Available online at www.sciencedirect.com
Molecular Phylogenetics and Evolution 45 (2007) 693–705 www.elsevier.com/locate/ympev
A phylogeny of Porella (Porellaceae, Jungermanniopsida) based on nuclear and chloroplast DNA sequences Jo¨rn Hentschel a, Rui-Liang Zhu b, David G. Long c, Paul G. Davison d, Harald Schneider a,1, S. Robbert Gradstein a, Jochen Heinrichs a,* a
Department of Systematic Botany, Albrecht von Haller Institute of Plant Sciences, Untere Karspu¨le 2, 37073 Go¨ttingen, Germany b Department of Biology, East China Normal University, 3663 Zhong Shan North Road, Shanghai 200062, China c Royal Botanic Garden Edinburgh, 20A Inverleith Row, Edinburgh EH3 5LR, Scotland, UK d Department of Biology, University of North Alabama, UNA Box 5232, Florence, AL 35632-0232, USA Received 1 March 2007; revised 13 April 2007; accepted 7 May 2007 Available online 25 May 2007
Abstract The cosmopolitan family Porellaceae includes about 60 species in two or three genera: the large genus Porella and the monospecific Ascidiota and Macvicaria (alternatively Porella subg. Macvicaria). Maximum parsimony, maximum likelihood and Bayesian inference of phylogeny of a dataset including three markers (rbcL, trnL–trnF region of cp DNA, nrITS region) of 96 accessions resulted in similar topologies supporting the generic status of Ascidiota. Macvicaria is nested in a subclade of Porella. Relationships among species of Porella are in general well resolved and many terminal nodes achieve good statistical support whereas basal relationships are at best moderately supported. Multiple accessions of single species are usually placed in monophyletic lineages. Accessions of P. platyphylla split into a European and a North American clade with one accession from North America embedded within the European samples. The Macaronesian endemic P. inaequalis is closely related to the Asian species P. grandiloba. Porella obtusata and P. canariensis cannot be separated on the basis of the sequence data presented in this study. The molecular topologies indicate a range extension of the Asian P. gracillima subsp. urogea to Eastern North America and of the Neotropical P. swartziana to South Africa. Current supraspecific classifications of Porella are not reflected in the molecular topologies with a correlation between genetic variation and the geographical distribution of the related accessions rather than a correlation between genetic variation and morphology. Ó 2007 Elsevier Inc. All rights reserved. Keywords: Liverwort; Jungermanniopsida; Porellales; Porellaceae; Porella; Ascidiota; Macvicaria; Molecular phylogeny; Cryptic speciation
1. Introduction Porella L. with over 200 published binomials (Hattori, 1989) and an estimated number of 50–60 species (Schuster, 1980) is characterised by bilobed, incubously inserted leaves with a vestigial, often recurved keel, large underleaves that are similar to the lobules in shape, terminal gynoecia on short lateral branches, large, flattened perianths, and a capsule that is barely exserted from the peri*
Corresponding author. Fax: +49 551 39 2329. E-mail address:
[email protected] (J. Heinrichs). 1 Present address: Department of Botany, Natural History Museum, London SW7 5BD, UK. 1055-7903/$ - see front matter Ó 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.ympev.2007.05.005
anth and dehisces into numerous irregular valves. Two monospecific genera have been placed alongside cosmopolitan Porella, the Asian–North American Ascidiota C.Massal. with leaf lobes developed as water-sacs, and the Asian Macvicaria W.E.Nicholson with crisped undulate leaves and inflated perianths, the latter sometimes treated as a subgenus of Porella (Inoue, 1976; Hattori, 1978; Schuster, 1980). Together these genera make up family Porellaceae (Schuster, 1980). Porella has not only been studied morphologically (e.g., Hattori, 1970, 1978, 1986; Swails, 1970; Schuster, 1980; So, 2002, 2005) but also in terms of secondary metabolite composition (e.g., Asakawa et al., 1978; Buchanan et al., 1996; Bungert et al., 1998; Hashimoto et al., 2000). As a result,
694
J. Hentschel et al. / Molecular Phylogenetics and Evolution 45 (2007) 693–705
Porella species are now well known for their production of terpenoids and aromatic compounds with antitumor, antimicrobial, and antifungal activity (Asakawa, 1998). Additionally, several studies have been carried out that utilized DNA fingerprint methods or isozyme data (e.g., Boisselier-Dubayle and Bischler, 1994; Boisselier-Dubayle et al., 1998; Therrien et al., 1998; Freitas and Brehm, 2001; Bischler et al., 2006). These studies pointed at incongruence of molecular and morphological variation (Therrien et al., 1998) and led to the discovery of the allopolyploid status of P. baueri (Schiffn.) C.E.O.Jensen, which is possibly the interspecific hybrid of P. platyphylla (L.) Pfeiff. and P. cordaeana (Huebener) Moore (Boisselier-Dubayle et al., 1998). Taxonomy of Porella has been regarded as notoriously difficult. Schuster (1980): 666) notes ‘‘the actual number of valid taxa is problematical, since the genus seems to be in an active state of evolution with species boundaries ill defined and the species showing phenomenal plasticity’’. Rampant intraspecific variation and extensive lack of stable morphological characters hampers not only species taxonomy but complicates also supraspecific classification. Schuster proposed the first comprehensive infrageneric classification as recently as 1980. Based primarily on leaf lobe and lobule insertion, he (Schuster, 1980) distinguished three subgenera [P. subg. Macvicaria (W.E.Nicholson) Inoue, P. subg. Protoporella R.M.Schust, P. subg. Porella] and six sections. However, Schuster (1980) considered his subdivision provisional due to high infraspecific variation and uniformity in basic architecture of Porella. In contrast to other authors, he treated the genus Macvicaria as a subgenus of Porella. DNA-based investigations have already been applied in several studies focussing on various families of liverworts (e.g., Schaumann et al., 2005; Heinrichs et al., 2006; Hentschel et al., 2006a; Wilson et al., 2007). These studies have greatly improved our understanding of species and genus relationships, and often recovered conflicts with previous morphology-based classifications. The principal objective of this study is to recover a first comprehensive phylogeny of Porella based on a worldwide sampling and nuclear as well as chloroplast DNA markers. A further goal is to evaluate species concepts by sequencing of multiple accessions as well as the supraspecific classification of Porella proposed by Schuster (1980). 2. Materials and methods 2.1. Taxon sampling and outgroup selection Two hundred and eighty-three new sequences generated from 96 specimens were used in this study, including 96 rbcL sequences, 95 trnL-F sequences as well as 92 nrITS sequences. Taxa studied are listed in Table 1, with GenBank accession numbers and voucher details. The determinations of the voucher specimens were carefully examined and original identifications were corrected when necessary.
Ingroup species were selected according to availability and to represent the morphological variation and geographical distribution of Porellaceae. Multiple accessions of several species were used to explore intraspecific genetic variation. Two datasets were compiled. Dataset I includes representatives of Lepidolaenaceae, Goebeliellaceae, and Porellaceae (Ascidiota, Macvicaria, Porella div. sp.). It was mainly designed to determine the sister relationships of Ascidiota and Porella. Dataset II includes only Porellaceae and allows for a more comprehensive alignment of spacer regions. Based on the multi-gene, multi-taxon analyses of Forrest et al. (2006), He-Nygre´n et al. (2006) and Heinrichs et al. (2007) Lepidolaenaceae and Goebeliellaceae were designated as outgroups in dataset I. Based on results of the phylogenetic analyses of dataset I Ascidiota blepharophylla C.Massal. was chosen as outgroup of dataset II. 2.2. DNA extraction, PCR amplification and sequencing Plant tissue from the distal portions of a few shoots was isolated from up to 48 years old herbarium collections or specimens dried in silica gel. Total genomic DNA was purified using Invisorb Spin Plant Mini Kit (Invitek, Berlin, Germany) prior to amplification. Protocols for PCR were carried out as described in previous publications: trnL-F region from Gradstein et al. (2006), rbcL from Hentschel et al. (2006a), and nrITS region from Hentschel et al. (2006b). Bidirectional sequences were generated using a MegaBACE 1000 automated sequencing machine using DYEnamic ET Primer DNA Sequencing Reagent (Amersham Biosciences, Little Chalfont, UK). Sequencing primers were those used for PCR. For all 96 accessions at least two from three markers were obtained. 2.3. Phylogenetic analyses All sequences were aligned manually in Bioedit version 7.0.5.2 (Hall, 1999). Ambiguous positions were excluded from all alignments and lacking data were coded as missing. Alignment of dataset I resulted in a rbcL alignment with 1324 positions, trnL-F 525, and an nrITS alignment with 830 putatively homologous sites (dataset I); alignment of dataset II resulted in a rbcL alignment with 1324 positions, trnL-F 550, and a nrITS alignment with 846 positions. Maximum parsimony (MP) and maximum likelihood (ML) analyses were carried out with PAUP* version 4.0b10 (Swofford, 2000). MP heuristic searches were conducted with the following options implemented: heuristic search mode, 1000 random-addition-sequence replicates, tree bisectionreconnection (TBR) branch swapping, MULTrees option on, and collapse zero-length branches off. All characters were treated as equally weighted and unordered. Nonparametric bootstrapping values (BS) were generated as heuristic searches with 1000 replicates, each with ten
J. Hentschel et al. / Molecular Phylogenetics and Evolution 45 (2007) 693–705
695
Table 1 Taxa used in the present study, including information about the origin of the studied material, voucher information, and the herbarium where the voucher is deposited, as well as GenBank accession numbers Taxon
Origin
Ascidiota blepharophylla C.Massal. subsp. alaskana Steere & R.M.Schust. (I) A. blepharophylla subsp. alaskana (II) Gackstroemia ljungneri (Herzog) Grolle Gackstroemia magellanica (Lam.) Trevis. Gackstroemia schwabei (Herzog) Grolle Gackstroemia weindorferi (Herzog) Grolle Goebeliella cornigera (Mitt.) Steph. Lepidolaena brachyclada (Taylor ex Lehm.) Trevis. Lepidolaena clavigera (Hook.) Dumort. ex Trevis. Lepidolaena taylorii (Gottsche) Trevis. Porella abyssinica var. abyssinica Trevis. Porella abyssinica var. hoehnelii (Steph.) Po´cs Porella acutifolia (Lehm. & Lindenb.) Trevis. Porella acutifolia
rbcL
trnL–F
nrITS
Long, 11186 (GOET)
EF545281
EF545376
EF545468
Lewis, 514 (F) Drehwald & Drehwald, 910052 (JE) Frahm, 27-2 (GOET) Frey & Schaumann, 01-372g (JE) Streimann, 51752 (JE) von Konrath, 27327 (GOET) Scott, MUCV 5228 (JE) Frahm, 22-3 (GOET) Frahm, 1-34 (GOET) Po´cs et al., 90079/L (JE) Po´cs et al., 9210/V (JE) Gradstein, 10311 (GOET) Toia, 212 (JE)
EF545280 EF545277 EF545276 EF547187 EF545275 EF545274 EF547188 EF545278 EF545279 EF545322 EF545323 EF545309 EF545310
EF545375 EF545371 EF545368 EF545369 EF545370 EF545367 EF545372 EF545373 EF545374 EF545417 EF545418 EF545404 EF545405
EF545467 EF545464 EF545463 — EF545462 — — EF545465 EF545466 EF545509 EF545510 EF545496 EF545497
EF545364 EF545363 EF545362 EF545365 EF545366 EF545333 EF545327
EF545459 EF545458 EF545457 EF545460 EF545461 EF545428 EF545422
EF545551 EF545550 EF545549 EF545552 EF545553 EF545520 EF545514
EF545326 EF545318 EF545317 EF545314 EF545311 EF545308 EF545352
EF545421 EF545413 EF545412 EF545409 EF545406 EF545403 EF545447
EF545513 EF545505 EF545504 EF545501 EF545498 EF545495 EF545539
EF545353
EF545448
EF545540
EF545325 EF545284
EF545420 EF545379
EF545512 EF545471
EF545285 EF545283 EF545312 EF545287 EF545286 EF545288 EF545319 EF545320 EF545330 EF545328 EF545344
EF545380 EF545378 EF545407 EF545382 EF545381 EF545383 EF545414 EF545415 EF545425 EF545423 EF545440
EF545472 EF545470 EF545499 EF545474 EF545473 EF545475 EF545506 EF545507 EF545517 EF545515 EF545531
Japan China China Bhutan
Long, 29768 (GOET) Caspari, 4/16 (GOET) Eckstein, 2323 (GOET) Zu¨ndorf, 21671 (JE) Eckstein, 4261 (GOET) Churchill et al., 22001 (GOET) Scha¨fer-Verwimp & Verwimp, 6997 (GOET) Churchill et al., 23218 (GOET) Li & Wang, 1301 (GOET) Huneck, K 86-12 (JE) Koponen, 46872 (JE) Long, 10649 (JE) Long, 34391 (E) Scha¨fer-Verwimp & Verwimp, 24763 (GOET) Scha¨fer-Verwimp & Verwimp, 24760 (GOET) Hedderson, 14779 (BOL) Scha¨fer-Verwimp & Verwimp, 7960 (GOET) Hyvo¨nen, 5572 (JE) Drehwald & Mues, 970001 (GOET) Gambaryan, VLA-h-1865 (GOET) Hentschel, Bryo 01730 (GOET) Long, 34092 (E) Schofield, 117519 (UBC) Streimann, 49240 (JE) Streimann, 27018 (JE) Churchill et al., 22864 (GOET) Churchill et al., 23663 (GOET) Inoue, Bryophyta Selecta Exsiccata 870 (JE) Ohnishi, 3407 (HIRO) Long, 35219 (GOET) Zhu, 20060422-1 (GOET) Long, 10823 (JE)
EF545343 EF545340 EF545342 EF545346
EF545439 EF545435 EF545437 EF545436
EF545530 EF545527 EF545529 EF545533
Japan China Russia USA
Tanaka, 7339 (HIRO) Li & Wang, 4285 (GOET) Gambaryan, VLA-h-1955 (GOET) Davison, 6719 (GOET)
EF545321 EF545355 EF545356 EF545354
EF545416 EF545450 EF545451 EF545449
EF545508 EF545542 EF545543 EF545541
China
Long, 34500 (E)
EF545358
EF545453 EF545545 (continued on next page)
Porella Porella Porella Porella Porella Porella Porella
arboris-vitae (With.) Grolle arboris-vitae arboris-vitae arboris-vitae (I) arboris-vitae (II) brachiata (Taylor) Spruce brasiliensis (Raddi) Schiffn.
Porella Porella Porella Porella Porella Porella Porella
brasiliensis caespitans var. cordifolia (Steph.) S.Hatt. caespitans var. cordifolia caespitans var. nipponica S.Hatt. campylophylla (Lehm. & Lindenb.) Trevis. cf. campylophylla canariensis (F.Weber) Underw. (I)
Bolivia China North Korea China Bhutan China Canary Islands
Porella canariensis (II)
Canary Islands
Porella capensis (Gottsche) Steph. Porella chilensis (Lehm. & Lindenb.) Trevis. (I)
South Africa Argentina
Porella Porella Porella Porella Porella Porella Porella Porella Porella Porella Porella
Argentina Argentina Russia Germany British Isles Alaska Australia Australia Bolivia Bolivia Japan
Porella densifolia (II) Porella densifolia (I) *Porella densifolia (II) *Porella densifolia var. appendiculata (Steph.) S.Hatt. Porella fauriei (Steph.) S.Hatt. Porella gracillima Mitt. Porella gracillima Porella gracillima subsp. urogea (C.Massal.) S.Hatt. & M.X.Zhang Porella cf. grollei S.Hatt.
GenBank accession
Alaska Alaska Argentina Chile Chile Australia New Zealand Tasmania New Zealand New Zealand Tanzania Kenya Indonesia Papua New Guinea British Isles Madeira Canary Islands Germany Germany Bolivia Brazil
chilensis (II) chilensis (III) chinensis (Steph.) S.Hatt. cordaeana (Huebener) Moore cordaeana cordaeana cranfordii Steph. (I) cranfordii (II) crispata (Hook.) Trevis. (I) crispata (II) densifolia (Steph.) S.Hatt. (I)
Voucher (Herbarium)
696
J. Hentschel et al. / Molecular Phylogenetics and Evolution 45 (2007) 693–705
Table 1 (continued) Taxon
Origin
Porella Porella Porella Porella Porella Porella
grandiloba Lindb. grandiloba inaequalis Perss. japonica (Sande Lac.) Mitt. leiboldii (Lehm. & Lindenb.) Trevis. (I) leiboldii (II)
North Korea Japan Madeira Japan Costa Rica Costa Rica
Porella Porella Porella Porella Porella Porella Porella Porella Porella Porella Porella
macroloba (Steph.) S.Hatt. & Inoue (I) macroloba (II) madagascariensis (Nees & Mont.) Trevis. navicularis (Lehm. & Lindenb.) Pfeiff. (I) navicularis (II) nitens (Steph.) S.Hatt. nitens oblongifolia S.Hatt. obtusata (Taylor) Trevis. obtusata obtusata
China China Sri Lanka USA USA Bhutan China Bhutan Italy France Canary Islands
Porella perrottetiana (Mont.) Trevis. Porella perrottetiana (I)
Bhutan Japan
Porella Porella Porella Porella Porella Porella Porella Porella Porella Porella Porella Porella Porella Porella
Japan USA Germany Bulgaria Italy Canada Canada USA USA USA Canada Canada Ecuador Japan
perrottetiana (II) pinnata L. platyphylla (L.) Pfeiff. platyphylla platyphylla platyphylla (I) platyphylla (II) platyphylla (I) platyphylla (II) platyphylla (III) roellii Steph. (I) roellii (II) saccata M.L.So spinulosa (Steph.) S.Hatt.
Porella squamulifera (Taylor) Trevis. Porella squamulifera
Bolivia Ecuador
Porella stephaniana (C.Massal.) S.Hatt.
Japan
Porella subdentata (Mitt.) E.W.Jones Porella subobtusa (Steph.) S.Hatt.
DR of Congo Japan
Porella Porella Porella Porella Porella Porella
South Africa Bolivia Bolivia China Japan Russia
swartziana (F.Weber) Trevis. swartziana (I) swartziana (II) ulophylla (Steph.) S.Hatt. vernicosa Lindb. vernicosa
Porella vernicosa
North Korea
Voucher (Herbarium)
Huneck, K 86-30 (JE) Ohnishi, 4741 (HIRO) Caspari, 5/18b (GOET) Deguchi, 36624 (HIRO) Holz, CR 99-0539 (GOET) Scha¨fer-Verwimp & Holz, SV/ H-0021 (GOET) Koponen, 45262 (JE) Li & Wang, 4429 (GOET) Eggers, SL 7,05 (JE) Whittemore, 4393 (GOET) Whittemore, 4405 (GOET) Long, 8482 (JE) Long, 34415 (GOET) Long, 10726 (JE) Eckstein, 4681 (GOET) Long, 35373 (GOET) Scha¨fer-Verwimp & Verwimp, 24754 (GOET) Long, 10727 (JE) Deguchi, Bryophytes of Asia 4, 96 (GOET) Yamaguchi 12583 (HIRO) Brant 2991 (GOET) Hentschel, Bryo 01567 (GOET) Hentschel, Bryo 0763 (GOET) Eckstein 4632 (GOET) Williams & Cain, s.n. (GOET) Schofield, 106589 (UBC) Worthington, 32690 (GOET) Holmberg & Darigo, 22 GOET) Schofield et al., 114561 (UBC) Schofield, 77156 (GOET) Schofield, 112212 (UBC) Frahm et al., 978 (GOET) Inoue, Bryophyta Selecta Exsiccata 570 (JE) Churchill et al., 22570 (GOET) Scha¨fer-Verwimp et al., 24401 (GOET) Inoue, Bryophyta Selecta Exsiccata 920 (JE) Mu¨ller, Z 620 (GOET) Deguchi, Bryophytes of Asia 4, 97 (GOET) Arts, RSA 21/20 (JE) Churchill et al., 23302 (GOET) Churchill et al., 22662 (GOET) Zhu, 20060122-1, (GOET) Mori, 678 (HIRO) Bakalin, Hepaticae Rossicae Exsiccatae 43 (GOET) Huneck, s.n. (JE)
GenBank accession rbcL
trnL–F
nrITS
EF545301 EF545299 EF545298 EF545300 EF545334 EF545335
EF545395 EF545394 EF545393 EF545396 EF545429 EF545430
EF545488 EF545486 EF545485 EF545487 EF545521 EF545522
EF545336 EF545337 EF545315 EF545293 EF545294 EF545338 EF545339 EF545341 EF545351 EF545350 EF545349
EF545431 EF545432 EF545410 EF545388 EF545389 EF545433 EF545434 EF545438 EF545446 EF545445 EF545444
EF545523 EF545524 EF545502 EF545480 EF545481 EF545525 EF545526 EF545528 EF545538 EF545537 EF545536
EF545306 EF545307
EF545401 EF545402
EF545493 EF545494
EF545305 EF545282 EF545289 EF545290 EF545291 EF545296 EF545297 EF545292 EF545295 EF547189 EF545347 EF545348 EF545329 EF545357
EF545400 EF545377 EF545384 EF545385 EF545386 — EF545391 EF545387 EF545390 EF545392 EF545442 EF545443 EF545424 EF545452
EF545492 EF545469 EF545476 EF545477 EF545478 EF545483 EF545484 EF545479 EF545482 — EF545534 EF545535 EF545516 EF545544
EF545331 EF545332
EF545426 EF545427
EF545518 EF545519
EF545345
EF545441
EF545532
EF545324 EF545316
EF545419 EF545411
EF545511 EF545503
EF545303 EF545302 EF545304 EF545313 EF545359 EF545361
EF545398 EF545397 EF545399 EF545408 EF545454 EF545456
EF545490 EF545489 EF545491 EF545500 EF545546 EF545548
EF545360
EF545455
EF545547
Herbarium acronyms follow Holmgren et al. (1990). Sequences marked with an asterisk (*) were not included in combined analyses (for details see Section 4).
random-addition replicates. The number of rearrangements was restricted to 10 millions per replicate. Where more than one most parsimonious tree was found, trees were summarised in a strict consensus tree. For the BS, clades were considered robust or strongly supported if
BS P 90%, moderately supported if 10) to the data than the homogeneous model ( ln L = 12,867.572; (2loge(B10) = 195.174). The resulting topologies were identical and similar to the ML and MP trees. They will not be
698
J. Hentschel et al. / Molecular Phylogenetics and Evolution 45 (2007) 693–705
Fig. 1. A single most likely phylogram ( ln L = 14,445.061) resulting from maximum likelihood analysis of combined molecular dataset I. BS P 50% is indicated at branches. Gackstroemia, Lepidolaena, and Goebeliella designated as outgroups.
discussed further but Posterior probabilities (BPP P 0.95) based on the heterogeneous two model inference are highlighted in the ML tree (Fig. 4). The MP and ML analyses reveal similar topologies and divide Porella into two main lineages, although the MP analysis resolves P. cranfordii Steph. in an unsupported sister relationship to the remainder of Porella. One main lineage is moderately supported in the BS analyses (BS MP = 71, ML = 87) and achieves significant Bayesian support. The other main lineage is unsupported. Whereas the branching within several subclades is largely resolved, the phylogenetic relationships between them are mostly unresolved or poorly supported in the BS analyses. Several sister relationships with at best
moderate BS support achieve significant Bayesian clade credibility values. One clade within the moderately bootstrap supported main lineage contains a strongly supported clade (MP BS = 98, ML BS = 99) with accessions predominantly from Asia. Within this clade different accessions of P. caespitans s.l. are nested in a polyphyletic lineage with P. ulophylla (Steph.) S.Hatt. [=Macvicaria], P. madagascariensis (Nees & Mont.) Trevis., P. chinensis (Steph.) S.Hatt., and P. subobtusa (Steph.) S.Hatt. Another well-supported clade (MP BS = 97, ML BS = 94) consists of the holarctic species P. platyphylla, P. cordaeana, and P. navicularis (Lehm. & Lindenb.) Pfeiff. The accessions of P. platyphylla split into a European and a North American clade with one
J. Hentschel et al. / Molecular Phylogenetics and Evolution 45 (2007) 693–705
699
Fig. 2. Clade with incongruence evident from chloroplast and nuclear partitions of dataset II. Incongruent accessions have been removed from partitions of dataset II before combination.
Fig. 3. Rooted strict consensus of 45,337 equally parsimonious trees recovered during heuristic searches of combined molecular dataset II. Subgenus and section affiliation of Porella species according to Schuster (1980).
700
J. Hentschel et al. / Molecular Phylogenetics and Evolution 45 (2007) 693–705
Fig. 4. One of two most likely phylograms ( ln L = 12,721.816) resulting from maximum likelihood analysis of combined dataset II. BS P 50% is indicated at branches. Support (P0.95) from Bayesian searches is indicated by thickened branches.
accession from North America embedded within the European samples. The relationships between these four subclades are largely unsupported, however, the monophyly of P. platypylla achieves a MP BS of 73. Another example for a geographical split between several accessions of one taxon is the robust P. arboris-vitae (With.) Grolle clade with a sister relationship of accessions from Macaronesia and Europe. One accession of P. swartziana (F.Weber) Trevis. from South Africa is nested in a robust lineage that contains two other accessions of P. swartziana from Bolivia. All other African accessions are resolved in a separate wellsupported lineage. Three lineages containing members of
the P. pinnata–swartziana complex form a grade with poor bootstrap support. Porella inaequalis Perss. is placed in a moderately (MP BS = 81) or strongly supported (ML BS = 91) sister relationship to P. grandiloba Lindb. Porella pinnata L. is resolved sister to three accessions of P. chilensis (Lehm. & Lindenb.) Trevis. from Argentina (MP BS = 98, ML BS = 99). The molecular backbone of the second main clade is largely unresolved and relationships between several subclades as well as the positions of P. fauriei, P. japonica, and P. cranfordii are uncertain. Several of the taxa sampled with more than one accession are not monophyletic. Porella spinulosa (Steph.) S.Hatt. is resolved within a robust
J. Hentschel et al. / Molecular Phylogenetics and Evolution 45 (2007) 693–705
lineage of P. gracillima Mitt. accessions. Porella obtusata (Taylor) Trevis. and P. canariensis (F.Weber) Underw. form a monophyletic lineage with strong bootstrap support; both taxa cannot be separated on the basis of the presented molecular sequence data. Close relationships between the P. obtusata/canariensis-clade and accessions of P. roellii Steph. are poorly (MP BS = 59) or moderately (ML BS = 71) supported. Porella stephaniana and P. oblongifolia are nested within a strongly supported (MP and ML BS = 98) clade with several accessions of P. densifolia s.l. 4. Discussion 4.1. Supraspecific classification concepts and the positions of Porella [Macvicaria] ulophylla and Ascidiota blepharophylla The phylogenetic analyses (Fig. 1) strongly support the separation of the Asian–North American Ascidiota from Porella at the genus level, which is also obvious from morphology. Ascidiota is well separated from Porella by the presence of well-developed water-sacs originating by infolding of the margins of lobe and lobule bases, and by having no or very weakly developed cilia on leaf and underleaf margins (Schuster, 1980). Our study also shows that the Asian Macvicaria ulophylla is nested in Porella, in a robust subclade together with P. caespitans s.l., P. madagascariensis, P. chinensis and P. subobtusa. Previously, Macvicaria was treated as a monospecific genus (Hattori, 1978; Bai, 2000), as a subgenus of Porella (Inoue, 1976; Schuster, 1980) or as a section of Madotheca Dumort. (=Porella) (Schiffner, 1934). Porella ulophylla differs from all other Porellaceae by the inflated perianths and the often crisped-undulated leaves. Schuster (1980) proposed the subgenus Protoporella represented solely by Porella fauriei, based on the broad connection of leaf lobule and stem, with the insertion line pointing towards stem apex, and the long, more or less convex leaf keel in the latter species. In our molecular topologies, however, deeper nodes are poorly resolved and sister relationships of P. fauriei to other sampled taxa remain unclear. Based on leaf and perianth shape as well as distribution of paraphyllia, Schuster (1980) accepted several sections of Porella. He erected P. sect. Platyphyllae R.M.Schust. to include several holarctic and temperate Asian taxa with decurved-involute, decurrent lobes, and lobules with a Jshaped lobule insertion such as P. platyphylla, P. arborisvitae, and P. obtusata. However, species assigned to P. sect. Platyphyllae are placed in both main lineages of Porella (Fig. 3). Schuster (1980) furthermore proposed a section Paraphyllae R.M.Schust. for neotropical P. squamulifera (Taylor) Trevis., P. brachiata (Taylor) Spruce, and P. leiboldii (Lehm. & Lindenb.) Trevis., species that frequently produce stem paraphyllia. The recently described P. saccata M.L.So (So, 2005) also belongs to this complex. These
701
species form a robust monophylum, however, P. crispata (Hook.) Trevis. that lacks paraphyllia (Swails, 1970) is nested in the sect. Paraphyllae clade. Porella sect. Porella is characterised by the flattened, narrowly lingulate lobules which are not decurrent below the level of the keel, narrow, flat underleaves, and narrowly ovate to oblong dorsal lobes. Schuster (1980) included the P. pinnata-swartziana complex, as well as P. oblongifolia, P. japonica, and P. madagascariensis. Porella nitens (Steph.) S.Hatt. and P. grandiloba were regarded as being related to the P. pinnata– swartziana complex (Schuster, 1980; Persson, 1955). These species are placed in several lineages, hence Porella sect. Porella is polyphyletic. A well-supported subclade in the presented phylogenies is an assemblage of Asian species of which the majority are currently assigned to the sections Acutifoliae R.M.Schust. or Obtusilobae R.M.Schust. The subclade contains accessions of P. acutifolia, P. campylophylla, P. perrottetiana, and the P. caespitans complex. The taxa are mainly characterised by acute, pilose, or rarely subacute lobes, and obliquely patent lobules and have been regarded closely related based on morphological similarities (Hattori, 1976, 1979; Shaheen and Srivastava, 1989). Section Acutifoliae is polyphyletic because P. madagascariensis and P. chinensis are currently assigned to sections Porella or Platyphyllae, respectively, and P. ulophylla to subgenus Macvicaria. Schuster (1980) stated that his classification is a preliminary one that may be helpful for the separation of morphologically closely related groups. However, all sections represented by multiple accessions in our study were either paraphyletic or polyphyletic (Fig. 3), indicating that the distinguishing morphological characters must be re-evaluated. We assume that characters of the Porella gametophyte are not suitable to define monophyletic entities above species level as the result of putatively rapid transformation of these characters in response to environmental stresses. It would be a worthwhile undertaking to investigate sporophyte variation in Porella, however, present knowledge is insufficient for an evaluation of these characters. Problems in defining supraspecific entities in larger genera were reported for other leafy liverworts such as Plagiochila (Dumort.) Dumort., whose sections can at best be defined by a combination of gametophytical and sporophytical characters, as well as geographical distribution (Heinrichs et al., 2005, 2006). 4.2. Implications for species taxonomy Molecular data may allow for a test of the monophyly of morphological species concepts. In mosses, several recent studies revealed rampant parallel morphological evolution and many species were resolved as polyphyletic (Shaw and Allen, 2000; Werner and Guerra, 2004; Stech and Wagner, 2005; Vanderpoorten and Goffinet, 2006). In liverworts such high levels of incongruence have not yet been found (Heinrichs et al., 2005; Schaumann et al., 2005; Hartmann
702
J. Hentschel et al. / Molecular Phylogenetics and Evolution 45 (2007) 693–705
et al., 2006; Hentschel et al., 2006b) and Porella does not seem to make an exception. Sequencing of multiple accessions of morphologically defined species of Porella usually resulted in robust monophyletic lineages. Morphological characters appear well suited to discriminate species in Porella. However, in some cases our study indicates a need for modification of current species circumscriptions. Additionally, some novel relationships are uncovered which allow for resolving controversial treatments based on morphology. For example, the synonymy of Neotropical P. swartziana and African P. capensis, advocated by Gottsche et al. (1844–1847) and Swails (1970) but opposed by Jones (1963), is not supported by our molecular topologies. The molecular data also show that P. pinnata and P. inaequalis, considered closely related by Persson (1955) based on morphological evidence but distantly so by Bischler et al. (2006) using isozyme data besides morphology, are members of different clades and obviously not closely related. The Asiatic P. spinulosa, P. cf. grollei S.Hatt. and P. vernicosa Lindb. are nested within the P. arboris-vitae complex. Porella spinulosa was earlier described as forma spinulosa (Steph.) S.Hatt. of P. vernicosa (Hattori, 1970). The molecular topologies indicate a close relationship of both taxa but clearly support species rank. Porella vernicosa also resembles P. fauriei of the monospecific subgenus Protoporella. Several authors regarded P. fauriei as a subspecies or variety of. P. vernicosa (e.g., Stotler and Crandall-Stotler, 1977). Porella vernicosa is placed sister to P. arboris-vitae (Figs. 1, 3, and 4) and hence the separation at the species level is confirmed. The poorly supported P. arboris-vitae complex is otherwise made up by P. obtusata, P. canariensis, P. roellii and P. gracillima. Porella canariensis has often been considered to be conspecific with P. arboris-vitae. In both species the lobules and underleaves are usually dentate or spinose-ciliate, whereas in P. obtusata the margins are almost entire. However, a conspecifity of P. canariensis and P. arborisvitae is not supported in our analysis due to the sister relationship of P. roellii and the P. obtusata/canariensis clade. Within the P. arboris-vitae complex, the sequence similarities of the accessions of P. obtusata and P. canariensis are remarkable. Recent studies using isozymes or RAPDs (Random Amplified Polymorphic DNAs) (BoisselierDubayle and Bischler, 1994; Freitas and Brehm, 2001; Bischler et al., 2006) suggested high levels of intraspecific and interspecific polymorphisms of these taxa. These polymorphisms are not confirmed in our study. Porella obtusata and P. canariensis cannot be separated on the basis of the molecular sequence data presented in this study. Boisselier-Dubayle and Bischler (1994) and Bischler et al. (2006) also showed that P. obtusata and P. canariensis are morphologically less clearly separated than previously believed. Our sequence data do not support the separation of P. obtusata and P. canariensis. Based on the morphological overlap of both taxa (Bischler et al., 2006) and the extensive sequence similarities we suggest that the taxa
are combined under the name P. obtusata and separated at the variety level. Our chloroplast and nuclear datasets are largely congruent and do not provide evidence for frequent hybridisation in Porella. Putative incongruence has been detected between the nuclear and chloroplast partitions of a clade that includes three Porella densifolia accessions, P. macroloba (Steph.) S.Hatt. & Inoue, P. nitens, and members of the P. arboris-vitae complex (Fig. 2). For this reason, sequences of P. densifolia (II) from Japan and P. densifolia var. appendiculata were excluded from phylogenetic analyses of the combined datasets. These data point towards the need of a broader species concept in this clade rather than to hybridisation. The putative incongruence may indicate intraspecific recombination and not interspecific introgression. Hattori (1970) based the separation of P. densifolia, P. stephaniana and P. oblongifolia on the shape of the lobes, lobules and underleaves, and their dentition. It is already known that these characters are often unreliable (Boisselier-Dubayle and Bischler, 1994; Bischler et al., 2006). Although the molecular data seem to support the broad species concept in Porella advocated by So (2002, 2005) some results indicate putative speciation processes that were overlooked until now. Accessions of the widespread holarctic P. platyphylla [including P. platyphylloidea (Schwein.) Lindb., Therrien et al. (1998)] split into a European and a North American clade with one accession from North America embedded within the European samples (Figs. 3 and 4). The relatively long branches leading to the tips (Fig. 4) are conspicuous when compared to that of other Porella species represented by multiple accessions. High levels of isozyme or RAPDs variation have been reported for P. canariensis, P. obtusata, P. arboris-vitae, and P. platyphylla (Therrien et al., 1998; Freitas and Brehm, 2001; Wyatt et al., 2005; Bischler et al., 2006) but this variation is reflected in notable sequence divergence leading to long branches only in the last species (Fig. 4). Genetic differentiation without or with minute morphological differentiation has sometimes been interpreted as cryptic speciation (e.g., Odrzykoski and Szweykowski, 1991; Fiedorow et al., 2001; McDaniel and Shaw, 2003). It would be a worthwhile undertaking to extend the accession set of P. platyphylla and search for correlation of morphological characters and the observed tree topologies. Since P. platyphylloidea and P. platyphylla have been separated in terms of elater characters (Schuster, 1980), special attention should be drawn to populations with sporophytes. 4.3. Implications for biogeography and extensions of range Several case studies in bryophytes demonstrated morphological homoplasy above species level that is also evident in Porella (see Section 4.1). These studies suggested a correlation between genetic variation and the geographical distribution of the related accessions rather than between genetic variation and morphology (Shaw and
J. Hentschel et al. / Molecular Phylogenetics and Evolution 45 (2007) 693–705
Allen, 2000; Stech and Dohrmann, 2004; Grundmann et al., 2006; Hartmann et al., 2006). This pattern is also seen in Porella with several supraspecific Neotropical, African, Asian, American or Holarctic clades. The molecular topologies support several disjunct species ranges. Extensive sequence similarities of Alaskan and European P. cordaeana as well as of one North American and several European P. platyphylla point to ongoing or rather recent intercontinental gene flow. On the other hand the P. arboris-vitae clade shows clear structure and a subdivision in a Macaronesian subclade and a subclade from Continental Europe and the British Isles. A similar intraspecific variation related to a geographical rather than a morphological pattern has already been demonstrated for the liverwort genera Bryopteris (Nees) Lindb. (Hartmann et al., 2006), Herbertus Gray (Feldberg et al., 2007) and Plagiochila (Heinrichs et al., 2005). The Asian P. gracillima [subsp. urogea (C.Massal.) S.Hatt. & M.X.Zhang] has been observed in a single North American locality within Great Smoky Mountains National Park for almost 20 years, and has already been correctly identified based on morphology by the late S.Hattori. The isolated occurrence of P. gracillima in Eastern North America could be the result of a quite recent range extension, as indicated by the extensive sequence similarities of the North American and Asian accessions. The common neotropical species P. swartziana is here newly reported for Africa based on a specimen from South Africa. Recent molecular studies have already lent support to similar disjunct African-Neotropical species ranges within the leafy liverwort genera Herbertus (Feldberg et al., 2007), Tylimanthus Mitt. (Stech et al., 2006), and Plagiochila (Heinrichs et al., 2005). The extensive sequence similarities of the related neotropical and African accessions add to the growing evidence that many Afro-American ranges could be the result of long distance dispersal (Heinrichs et al., 2005; Schaumann et al., 2005; Hartmann et al., 2006). Heinrichs et al. (2006) and Vanderpoorten and Long (2006) provided molecular evidence for close relationships of the tropical American and Macaronesian liverwort floras. However, the Macaronesian endemic Porella inaequalis is closely related to the Asian P. grandiloba rather than to Neotropical species. Other Porella species occurring in Macaronesia (P. arboris-vitae, P. cordaeana, P. pinnata, P. platyphylla, P. obtusata/canariensis) are further on distributed in the Holarctic or Asia. A close relationship of these species and Neotropical ones is not observed based on the current data set. 4.4. Porellaceae in a phylogenetic context, and resolution within the family Lepidolaenaceae and Goebeliellaceae have been resolved as closest relatives of Porellaceae in recent molecular studies (Forrest et al., 2006; He-Nygre´n et al., 2006; Heinrichs et al., 2007). Extension of the Porellaceae sam-
703
pling of Forrest et al. (2006) in the present study confirms the sister relationship of Ascidiota and Porella. The phylogram resulting from the ML analysis of dataset I shows long branches, and hence numerous mutations, separating Ascidiota and Porella. This pattern possibly indicates a long period of separation of these two lineages. Currently no other taxa beside Ascidiota are known that may serve as outgroup for Porella and could shorten the remarkably long branch leading to Porella. Porellales possibly originated in the Early Permian whereas extant species of Porella branch off in the Tertiary (Heinrichs et al., 2007). A relatively recent origin of extant species of Porella would be in accordance with the limited resolution and relatively short branches of the Porella crown group (Figs. 1 and 4). The split of Ascidiota and Porella, however, must date back at least to the Early Tertiary as indicated by an Eocene Baltic amber fossil described as the extinct P. subgrandiloba Grolle & M.L.So (Grolle and So, 2004). No conflicting phylogenetic signals between the investigated molecular markers affected the Porella backbone. Thus, possibly a series of initial diversification occurred nearly simultaneously in Porella resulting in subsequent weak phylogenetic signals. Similar scenarios have been postulated for certain groups of ferns (Schneider et al., 2004), angiosperms (Fishbein et al., 2001; Pennington et al., 2004; Ho¨randl et al., 2005; Pellmyr et al., 2007) and animals (e.g., Waits et al., 1999; Poe and Chubb, 2004; Mendelson and Shaw, 2005). Acknowledgments We thank Jan Eckstein (Go¨ttingen), Steffen Caspari (St. Wendel), Matt von Konrat (Chicago), Ta´mas Po´cs (Eger), and Alfons Scha¨fer-Verwimp (Herdwangen-Scho¨nach) as well as the curators and directors of Herbarium Haussknecht, Jena (JE), Bolus Herbarium, Cape Town (BOL), Field Museum, Chicago (F), University of British Columbia Herbarium, Vancouver (UBC), Hiroshima Museum Herbarium (HIRO), and Alaska Museum of the North, Fairbanks (ALA) for the loan of specimens and the permission to extract DNA. Thanks are also due to the curators of the herbaria of the Far Eastern Branch of the Russian Academy of Sciences, Vladivostok (VLA), Duke University, Durham (DUKE) and New York Botanical Garden (NY) for their search of Ascidiota specimens and to Alain Vanderpoorten (Lie`ge) for comments on the manuscript. The late Sinske Hattori kindly identified Porella gracillima subsp. urogea new to USA. Financial support from the German Research Foundation (DFG Grant HE 3584/2 to J.H., S.R.G., and H.S.) and the National Natural Science Foundation of China (No. 30470142 to R.L.Z.) is gratefully acknowledged. References Asakawa, Y., 1998. Biologically active compounds from bryophytes. J. Hattori. Bot. Lab. 84, 91–104.
704
J. Hentschel et al. / Molecular Phylogenetics and Evolution 45 (2007) 693–705
Asakawa, Y., Toyota, M., Takemoto, T., 1978. Sesquiterpenes from Porella species. Phytochemistry 17, 457–460. Bai, X., 2000. Porellaceae. In: Gao, C., Cao, T. (Eds.), Flora Yunnanica, Tomus 17 (Bryophyta: Hepaticae, Anthocerotae). Science Press, Bejing, pp. 352–379. Bischler, H., Boisselier-Dubayle, M.C., Fontinha, S., Lambourdie´re, J., 2006. Species boundaries in European and Macaronesian Porella L. (Jungermanniales, Porellaceae). Cryptog. Bryol. 27, 35–57. Boisselier-Dubayle, M.C., Bischler, H., 1994. A combination of molecular and morphological characters for delimitation of taxa in European Porella. J. Bryol. 18, 1–11. Boisselier-Dubayle, M.C., Lambourdiere, J., Bischler, H., 1998. The leafy liverwort Porella baueri (Porellaceae) is an allopolyploid. Pl. Syst. Evol. 210, 175–197. Buchanan, M.S., Connolly, J.D., Rycroft, D.S., 1996. Pinguisane and sacculatane terpenoids from the liverwort Porella platyphylla. Phytochemistry 43, 1249–1253. Bungert, M., Gabler, J., Adam, K.-P., Zapp, J., Becker, H., 1998. Pinguisane sesquiterpenes from the liverwort Porella navicularis. Phytochemistry 49, 1079–1083. Feldberg, K., Hentschel, J., Wilson, R., Rycroft, D.S., Glenny, D., Heinrichs, J., 2007. Phylogenetic biogeography of the leafy liverwort Herbertus (Jungermanniales, Herbertaceae) based on nuclear and chloroplast DNA sequence data: correlation between genetic variation and geographical distribution. J. Biogeogr. 34, 688–689. Fiedorow, P., Odrzykoski, I., Szweykowski, J., Szweykowska-Kulinska, Z., 2001. Phylogeny of the European species of the genus Pellia (Hepaticae; Metzgeriales) based on the molecular data from nuclear tRNA(CAA)(LEU) intergenic sequences. Gene 262, 309– 315. Fishbein, M., Hibsch-Jetter, C., Soltis, D.E., Hufford, L., 2001. Phylogeny of Saxifragales (Angiosperms, Eudicots): analysis of a rapid, ancient radiation. Syst. Biol. 50, 817–847. Forrest, L.L., Davis, E.C., Long, D.G., Crandall-Stotler, B.J., Clark, A., Hollingsworth, M.L., 2006. Unraveling the evolutionary history of the liverworts (Marchantiophyta): multiple taxa, genomes and analyses. Bryologist 109, 303–334. Freitas, H., Brehm, A., 2001. Genetic diversity of the Macaronesian leafy liverwort Porella canariensis inferred from RAPD markers. J. Heredity 92, 339–345. Gottsche, C.M., Lindenberg, J.B.G., Nees ab Esenbeck, C.G., 1844–1847. Synopsis Hepaticarum. Meissner, Hamburg. Gradstein, S.R., Wilson, R., Ilkiu-Borges, A.L., Heinrichs, J., 2006. Phylogenetic relationships and neotenic evolution of Metzgeriopsis (Lejeuneaceae) based on chloroplast DNA sequences and morphology. Bot. J. Linn. Soc. 151, 293–308. Grolle, R., So, M.L., 2004. A fossil Porella (Porellaceae, Hepaticae) in Eocene Baltic Amber. J. Linn. Soc. Bot. 145, 485–488. Grundmann, M., Schneider, H., Russel, S.J., Vogel, J.C., 2006. Phylogenetic relationships of the moss genus Pleurochaete Lindb. (Bryales: Pottiaceae) based on chloroplast and nuclear genomic markers. Org. Divers. Evol. 6, 33–45. Hall, T.A., 1999. BIOEDIT: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41, 95–98. Hartmann, F.A., Wilson, R., Gradstein, S.R., Schneider, H., Heinrichs, J., 2006. Testing hypotheses on species delimitations and disjunctions in the liverwort Bryopteris (Jungermanniopsida: Lejeuneaceae). Int. J. Pl. Sci. 167, 1205–1214. Hashimoto, T., Irita, H., Tanaka, M., Takaoka, S., Asakawa, Y., 2000. Pinguisane and imeric pinguisane-type sesquiterpenoids from the Japanese liverwort Porella acutifolia subsp. tosana. Phytochemistry 53, 593–604. Hattori, S., 1970. Studies of the Asiatic species of the genus Porella (Hepaticae). III. J. Hattori Bot. Lab. 33, 41–87. Hattori, S., 1976. Studies of the Asiatic species of the genus Porella (Hepaticae). VI. J. Hattori Bot. Lab. 40, 121–138.
Hattori, S., 1978. Studies on the Asiatic species of the genus Porella (Hepaticae). VII. A synopsis of Asiatic Porellaceae. J. Hattori Bot. Lab. 44, 91–120. Hattori, S., 1986. New Zealand Porella species. Bull. Natl. Sci. Mus. Tokyo B 12, 31–36. Hattori, S., 1989. Porella. In: Geissler, P., Bischler, H. (Eds.), Index Hepaticarum, vol. 11. Cramer, Berlin, Stuttgart, pp. 91–311. Heinrichs, J., Hentschel, J., Wilson, R., Feldberg, K., Schneider, H., 2007. Evolution of leafy liverworts (Jungermaniidae, Marchantiophyta): estimating divergence times from chloroplast DNA sequences using penalized likelihood with integrated fossil evidence. Taxon 56, 31–44. Heinrichs, J., Lindner, M., Gradstein, S.R., Groth, H., Buchbender, V., Solga, A., Fischer, E., 2005. Origin and subdivision of Plagiochila (Jungermanniidae: Plagiochilaceae) in tropical Africa based on evidence from nuclear and chloroplast DNA sequences and morphology. Taxon 54, 317–333. Heinrichs, J., Lindner, M., Groth, H., Hentschel, J., Feldberg, K., Renker, C., Engel, J.J., von Konrat, M., Long, D.G., Schneider, H., 2006. Goodbye or welcome Gondwana?—insights into the phylogenetic biogeography of the leafy liverwort Plagiochila with a description of Proskauera, gen. nov. (Plagiochilaceae, Jungermanniales). Pl. Syst. Evol. 258, 227–250. Hentschel, J., Wilson, R., Burghardt, M., Zu¨ndorf, H.-J., Schneider, H., Heinrichs, J., 2006a. Reinstatement of Lophocoleaceae (Jungermanniopsida) based on chloroplast gene rbcL data: exploring the importance of female involucres for the systematics of Jungermanniales. Pl. Syst. Evol. 258, 211–226. Hentschel, J., Zu¨ndorf, H.-J., Hellwig, F.H., Scha¨fer-Verwimp, A., Heinrichs, J., 2006b. Taxonomic studies in Chiloscyphus Corda (Jungermanniales: Lophocoleaceae) based on nrITS sequences and morphology. Pl. Syst. Evol. 262, 125–137. He-Nygre´n, X., Jusle´n, A., Ahonen, I., Glenny, D., Piippo, S., 2006. Illuminating the evolutionary history of liverworts (Marchantiophyta)—towards a natural classification. Cladistics 22, 1–31. Ho¨randl, E., Paun, O., Johansson, J.T., Lehnebach, C., Armstrong, T., Chen, L., Lockhart, P., 2005. Phylogenetic relationships and evolutionary traits in Ranunculus s.l. (Ranunculaceae) inferred from ITS sequence analysis. Molec. Phylogen. Evol. 36, 305–327. Holmgren, P.K., Holmgren, N.H., Barnett, L.C., 1990. Index Herbariorum. Part I: The Herbaria of the World. Regnum Veg, eighth ed. New York Botanical Garden, New York. Huelsenbeck, J.P., Ronquist, F., 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17, 754–755. Inoue, H., 1976. Illustrations of Japanese Hepaticae, vol. 3. Tsukiji Shokan Publ. Co., Tokyo. Jones, E.W., 1963. African Hepatics XVI. Porella in Tropical Africa. Trans. Brit. Bryol. Soc. 4, 446–461. Kass,R.E., Raftery,A.E., 1995.Bayesfactor.J.Am. Stat.Assoc.90,773–795. Mason-Gamer, R.J., Kellog, E.A., 1996. Testing for phylogenetic conflict among molecular data sets in the tribe Triticeae (Gramineae). Syst. Biol. 45, 524–545. McDaniel, S.F., Shaw, A.J., 2003. Phylogeographic structure and cryptic speciation in the trans-antarctic moss Pyrrhobryum mnioides. Evolution 57, 205–215. Mendelson, T.C., Shaw, K.L., 2005. Sexual behaviour: rapid speciation in an arthropod. Nature 433, 375–376. Nylander, J.A.A., Ronquist, F., Huelsenbeck, J.P., Nives-Aldrey, J.L., 2004. Bayesian phylogenetic analysis of combined data. Syst. Biol. 53, 47–67. Odrzykoski, I.J., Szweykowski, J., 1991. Genetic differentiation without concordant morphological divergence in the thallose liverwort Conocephalum conicum. Pl. Syst. Evol. 178, 135–151. Pedersen, N., Holyoak, D.T., Newton, A.E., 2006. Systematics and morphological evolution within the moss family Bryaceae: a comparison between parsimony and Bayesian methods for reconstruction of ancestral character states. Molec. Phylogen. Evol. 43, 891–907. Pellmyr, O., Segraves, K.A., Althoff, D.M., Balca´zar-Lara, M., LeebensMack, J., 2007. The phylogeny of yuccas. Molec. Phylogen. Evol. 43, 493–501.
J. Hentschel et al. / Molecular Phylogenetics and Evolution 45 (2007) 693–705 Pennington, R.T., Lavin, M., Prado, D.E., Pendry, C.A., Pell, S.K., Butterworth, C.A., 2004. Historical climate change and speciation: neotropical seasonally dry forest plants show patterns of both Tertiary and Quaternary diversification. Philos. Trans., Ser. B 359, 515–537. Persson, H., 1955. Remarks on the Porella pinnata group. Arch. Soc. Zool. Bot. Fenn. ‘‘Vanamo’’ 9, 225–231. Poe, S., Chubb, A.L., 2004. Birds in a bush: five genes indicate explosive evolution of avian orders. Evolution 58, 404–415. Posada, D., Crandall, K.A., 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14, 817–818. Rambault, A., Drummond, A., 2003–2005. Tracer. MCMC Trace Analysis Tool version 1.3, Available from http://evolve.zoo.ox.ac.uk/. Schaumann, F., Frey, W., Pfeiffer, T., Stech, M., 2005. Molecular circumscription, intrafamilial relationships and biogeography of the Gondwanan liverwort family Pallaviciniaceae (Hepaticophytina, Bryophyta). Pl. Syst. Evol. 252, 27–48. Schiffner, V., 1934. Macvicaria und Madotheca ulophylla. Ann. Bryol. 2, 87–106. Schneider, H., Smith, A.R., Cranfill, R., Hildebrand, T.J., Haufler, C.H., Ranker, T.A., 2004. Unravelling the phylogeny of polygrammoid ferns (Polypodiaceae and Grammitidaceae): exploring aspects of the diversification of epiphytic plants. Molec. Phylogen. Evol. 31, 1041–1063. Schuster, R.M., 1980The Hepaticae and Anthocerotae of North America East of the Hundredth Meridian, vol. 4. Columbia University Press, New York. Shaheen, F., Srivastava, S.C., 1989. Porella campylophylla (Lehm. & Lindenb.) Trev. complex in India. Geophytology 19, 34–48. Shaw, A.J., Allen, B., 2000. Phylogenetic relationships, morphological incongruence, and geographic speciation in the Fontinalaceae (Bryophyta). Molec. Phylogen. Evol. 16, 225–237. So, M.L., 2002. The genus Porella (Porellaceae, Hepaticae) in Australasia and the South Pacific. Syst. Bot. 27, 4–13. So, M.L., 2005. Porella (Porellaceae, Marchantiophyta) in Latin America. New Zealand J. Bot. 43, 301–321. Stech, M., Dohrmann, J., 2004. Molecular relationships and biogeography of two Gondwanan Campylopus species, C. pilifer and C. introflexus (Dicranaceae). Monogr. Syst. Bot. Missouri Bot. Gard. 98, 415–431. Stech, M., Osman, S., Sim-Sim, M., Frey, W., 2006. Molecular systematics and biogeography of the liverwort genus Tylimanthus (Acrobolba-
705
ceae). Studies in austral temperate rain forest bryophytes 33. Nova Hedwigia 83, 17–30. Stech, M., Wagner, D., 2005. Molecular relationships, biogeography, and evolution of Gondwanan Campylopus species (Dicranaceae, Bryopsida). Taxon 54, 377–382. Stotler, R., Crandall-Stotler, B., 1977. A checklist of the liverworts and hornworts of North America. Bryologist 80, 405–428. Swails, L.F., 1970. The genus Porella in Latin America. Nova Hedwigia 19, 201–291. Swofford, D.L., 2000. PAUP*, phylogenetic analyses using parsimony (*and other methods), version 4.01b10. Sinauer Associates, Sunderland, Massachusetts. Tamura, K., Nei, M., 1993. Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Molec. Biol. Evol. 10, 512–526. Therrien, J.P., Crandall-Stotler, B.J., Stotler, R.E., 1998. Morphological and genetic variation in Porella platyphylla and P. platyphylloidea and their systematic implications. Bryologist 101, 1–19. Vanderpoorten, A., Goffinet, B., 2006. Mapping uncertainty and phylogenetic uncertainty in ancestral character state reconstruction: an example in the moss genus Brachytheciastrum. Syst. Biol. 55, 957–971. Vanderpoorten, A., Long, D.G., 2006. Budding speciation and neotropical origin of the Azorean endemic liverwort, Leptoscyphus azoricus. Molec. Phylogen. Evol. 40, 73–83. Waits, L.P., Sullivan, J., O’Brien, S.J., Ward, R.H., 1999. Rapid radiation in the family Ursidae indicated by likelihood phylogenetic estimation from multiple fragments of mtDNA. Molec. Phylogen. Evol. 13, 82– 92. Werner, O., Guerra, J., 2004. Molecular phylogeography of the moss Tortula muralis Hedw (Pottiaceae) based on chloroplast rps4 gene sequence data. Pl. Biol. 6, 147–157. Wilson, R., Gradstein, S.R., Schneider, H., Heinrichs, J., 2007. Unravelling the phylogeny of Lejeuneaceae (Jungermanniopsida): evidence for four main lineages. Molec. Phylogen. Evol. 43, 270– 282. Wyatt, R., Odrzykoski, I.J., Cronberg, N., 2005. High levels of genetic variation in the haploid leafy liverwort Porella platyphylla from the southeastern United States. J. Bryol. 27, 247–252.
