00S W QO0CC-O08 I W 00ET

W (08-009 WQ0cI-O001 W (YOT

W QOOT

W QOO0C-O008 I W Q00C-O00F I W OOLI

W Q00T

W (QOET WOQO0cI-O0001 W (Q00T

wnyjOH (41YyD'D) HsnwiAUOsoIY KX “Ep ‘Jadoy (sag) suspiusayye suaidoydiy “Zp SILIVG (SLUR) ISUDYBMPBILS “| | P

sued (Jado) asuduosjey wuNIaWaYy_] “Op “MIOpLy (uyNy) WiNnwes[Ns “Ss ‘“6¢

‘MJaplV (auiN|g) uiNnifisnd “s “gE

“POUl SLUR WINLUT]IOBIS “S “7 ¢

IYO (dye) WiNOFIsseid WINssO[SOII[IS “OE

[SAlg’D (1S104'H) eNSNUOd ‘g “SE ‘SIEM SY AW BMEBL] (DUNG) BdIGaI{29 ‘g “PE ‘Poul SLUBY SISUIDUIOG ‘q “€¢

dd DW (oyeg) ey Aydosyreseg “g “7E SYD H (ouinjg) eieye endesolg “1 ¢ SLUR (IY DD) BYOJIENIA D “OE ‘jadog (s9yeg) BuRIRUUNS FD “67

UN] BSOIdS ‘FD “BZ

‘jadoD saplonpsieMuldal D * LZ

oe eee a

auinyg eynosnisojid y “SZ

‘jadoy (4ayeg) sisuosueped 1 “PZ ‘jadoy (49y4Bg) BIR[OIDURIGO FD “¢Z ‘jadoD (A9yeg) BuRIPIOJsINUY DH “ZZ

FERN GAZ. 17(4): 183-203. 2005

200

eV %00 “0 ILL %IT 6 %IS “CT %CI “Ss

nny sununsy

OL Ss OS [B10] salads %Cl “6 %91 "6 %0L MWapus AY]EIO] *¢ %OL “S %0T “I %IE “OT 1Wapus ANUNO) “p %el “OL %9I “6 %0L “YP (WD “MID BISoTRI MA JO CU PIN) BISOTRI AE “8S “Pr 49S “TE Ep “PT OUI] S,JISUL]JOZ SsO1OK JsRa] Iv JO ‘puOAd ATIYSIS JO apIM-vISaTR *Z %I1 ‘8 %06 ‘S %OL ‘Pb ByUR] LIS 1O/pue eISOUATOd 0} “BISaTRIY puoOAdQ peaidsapiAA *| nyeqeury yA, OSpry apoyany, =—s BARE JUNOT uOnNLysip s1ydess005

sa1dads [e}0) JO 9dv)UIII0d puK ‘ou SaIdads

npn sununy pue nyeqeury nop) ‘aspry spayinypy “VAs JUNOPY JO svadepHIUWeIDH JO UONGLASIG d1Yydeis0ayH “s aquy

PARRIS: GRAMMITIDACEAE OF MOUNT JAYA, NEW GUINEA 201

DISCUSSION

Compared with Murkele Ridge, Mount Kinabalu and Gunung Mulu, Mount Jaya has many more country endemics. This reflects the explosive speciation of Grammitidaceae in New Guinea (together with many other groups of plants) and their distribution along the central ranges of the island. The wide distribution of species along the central ranges of New Guinea is probably also an explanation for the low number of endemic species compared with Mount Kinabalu and Murkele Ridge. While Mount Jaya is the highest peak in a series of high altitude peaks on the ranges that form the backbone of New Guinea, Murkele Ridge and Mount Kinabalu are isolated from other areas at high altitudes. Mount Jaya and Gunung Mulu have fewer species that are widespread in Malesia and beyond than do Murkele Ridge and Mount Kinabalu. Gunung Mulu, being relatively low altitude, does not have the array of habitats or even comparable areas of the same habitats as do the higher Murkele Ridge and Mount Kinabalu. The reason for Mount Jaya having fewer of the widespread species is less obvious. In the discussion of Murkele Ridge (see above), seven widespread species are mentioned that are absent from Mount Jaya, but present in New Guinea and thus to be expected. Another two species that occur on Mount Kinabalu and are disjunct with Papua New Guinea are also expected on Mount Jaya. They are Calymmodon atrichus and Grammitis clemensiae.

There is a group of 10 grammitid species that have been collected on all of the four montane Malesian localities documented here: they are Calymmodon cucullatus, Chrysogrammitis musgraviana, Ctenopteris blechnoides, C. nutans, Grammitis knutsfordiana, G. padangensis, G. sumatrana, Prosaptia contigua, Scleroglossum pusillum and S. sulcatum. Two other widespread high altitude subalpine species are not found on Gunung Mulu (2400 m alt.) because it is too low: Grammitis locellata (2800-3800 m) and Prosaptia davalliacea (2800-3800 m). Another group of species, Calymmodon clavifer, Ctenopteris denticulata ,C. millefolia, C. subsecundodissecta, Grammitis dolichosora, G. intromissa, G. scabristipes, Prosaptia obliquata, P. venulosa and Xiphopteris subpinnatifida occur on Mount Jaya, Murkele Ridge and Mount Kinabalu within the altitudinal range of Gunung Mulu and may be expected there. The widespread Grammitis reinwardtii is absent from Mount Jaya, but occurs in the other three localities. The species that are absent from Gunung Mulu and from Mount Jaya, together with the 10 listed above, form a core of 23 grammitid species that are widespread and common throughout Malesia. Examination of herbarium collections during preparation of the account of Grammitidaceae for Flora Malesiana has provided much detailed data on species distribution that has been collated for over 500 localities. The core group of 23 grammitid species is well-represented in the following localities, each of which has 20 or more species of Grammitidaceae. The altitude range over which Grammitidaceae have been collected and the number of species currently known is given for each locality. Peninsular Malaysia: Cameron Highlands, 1400-2000 m, 23 species; Gunung Tahan, 600-2200 m, 21 species; Sumatra, Gunung Kerinci. 300-3800 m, 22 species: Java, Gunung Gede, 1200-3000 m, 39 species; Sulawesi, Mount Roroka Timbu, 1200-2400 m, 25 species: Philippines, Mount Apo, 1000-3000 m, 25 species: Papua, Mount Trikora, 2800-4200 m, 36 species, Idenberg River, 700-2200 m, 22 species; Papua New Guinea, Aseki, 1300-1800 m, 20 species, Ekuti Range at Watut-Aseki Divide, 2000-2100 m, 22 species, Mount Giluwe, 2100-4100 m, 38 species, Mount Hagen, 2400-4000 m, 34 species, Rawlinson Range, 1500-4000 m, 33 species, Saruwaged Range, 1200-4000 m, 40 species, Star Mountains, 1400-4000 m, 24 species, Mount Suckling, 400-3700 m, 20 species, Mount Wilhelm, 2700-4500 m, 26

202 FERN GAZ. 17(4): 183-203. 2005

species.

The identification of pteridophytes below the subalpine and alpine areas of Mount Jaya is not yet complete (R.J. Johns pers. comm.), so a complete evaluation of the relationships of the pteridophyte flora is not yet possible. Ridley (1916) records a number of pteridophytes from Mount Jaya, however, and this is sufficient to establish that there is a core of widespread pteridophytes i in families other than Grammitidaceae that are also present on Mount Kinabalu (Parris ef a/., 1992), Murkele Ridge (Parris, 1993) and Gunung Mulu (Parris, unpublished). They include Asplenium tenerum G. Forst., Diplazium cordifolium Blume, Lindsaea rigida J. Sm. (Kato, 1992), Lycopodiella cernua (L.) Pic. Serm., Ophioglossum pendulum L. (Kato, 1989b), Selliguea enervis (Cav.) Ching and Trichomanes pallidum Blume.

CONCLUSIONS

Mount Jaya has some unique features compared with the other Malesian mountains considered here: the presence of a rich grammitid flora in the heath forests between 200 and 600 m alt. is unexpected, and there is a good representation of widespread New Guinea endemic Grammitidaceae. The absence of high altitude endemics in the family is surprising, given that several other high New Guinea mountains have one or more endemic species. Mount Jaya is much less species-rich than Mount Kinabalu, and this may be explained at least in part by the much longer history of collecting on the latter and the paucity of collections from the largely inaccessible upper montane forest of the former. Compared with Mount Jaya, the Murkele Ridge, 2000 m lower, supports a rich flora of Grammitidaceae, but the reason for this is not evident.

Zollinger’s line does not present an insuperable barrier to the distribution of Grammitidaceae, as 59 (45 %) of the 130 species listed in Tables | to 4 occur on both sides of it.

The concept of a core Malesian-wide pteridophyte flora present in all suitable areas has not been suggested previously, but there is ample evidence for it in 23 species of Grammitidaceae and numerous species in other pteridophyte families.

ACKNOWLEDGEMENTS I wish to thank Robert Johns and Peter Edwards (K) for organising the loans of Kew specimens, Alison Paul (BM) for seeking out Kloss’ collections, the Keepers of BM and K for the loan of material and Ewen Cameron (AK) for facilitating the loans.

REFERENCES

JOHNS, R.J. et al. in prep. A guide to the subalpine and alpine flora of Mount Jaya, New Guinea. Royal Botanic Garden, Kew.

KATO, M. 1989a. The fern flora of Seram. In: BAAS, P., KALKMAN, K. & GEESINK, R. (Eds) The Plant Diversity of Malesia, pp. 225-234. Kluwer, Dordrecht.

KATO, M. 1989b. Taxonomic Studies of Pteridophytes of Ambon and Seram (Moluccas) Collected by Indonesian- Japanese Botanical Expeditions. III. Eusporangiate and Some Lower Leptosporangiate Families. Acta Phytotax. Geobot. 0: 77-92.

KATO, M. 1992. Taxonomic Studies of Pteridophytes of Ambon and Seram (Moluccas) Collected by Indonesian-Japanese Botanical Expeditions. IX. Woodsiaceae, Lindsaeaceae, and Adiantaceae. J. Fac. Sci. Univ. Tokyo Sec. III, 15: 135-152.

PARRIS: GRAMMITIDACEAE OF MOUNT JAYA, NEW GUINEA 203

KATO, M. & PARRIS, B.S. 1992. Taxonomic Studies of Pteridophytes of Ambon and Seram (Moluccas) Collected by Indonesian-Japanese Botanical Expeditions. VII. Grammitidaceae. J. Fac. Sci. Univ. Tokyo Sec. III, 15: 111-133.

PARRIS, B.S. 1983. A taxonomic revision of the genus Grammitis Swartz (Grammitidaceae: Filicales) in New Guinea. Blumea 29: 13-222.

PARRIS, B.S. 1993. Studies on the Phytogeography and Altitudinal Zonation of the Pteridophyte Flora of Seram. In: EDWARDS, I.D., MACDONALD, A.A. & PROCTOR, J. (Eds) Natural History of Seram, Maluku, Indonesia, pp. 75-89. Intercept, Andover.

ARRIS, B.S. 2001. Taxonomy of Malesian Grammitidaceae in relation to ecology and phytogeography. In SAW, L.G., CHUA, L.S.L. & KHOO, K.C. (Eds) Taxonomy, the Cornerstone of Biodiversity, pp 155-160. Forest Research Institute Malaysia, Kuala Lumpur.

PARRIS, B.S. unpublished. A Checklist of Pteridophytes from Gunung Mulu National Park.

PARRIS, B.S., BEAMAN, R.S. & BEAMAN, J,H. 1992. The Plants of Mount Kinabalu. 1. Ferns and Fern Allies. Royal Botanic Gardens, Kew.

RIDLEY, H.N. 1916. Report on the botany of the Wollaston Expedition to Dutch New Guinea. Trans. Linn. Soc. London (Bot.) 2, 9: 1-269.

WALLACE, A.R. 1869. The Malay Archipelago. Macmillan, London.

204 FERN GAZ. 17(4). 2005

MORPHOGENETIC EVENTS IN THE CERATOPTERIS RICHARDIT (PARKERIACEAE: PTERIDOPHYTA) SHOOT APEX

M. ROMANOVA! & J. JERNSTEDT”

Department of Agronomy and Range Science, University of California, David, CA 95616, USA. (Email: 'maromanova@ucdavis.edu; “jernstedt(@davis.edu)

Keywords: Shoot apex meristem, shoot apical cell, leaf apical cell, leaf primordium, plasmodesmata, morphogenesis, development, dictyostele, protostele

The discovery of developmentally important genes, combined with remarkably regular cell division patterns, has made the Ceratopteris richardii sporophyte an advantageous system for developmental studies. Here we report results of an investigation of the ontogenetic changes in shoot apical meristem (SAM) structure associated with leaf initiation and histogensis in C. richardii. As described for a number of other ferns, the shoot apical cell (AC) and leaf apical cell (LAC) of the first leaf of C. richardii embryos appear simultaneously; all subsequent sporophyte leaves originate from the SAM. An LAC originates in every immediate derivative of the tetrahedral AC. After this segment is displaced from the AC by two, more recent derivatives, it undergoes two anticlinal divisions and one transverse division, setting up the LAC. The pattern of LAC origin does not change during sporophyte ontogeny, but the rhythm of leaf development does. For the first leaves of a young sporophyte, the subsequent division of the AC is correlated with further proliferation of cells of the incipient leaf primordium (LP), and the AC does not undergo further divisions until the LP is established. Development of the first 6-9 leaves is followed by a developmental pause of 12-20 days, during which the shoot apex changes its structure and morphogenetic pattern. LPs are arrested at the 4- or 5-cell stage until displaced some distance from the AC, which continues to segment and creates an increasing “pool” of early stage LPs. This results in an elongated shape for the shoot apex. These changes are correlated with changes in the shoot vascular system. A protostele is found in the young sporophytes with simple leaves developing without pause; each leaf has a unibundle leaf trace (LT). Dictyosteles occur in older sporophytes bearing compound leaves; these leaves start to develop when separated from the AC by a number of incipient LPs, and have 2-5 bundles per LT. Procambium differentiates simultaneously with LP development: no signs of vascular differentiation are observed in either the juvenile or the adult shoot apex. Ontogenetic changes in stelar type might be explained by the loss of meristem identity and competence to differentiate into procambium by the central cells of the apex. This is in agreement with Sano et a/. (unpublished observation) that KNOX class | genes are expressed in the AC and procambium of C. richardii adult sporophytes (but not in the other cells of the elongated apex). Because plasmodesmata (PD) provide selective routes for signaling within the shoot apex, we examined PD architecture and distribution in the C. richardii SAM. All PD in the C. richardii SAM are primary. PD density in the AC and its immediate derivatives does not change significantly in sporophyte ontogeny but is about ten-fold higher than reported for dicot SAMs. High PD density and the resulting intercellular connectedness may compensate for the lack of ability to form secondary PD. Only the AC and its two immediate derivatives, the latter being constantly displaced, are interpreted to be undifferentiated in the C. richardii SAM.

FERN GAZ. 17(4): 205-215. 2005 205

REPRODUCTIVE STRATEGIES AND POPULATION STRUCTURE IN THE ENDANGERED PTERIDOPHYTE TRICHOMANES SPECIOSUM (HYMENOPHYLLACEAE: PTERIDOPHYTA)

FJ. RUMSEY’, J.A. BARRETT”, M. GIBBY’, S.J. RUSSELL! & J.C. VOGEL!

'Dept. of Botany, The Natural History Museum, Cromwell Road, London, UK. “Dept. of Genetics, University of Cambridge, Cambridge, UK; *Royal Botanic Garden Edinburgh, 20A Inverleith Row, Edinburgh, UK. (Email: F.Rumsey@nhm.ac.uk)

Keywords: Glacial refugia, gametophyte, historical biogeography, dispersal, filmy-

ABSTRACT

Trichomanes speciosum Willd. (Hymenophyllaceae) is unique amongst European ferns in that the gametophyte generation can survive indefinitely in the absence of the sporophyte, propagate vegetatively and disperse locally. Reproductive success, both in terms of spore production and sporophytic recruitment currently vary widely across the species’ broad, disjunct range, declining from south to north and west to east. Reproductive success would appear to be controlled largely by climatic factors, although genetic components also must be considered. Detailed study of populations throughout the species’ range, over a 15 year period, have led to a greater understanding of growth rates, powers of dispersal and the reproductive strategies currently operating. Using this knowledge, the extent and pattern of genetic variation regionally and locally, i.e within sites, can be used to infer the routes and mechanisms of colonisation and subsequent reproductive history.

Many sites for this species have been considered as glacial refugia, 1.e., supporting relictual populations through cycles of glaciation throughout the Tertiary. The validity of these claims is tested using molecular and other data.

INTRODUCTION

Trichomanes speciosum Willd. (syn. Vandenboschia speciosa (Willd.) Kunkel), the Killarney Fern, is the sole native European representative of the genus 7richomanes sensu lato (Tutin et al., 1993) and the largest of the region’s filmy-ferns (Hymenophyllaceae). It has been considered one of the continent’s most vulnerable lant species, threatened both by habitat destruction and long the victim of over-collection, and has been accorded legal protection throughout its range under the

Bern Convention (Anon, 1979) and E.C. Habitats Directive (Anon, 1992). Trichomanes speciosum is unique amongst European ferns in that its gametophytic generation is perennial, gemmiferous and capable of persisting and dispersing in the absence of a sporophyte. This “gametophytic independence’ is apparently a very rare condition amongst homosporous ferns, and known in three families, all widespread as epiphytes in tropical regions: the Vittariaceae, the Grammitidaceae, and the Hymenophyllaceae. However, gametophytic independence is apparently only well developed in those rare species found at the temperate extremes of these families’

206 FERN GAZ. 17(4): 205-215. 2005

ranges (Rumsey & Sheffield, 1996).

Independent gametophytes of a Trichomanes species were first tentatively identified from eastern North America by Wagner & Evers (1963) and confirmed by Farrar (1967, 1985, 1992) who identified a distinct taxon, 7 intricatum Farrar (Farrar, 1992) represented solely by gametophyte populations.This taxon was found to exist far to the north of the line of the last glacial maximum, indicating it clearly had the ability to disperse over some distance, even if sporophytes and hence spores were currently not produced. The other widespread North American species of Trichomanes have been found as gametophyte populations but always in close proximity to sites in which sporophytes are present (Farrar, 1993). The situation in Europe falls between these extremes and differs in scale, in that both generations of 7. speciosum are known, but the gametophyte has an extensive range (hundreds of kilometres) beyond that of the sporophyte.

Reproductive Potential and Strategies

The sporophyte of 7 speciosum has a long-running, branching rhizome, at intervals bearing fronds which are functional for 4 or more years, and may persist for longer. Rhizome growth rates are heavily dependant on the plant’s nutrient status and climatic environment but main axes may elongate by over 30 cm per annum, although usually by much less. Suppressed side branches borne in the sporophyll axils allow for steady patch formation should the leader not find suitable habitat and die. This growth form allows a guerrilla-type growth strategy similar to that employed by species of Lycopodium to exploit patchy environments (Callaghan et al.,1986). Specialised mechanisms of vegetative propagation are absent (viviparous plantlets have been recorded twice in cultivated material over two centuries of observation (Druery, 1910; Ensoll, pers comm.)). Rhizome fragmentation, dislodgement and re-growth are likely to be infrequent but significant events. Extensive, intricate and usually dense patches form with time. Within these it is generally impossible to determine the number of individuals present without recourse to molecular techniques. Individual clones covering several hundred square metres have been recorded by the authors in the Azores, obviously indicative of considerable antiquity. However, colony size is not necessarily an accurate indication of age. Several British colonies known for over a century are very restricted in size, their maximal extent presumably determined by the extent of the micro-features that they grow in.

Unlike many European ferns Trichomanes speciosum is not fertile annually once mature. Indeed Ratcliffe er al., (1993) reported that the majority of British sporophytes had never been seen to be fertile. Detailed investigation of fertility is difficult and in many cases not desirable because of the nature of the sites occupied. However, during the last decade four Welsh and two English sporophytes known to the authors have at least attempted to produce spores. Involucres have formed but the receptacular bristle upon which the sporangia are spirally borne has, in at least two of the sites, not extended and no spores have formed as a consequence. This receptacular bristle structure has a basal meristem and presumably continues to elongate and produce new sporangia until a certain trigger, perhaps desiccation causing sporangial dehiscence, causes it to cease. Its failure to develop in marginal British sites may have climatic cause, such as exposure to cold. This and the factors determining the onset of fertility are almost certainly complex and have not been established for 7. speciosum. Anecdotal evidence would seem to suggest that sporophytes in extreme shade are less regularly fertile and,

RUMSEY et al.: REPRODUCTIVE STRATEGIES IN T: SPECIOSUM 207

as most plants at the more northern and marginal extremes of the species range are progressively restricted to more sheltered and hence usually darker microsites, this alone may account for the reduction in northern fertility. Considerable variation exists in the degree of fertility, i.e. the number of involucres produced — dictated by frond stature and dissection, and the number of sporangia produced per involucre, as outlined above. Sporophytes in northern and more easterly British sites are invariably small and relatively poorly dissected, reducing sporing potential. Calculations based on observation of the number of fertile fronds known in all British populations for one year, an estimated mean number of involucres per frond and numbers of sporangia per involucre, suggested that the total British spore production per decade would be considerably less than that produced by a single healthy adult Athyrium filix-femina in one year (Rumsey, 1994) ! The species is thus likely to be a poor coloniser by spore or vegetative means.

Sporophyte recruitment parallels the pattern of spore production in that there is a marked reduction in the production of sporophytes northwards, 1.e. into northern France and the British Isles and even more markedly, eastwards into continental Europe. Thus within the wetter portions of Macaronesia the species functions as a “normal” pteridophyte, completing its life cycle from spore to spore, whereas elsewhere the cycle progressively breaks down, until over much of the species extensive distribution in northern and continental Europe the gametophyte is effectively functionally independent. Immature sporophytes have rarely been noted within the British Isles and show high levels of mortality (Rumsey, 1994), although recent years have perhaps shown an increased incidence in their production (“Sentinel”, 2003; Rumsey, unpubl.) which may be a reflection of the run of warm but wet summers and milder winters experienced in the British Isles. Immature sporophytes have also been reported in central Europe, eg. Rasbach er al., 1994 which indicates that gametangial production and normal function are occurring in at least some “independent” gametophytic clones. Given a potentially indefinite gametophytic lifespan the probability must be that sporophytes would form, although perhaps extremely infrequently, in the vast majority of populations.

Sporophytic production and recruitment almost certainly is predominantly a reflection of the rate and extent of gametangial production. The archegoniophore is a comparatively massive, differentiated, multicellular structure unlike the simple irregular branching filamentous structure it is borne on. The processes triggering its initiation are unknown but rates of cell growth and light/nutrient availability might be implicated. The smaller antheridia are more regularly produced and may persist in a non-dehisced state for many months. Rumsey & Sheffield (1996) summarised knowledge on the in situ and ex situ production of gametangia by different British gametophyte clones over a 5 year period. They identified consistent sterility, or super-fertility, of some clones, even when sub-divided and grown apart under similar but not rigorously controlled growth conditions, suggesting a significant genetic component to gametangial production, but with no clear geographical, ecological or discernible historical basis. Locally this may complicate considerations of the effect of climatic and other external factors on gametangial production but broad regional differences, which are reflected in sporophyte recruitment, are still apparent.

Understanding the Current Distribution- Refugia or Recent Colonisation? First detected as recently as 1989 (Rumsey ef al., 1990; 1991). subsequent survey has

208 FERN GAZ. 17(4): 205-215. 2005

revealed the gametophyte to be widespread but extremely localised within the British Isles, significantly occurring in many areas from which the sporophyte had never been reported (Rumsey et al., 1998). Similarly, as the distinctive habit and habitat of the gametophyte became known, discoveries across northern continental Europe rapidly followed, e.g. Vogel et a/., 1993; Rasbach ef al., 1994; Bennert et al, 1994; Bujnoch & Kottke, 1994: Kirsch & Bennnert, 1996; Horn & Elsner, 1997; Huck, 1997; Reichling, 1997; Boudrie, 2001; Krippel, 2001, Loriot et a/., 2002, Kottke, 2003, etc.

Initial reports were from deeply dissected sandstone massifs, such as the Petite Suisse of Luxembourg, the northern Vosges and the Elbsandsteingebirge, all areas noted for the occurrence of other rare, disjunct ‘Atlantic’ taxa (sensu Ratcliffe, 1968) but particularly the filmy-fern Hymenophyllum tunbrigense (L.) Sm. (Richards & Evans, 1972). In these disjunct areas H. tunbrigense has been considered to be a Tertiary relict (Drude, 1902; Klein, 1926). Subsequent examination has revealed the gametophyte of Trichomanes speciosum to be more widely distributed and in less “exceptional” landforms and habitats in intervening regions in western and Central Europe, such that the recorded distribution of the species is now much less discontinuous and more extensive than previously realised. This raises interesting questions as to when this broader gametophyte distribution was achieved and from where?

The broad continental and northern-European distribution of Trichomanes speciosum may have been achieved as a result of a wave of post-glacial migration from southern and western refugia, or conversely from limited diffusion from a range of more northerly peri-glacial refugia, or indeed result from an amalgam of both. Sites containing gametophytes alone may thus represent recent arrivals, or the tenacious survivors of a species uniquely suited to coping with a sub-optimal climate. Sadly, few relevant sub-fossil or palynological finds are known, or are realistically to be expected. Comparison with the distribution of other taxa can be informative, but elucidation of the species quarternary history is largely reliant on molecular studies of the pattern, extent and structure of genetic variation throughout the species range. Knowledge of the comparative reproductive and dispersive capabilities of the different generations can then be used to infer whether sporophytes were likely to be present historically in areas from which they are now unknown.

Generally our expectation must be that sites which have been suitable for the peri-glacial survival of a particular species are likely to have allowed the survival of a Suite of taxa, and that this community, or many of its distinctive components may well recur in other disjunct refugial sites. The peculiar biology of Trichomanes speciosum and its ability to grow in conditions that are not conducive to the survival of all other vascular plants may, however, mean that it is a special case.

We also expect refugial sites to support greater genetic diversity than those which have been recently colonised. A potential confounding factor is that of severe bottlenecking — low or no diversity may result from loss over time/genetic drift, as well as by founder effects. We must also consider that an_ efficiently vegetatively-reproducing organism of narrow ecological amplitude might effectively block niches and thus prevent the arrival of other individuals (“the first in wins”). Variation may then be reliant on the rates of mutation in the various ramets, although these would be expected to accrue with time, meeting our original expectation. Ancient refugial sites, if isolated, might be expected to show private variation, i.e. unique to them, although post-glacial dispersion, if efficient, will obscure such patterns.

RUMSEY et al.: REPRODUCTIVE STRATEGIES IN T. SPECIOSUM 209

Refugia or Not?

Several disjunct areas are now known to support very small populations of Trichomanes speciosum, many of which have been discovered comparatively recently. Within these the relative proportion of sporophytes vs. gametophytes differs but many show comparably low levels of sporophytic presence and association of a suite of Atlantic species. Do they and their associated species show similar levels and patterns of genetic diversity and what can this reveal as to their history?

Southern Spain

Trichomanes speciosum was first discovered in Southern Spain by Richards (1934) and is now known from a very limited range of sierras leading inland from the coast near Algeciras, northwards to near Cortes de la Frontera (Rumsey & Vogel, 1998; Garrido Diaz & Hidalgo Maqueda, 1998). Currently nine populations are known to be extant, more than half (5) of which are thought to support fewer than 10 discrete sporophyte clumps each. Only one site (Sierra de Montecoche) currently contains gametophytes in the absence of sporophytes. The species main concentration is in the Sierra de Ojén and its westward continuation, the Sierra de Saladavieja. Here a total of approaching 200 distinct mature regularly fertile sporophytic clumps, closely associated with small localised areas of gametophyte and numerous juvenile sporophytes, occur in 12 distinct sub-populations by parallel small watercourses. High levels of mortality, with extreme reductions, at least in sporophyte populations, linked with the cycles of cork-oak harvesting and climatic variation, have been documented in the recent past (Molesworth-Allen, 1977).

Allozymic study of material from 3 sites: Sierra de Ojén (4 sub-populations), the Valle del Rio de la Miel and the Sierra de Montecoche, indicates that each site ha alleles not shared with the other sites (Table 1) and that the large Sierra de Ojeén site contained at least 16 different multilocus phenotypes (MLPs), 10 of which were present in the largest sampled sub-population. The observed differences between closely adjacent watercourses may indicate low levels of gene-flow even over such restricted ranges but patterns indicate the likelihood of intergametophytic mating within these sub-populations, with the fern fully completing its life cycle and behaving ina “normal” pteridophytic fashion, unlike populations further north in its range.

These small populations co-exist with a limited range of Macaronesian bryophytes unknown elsewhere in Mainland Europe (Rumsey & Vogel, 1999) as well as other pteridophyte taxa, including Culcita macrocarpa, Diplazium caudatum, Pteris incompleta and previously Drvopteris guanchica — the latter suggesting a closest affinity with the Canarian flora. This floristic enclave, with fragmentary representation on the adjacent North African coast and into the Rif mountains, either represents the vestige of a once wider coastal African/C anarian/Iberian distribution fragmented by aridification and the increase of the Saharan desert area, or, and less likely, represents independent post-glacial (re-)colonisation by the many and varied Macaronesian floral elements. All potentially are highly dispersible given the small size of their diaspores.

Alpi Apuane, Italy

Trichomanes speciosum was found, new to the Italian flora, only as recently as 1976 (Ferrarini, 1977). Subsequently a total of five “populations” have been detected growing within the stream gorge of the Valle de Serra and its side valleys, close to the town of Serravezza (Marchetti, in Ferrarini et al., 1986). Few discrete individual mature

210 FERN GAZ. 17(4): 205-215. 2005

Table 1. Distribution of multilocus phenotypes (MLPs) in southern Spanish Trichomanes speciosum populations (figures represent the number of apparently discrete individuals possessing that MLP).

Allozyme Phenotype Site

IDH | 6-PGD| MDH | Miel | Ojén 1 | Ojén 2 | Ojén 3 | Ojén 4 sina -cocne

Z

O =

>

og too 4 (Ml of ne 3) Fa > fo] oo a OO bo No Ww A ee = |

wl) wl] ml] wm] wm} Pm] PS] SY] SY] SE] YP] PS S| S| CY CF Co] we me) me] el yO me Om Oe Oe Oe >| A] DW] Wl S| wl] Se] UO] WW] wD] W| BD! S| | &

| _

sporophytes are known but at least one is extensive and regularly highly fertile. The known sporophytes share a distinctive and apparently unique morphology.

Initial researches indicate a wider distribution of the gametophyte (Rumsey & Vogel, unpubl.), perhaps paralleling that of another disjunct Atlantic element, the filmy-fern Hymenophyllum tunbrigense, for which ca. 70 sites are known in the communes of Massa, Montignoso, Stazzema, Pietrasanta, as well as Serravezza (Pichi-Sermolli, 1936; Marchetti, 1992). Sporophytic recruitment has been observed in sites containing mature sporophytes but not in sites currently only known to support gametophyte populations.

Unfortunately only a limited range of material has been available for analysis. Both localities from which multiple samples have been taken support more than one MLP (3

) with patterns consistent with the occurrence of intergametophytic mating.

RUMSEY et al.: REPRODUCTIVE STRATEGIES IN T. SPECIOSUM 211

Sporophytes are currently unknown in the most genetically diverse of these sites. One allele in the enzyme PGM has not to date been recovered in any other Trichomanes speciosum population throughout its global range.

All the evidence would thus tend to argue that this area must be considered refugial.

Southwest Scotland

In contrast to the Iberian and Italian sites this area was heavily glaciated and the species occurrence here must result from post-glacial (re-) colonization. It is thus instructive to compare the patterns of distribution and extent of genetic variation with the almost certainly refugial areas above and more equivocal areas.

Fewer than 20 discrete sporophyte plants have been detected in Scotland since its first discovery in 1863 (Babington, 1863) and less than half this number are known to be extant. Only one of these colonies has ever been seen to produce spores and most are, or were, very limited in size and extent. However, the gametophyte generation is widespread and may be locally abundant. Recent recruitment of sporophytes has been seen to occur both in sites containing mature sporophytes and also in at least one site were these were absent.

Genetic diversity of the species at this the northern-most limit of its distribution was studied by Rumsey et al., (1999). 121 samples from 37 populations in 22 sites were examined and 7 allozyme MLPs were detected. Two of these MLPs covered large areas (>30 km in total extent), while the others were restricted to one, or rarely few, localities. No site was found to contain more than one MLP. Whether this is evidence for blocking of available niches by the first arrival, or merely a reflection of the extreme rarity of propagules of other MLPs arriving in occupied sites is unclear and may require experimental verification through deliberate introduction. The extent of MLPs was not correlated with presence /absence of sporophytes. Where they extend over many kilometres, in some cases on separate islands, the only credible means of dispersal is by spore. Where sporophytes are currently unknown such a distribution indicates their past presence, and a greater fertility than in the recorded past. Sporophytes may have occurred as long ago as in the sub-atlantic period, or during brief warm/wet phases subsequently — such as we appear to be experiencing currently. Very broad distributions must indicate considerable age given the poor dispersal powers shown by both generations of this organism. It is tempting to speculate that geographical extent could be correlated with arrival time but many conflicting factors have also to be considered.

Communities in glaciated and un-glaciated areas share many rare and disjunct taxa and their presence alone is un-informative as to refugial status, but the population structure and genetic make-up of these taxa is much more revealing. Many of the Macaronesian bryophytic taxa show disjunctions to southwest Scotland, but the majority of these diocious species are represented by one sex alone in the British Isles, yet exist in mixed sex populations and fruit in southern (refugial?) areas. Such a pattern might be expected given extremely rare long-range post-glacial dispersal events to the nort

Central Europe

Rumsey er al., (1998) examined allozymic diversity in 35 sites from the then easternmost known extent of the species distribution in the Elbsandsteingebirge on the Czech- German border, through Germany to the French border in the northern Vosges mountains. The extensive distribution of some MLPs, as in Scotland, indicated that

212 FERN GAZ. 17(4): 205-215. 2005

sporophytes almost certainly must have been present in the past. The majority of sites supported single MLPs ; however some sites in the pfalzerwald supported multiple MLPs, some of which are indicative of past inter-gametophytic mating and thus by inference sporophyte production. From the allozyme data alone Rumsey ef al., (1998) suggested that post-glacial spread was more likely but were unable to exclude the possibility that some at least of these locations may have acted as refugia for the species. The levels of intra-population variation, even though low, and the presence of unique simple banding patterns for two enzyme systems (ACON, 6-PGD), gave support to this notion but the proximity of glacial forelands to many of the current localities suggested survival was less likely (Vogel ef al., 1993). Subsequent work by Rumsey ef al. (in prep.), sequencing of the chloroplast DNA haplotypes first reported by Rumsey et al. (1996) has provided additional evidence which would appear to strengthen the case for, at least limited, peri-glacial survival. Macaronesian island refugia show a major dichotomy between the Azores in the north, and Madeira and the Canaries to the south. Minor haplotype variants link some Canarian populations with those from southern Iberia, supporting the broad floristic links already known. Investigated areas of Europe on the western Atlantic seaboard from Brittany northwards show a mix of the haplotypic variants found in the southern refugial sites, although not an intimate one (i.e. sites contain single haplotypes in all cases). The overall picture created is in contrast with the usual pattern of Northern Purity — Southern Richness reported by Hewitt (1999) and others. Interestingly, central Europe is uniformly of the Azorean, believed to be ancestral, type, in spite of the fact that all of the nearest spore sources (Spain, Italy) are of the contrasting major haplotype. If the distribution were the result of post-glacial immigration one might expect a similar pattern to that shown in northwest Europe, with all of the refugia and particularly the most proximal to be involved. Unique minor variants also exist within central Europe, for example in Luxembourg, where they cover an extensive area indicative of a long period of occupation and/or past sporophyte presence. Extremely severe conditions in proximity to sites and the growth and reproductive strategies of Trichomanes speciosum may explain the lack of expected diversity in these putatively refugial areas.

REFERENCES

ANON. 1979. The Convention on the Conservation of European Wildlife and Natural Habitats. Strasbourg, Council of Europe.

ANON. 1992. Council directive 92/43/EEC of 21 May 1992 On the conservation of natural habitats and of wild faunas and floras. Official Journal of the European Communities 206: 7-49.

BABINGTON, C. C. 1863. Trichomanes radicans. J. Bot. (London) |: 293-4.

BENNERT, H. W., JAGER, W., LEONHARDS, W., RASBACH, H. & RASBACH, K. 1994 Prothallien des Hautfarns Trichomanes speciosum (Hymenophyllaceae) auch in Nordrhein- Westfalen. Flor. Rundbr. 28: 80.

BOUDRIE, M. 2001, Premiére découverte de gametophytes de Trichomanes speciosum Willd. (Hymenophyllaceae, Pteridophyta) dans le Massif Central francais. Bull. Soc. Bot. du Centre-OQuest 32: 73-78.

BUJNOCH, W. & KOTTKE, U. 1994. Der Gametophyt von Trichomanes speciosum Willd. im Regierungsbezirk Trier. Dendrocopos 21: 225-230.

DRUDE, O. 1902. Der Hercynische Florenbezirk. (Die Vegetation der Erde Bd. V1). W. Engelmann, Leipzig.

RUMSEY et al.: REPRODUCTIVE STRATEGIES IN T. SPECIOSUM 213

DRUERY, C. T. 1910. British ferns and their variations. George Routledge & Sons, Londo

n.

FARRAR, D. R. 1967. Gametophytes of four tropical fern genera reproducing independently of their sporophytes in the southern Appalachians. Science 155: 1266-1267.

FARRAR, D. R. 1985. Independent fern gametophytes in the wild. Proc. R. Soc. Edinb. 86B: 361-369.

FARRAR, D. R. 1992. Trichomanes intricatum: the independent Trichomanes gametophyte in the Eastern United States. Amer. Fern J. 82: 68-74.

FARRAR, D. R. 1993. Hymenophyllaceae in: Morin, N.R. ef al., eds. Flora of North America Vol.2. Pteridophytes and Gymnosperms, pp. 190-197. Oxford University Press, New York.

FERRARINI, E. 1977. Un antico relitto atlantico ai piedi delle alpi Apuane: Trichomanes speciosum Willd., entita nuova per la flora italiana. Giorn. Bot. Ital. LT: b7ieiy 7.

FERRARINI, E., CIAMPOLINI, F., PICHI SERMOLLI, R.E.G. & MARCHETTI, D. 1986. Iconographia Palynologica Pteridophytorum Italiae. Webbia 40: 1-202.

GARRIDO DIAZ, B. & HIDALGO MAQUEDA, R. 1998. Nueva localidad de Vandenboschia speciosa (Willd.) Kunkel. Acta Bot. Malacit. 23: 229-232.

HORN, K. & ELSNER, O. 1997. Neufunde von Gametophyten des Hautfarns Trichomanes speciosum Willd. (Hymenophyllaceae) in Unter- und Oberfranken. Ber. Naturf. Ges. Bamberg 71: 53-68.

HUCK, S. 1997. Die Prothallien des Hautfarns Trichomanes speciosum Willd. (Hymenophyllaceae) in Hessen. Hess. Flor. Briefe. 46: 28-30.

KIRSCH, H. & BENNERT, H. W. 1996. Erstnachweis von Gametophyten des Hautfarns Trichomanes speciosum Willd. (Hymenophyllaceae) in Bayern. Nachr. Naturwisss. Mus. Aschaffenburg 103: 119-133.

KLEIN, E. J. 1926. Hymenophyllum tunbrigense (L.) Sm., das Juwel des Begone Sandsteins. Cahiers Luxembourgeois |: 3-33; 2: 97-125.

KOTTKE, U. 2003. Die Verbreitung der Gametophyten des Hautfarns 7richomanes speciosum (Hymenophyllaceae) im siidlichen Rheinischen Schieferbirge. Dendrocopos 30: 169-174.

KRIPPEL, Y. 2001. Aire de repartition et statut de Trichomanes speciosum Willd. (Hymenophyllaceae) au Luxembourg. Bull. Soc. Nat. luxemb. 102: 3-13.

LORIOT, S., BLANCHARD, F., LAMOTTE, T. & COTTIGNIES, A. 2002. Présence du gametophyte independent de Trichomanes speciosum — Willd. (Hymenophyllaceae, Pteridophyta) dans les vallées du Pays Basque. Premieres données concernabnt sa distribution dans les Pyrénées-Atlantiques. Le Monde des Plantes 477: 23.

MANTON, I. 1950. Problems in cytology and evolution in the Pteridophyta. Cambridge University Press, Cambridge.

MARCHETTI, D. 1992. Le pteridofite indigene della regione apuana. Mem. Accad. Lunig. Sci. 60-61: 419-420.

MOLESWORTH-ALLEN, B. 1977. Observations on some rare Spanish ferns in Cadiz province, Spain. Fern Gaz. 11: 271-275.

PICHI-SERMOLLI, R. 1936. Sulla presenza nel Monte Pisano dell’ Hymenophyllum tunbrigense Smith. Nuovo Giorn. Bot. Ital. 43: 764-768.

REICHLING, L. 1997. Trichomanes speciosum Willd., un mystérieux passager

214 FERN GAZ. 17(4): 205-215. 2005

clandestine. Adoxa 15-6: 1-4.

RASBACH, H., RASBACH, K. & JEROME, C. 1993. Uber das Vorkommen des Hautfarns Trichomanes speciosum (Hymenophyllaceae) in den Vogesen (Frankreich) und dem benachbarten Deutschland. Carolinea 51: 51-52.

RATCLIFFE, D. A. 1968. An ecological account of Atlantic bryophytes in the British Isles. New Phytol. 67: 231-247.

RATCLIFFE, D. A., BIRKS, H. J. B. & BIRKS, H. H. 1993. The ecology and conservation of the Killarney fern Trichomanes speciosum Willd. in Britain and Ireland. Biol. Cons. 66: 231-247.

RICHARDS, P. W. 1934. Trichomanes radicans Sw. in the south of Spain. J. Bot. (London) 72: 351.

RICHARDS, P. W. & EVANS, G. B. 1972. Biological Flora of the British Isles: Hymenophyllum.. J. Ecol. 60: 245- 268.

RUMSEY, F. J. 1994. The distribution, ecology and population biology of the Killarney fern Trichomanes speciosum Willd. Unpublished Ph.D. thesis. University of Manchester, England.

RUMSEY, F. J., HEADLEY, A. D., FARRAR, D. R. & SHEFFIELD, E. 1991. The Killarney Fern (Trichomanes speciosum) in Yorkshire. Naturalist 116: 41-43.

RUMSEY, F. J., JERMY, A. C. & SHEFFIELD, E. 1998. The independent gametophytic stage of Trichomanes speciosum Willd. (Hymenophyllaceae), the Killarney Fern, and its distribution in the British Isles. Watsonia 22: 1-19.

RUMSEY, F. J., RAINE, C. A. & SHEFFIELD, E. 1992. The reproductive capability of ‘independent’ Trichomanes gametophytes. In IDE, J. M., JERMY, A. C. & PAUL, A. M., (eds.) Fern a Past, Present and Future Objectives, pp.299- 304. Intercept, Andov

RUMSEY, F. J, RUSSELL, S. J., JI, J.. BARRETT, J. A. & GIBBY, M. 1996. Genetic variation in the endangered filmy fern Trichomanes speciosum Willd. In Camus, J. M., Gibby, M. and Johns, R. J., (eds.) Pteridology in Perspective, pp.161-165. Royal Botanic Gardens, Kew.

RUMSEY, F. J., SHEFFIELD, E. & FARRAR, D.R. 1990. British filmy-fern gametophytes. Pteridol. 2: 40-42.

RUMSEY, F. J. & SHEFFIELD, E. 1996. Inter-generational ecological niche separation and the "independent gametophyte" phenomenon. In Camus, J. M., Gibby, M. and Johns, R. J., (eds.) Pteridology in Perspective, pp.563-570. Royal Botanic Gardens, Kew .

RUMSEY, FJ. & VOGEL, J.C. (1998) Trichomanes speciosum willd. (Hymenophyllaceae) in southern Spain. Fern Gaz. 15: 197-203.

RUMSEY, F.J., VOGEL, J.C., RUSSELL, S.J., BARRETT, J.A. & GIBBY, M. 1998. Climate, colonisation and celibacy: population structure in Central European Trichomanes speciosum (Pteridophyta). Botanica Acta 111; 481-489.

RUMSEY, F. J. & VOGEL, J. C. 1999. Tetrastichium fontanum (Mitt.) Card. new to mainland Europe. J. Bryol. 21: 72-73.

RUMSEY, F. J., VOGEL, J. C., RUSSELL, S. J., BARRETT, J. A. & GIBBY, M. 1999. Population structure and conservation biology of the endangered fern Trichomanes speciosum Willd. (Hymenophyllaceae) at its northern distributional limit. Biol. J. Linn. Soc. 66 333-344.

“SENTINEL” 2003. Killarney fern conservation. Pteridologist 4: 58-61.

TUTIN, T. G, BURGES, N. A., CHATER, A. O., EDMONDSON, J. R., HEYWOOD,

RUMSEY et al.: REPRODUCTIVE STRATEGIES IN T. SPECIOSUM 215

V. H., MOORE, D. M., VALENTINE, D. H., WALTERS, S. M. & WEBB, D. A. 1993. Flora Europaea Vol. 1, 2nd. Ed. Cambridge University Press, Cambridge. VOGEL, J. C., JESSEN, S., GIBBY, M., JERMY, A. C. & ELLIS, L. 1993. Gametophytes of 7richomanes speciosum (Hymenophyllaceae: Pteridophyta) in Central Europe. Fern Gaz. 14: 227-232.

WAGNER, W. H. JR. & EVERS, R.A. 1963. Sterile prothallial clones (7richomanes) locally abundant on IIlinois sandstones. Amer. J. Bot. 50: 623

216 FERN GAZ. 17(4). 2005

PTERIDOPHYTE DIVERSITY IN LUXEMBOURG Y. KRIPPEL

Research associate of the Natural History Museum of Luxembourg Rue de Rollingen, 18A; L - 7475 Schoos, Luxembourg (Email: yves.krippel@mnhn.lu)

The Grand-Duchy of Luxembourg is internationally known for its small relictual population of Tunbridge filmy-fern (Hymenophyllum tunbrigense (L.) Sm.), first discovered here in 1821. In recent decades, however, ferns and allied plants - especially subspecies and hybrids - have often been neglected by botanists in Luxembourg, so that reliable data about distribution and status of pteridophyte species is often lacking. Over the last few years, the study of the pteridophytes of Luxembourg has been relaunched, and previous deficiencies in information are being reduced progressively.

Despite the small geographic area of Luxembourg (2586 km’), the diversity of pteridophytes is remarkable. A great number of geological substrates, special geomorphological features with particular microclimatic conditions and a great variety of habitats, ranging from acidic to rather base-rich, have favoured the luxuriant development of a certain number of ferns and fern allies. Not less than 57 pteridophytes (species, subspecies, nothosubspecies and hybrids) are known for Luxembourg. Regrettably a certain number of species have not been observed in the recent decades and can be considered as extinct in wild. Fortunately new taxa have been discovered recently, such as Trichomanes speciosum Willd. (gametophytes), Asp/lenium trichomanes L. subsp. pachyrachis (Christ) Lovis & Reichstein, Asplenium trichomanes L. nothossp. staufferi Lovis & Reichstein, Asplenium x murbekii Dorfler and Equisetum x litorale Kiihl. ex Rupr. At present the pteridophyte flora of Luxembourg includes five clubmosses, seven horsetails and 45 ferns. The Red List, based on the checklist of the pteridophytes of Luxembourg, applies the revised IUCN Red List categories (IUCN 2001) at a national level. Currently 14.7% of the taxa fall in the category Regionally Extinct (RE), 6.5% are Critically Endangered (CR), 6.5% Endangered (EN), 8.2% Vulnerable (VU) and 18% Extremely Rare (R). Only c. 40% of all pteridophyte taxa are considered not to be threatened (*) for Luxembourg.

Despite the recent discoveries of new taxa, the pteridophyte flora of Luxembourg has experienced a significant loss of species. This is predominantly due to habitat changes and resulting climatic variations (habitat destruction, deforestation, changes in drainage patterns etc.). Management plans for rare and endangered species may prevent further loss. For critical species and extremely threatened sites, however, it is advisable to envisage securing the collection through ex-situ conservation and storage of representative samples for future scientific purposes.

REFERENCES IUCN 2001. - IUCN Red List Categories and Criteria: Version 3.1. IUCN Species Survival Commission. IUCN, Gland, Switzerland and Cambridge, UK. ii + 30 pp.

FERN GAZ. 17(4): 217-222. 2005 217

SOME NOTES ON THE TAXONOMY, BIOGEOGRAPHY AND ECOLOGY OF DANAEA (MARATTIACEAE)

M.J.M. CHRISTENHUSZ' & H. TUOMISTO?’

Department of Biology, University of Turku, FI-20014 Turku, Finland, (Email: 'maarten.christenhusz@utu.fi, *hanna.tuomisto@utu.fi)

Key words: Biogeography, Danaea, ecology, Marattiaceae, Neotropics, phylogeny, taxonomy.

ABSTRACT

The neotropical genus Danaea is in revision and some notes on the genus are presented here. We find that the genus can be divided into three subgroups, based on morphological characters of the rhizome, stipe articulation and pinna margin serration. The tree groups are discussed and some taxonomic issues are addressed. The occurrence of bipinnate D. nodosa is reported from Jamaica, and trifoliate specimens of D. simplicifolia are reported from French Guiana. The identities of D. alata, D. jamaicensis, D. jenmanii, D. mazeana, D. media, D. nodosa, D. trifoliata and D. ulei and their synonyms are discussed. Further studies on the phylogeny, taxonomy and ecology of Danaea are needed.

INTRODUCTION

The genus Danaea Sm. is one of the few genera of ferns confined to the Western Hemisphere. It has been repeatedly observed that species limits within the genus are poorly understood, and that the genus is in great need of monographic study (e.g., Morton, 1951, Kramer, 1978). De Vriese & Harting (1853) monographed the Marattiaceae but excluded Danaea, which was at the time, placed in a separate family: the Danaeaceae. Underwood (1902) revised the North American species, and the Ecuadorian species were recently revised by Tuomisto and Moran (2001), but many taxonomic problems remain. In the recent monograph by Rolleri (2004) a large degree of synonymization was suggested, with which we do not agree.

In recent years, it has become obvious that information on characters such as plant habit, rhizome morphology and leaf colour are useful, and often necessary, in identifying Danaea specimens to the correct species (Tuomisto & Moran, 2001). Unfortunately, characters like habit and colour are not easily preserved through the process of pressing and drying, which complicates the identification of herbarium specimens of Danaea. Furthermore, the rhizome is cumbersome to collect and difficult to divide between several duplicates. Collecting the rhizome kills the plant in any case. Consequently, many herbarium specimens, especially those of the larger species, do not include enough rhizome to allow reconstructing its morphology. Especially historical specimens, many of which are types, do not include sufficient information about these characters, either on the specimen itself or on the accompanying label. This has made it difficult to establish the identity of various species with certainty.

To solve this problem, we have carried out field work in different parts of the neotropics. As many type localities as possible were visited to establish the identity of the type specimens. Observing the populations in the field also yields information about the natural variation in morphological characters, and hence helps in circumscribing the

218 FERN GAZ. 17(4): 217-222. 2005

species. So far, field work has been carried out in Colombia, Costa Rica, Ecuador, French Guiana, Jamaica, Peru, Puerto Rico, Suriname and the Lesser Antilles. Future phylogenetic and systematic studies will hopefully give a greater insight in the diversity and relationships between the species of Danaea.

DISTRIBUTION AND ECOLOGY OF DANAEA Danaea occurs solely in the New World tropics. It can be found from Mexico, through all of Central and South America, south to Bolivia. It also occurs on all Greater Antilles, and the wetter, higher islands of the Lesser Antilles. There are isolated populations in Atlantic south-eastern Brazil, southern Paraguay and Cocos Island (Costa Rica). The biogeographical patterns of Danaea are presently being studied.

Danaea occurs mainly in permanently wet forests on well drained soils. Several species seem to prefer steep slopes or creek banks. Danaea is mostly found in undisturbed rainforests, although some species tolerate disturbance, and some species even seem to favour recently disturbed places like shady wet roadside banks. Danaea species can be found in lowland rainforests, up to an elevation of about 2000 m in montane cloud forest and elfin woodland, but most species are found between about 300 and 1000 m elevation. In Amazonian lowland rain forests, it has been found that different species grow on soils with different nutrient status (Tuomisto & Poulsen 1996).

THE THREE MAJOR GROUPS IN DANAEA Based on morphological characters, we divide Danaea into three major groups. This division is still tentative, as the groups overlap in some of the characters, and some species cannot be unambiguously allocated to a group. This division does not follow any of the sections proposed by Presl (1845), and the phylogenetic relationships between the species are not yet known although they are being studied.

The first group consists of species that resemble Danaea nodosa and are often identified as such in herbaria. These are large-statured species whose adults often exceed 1.5 m in height. Most species are characterized by a creeping rhizome (except D. erecta Tuomisto & R.C.Moran), the absence of nodes on the stipes (except D. media Liebm.), and the presence of dentations at pinna apices (except D. erecta and D. grandifolia Underw.).

The second group consists of species supposedly related to Danaea leprieurit Kunze. This group of species has previously been united under the name D. elliptica (see below). These are species of intermediate stature, 0.4-1.5 m in height. This group is characterized by erect, radially symmetric rhizomes, stipes articulated with swollen nodes (nodes sometimes lacking in D. bipinnata Tuomisto) and entire pinna apices. The apical pinna is always present, never replaced by an apical bud.

The third group consists of species resembling Danaea alata Sm, D. moritziana C.Presl and D. mazeana Underw.. These are species of small to intermediate stature, 0.1-1.5 m in height, and are characterized by radially symmetric (rarely dorsiventral) rhizomes that can be erect, creeping or ascending. The stipes are nodose, the blade usually bares many crowded pinnae, and the pinna apices are (coarsely) serrate. Some of the smaller species have translucent pinnae (D. trichomanoides T.Moore ex Spruce, D. imbricata Tuomisto & R.C.Moran, D. tenera C.V.Morton), and many species have bicolorous fronds, with the abaxial side pale green or whitish. The apical pinna can be replaced by a proliferous bud, but this does not seem to be a distinguishing feature

CHRISTENHUSZ & TUOMISTO: SOME NOTES ON DANAEA Pah

between species. To distinguish species within this group, Underwood used mostly characters of the venation, which is, in most cases, also highly variable.

Notes on the Danaea nodosa group

Danaea nodosa, the type species of the genus, has always been considered to be widespread, occurring throughout the range of the genus. However, recent field studies and close study of herbarium material have revealed, that the name has been applied to several clearly separate species. Often these species have different ecological preferences.

The lectotype of Danaea nodosa is a plate in Plumier (1705), which is based on material either collected “between Morne Rouge and Saint Pierre, Martinique” or ‘‘Port- de-Paix, Haiti’. The latter locality was selected as the type locality by Underwood (1909). This was a fortunate choice, because the material we have seen from Haiti agrees with the type illustration very well: the plants have slender, dorsiventral rhizomes with two rows of leaf scars, and the pinnae are long and parallel-sided. In contrast, the plants that grow on Martinique do not match the type illustration: they have very bulky rhizomes with fronds arranged in several rows, and their pinnae are narrower and more tapering. It is obvious that the populations on the two islands represent two different species (Christenhusz, in press).

In Jamaica, most of the plants we have seen agree very well with the material from Haiti, and can thus be considered true Danaea nodosa. However, we found that leaf dissection in Jamaican plants varied from the normal once pinnate to fully bipinnate (Christenhusz & Tuomisto 3194 (IJ, TUR, UCWI). In some populations, the tendency to bipinnate fronds was stronger than in others, but in all populations where bipinnate fronds were found, once pinnate fronds were also present. No other differences were found between the once and twice pinnate fronds than the degree of leaf dissection. It seems obvious, therefore, that this is a case of within-species variation. An interesting question concerns the identity of the South American material. We have observed some differences in pinna shape between the Jamaican and South American populations, but their rhizomes and habit are similar. Even though we have never observed any tendency towards bipinnate leaf dissection in any South American population, for the time being we consider these as D. nodosa.

Field work in Western Amazonia has shown that Danaea nodosa grows on relatively nutrient-rich, clayey soils. Its juveniles have simple fronds only when very small, and the fronds become pinnate usually when they are still less than 10 cm long. A very similar plant but with cartilaginous pinna margins has been found on poorer. more loamy soils. This species has larger pinnae, especially at the juvenile stage when fronds often remain simple until they exceed 30 cm in length (Tuomisto & Groot, 1995, Tuomisto & Moran, 2001). On the basis of a photograph of one of the isotypes, the latter species was identified as D. u/ei H.Christ. Unfortunately, other isotypes of D. ulei show clearly that the rhizomes in this species are erect, indicating that it belongs to the D. leprieurii group. Its pinna margins are not cartilaginous, either. Consequently, the poor-soil segregate of D. nodosa needs a new name (Christenhusz & Tuomisto, in press).

In Costa Rica, populations of supposed Danaea nodosa were found to be variable in whether the stipe is nodose or not. Juveniles usually had one or two nodes, but older fronds more often had none. The presence of nodes was variable even in fronds on the same rhizome, which shows that the two forms belong to the same species. These plants

220 FERN GAZ. 17(4): 217-222. 2005

have previously been identified either as D. nodosa or as D. elliptica, depending on whether nodes were present or not, as this has traditionally been the most important diagnostic character between the two species. Since true D. nodosa never has nodes, the names that can be applied to this species are D. media Liebm. and D. elata Liebm. Danaea media was described on the basis of small plants (with nodes) and D. e/ata on the basis of a large specimen (without nodes) from the same population in Mexico. These names have earlier been synonymized with D. elliptica and D. nodosa, respectively, but our observations indicate that they represent another taxon worthy of recognition at species level. We propose that the name D. media is used as the valid name for this species, because this epithet refers to the species having intermediate characters between the D. nodosa and D. leprieurii groups.

Notes on the Danaea leprieurii group

Most of the species in this group have at some stage been identified as Danaea elliptica Sm. in Rees, which has been considered a wide-spread and common species. The identity and typification of D. elliptica has however long been in turmoil. The lectotype designated by Proctor (1977) is: Jamaica, Mount Diablo, Herb. Sloane 1: 85 (BM-SL). It has been observed repeatedly that this specimen is actually a juvenile of D. nodosa (L.) Sm (e.g., Lellinger 2000, Tuomisto & Moran 2001). We have confirmed this by visiting Mount Diablo, where we only found individuals of D. nodosa. The issue of the synonymization of D. elliptica is discussed in more detail in Christenhusz & Tuomisto (in press).

As mentioned above, Underwood selected Haiti rather than Martinique as the type locality of Danaea nodosa. \t is worth mentioning that since the two islands have different species of the D. nodosa group, the situation would now be very different had Underwood selected the other way round. If Martinique were the type locality, D. nodosa would be represented by a type illustration that matches the species occurring on that island rather poorly, and D. elliptica would become the valid name for the species that has so far been known as D. nodosa in the Greater Antilles and South America, which would be very confusing.

Due to the problems with the name D. elliptica, we call the present group of species the D. leprieurii group, as this is one of the oldest accepted names in this group. This group includes many apparently closely related species that are morphologically quite similar, but grow on different soil types and/or altitudes.

Danaea leprieurii was described from French Guiana, but recently also material from western Amazonia has been allocated to this species. (Tuomisto & Moran 2001). In western Amazonia, D. leprieurii grows on relatively poor substrates, whereas the recently described and morphologically rather similar D. bipinnata grows on richer soils. The two can be distinguished by size (D. /eprieurii is smaller), the number of nodes on the stipes (2-3 in D. leprieurii, 0-1 in D. bipinnata), and usually the degree of lamina dissection. Danaea bipinnata was described from lowland Amazonia, but it is very similar to D. elliptica Sm. var. crispula Rosenst, which was described from lower montane elevations in the Peruvian Andes. However, D. elliptica var. crispula is never bipinnate, and it is usually larger than D. bipinnata.

Danaea simplicifolia Rudge is confined to the Guiana shield, and can easily be recognized by its simple fronds. The only other species with simple fronds is D. carillensis H.Christ, an endemic of Costa Rica, with a nodose stipe, creeping rhizome and apical dentations, and therefore belonging to the D. alata group. D. simplicifolia

CHRISTENHUSZ & TUOMISTO: SOME NOTES ON DANAEA #21

only has a single node below its simple blade. From this node, sometimes one or two lateral pinnae are produced (French Guiana: Christenhusz 2325 (CAY, TUR), Christenhusz 2415 (CAY, TUR)). It then resembles D. trifoliata Rchb., a species that can be distinguished by its nodose stipe. Another character that distinguishes D. simplicifolia from juvenile Danaea that have not developed lateral pinnae yet, are the white abaxial surfaces of the blades, an uncommon feature in this group. It is only found in some representatives of the D. alata group and in the Ecuadorian D. bicolor Tuomisto & R.C.Moran.

Notes on the Danaea alata group

Danaea alata Sm. is a species confined to the Lesser Antilles (lectotype from Martinique, Plumier 1705). It is recognized by simple, widely spaced veins, nodose stipes, creeping, radial rhizomes and coarse apical dentations. Its synonyms are D. fendleri Underw. (Trinidad) and D. stenophylla Kunze (Guadeloupe).

At a first glance the closely related Danaea mazeana Underw. appears to be confined to Guadeloupe, but material from other Antilles is very variable. D. mazeana fits well within the variation of D. jamaicensis Underw., and cannot be consistently separated from it. However, D. jamaicensis sometimes has apical proliferations, which are never found in D. mazeana. This might be due to possible hybridization with D. jenmanii Underw. on Jamaica. D. mazeana differs in ecology from D. alata, as the former is confined to cloud forests above 600 m, whereas the latter is found only in rainforests at lower altitudes.

On present evidence, Danaea jenmanii and D. wrightii Underw. are conspecific. The type of D. wrightii is somewhat scalier, but all other characters overlap. Field work in Jamaica showed that the scalyness of D. jenmanii is variable. Because D. jenmanii has been applied more widely than D. wrightii, the first is proposed here for conservation over the latter. Both species were described in the same publication.

CONCLUSIONS

Field work and herbarium studies have shown that the diversity of Danaea is much greater than previously assumed. The first indication of this tendency was the Ecuadorian treatment (Tuomisto and Moran 2001) where 18 species were included, eight of which were described as new. The same tendency is continuing in other areas, and we now have several species in publication (Christenhusz, in press; Christenhusz & Tuomisto, in press). It seems obvious that the concepts of the geographical distribution ranges of the species will have to be changed once the taxonomy of the genus has been sorted out. More species than expected seem to have rather narrow ranges.

REFERENCES

CHRISTENHUSZ, M.J.M. in press. Three new species of Danaea (Marattiaceae) from French Guiana and the Lesser Antilles. Ann. Bot. Fennici.

CHRISTENHUSZ, M.J.M. & TUOMISTO, H. in press. Five new species of Danaea (Marattiaceae) from Peru and a new status for D. elliptica. Kew Bulletin.

KRAMER, K.U. 1978. The Pteridophytes of Suriname, an enumeration with keys of the ferns and fern-allies. Uitgaven “Natuurwetenschappelijke studiekring voor Suriname en de Nederlandse Antillen” 93:17-18, Utrecht.

LELLINGER, D.B. 2000. On the lectotypification of Danaea elliptica. Amer. Fern J.

poe FERN GAZ. 17(4): 217-222. 2005

90(3): 100-103

MORTON, C.V. 1951. A new fern of the genus Danaea from Colombia. J. Wash. Acad. Sci. 41(8): 276.

PLUMIER, C. 1705. Lingua cervina nodosa major. Traité des Fougéres de l’ Amériques, t. 108, page 90

PRESL, C.B. 1845. Genera Filicacearum. Supplementum Tentaminis Pteridographiae: 34-39. Praha.

PROCTOR, GR. 1977. Marattiaceae. In: HOWARD, R.A. Flora of the Lesser Antilles, Leeward and Windward Islands, vol. 2, Pteridophyta: 45-49. Arnold Arboretum, Harvard University, Jamaica Plain, MA.

ROLLERI, C.H. 2004. Revisién del Género Danaea (Marattiaceae-Pteridophyta). Darwiniana 42 (1-4): 217-301

ROSENSTOCK, E. 1907. Beitrage zur Pteridophytenflora Siidbrasiliens. Hedwigia 46: 162-164

SMITH, IE. 1808. Danaea elliptica, in REES, Cycl. 11: Danaea No. 2.

TUOMISTO, H. & GROOT, A. T. 1995. Identification of the juveniles of some ferns from western Amazonia. Amer. Fern J. 85(1): 1-28.

TUOMISTO, H. & MORAN, R.C. 2001. Marattiaceae. In: HARLING, G & ANDERSON, L. (eds.) Fl. Ecuador 66: 23-68.

TUOMISTO, H. & POULSEN, A. D. 1996. Influence of edaphic specialization on pteridophyte distribution in neotropical rain forests. Journal of Biogeography 23: 283-293.

UNDERWOOD, L.M. 1902. ens Ferns — V. A Review of the Genus Danaea. Bull. Torrey Bot. Club 29 (12): 669-679.

UNDERWOOD, L.M. 1909. Family 1. Marattiaceae. In: BRITTON, N., MURRILL, W. & BARNHART, J. (eds.) N. Amer. Fl. 16 (1): 17-23.

VRIESE, W.H. DE & HARTING, P. 1853. Monographie des Marattiacées. Noothoven van Goor, Leiden & Arnz, Diisseldorf.

FERN GAZ. 17(4): 223-234. 2005 223

TEMPORAL VARIATION IN SPOROPHYTE FERTILITY IN DRYOPTERIS INTERMEDIA AND POLYSTICHUM ACROSTICHOIDES (DRYOPTERIDACEAE: PTERIDOPHYTA)

J.M. SHARPE SHARPLEX Services, PO Box 499, Edgecomb, Maine 04556 USA

Key words: annual variation, demography, fern, Dryopteris intermedia, crown fertility, plant fertility, growth, Polystichum acrostichoides, pteridophyte, temperate zone

ABSTRACT

A long-term demographic study to assess temporal variation in two common New England ferns, Dryopteris intermedia (Evergreen wood fern) and Polystichum acrostichoides (Christmas fern) focused on fertile leaf production by the reproductively mature sporophyte. Over an eight-year period only 29.9% of the leaves in a D. intermedia crown were fertile, while for P. acrostichoides significantly more (48.0%) of the leaves in the crown were fertile. Annual values of the percentage of plants with fertile leaves for D. intermedia ranged from 42% to 77% and for P. acrostichoides from 57% to 93% reflecting significant annual variation, possibly related to winter weather conditions. Only 16% of D. intermedia and 14% of P. acrostichoides sporophytes were fertile every year of the study. Transitions from fertile plant status to sterile plant status occurred in 15% of the D. intermedia observations and 14% of the P acrostichoides observations. For both species, when a sterile year followed a fertile year the number of leaves in the crown decreased by approximately half a_ leaf. Determining the causes of such high levels of variability in plant fertility will be necessary before the role of ferns in any ecosystem can be fully understood.

INTRODUCTION

Long-term monitoring of marked individuals can provide insight into the life history of a species (Harper, 1977) as well as providing a solid basis for future experimental research (Noss, 1999). Werth & Cousens (1990) emphasized the importance of long-term studies of ferns in order to understand mortality and differential fitness, and noted that very little information is available on the life spans and generation times of ferns. There have been several short-term studies of the life history of sporophytes of specific temperate understory fern species. Field research on various aspects of the temperate fern life cycle has been reported for Blechnum spicant (L.) Smith (Cousens, 1973, 1981), Thelypteris dentata (Forsskal) E. P. St. John, Woodwardia virginica (L.) Smith, Osmunda regalis L. (Bartsch & Lawrence, 1997) and Polystichum acrostichoides (Michaux) Schott (Greer & McCarthy, 2000). Studies exceeding 12 years in length have been conducted on selected species of Botrychium (Montgomery, 1990: Johnson-Groh, 1999). Sato (1990) followed leaf production for more than nine years in Polystichum braunii (Spenner) Fée, P. tripteron (Kunze) Pr.and Drvopteris crassirhizoma Nakai. Few other long-term studies have considered annual variation in temperate fern life history characteristics.

Of the seven fern life history classes identified by Cousens ef al. (1988) the reproductively mature sporophyte class has the greatest impact on ecosystem processes. Spore production, controlled by the mature sporophyte, is a significant element of the

224 FERN GAZ. 17(4): 223-234. 2005

reproductive effort for ferns (Greer & McCarthy, 2000) and may also limit establishment of the next generation (Peck, 1980). Though few quantitative studies have addressed temporal variation in fertile leaf production it is assumed that most mature ferns reliably release spores every year (Sheffield 1996). Siman ef al. (1999) state that fertility is the rule for mature individuals of most ferns but note that the genus Pteridium is an exception. A few short-term studies have also noted variation from this generally accepted assumption. Cousens (1988) reported that marked adult fertile plants had been observed reverting to adult sterile plant status. This was based on a study of Woodwardia areolata (L.) T. Moore (= Lorinseria areolata (L.) C. Presl.) which lasted 30 months (Cousens, ef al., 1988). Sato (1990) used counts of number of veins off the leaf midrib (NV) to assess chronological age. He observed a year-to year decrease in NV in about 20% of comparisons of overwintered leaves to the newly emerged leaf cohort in individuals of D. crassirhizoma and P. tripteron, but not in P. braunii. Greer & McCarthy (2000) found that not all reproductively mature individuals of P. acrostichoides produced at least one fertile leaf in each of the two years of their study.

This exploratory study of the reproductively mature sporophyte of two temperate species was undertaken to extend our knowledge of annual variation in plant fertility in a long-term study lasting eight years. The goals were to 1) determine what percentage of the leaves produced by an individual each year were fertile, 2) assess the number of plants in a sample population that produced at least one fertile leaf from year to year, 3) examine levels of consistency in the annual sequence of fertile and sterile leaf production, 4) compare patterns of temporal variation for two different species, and 5) compare any observed temporal patterns to relatively short term local climate parameters.

STUDY SPECIES

Two temperate ferns common in northeastern forests of the USA were chosen for comparative study: Dryopteris intermedia (Willdenow) A. Gray, the Evergreen wood fern, and Polystichum acrostichoides, the Christmas fern. Leaves of D. intermedia are described as monomorphic by Montgomery & Wagner (1993) while those of P achrostichoides exhibit strong sterile-fertile leaf dimorphism, with the fertile leaf exhibiting a distinctive terminal segment of spore-bearing leaflets that are greatly reduced in size compared to the proximal sterile leaflets (Wagner, 1993). Both species have short internodes so that the leaves form a distinctly tufted crown around the shoot apex.

MATERIALS AND METHODS Sporophyte fertility was observed over an eight-year period from 1993 to 2000 in a secondary, mixed-hardwood forest approximately 15 m above sea level in the town of Dresden, Lincoln County, Maine (USA) using methods developed for long-term studies of fern demography in a tropical forest at the Luquillo Long Term Ecological Research site in Puerto Rico (Sharpe, 1997). The sample population consisted of 31 individuals of D. intermedia and 14 individuals of P acrostichoides which were considered reproductively matures as they had produced spores at least once during the eight-year study period. Of those monitored, 24 individuals of D. intermedia and nine individuals of P. acrostichoides were fertile in 1993, the first year of the study. Each plant was identified by a numbered plastic stake and a small segment of colored, plastic-covered wire attached to a leaf base and transferred annually to one of the leaves in the next

SHARPE: TEMPORAL VARIATION IN SPOROPHYTE FERTILITY 225

year’s cohort of leaves. Annual observations of sporophytes were made during the last week in August to insure that all leaves had fully expanded.

All leaves in the current year cohort were counted and each leaf was classified as sterile or fertile. To calculate a crown fertility perecentage for each sporophyte, the number of fertile leaves was compared to total number of leaves in the crown. Percentage of the plants fertile each year was determined by comparing the number of plants with at least one fertile leaf present in the crown to the total number of plants observed each year. Temporal patterns of fertile leaf production were further evaluated by noting the sequence of changes in plant fertility status, observed seven times within the eight-year monitoring period for each plant. Year-to-year transitions in plant fertility status were classified as FF (fertile to fertile), SS (sterile to sterile), FS (fertile to sterile) and SF (sterile to fertile). These categories are illustrated with an example of an individual plant which happened to experience all four types of plant fertility transitions (Figure 1). The number of continuous years a plant was sterile following and preceding a fertile year was also noted.

In order to evaluate whether relatively short term weather patterns could have an effect on plant growth, values for mean daily minimum and maximum temperatures, precipitation and snowfall were taken from records of the NOAA weather station located within 100 km of the study site in Portland Maine. Climate data were analyzed separately for summer (May through October) and winter (November through April). These analyses did not show significant variation in mean daily summer temperatures and precipitation. However, winter temperatures had significantly lower mean daily minimums in 1994 and 1996. Snowfall was significantly higher in the winter of 1996. Furthermore, there was a major cyclic disturbance in the form of an ice storm in 4

INDIVIDUAL PLANT FERTILITY: DRYOPTERIS INTERMEDIA #17

6 = | FERTILE LEAVES 5 | M STERILE LEAVES | __ —— =

> n

NUMBER OF LEAVES No Ww

1993 1994 1995 1996 1997 1998 YEAR

Figure 1. Profile of the fertile and sterile leaves present in the crown of a sample individual (#17) of Drvopteris intermedia for each of the eight years the plant was monitored. Plant fertility status transitions are noted between bars (SS = Sterile Sterile, SF = Sterile/Fertile, FF ~ Fertile/Fertile, FS = Fertile/Sterile) for this sporophyte which happened to experience all four of the possible year-to-year transitions.

226 FERN GAZ. 17(4): 223-234. 2005

February 1998 (Darwin ef al., 2004).

Data analysis

Means and standard error of the means (SE) were calculated for crown fertility for each plant using data collected each year (N = 8). Standard error of the means were chosen to show variation because of the small sample sizes. All percentages were arcsine transformed before statistical tests were applied. To factor out variability among marked individuals, Repeated Measures Analysis of Variance (RMA) was used to evaluate year-to-year differences in the percent of crown fertile. A one sample t-test was used to compare plant fertility and transition type percentages from year to year for each species. Paired t-tests were used to compare the different fertility characteristics for the two species. Data were analyzed using Statistix 8 (Analytical Software, 2003).

RESULTS AND DISCUSSION

Crown fertility In eight years of monitoring these populations, observations of crowns that were composed entirely of fertile leaves were unusual for both species. Of 360 individual annual flushes of leaves observed in this study, only eight individual leaf cohorts of Dryopteris intermedia and 18 of Polvstichum acrostichoides were completely fertile. Fewer than half the leaves produced by these reproductively mature plants were fertile and the mean percentage of leaves in the crown that were fertile is significantly lower for D. intermedia than for P. acrostichoides (Table 1). Neither species in this study approached 100% fertile leaf production. This could be a function of the chronological age of these plants, which is unknown. If they are all relatively young plants, perhaps the ability to produce a full cohort of fertile leaves had not yet occurred. If they are all relatively old plants, then possibly the ability to produce a full cohort of fertile leaves had been lost. Although no increasing or decreasing pattern can be seen by comparing the percentages from year to year, perhaps eight years is too small a segment of the total life span to observe such patterns. It is also possible that full fertile leaf production is not possible in the specific habitat where these plants were growing. It is possible that each species has an intrinsic limit on fertile leaf production by reproductively mature plants which integrates its habitat requirements and need for vegetative support tissue. While there are reports of high percentages of crown fertility in some temperate ferns, for example in Leptopteris hvmenophylloides (Sussex, 1958), low crown fertility has also been observed in Matteuccia ol si (Von Aderkas & Green, 1986). In a study undertaken in southern eria, Gureyeva (2001) found that environmental factors may severely limit spore nigra ae by ferns and Cousens (1973) noted population differences in fertile leaf production by Blechnum spicant.

There were year-to-year differences in crown fertility. The annual mean percentage of leaves that were fertile within the crown ranged from 21.9% to 38.9% for D. intermedia and from 37.6% to 60.6% for P. acrostichoides with little concordance between species in the pattern of annual increases and decreases (Table 1). Dryvopteris intermedia had the highest total fertile leaf production in 1994 following one of the two coldest winters of the study and had the lowest total fertile leaf production following the ice storm in 1998. In contrast, P acrostichoides had the lowest fertile leaf production in 1996 following one of the two coldest winters of the study and exhibited no unusual response after the ice storm. Investigations into the specific triggers for fertile leaf production have been recommended by Wagner & Wagner (1977) and could identify potential environmental factors involved. For example, it has been

SHARPE: TEMPORAL VARIATION IN SPOROPHYTE FERTILITY 22]

TABLE 1. Summary of annual variation in crown and plant fertility for reproductively mature sporophytes of Dryopteris intermedia and Polystichum acrostichoides. N is the number of plants measured. One-sample t-tests are used to compare years for percentage of plants and transition type percentages. Repeated measures analysis of variance is used to compare years for individual percentage of crowns fertile. Paired t-test is used to compare the year-to-year differences between species. Significant differences are indicated by * (P < 0.0500), ** (P < 0.0100) and *** (P < 0.0010).

Year Fertile leaves in crown Percentage — Transition type percentages N_ Percentage of of plants crown (mean) fertile SS SF FF FS

Dryopteris intermedia

1993 1 35.1+4.87% 77% - - - - 1994 31 38.945.88% 71% 10% 13% 58% 19% 1995 31 33.9+5.09% 71% 13% 16% 55% 16% 1996 31 22.945.11% 48% 23% 7% 42% 29% 1997 31 35.1+4.63% 77% 23% 29% 48% 0% 1998 31 21.945.35% 42% 23% 0% 42% 36% 1999 31 23.8+4.58% 55% 39% 19% 36% 7% 2000 31 27.344.12% 68% 26% 19% 48% 7% All 248 29.9+1.68% 64% 22% 15% 47% 15% F/t F = 1.88 t= 11.52 t=6.11 4.10 1449 3.28 df 7 7 6 6 6 6 P 0.0748 <0.0001 0.0009 0.0063 <0.0000 0.1637 EK RK *K RK Polystichum acrostichoides 199 49 .6+10.64% 64% - ~ as ~ 1994 14 41.4+9.73% 71% 23% 23% 46% 23% 1995 14 44.2+10.85% 71% 15% 15% 54% 15% 1996 14 37.6£11.99% 57% 33% 0% 58% 8% 1997 14 50.4411.28% 71% 17% 25% 50% 8% 1998 14 52.7+10.27% 79% 14% 21% 57% 7% 1999 14 60.6+9.14% 93% 0% 21% 71% 7% 2000 14 47.4+9.01% 71% 0% 7™% 64% 29% All 112 48.0+1.71% 72% 12% 16% 58% 14% F/t F = 0.56 t= 12.97 t=284 457 322 4.19 df 7 | 6 6 6 6 F 0.7891 <0.0001 0.0295 0.0038 <0.0000 0.0057 kK * *x RK ** Species comparisons t -4.57 =].22 1.22 -0.31 -1.76 0.38 df 6 6

| ey 6 6 P 0.0026 0.2603 0.2696 0.7702 0.1298 0.7186 **

228 FERN GAZ. 17(4): 223-234. 2005

demonstrated in greenhouse studies (Siman & Sheffield, 2002) that under field conditions individuals do not necessarily produce the maximum number of fertile leaves possible each year. In a year-long study of Polypodium vulgare L. in the United Kingdom, field-grown and greenhouse-grown plants from identical genetic stock were compared by Siman & Sheffield (2002). The indoor population showed a four-fold increase in new leaves, mostly fertile, emerging in a wave pattern thought to represent three annual cohorts for field grown plants as compared to the outdoor population. There do appear to be limits to fertile leaf production which may be consistent within species over an extended period of time. Experiments similar to this one done by Siman & Sheffield (2002) would confirm these limitations.

Plant fertili

Of 248 individual leaf cohorts of D. intermedia observed during the eight years of the study, 64% included at least one fertile leaf, while for 112 cohorts of P. acrostichoides the percentage was 72% (Table 1). The percentage of plants in the population that produced at least one fertile leaf each year ranged from a low of 42% in 1998 to a high of 77% in 1993 for D. intermedia, a range from the lowest year to the highest of 35%, reflecting significant year-to-year variation. For P. acrostichoides there were also significant year-to-year differences, with a range from the lowest year (1996) to the highest year (1999) of 36% (Table 1). For D. intermedia, the lowest percentage of plant fertility occurred following the ice storm. A paired comparison of annual plant fertility percentages did not detect a significant difference between the two species (Table 1). Annual sequence of plant fertility status

There are some elements of life history than can only be determined by making repeated observations of marked individuals over a long period of time. One of those elements is the year-to-year change in an individual sporophyte’s fertility status. Overall, a plant was sterile one year as well as the next (SS) 48 times in the 217 observations of transition for D. intermedia (Table 1). Of these SS transitions, 9 (3 plants) occurred before a plant produced its first fertile leaf while for all others (81%) the SS transition occurred after the plant had produced at least one fertile leaf. For P acrostichoides, SS transitions occurred 11 times out of 98 transition observations, with one plant (of 14) accounting for all of the SS transitions that occurred before a fertile leaf was produced. Thus 89% of the SS transitions for P. acrostichoides occurred after a fertile leaf had been produced. Sterile to fertile (SF) transitions occurred in 15% of the D. intermedia observations and 16% of the P. acrostichoides observations (Table 1).

Fertile to fertile (FF) transitions occurred in 47% of the D. intermedia observations and 58% of the P. acrostichoides observations (Table 1). A small percentages of all D. intermedia plants observed produced fertile leaves only once or twice, while all P acrostichoides individuals produced at least one fertile leaf during three or more years (Table 2A). Although all plants in this study were reproductively mature, only 16% of the individuals of D. intermedia and 14% of the individuals of P. achrostichoides produced at least one fertile leaf during each of the eight years of the study (Table 2A). Of the sporophytes of D. intermedia which were fertile at the start of the study in 1993, 75% were fertile in 1994 while only 21% were consistently fertile through 2000, the last year of the study (Table 2B). Of the sporophytes of P. acrostichoides which were fertile in 1993, 67% were fertile in 1994, compared to the 60.8% sequential reproducers observed in a two-year study of this species in southeastern Ohio by Greer & McCarthy (2000). By 2000, only 22% of the P. acrostichoides the individuals that were fertile in 1993 could be called sequential reproducers. Thus, as the number of years a plant is

SHARPE: TEMPORAL VARIATION IN SPOROPHYTE FERTILITY Zao

monitored increased, the percentage of sequential reproducers decreased at approximately the same rate for both species (Table 2B).

Fertile to sterile transitions (FS) reverse the more common trend of fertile-to-fertile transitions in the chronological maturation of a reproductively mature sporophyte described by Sheffield (1996). However, plants that were fertile one year and produced

TABLE 2. Comparison of fertility and sterility frequencies for reproductively mature sporophytes of D. intermedia and P. acrostichoides observed for an eight-year period from 1993 to 2000. Table A includes all plants which became fertile during the eight-year monitoring period. Table B includes only plants which were fertile in 1993.

Years D. intermedia P. acrostichoides

A. Number of years that individuals are fertile

l 7% 0% 2 3% 0% a 16% 21% + 12% 7% a 16% 21% 6 19% 14% 7 9% 21% 8 16% 14%

B. Number of consecutive years that fertile individuals remain fertile

l 100% 100% = 75% 67% 5 63% 67% 4 38% 56% 5 38% 56% 6 29% 44% 7 25% 44% 8 21% 22% C. Number of fertile/sterile transitions per individual l 54% 50% 2 32% 37% E 14% 13%

D. Span of sterile years after and before a fertile year

Ap WN xe oS S =

230 FERN GAZ. 17(4): 223-234. 2005

no fertile leaves the next year (FS) were seen in 15% of the D. intermedia observations and 14% of the P. acrostichoides transition observations. For D. intermedia, 70% of the individuals reverted to sterile status from fertile status (FS) at least once during the eight-year monitoring period, while 57% of P. acrostichoides individuals had FS transitions. For approximately half of the plants of both species an FS transition occurred only once (Table 2C). Given that a plant must transition back to fertile status in order to become sterile again, the maximum number of FS transitions possible in an eight-year study is three. This maximum number of FS transitions was noted for 14% of the observations for D. intermedia and 13% of the FS observations for P. acrostichoides. The span of years that a previously fertile sporophyte remained sterile ranged from one to five for D. intermedia and one to four for P. acrostichoides (Table 2D). For both species the majority (60%) remained sterile for only one year (Table 2D). Of the FS transitions for Drvopteris intermedia (Table 1), most occurred in 1998 (36%), following the ice storm and in 1996 (29%), following a winter with very low daily minimum temperatures (Table 1). For P. acrostichoides most FS transitions (29%) occurred in 2000 (Table 1).

Crown size and changes plant fertility status

Changes in crown size can be related to the annual transitions to and from fertile leaf production. Plants maintained a stable crown size following FF transitions (Table 3). Crown size was also stable for SS transitions for D. intermedia, though it increased for P. acrostichoides. However, following FS transitions, individual plants produce a significantly smaller sterile crown by over half a leaf (Table 3). Following SF transitions, the increase in the size of the fertile crown is also significant for D. intermedia, with an increase in P. acrostichoides as well (Table 3). Thus, even for the same individuals, the number of leaves in the crown increases with a shift from plant sterility to fertility. On a smaller scale than that observed experimentally by Siman & Sheffield (2002), it appears that when field conditions favor above-average fertile leaf production for D. intermedia and P. acrostichoides, more leaves are produced. TABLE 3. Comparison of changes in crown sizes for transitions in plant fertility sta- tus from one year to the next. Significant differences are indicated by * (P < 0.0500) and ** (P< 0.0100).

Transition Dryopteris intermedia Polystichum acrostichoides ean t F N ls difference difference

Fertile/Fertile 101 -0.04040.186 -0.21 0.8314 53 -0.07640.294 0.26 0.7982 Sterile/Sterile 48 -0.02140.210 -2.25 0.0308* — 11 +0.727+40.304 -1.13 0.2793 Fertile/Sterile 35 -0.57140.254 3.20 0.0031** 13 -0.53940.475 1.16 0.2654 Sterile/Fertile 33 +0.727+0.227 -0.10 0.9212 15 +0.33340.287 2.39 0.0379*

Min/Max -4/+6 -7/+5 F 3.57 bore P 0.0149* 0.3438

SHARPE: TEMPORAL VARIATION IN SPOROPHYTE FERTILITY 231

CONCLUSIONS

The most striking observations to emerge from this temporal study of annual variation in individual plant fertility are 1) the overall low level of fertile leaf production in both species with very few observations of whole crown fertility, 2) the inconsistency of the percentage of the crown which is fertile from year to year, 3) significant differences in the number of plants that are fertile from year to year for each species, a phenomenon possibly linked to winter weather conditions, 4) the difference between species in the overall percentage of plants that are fertile 5) the similarities between species with respect to the annual patterns in changes of the percentages of fertile plants and transitions in plant fertility status 6) the significant year-to-year variation in the percent of the population that experiences a fertile to fertile plant status transition, 7) expansion of individual crown size when fertile leaves are produced, and reduction of crown size when a fertile plant fails to produce fertile leaves the following year. Most of these characteristics of fertile leaf production could not have been known without annual monitoring of marked individuals. Though eight years seems to be a relatively short time compared to the life span of long-lived sporophytes, it was possible to quantify several previously untested assumptions about sporophyte growth.

The original intention of this study was to determine if annual variability in plant fertility could be detected in field populations of sporophytes of temperate species. With the exception of general observations relating to significant local weather patterns, no attempt was made to determine the cause of any observed variation. It is certainly clear from these results that there can be a high level of annual variability in fertile leaf production in temperate ferns. It is also clear that observing a sporophyte only once can lead to a number of erroneous assumptions about its past life history status. A plant with no fertile leaves cannot automatically be assumed to be an immature plant. A plant with one fertile leaf may or may not have been producing fertile leaves longer than one with a 100% fertile crown. A plant that is fertile may or may not be fertile the following year. A population that has the same proportion of fertile plants from one year to the next may still reflect numerous changes due to independent changes in the fertility status of individual plants. While the two species chosen for observation in this study may differ in the overall number and timing of fertile leaves produced, there was a striking similarity in the frequency of the different types of year-to-year transitions in plant fertility status.

Fundamental to understanding of all of these observations is a knowledge of the specific process that determines whether an emergent leaf will be fertile or sterile (White 1971). The current exploratory study, limited as it was to simply observing the reproductive sporophyte, has provided few clues about the stimuli behind the observed patterns. The differences between species detected in this long-term field study of just two species indicate that for each species the fertility triggers may be different. Significant year-to-year variation in plant fertility suggests that environmental factors may be important. Annual variation may be related to climate factors but the relatively short time span of the study and the many potential microhabitat influences require experiments to properly assess these observations.

More long-term field studies are needed as the differences between these two species suggest that the patterns of leaf and plant fertility can vary widely. Because the limitations on observing plants in the field were unknown at the start of this study, sample sizes of plants for which data was available for all eight years were small by the

aan FERN GAZ. 17(4): 223-234. 2005

end of the study, especially for P acrostichoides. The number of plant fertility transitions from fertile to sterile taking place after a plant had become reproductively mature could only be observed a maximum of three times in an eight-year period, limiting the potential for identifying underlying causes. Only one large cyclic event occurred during the study, limiting the potential for generalizations about disturbance effects.

The length of study was limited because of the choice of study site. It was conducted on private land which changed hands after eight years and was no longer accessible for research. For long-lived perennials, a long-term study should encompass a large part of the life span of a sporophyte, and should ideally extend to 100 years or more. As indicated by Werth & Cousens (1990), a long-term study should be conducted at a site guaranteed to be in place for the long-term. The Long Term Ecological Research (LTER) program (Van Cleve & Martin 1991) which has extended world-wide from its original designation of sites in the United States can provide such stable study locations for future long-term studies. The LTER sites also provide for maintenance of standardized data sets long-term which allows for continual monitoring of marked individuals beyond the lifetime of a single individual researcher. The availability of LTER data on the internet also facilitates supplemental short-term studies and experiments based on earlier observations, another recommendation made by Werth & Cousens (1990). Long-term natural experiments such as this one rely on inference from uncontrolled climate data. Although this approach does not have the rigor of a controlled experiment, it does document variation under real environmental conditions. Another advantage of conducting both long and short term studies at an LTER is the continual monitoring of climate parameters as well as other abiotic and _ biotic environmental characteristics which can be related to an individual researcher’s observations about ferns.

This long-term study of fertility in two wintergreen fern species from the same forest showed different patterns of annual variation, suggesting that any future ecological assumptions about ferns must account for differences among fern species. Studies of populations of both D. intermedia and P. acrostichoides throughout their ranges are needed and other life history stages could exhibit such amplitudes of annual variability as well. Long-term field monitoring and additional experimentation with other fern species will be necessary before the ecological generalizations hoped for by Harper (1982) can be made.

ACKNOWLEDGMENTS I am grateful for the encouragement given to me by Mike Cousens many years ago. His pioneering work in fern demographic studies inspired many of the questions I have addressed. Thought-provoking discussions of fern ecology with Herb Wagner and Charlie Werth further stimulated my enthusiasm for field studies of ferns. This paper is dedicated to the memory of these three pteridologists whose insights based on field studies were incomparable.

REFERENCES ANALYTICAL SOFTWARE, INC. 2003. Statistix 8. Analytical Software, Inc., Tallahassee FL. BARTSCH, I. & LAWRENCE, J. 1997. Leaf size and biomass allocation in Thelypteris dentata, Woodwardia virginica, and Osmunda regalis in central Florida. Amer. Fern

SHARPE: TEMPORAL VARIATION IN SPOROPHYTE FERTILITY 233

J. 87:71-76.

COUSENS, M.I. 1973. Reproductive biology and autecology of Blechnum spicant. Ph.D. Dissertation, Washington State University, Pullman, WA. .

COUSENS, M.I. 1981. Blechnum spicant: habitat and vigor of optimal, marginal, and disjunct populations, and field observations of gametophytes. Bot. Gaz. 142:251-

COUSENS, M.I. 1988. Reproductive strategies of pteridophytes. In: Doust, J.L. & Doust, L.L. (Eds) Plant Reproductive Ecology, Patterns and Strategies, pp. 307-328. Oxford University Press, New York.

COUSENS, M.I., LACEY, D.G. & SCHELLER, J.M. 1988. Safe sites and the ecological life history of Lorinseria areolata. Amer. J. Bot. 75:797-807.

DARWIN, A.T., LADD, D., GALDINS, R., CONTRERAS, T.A. & FAHRIG, L. 2004. Response of forest understory vegetation to a major ice storm. J. Torrey Bot. Soc. 131: 45-52.

GREER, GK. & MCCARTHY, B.C. 2000. Patterns of growth and reproduction in a natural population of the fern Polystichum acrostichoides. Amer. Fern J. 90:60-76.

GUREYEVA, I.I. 2001. Homosporous ferns of south Siberia. Tomsk State University Publishers, Tomsk, Russia.

HARPER, J.L. 1977. Population Biology of Plants. Academic Press, New York.

HARPER, J.L. 1982. After description. In: NEMAN, E.I. (Ed) The plant community as a working mechanism, pp 11-25. Blackwell Scientific, Boston MA.

JOHNSON-GROH, C. 1999. Population ecology of Botrychium (Moonworts): status report on Minnesota Botrvchium permanent plot monitoring. Gustavus Adolphus College, St. Peter MN.

MONTGOMERY, J.D. 1990. Survivorship and predation changes in five populations of Botrychium dissectum in eastern Pennsylvania. Amer. Fern J. 80:173-182.

MONTGOMERY, J.D. & WAGNER, W.H. JR., 1993. Dryopteris. In: Flora of North America Editorial Committee (Eds). Flora of North America north of Mexico, Volume 2, pp. 28-290. Oxford University Press, New York.

NOSS, REED F. 1999. Assessing and monitoring forest biodiversity: a suggested framework and indicators. For. Ecol. Man. 115:135-146.

PECK, J.H. 1980. Life history and reproductive biology of the ferns of Woodman Hollow, Webster County, Iowa. Ph.D. Dissertation, lowa State University, Ames,

SATO, T. 1990. Estimation of chronological age for sporophyte maturation in three semi-evergreen ferns in Hokkaido.Ecological Research 5: 55-62.

SHARPE, J.M. 1997. Leaf growth and demography of the rheophytic fern Thelypteris angustifolia (Willdenow) Proctor in a Puerto Rican rainforest. Plant Ecology 130:203-212.

SHEFFIELD, E. 1996. From pteridophyte spore to sporophyte in the natural environment. In: CAMUS, J.M., GIBBY, M. & JOHNS, R.J. (Eds) Pteridology in Perspective, pp. 541-549. Royal Botanic Gardens, Kew.

SIMAN, S.E. & SHEFFIELD, E. 2002. Polvpodium vulgare plants pe continuously in a non-seasonal glasshouse environment. Amer. Fern J. -3 SIMAN, S.E., POVEY, A.C. & SHEFFIELD, E. 1999. Human health ea a —

spores—A review. Fern Gazette 15:275-287. .

SUSSEX, I.M. 1958. A morphological and experimental study of leaf development in

Leptopteris hvmenophylloides (A. Rich) Presl. Phytomorphology 8:96-107.

234 FERN GAZ. 17(4): 223-234. 2005

VAN CLEVE, K. & MARTIN, S. (Eds). 1991. Long-term ecological research in the United States. University of Washington, College of Forest Resources, Seattle. VON ADERKAS, P. & GREEN, P.E.J. 1986. Leaf development of the ostrich fern

Matteuccia struthiopteris (L.) Todaro. J. Linn. Soc., Bot. 93: 307-321.

WAGNER, D.H. 1993. Polystichum. In: Flora of North America Editorial Committee (Eds) Flora of North America north of Mexico, Volume 2, pp. 290-299. Oxford University Press, New York.

WAGNER, W.H. JR. & WAGNER, F.S. 1977. Fertile-sterile leaf dimorphy in ferns. Gard. Bull. Singapore 30:251-267.

WERTH, C.R. & COUSENS, M.I. 1990. Summary: the contributions of population studies on ferns. Amer. Fern J. 80: 183-190.

WHITE, R. A. 1971. Experimental and developmental studies of the fern sporophyte. Bot. Rev. 37:509-540.

FERN GAZ. 17(4): 235-242. 2005 230

NEW ZEALAND ASPLENIUM (ASPLENIACEAE: PTERIDOPHYTA) REVISITED — DNA SEQUENCING AND AFLP FINGERPRINTING

L.R. PERRIE' & P.J. BROWNSEY Museum of New Zealand Te Papa Tongarewa, P.O. Box 467, Wellington, New Zealand (Email: 'leonp@tepapa.govt.nz)

Key words: New Zealand, Asplenium, A. hookerianum, A. bulbiferum, A. gracillimum, A. richardii, chloroplast DNA sequence, AFLP DNA-fingerprinting, polyploid, hybridisation.

This paper summarises our continuing study of the evolutionary history of Asplenium in New Zealand. Chloroplast DNA sequence data and AFLP DNA-fingerprinting have been used to examine the origins and relationships of the New Zealand Asplenium taxa, to test species boundaries, and to further investigate the ancestry of the octoploid Asp/enium taxa belonging to the Austral group.

INTRODUCTION

In 1863, after three years in New Zealand, Samuel Butler’s “we have one very stupid white gentian” (Butler, 1863) was assumedly a frustrated reflection of the difficulty of recognising distinct entities within New Zealand Gentaniella Moench (there are actually 30 species, and only half have white flowers; Glenny, 2003). Analogous sentiments are apparent from those who first dealt taxonomically with Asp/enium L. in New Zealand. Joseph Hooker (1855, p.33) wrote “The New Zealand kinds have defied all attempts to be limited by words”. Similarly, “[some] are extremely inconstant and pass into one another, thus rendering their classification a work of considerable difficulty” (Thomson, 1882, p.72), and “The New Zealand species present exceptional difficulties to the student, on account of their extreme variability and the manner in which several of them are connected by intermediate forms” (Cheeseman, 1906, p.987). Allan (1961, p.75) stated that the species of New Zealand Asplenium were “very ill- defined”, and that while many appeared to respond “markedly to environmental conditions...[,] there is also no doubt that hybridism plays an important part”.

Indeed, hybridisation has proved to be frequent amongst (some) New Zealand Asplenium, and although these hybrids blur the morphological discontinuities between taxa, they are sterile and present little opportunity for introgressive gene flow (Brownsey, 1977a). The identification of hybrids was an integral part of delimiting taxa in Brownsey’s (1977b) taxonomic revision of New Zealand Asplenium, in which 14 species and four subspecies were recognised. Since that treatment three species have been newly described (Brownsey & Jackson 1984; Brownsey, 1985; Brownsey & de Lange, 1997), some names have changed (Brownsey, 1979; Brownsey, 1999), and the appropriate rank for some taxa has been debated (Ogle, 1985; Brownsey, 1998; Perrie & Brownsey, 2005a), such that a present treatment (Table 1) might recognise 18 species and four subspecies; comprising approximately 10% of the New Zealand fern flora (Brownsey & Smith-Dodsworth, 2000).

This paper will summarise some of our recent discoveries in New Zealand Asplenium, but we emphasise that our understanding of the evolution of New Zealand

236 FERN GAZ. 17(4): 235-242. 2005

TABLE 1: A present taxonomic treatment of Asplenium in New Zealand compared with that of Brownsey (1977b).

Present Treatment As treated by Brownsey (1977b)

A. appendiculatum (Labill.) C.Presl A. terrestre Brownsey subsp. ferrestre

subsp. appendiculatum

subsp. maritimum (Brownsey) subsp. maritimum Brownsey Brownsey A. bulbiferum G.Forst. A. bulbiferum G.Forst. subsp. bulbiferum A. gracillimum Colenso subsp. gracillimum (Colenso) Brownsey A. chathamense Brownsey - A. cimmeriorum Brownsey et de Lange A. flabellifolium Cav. A, flabellifolium Cav. A. flaccidum G.Forst. subsp. flaccidum A. flaccidum G.Forst. subsp. flaccidum subsp. haurakiense Brownsey subsp. haurakiense Brownsey A. hookerianum Colenso A. hookerianum Colenso A. lamprophyllum Carse A. lamprophyllum Carse A. lyallii (Hook.f.) T.Moore A, lyallii (Hook.f.) T.Moore A. oblongifolium Colenso A. lucidum Forst.f. A. obtusatum G.Forst. subsp. obtusatum | A. obtusatum G.Forst. subsp. obtusatum subsp. northlandicum Brownsey subsp. northlandicum Brownsey A. pauperequitum Brownsey et P.J.Jacks. | - A. polvodon G.Forst. A, polyodon G.Forst. A, richardii (Hook.f.) Hook.f. A. richardii (Hook.f.) Hook.f. A. scleroprium Hombr. A. scleroprium Hombr. A. shuttleworthianum Kunze A. shuttleworthianum Kunze A. trichomanes L. subsp. quadrivalens A. trichomanes L.

D.E.Mey emend Lovis

subsp. nov. -

PERRIE & BROWNSEY: NEW ZEALAND ASPLENIUM REVISITED za7

Asplenium is far from complete RELATIONSHIPS AND BIOGEOGRAPHY

Brownsey (1977a) considered the Asp/enium species hybridising in New Zealand to comprise a closely related group, dubbed the ‘Austral’ group, and this has been supported by phylogenetic analyses of chloroplast DNA sequence data (Perrie & Brownsey, 2005a). The species in New Zealand that do not hybridise there - A. flabellifolium, A. trichomanes, A. pauperequitum, and A. polvodon - have closer affinities elsewhere (the former three with a primarily temperate, northern hemisphere group, and 4. polyodon with a group that includes 4. aethiopicum). Interestingly, the Austral group is closely related to the bird’s nest ferns (e.g., 4. australasicum (J.Sm.) Hook.) and the finely dissected A. theciferum (Kunth) Mett., which, as Loxoscaphe thecifera (Kunth) T.Moore, is the type of Loxoscaphe T.Moore.

Amongst the New Zealand members of the Austral group, three chloroplast sub- groups are clearly delimited: the Bulbiferum, Flaccidum, and Obtusatum chloroplast groups (Perrie & Brownsey, 2005a). However, it remains unclear how these groups are related to one another, as different DNA regions support conflicting relationships: {{Bul,Obt}Fla} with 71% bootstrap support from the rbcL gene, versus {{Bul,Fla}Obt} with 85% bootstrap support from the trnL-trnF intergenic spacer.

Molecular dating, using penalised likelihood of rbhcL DNA sequence data and a calibration of 140 million years ago for the divergence of Asplenium and Hymenasplenium Hayata, recovers dates of 45 million years ago or younger for the divergence between each New Zealand Asp/enium species and its closest non-New Zealand relative in the sample set analysed (Perrie & Brownsey, 2005a). The calibration age is probably conservatively old (for instance, Schneider e¢ al. (2004) calculated the Asplenium-Hymenasplenium split at just 55 million years ago), so it is likely that the estimated divergence ages are (considerably) younger than actually calculated. In any case, the estimated ages of the divergence are much more recent than the geological separation of New Zealand from Gondwana (c.80 million years ago), implicating multiple events of long-distance dispersal in the origins of New Zealand Asplenium (Perrie & Brownsey, 2005a). The entire Austral group appears to have arisen after New Zealand’s isolation. Therefore, the disjunct distributions of each of the seven Austral Asplenium taxa that occur in New Zealand and elsewhere requires inference of at least one dispersal event (unless the taxa are actually polyphyletic). Dispersal across the Tasman Sea (c.2000 km) is potentially so common that it would be of considerable interest to investigate the degree of genetic isolation between Australian and New Zealand populations of shared fern species.

Figure 1. Plants from the ec Ranges of rece ry ianum n that exhibit the broad-pinnuled and narrow-pinnuled morphology that has previously been referred to as A. hookerianum and A. colensoi, respectively.

238 FERN GAZ. 17(4): 235-242. 2005

ASPLENIUM HOOKERIANUM

AFLP DNA-fingerprinting has been used in conjunction with chloroplast DNA sequencing to investigate some groups in more detail. The morphologically distinct Asplenium hookerianum and A. colensoi were described by Colenso (1845) in the same publication; the ultimate segments are broad in the former, and narrow in the latter (Figure 1). It is quite common to find sites where both broad and narrow-pinnuled plants grow together, with few if any intermediates, thereby creating the impression of two morphologically discrete groups growing in sympatry. However, genetic variation as assayed by AFLP DNA-fingerprinting and chloroplast DNA sequencing is not concordant with the morphological variation (Perrie & Brownsey, 2005b), and indicates that only one species should be recognised (i.e., A. hookerianum).

ASPLENIUM BULBIFERUM

In part because of the apparent origins of Asplenium (bulbiferum subsp.) gracillimum discussed below, we believe that the taxa previously recognised as A. bulbiferum subsp. bulbiferum and A. bulbiferum subsp. gracillimum (Brownsey, 1977b) should be treated at the specific level as A. bulbiferum s.s. (hereafter simply A. bulbiferum) and A. gracillimum (Perrie & Brownsey, 2005a). Circumscribed as such, the name A.

A. hookerianum

>

>

rs)

3

>

oO

=,

pst)

3

J

3

. - ae Qe A. richardii > A. gracillimum $ 2

Jee ane eR CER PPS eat =

3

©

3 ba A. richardii

A. bulbiferum

A B

a A. bulbiferum

Figure 2. A: Maximum parsimony analysis of 469 AFLP characters for the tetraploids Asplenium bulbiferum and A. hookerianum, together with the octoploid A. richardii, a putative autopolyploid of A. hookerianum. Representatives from one of the A. gracillimum lineages are included for comparison. One of 55 most parsimonious trees, which differ in the internal arrangement of samples within 4. bu/biferum and A. hookerianum. Thickened branches indicate those with >80% bootstrap support. B: One of three most parsimonious trees, differing in the arrangement of the A. hookerianum and A. gracillimum samples, from chloroplast trnL-trnF region DNA sequence data. ae: using samples from the Flaccidum and Obtusatum chloroplast groups (not shown).

PERRIE & BROWNSEY: NEW ZEALAND ASPLENIUM REVISITED 239

cae ila is then restricted to tetraploid plants, and is endemic to New Zealand.

urther, it is quite distinct from the plant common in cultivation that is frequently hare “A. bulbiferum” (Perrie et al., 2005). Asplenium bulbiferum appears to be less genetically variable than its closest known tetraploid relative, A. hookerianum. Only one trnL-trnF region haplotype is presently known from A. bulbiferum, as opposed to eight in A. hookerianum. Also, only 38% of AFLP loci were polymorphic within A. bulbiferum, compared to 52% within A. hookerianum, despite more populations being sampled from the former.

ASPLENIUM GRACILLIMUM

The octoploid Asplenium gracillimum (previously A. bulbiferum subsp. gracillimum) appears to be an allopolyploid between the tetraploids A. bulbiferum and A. hookerianum. Asplenium gracillimum produces bulbils like A. bu/biferum (albeit with much less frequency), but has the chloroplast DNA sequence of A. hookerianum (Perrie & Brownsey, 2005a). Further, A. gracillimum is intermediate between its two putative

app.app, obt.nor lya scl app.mar shu

=

ee

ae a . . Pg

. . ao”

‘ ae

;

r we .

{ |

bul obt.obt obl flc.flc, flc.hau, lam cha

Figure 3. Hypothesised origins for the New Zealand species of the Asp/enium Austral group. Octoploid taxa subtended by a single arrow are thought to be autopolyploids. Probable allopolyploids are subtended by two arrows. Solid arrows re flect linkages indicated by chloroplast DNA sequences; dashed arrows are linkages inferred from morphology. Arrows not pointing directly to a tetraploid taxon reflect uncertainty. Taxa

are represented by the first three letters of their name in Table 1, except “fle” = flaccidum. Fronds are not to scale.

240 FERN GAZ. 17(4): 235-242. 2005

parents with regards to frond architecture, scale shape, and habitat.

Reminiscent of the results of Trewick ef al. (2002) with Asplenium ceterach L., chloroplast DNA sequence haplotypes shared by Asplenium hookerianum and A. gracillimum indicate that the latter has had multiple polyploid origins. Although this work is still preliminary, the different chloroplast DNA sequence haplotypes found in A. gracillimum appear to correspond to distinct groups detected with AFLP (Perrie & Brownsey, in prep.). Some of these apparently independently derived lineages of A. gracillimum appear to be sympatric, but it is unknown what happens reproductively where they meet.

ASPLENIUM CIMMERIORUM

The recently described Asplenium cimmeriorum has a conservation status of ‘sparse’, being known only from limestone regions in the north-west of the South Island and around Waitomo in the North Island (Brownsey & de Lange, 1997). Chloroplast DNA sequences indicate that the north and south populations of 4. cimmeriorum have independent derivations; one with an 4. hookerianum-like chloroplast and the other with an A. bu/biferum-like chloroplast (Perrie, de Lange, & Brownsey, in prep.). AFLP also recovers them as distinct groups (with both also distinct from 4. gracillimum). This suggests a need to provide independent conservation assessments, with the northern Waitomo plants being much less common than those in the south.

Both groups of Asplenium cimmeriorum appear to be allopolyploids between A. bulbiferum and A. hookerianum (like A. gracillimum). This situation presents the problem of how to deal taxonomically with multiple lineages of independent but effectively equivalent origins, some of which are ecologically and/or morphologically recognisable (e.g., A. cimmeriorum from A. gracillimum) while others are not (e.g., the cryptic lineages within A. gracillimum).

ASPLENIUM RICHARDII

The octoploid Asplenium richardii is possibly an (old) autopolyploid of a narrow- pinnuled plant of 4. hookerianum. In maximum parsimony analysis of AFLP data, A. richardii falls just outside the diversity found in extant 4. hookerianum, which contrasts with the putatively allopolyploid lineages of A. gracillimum that lie intermediate between 4. bulbiferum and A. hookerianum (Figure 2; see Perrie et a/. (2003a) for discussion of AFLP data and type of polyploidy). The chloroplast DNA sequences of A. richardii are allied to, but nevertheless distinct from, those found in extant 4. hookerianum (Figure 2). This suggests that if A. richardii is an autopolyploid from a narrow-pinnuled, A. hookerianum-like plant, then it is perhaps not of recent origin. That is, sufficient subsequent time has passed since its origin that the A. richardii chloroplast DNA sequences have (a) been lost from the extant populations of A. hookerianum and/or (b) acquired apomorphic changes.

AUSTRAL GROUP OCTOPLOID ORIGINS

Chloroplast DNA sequences have revealed the origins of the octoploid, New Zealand taxa from the Austral group with varying degrees of precision (Perrie & Brownsey, 2004, 2005a), and our present knowledge is summarised in Figure 3. For instance, Asplenium hookerianum and A. gracillimum actually share trnL-trnF region haplotypes (Perrie & Brownsey, 2005a; Perrie & Brownsey, in prep.), while more than 2000 base- pairs from four chloroplast loci are unable to differentiate between A. oblongifolium and

PERRIE & BROWNSEY: NEW ZEALAND ASPLENIUM REVISITED _ 241

A. obtusatum subsp. obtusatum as the chloroplast parents of A. /vallii and A. scleroprium (Perrie & Brownsey, 2004). Some of the octoploids are thought to have autopolyploid origins: A. appendiculatum (from an A. flaccidum-like or A. chathamense-like progenitor), A. obtusatum subsp. northlandicum (possibly from A. obtusatum subsp. obtusatum), and A. richardii (see above). Others are probably allopolyploids: A. cimmeriorum, A. gracillimum (see above), A. lvallii (between something in the Obtusatum chloroplast group and possibly 4. hookerianum), and A. scleroprium (between something in the Obtusatum chloroplast group and something like A. flaccidum). Asplenium shuttleworthianum has chloroplast DNA sequences that nest within the Flaccidum chloroplast group, but no close tetraploid is known.

CONCLUSION

The extant Asplenium of New Zealand are derived from disparate groups within the genus. Long-distance dispersal has seemingly played a major role in shaping the distributions of Asplenium in the south-west Pacific, as is the case for much of the flora of New Zealand (Winkworth et al. 2002: Perrie et al. 2003b). Genetic data has allowed the further unravelling of the histories of some of the Austral octoploids and a resolution of some taxonomic problems (e.g., 4. hookerianum), but it also has unveiled new avenues for investigation (e.g., the multiple, cryptic groups within 4. gracillimum). Many issues within New Zealand Asp/enium remain to be tackled, such as clarifying the boundaries between some taxa (e.g., Perrie & Brownsey, 2004) and establishing the most appropriate ranks for others (Ogle, 1987).

ACKNOWLEDGEMENTS We are grateful to Peter de Lange and Lara Shepherd who have collaborated with us on some aspects of this work, and to the organisers for the opportunity to contribute to the Ferns for the 21St Century conference and its proceedings. This work was funded by the Foundation for Research, Science and Technology (contract MNZX0201).

REFERENCES

ALLAN, H.H. 1961. Flora of New Zealand, Vol. 1. Government Printer, Wellington.

BROWNSEY, P.J. 1977a. Asplenium hybrids in the New Zealand flora. New Zealand J. Bot. 15: 601-637.

BROWNSEY, P.J. 1977b. A taxonomic revision of the New Zealand species of Asplenium. New Zealand J. Bot. 15: 39-86.

BROWNSEY, P.J. 1979. Asplenium lucidum Forst.f., an illegitimate name for the New Zealand shining spleenwort. New Zealand J. Bot. 17: 217-218.

BROWNSEY, P.J. 1985. Asplenium chathamense - a new fern species from the Chatham Islands, New Zealand. New Zealand J. Bot. 23: 135-140

BROWNSEY, P.J. 1998. Aspleniaceae. Fl. Australia 48: 295-327.

BROWNSEY, P.J. 1999. A new combination in Asplenium. New Zealand J. Bot. 37:

BROWNSEY, P.J., JACKSON, P.J. 1984. Asplenium pauperequitum — a new fern species from the Poor Knights Islands, New Zealand. New Zealand J. Bot. 22: 315-321.

BROWNSEY, P.J., DE LANGE, P.J. 1997. ig ou cimmeriorum, a new fern species from New Zealand. New Zealand J. Bot. 35: 283-292.

BROWNSEY, P.J., SMITH-DODSWORTH, 1c. 2000. New Zealand ferns and allied

242 FERN GAZ. 17(4): 235-242. 2005

plants. 2nd eq. David Bateman Ltd., Auckland.

BUTLER, S. 1863. A first year in Canterbury Settlement. Butler, London.

CHEESEMAN, T.F. 1906. Man. New Zealand Fl. Government Printer, Wellington.

COLENSO, W. 1845. A classification and description of some newly discovered ferns, collected in the Northern Island of New Zealand, in the summer of 1841-42. Tasmanian J. Nat. Sci. 2: 161-189.

GLENNY, D. 2003. Understanding the New Zealand gentians. Te Taiao 2: 4-5.

HOOKER, J.D. 1855. Flora Novae Zelandiae, Vol. Il. Reeve, London

OGLE, C.C. 1987. Taxonomic changes in Asplenium (Aspleniaceae; Filicales) in New Zealand. New Zealand J. Bot. 25: 591-593.

PERRIE, L.R., BROWNSEY, P. J., LOCKHART, P.J., LARGE, M.F. 2003a. Evidence for an allopolyploid complex in New Zealand Polystichum. New Zealand J. Bot. 41: 189-215.

PERRIE, L.R., BROWNSEY, P. J., LOCKHART, P.J., BROWN, E.A., LARGE, M.F. 2003b. Biogeography of temperate Australasian Polystichum ferns as inferred from chloroplast sequence and AFLP. J. Biogeogr. 30: 1729-1736.

PERRIE, L.R., BROWNSEY, PJ. 2004. #nL intron variation in New Zealand taxa of the Asplenium obtusatum Chloroplast Group. Tuhinga - Records of the Museum of New Zealand Te Papa Tongarewa 15: 1-5.

PERRIE, L.R., BROWNSEY, P.J. 2005a. Insights into the biogeography and polyploid evolution of New Zealand Asplenium from chloroplast DNA sequence data. Amer. Peo J, 95: 1-21,

PERRIE, L.R., BROWNSEY, P.J. 2005b. Genetic variation is not concordant with merevalonien! variation in the fern Asplenium hookerianum sensu lato (Aspleniaceae). Amer. J. Bot. 92: 1559-1564.

PERRIE, L.R., SHEPHERD, L.D., BROWNSEY, P.J. 2005. Asplenium x lucrosum

nothosp. nov.: a sterile hybrid widely and erroneously cultivated as "Asplenium Le ely a Syst. Evol. 250: 243-257.

SCHNEIDER, H., SCHUETTPELZ, E., PRYER, K.M., CRANFILL, R., verter S., LUPIA, R. 2004. Ferns divenittad in the shadow of angiosperms. Nature 428: 553-557.

THOMSON, GM. 1882. The ferns and fern allies of New Zealand. Wise, Dunedin.

TREWICK , S.A.,. MORGAN-RICHARDS, M., RUSSELL, S.J., HENDERSON, S., RUMSEY, F.J., PINTER, I., BARRETT, J.A., GIBBY, M., VOGEL, J.C. 2002. Polyploidy, phylogeography and Pleistocene refugia of the rockfern Asplenium ceterach: evidence from chloroplast DNA. Molec. Ecol. 11: 2003-2012.

WINKWORTH, R.C., WAGSTAFF, S.J., GLENNY, D., LOCKHART, P.J. 2002. Plant dispersal N.E.W.S from New Zealand. Trends Ecol. Evol. 17: 514-520.

243 FERN GAZ. 17(4). 2005

MOLECULAR EVIDENCE FOR MULTIPLE ORIGINS OF THE APOMICTIC TRIPLOID FERN CORNOPTERIS CHRISTENSENIANA (WOODSIACEAE)

C.-H. PARK! & M. KATO”

‘Laboratory of Ecology Science, Department of Biology, Faculty of Science, Kyushu University, 6-10-1, Hakozaki, Higashi-ku, Fukuoka 812-8581, Japan Corresponding author; (Email: cparkscbh@mbox.nc.kyushu-u.ac.jp) “Department of Biological Sciences, Graduate School of Science, University of Tokyo, 7-3-1 Hongo, Tokyo 113-0033, Japan

Key words: apogamy, apomixis, Cornopteris christenseniana, hybrid fern, multiple origins, triploid

Apomixis (apogamy in a broad sense, agamospermy, agamospory) is a reproductive pathway via chromosomally unreduced spores and gametophytes and the subsequent apogamous reproduction (in a strict sense), i.e. asexual production of a sporophyte from vegetative cells of a gametophyte. Here we use the term apomixis in a broad sense and apogamy in a strict sense to avoid confusion. In pteridophytes, about 10% of the world’s species and about 15% of Japanese species are apomictic. Apomixis is often associated with polyploidy, and three quarters of apomictic pteridophytes are triploid. It is often argued that apomixis, like polyploidy, plays a significant role as an escape from sterility caused by hybridization.

Although Cornopteris christenseniana had been believed to be a sterile triploid interspecific hybrid, our previous study showed that it produces viable spores at various frequencies and forms apogamous sporophytes at low or moderate frequencies in culture (Park & Kato, 2003). It also suggested that apomictic reproduction (in a broad sense) occurs naturally in an artificial environment in the Fern Garden of the Botanical Gardens, University of Tokyo. We have analyzed molecular variation of the nuclear and cpDNA in C. christenseniana and its closely related species. The results strongly support the suggestion that many plants of C. christenseniana propagate by apomixis in the artificial environment and not by in situ hybridization. We also found that plants of C. christenseniana that had been collected from various wild populations, had arisen independently on a markedly recurrent basis. We suggest that Cornopteris christenseniana is an incipient apomictic species of multiple and polytopic origin, and its apomixis is not so strongly regulated as in obligate apomicts. The multiple origins of incipient apomicts imply that the evolution of apomixis in C. christenseniana began with unreduced sporogenesis with a pleiotropic effect leading to subsequent apogamy.

REFERENCES PARK, C.-H. & KATO, M. 2003 Apomixis in the interspecific triploid hybrid fern Cornopteris christenseniana (Woodsiaceae). J. Plant Res. 116(2): 93-104.

244 FERN GAZ. 17(3). 2005 INSTRUCTIONS FOR AUTHORS

PAPERS should not usually exceed 20 printed pages and are generally expected to be considerably shorter. Review articles, as well as reports of original research, are encouraged. Short notes are acceptable e.g. new records. The senior author should supply a fax and email address to facilitate correspondence.

MANUSCRIPTS should be submitted in English (British) in electronic format (preferably) or hard copy (two copies), in 10-point Times New Roman font and double spaced. Electronic versions of text and tables should be compatible with WORD, with figures as pdf or jpg files, and sent as email attachments or CDroms. All manuscripts will be refereed

THE TITLE should reflect the content of the paper and be in BOLD CAPITALS (11- point) and centrally aligned. Generic and specific names should be in italics and any title containing a generic or specific name must be followed by the family and Pteridophyta in brackets e.g.

TRICHOMANES SPECIOSUM (HYMENOPHYLLACEAE: PTERIDOPHYTA) IN SOUTHERN SPAIN

AUTHOR ABBREVIATIONS should follow Pichi ee (1996) Authors of

scientific names in Pteridophyta, Royal Botanic Gardens,

MAIN HEADINGS: should be in BOLD CAPITALS (0- point) and centrally

aligned.

SUBSIDIARY HEADINGS: sori be in bold, the first letter of each word in capitals,

the rest in lower case and left-align

AUTHORS' NAMES AND FULL ADDRESSES: follow the title and are centrally

aligned.

KEY WORDS: up to ten.

ABSTRACT: should reflect the content of the paper.

FIGURES: there is no distinction between photographs and line drawings in

numbering. All should be presented in a form ready for reproduction, ideally in JPG

format (please contact editor with queries), with a scale bar where appropriate.

Lettering or numbers (Arabic) should be in the bottom left using uppercase Times

Roman and be sufficiently large to be legible if reduction is necessary during printing.

The number of photographs allowed in any one issue is limited by cost. Figure captions

should be on a separate sheet.

TABLES: can be printed in either portrait or landscape format. Authors should consider

this when preparing tables. Authors should ensure that tables fit the printed page size in

a legible form.

MEASUREMENTS: should follow the metric system.

CHECKLISTS: should follow the format of Baksh-Comeau, Fern Gaz. 16(1, 2): 11-

122.

REFERENCES: should follow the style of a recent issue of The Fern Gazette, e.g.:-

HOOKER, W.J. 1864. Species Filicum, 5. Dulau & Co., London

MORTON, C.V. 1947. The American species of Hymenophy llum, section Sphaeroconium. Contr. U.S. Natl. Herb. 29(3): 139-201.

STEVENSON, D.W. & LOCONTE, H. 1996. Ordinal and familial relationships of pteridophyte genera. In: CAMUS, J.M., GIBBY, M. & JOHNS, R.J. (Eds) Pteridology in perspective, pp. 435-467. Royal Botanic Gardens, Kew.

JOURNAL ABBREVIATIONS: should follow Botanico Periodicum Huntianum &

Supplements.

Alterations from the original text at proof stage will be charged for unless they are

minor points of detail. Twenty-five offprints will be supplied free to the senior author.

aS each year) <e Full Personal

THE BRITISH PTERIDOLOGICAL SOCII Registered Charity No. 1092399

Patron: HRH The Prince of Wales Officers aid Committee from Marth 1005

President: Dr A.F. Dyer, 499 Lanark Road West, Balerno, Edinburgh EH14 TAL. E-mail: President@eBPS. org.uk Vice-Presidents : M.H. Rickard, Prof. B.A. Thomas

Honorary General Secretary: Miss J.M. i: 42 Crown Woods tes Eltham, London SE9 2NN ail: Secretary@eBPS.org. uk

Treasurer: A. Leonard, |! Victory Roa, taka Hants.PO13DR il: mail@andrew-leonard.co.uk Membership Secretary: — MLS. Porter, 5 West Geet Wigton, Cumbria CA7 9LG © E-mail: Membership@eBPS.org.uk — Meetings Secretary: PJ. Acie) 13 Star es St erat, Cray, Kent BR5 3LJ _ il: Meetings@eBPS.org.uk _

Conservation Officer/Recorder: Dr H. Ss: seseimig 180 ee Ra Edinburgh, EHS [AH © = . Rumsey, Department of Botany, The Natural History Museum, Cromwell Road, Condon. $9 SW7 SBD; E-mail: Conservation@eBPS. org.uk Editor of the Bulletin: | Miss A.M. Paul, Department of Botany, The Natural History M fuseum, “Cromvell pak London SW7 SBD; E-mail: Bulletin@eBPS. ae uk

Editor of the Fern Gazett Prof. M. Gibby, Royal Botanic Garden ae 20A Tnverteith Row, Edinburgh EH3 SLR; E-mail: FernGazette@eBPS. org. ak % Editor of the Pteridologist: Dr J.W. Merryweather, "The Whins', Auchtertyre, by Kyle of Lochalsh [1V40 8EG; E-mail: Pteridologist@eBPs. org.uk Editor of BPS WWW Site - www.eBPS.org.uk: A.C. Pigott, Kersey's Farm, Mendlesham, __ Stowmarket, Suffolk IP14 5RB; E-mail: Webmaster@eBPS. org.uk —

Elected Committee Members: _R.G. Ackers, A.R. Busby, Dr - Golding, Dr M. Hayward, F. McGavigan & B.D. Smith _ Booksales Organiser: S.J. siees Agios 234 Harold Road, Hastings, East Sussex TN35 SNG _ E-mail: Booksales@eBPS. ormuk Horticultural Information Officer and Archivist: AR. Busby, 16 Raby Comer Road, Conky, = = Co ve entry T 4 8 “7 s ppc 1 yee

LT. +

GD; + E-mail:

Merchandise Organisers: Mr B.D. & Mrs G Smith, Rookwood, | Prospect Rd, Oulton Broad, o Laren Suffolk NR32 3PT; E-mail: Merchandise@eBPS.org.uk 3 Plant Exchange Organiser R.G. Ackers, Deersbrook, Horsham Road, Walliswood,

Surrey RH5 5RL; E-mail: PlantExchange@eBPS.org.uk vies

Spore Exchange Organisers: Mr B. & Mrs A. Wright, 130 Prince Rupert Drive, Tockwith, = York YO26 7PU; E-mail: Spores@ eBPS.orguk

_ Trustees of Greenfield & Conienary Funds: ae A.F. Dyer, Miss JM. Ide, A. Leonard es

The BRITISH PTERIDOLOGICAL SOCIETY was founded i in ike and today continues asa s focus a fae fern enthusiasts. It provides a wide range of informatio blications cae me once literature. It also organises formal talks, informal discussions, field ‘meetings, oe visits. : The S ; has a wide membership ' which sikaded ee ae ers, nurserymen and | botanists, both amateur and professi eelteeal The Society's journals, the Fern Gazette, Preridologist and ee are published annually. The Fern Gazette pee matter chiefly of specialist interest on inter dology, the Preridologist, topics of. wet serial catty |, and the e Bulletin, Society _ ses and ets eps www Sos restuc’y wEBrSo org.uk :

: We a PAA Doe. PRA

— Members He Siccotieg bs Institutions £33. Family menbeabin in any. aac ee _ Applications for membership should be sent to the .

: for a g Order forms are available from

"Back numbers of the Fem Gazette, Preridologist and Bulletin 2 able for purcha _ PY: Acock, 13 Star Lane, iS Mey oa Kent BRS 31. E-mail: searene ;

THE FERN GAZETTE :

Rel oat aes

2005

eta ts 5

; and Population structure in Racer pers Periph) 205-215