Molecular Phylogenetics and Evolution 59 (2011) 489–509 Contents lists available at ScienceDirect Molecular Phylogenetics and Evolution journal homepage: www.elsevier.com/locate/ympev A multi-locus molecular phylogeny of the Lepidoziaceae: Laying the foundations for a stable classification Endymion D. Cooper a,b,⇑, A. Jonathan Shaw c, Blanka Shaw c, Murray J. Henwood a, Margaret M. Heslewood b, Elizabeth A. Brown b a b c School of Biological Sciences, University of Sydney, NSW 2006, Australia National Herbarium of New South Wales, Mrs Macquaries Road, Sydney NSW 2000, Australia Department of Biology, Duke University, Durham, NC 27708, USA a r t i c l e i n f o Article history: Received 10 November 2010 Revised 31 January 2011 Accepted 2 February 2011 Available online 18 February 2011 Keywords: Lepidoziaceae Jungermanniopsida Molecular phylogeny Classification Neogrollea Zoopsidioideae a b s t r a c t The Lepidoziaceae, with over 700 species in 30 genera, is one of the largest leafy liverwort families. Despite receiving considerable attention, the composition of subfamilies and genera remains unsatisfactorily resolved. In this study, 10 loci (one nuclear 26S, two mitochondrial nad1 and rps3, and seven chloroplast atpB, psbA, psbT-psbH, rbcL, rps4, trnG and trnL-trnF) are used to estimate the phylogeny of 93 species of Lepidoziaceae. These molecular data provide strong evidence against the monophyly of three subfamilies; Lepidozioideae, Lembidioideae and Zoopsidoideae, and seven of the 20 sampled genera; Lepidozia, Telaranea, Kurzia, Zoopsis, Lembidium, Paracromastigum and Chloranthelia. Several robust clades are recognised that might provide the basis for a revised subfamily circumscription including a narrower circumscription of the Lepidozioideae and a more inclusive Lembidioideae. Neogrollea notabilis is returned to the Lepidoziaceae and Megalembidium insulanum is placed in the Lembidioideae. Ó 2011 Elsevier Inc. All rights reserved. 1. Introduction The Lepidoziaceae is one of the largest families of leafy liverworts (Jungermanniales) with over 2200 published binomials, but perhaps as few as 720 accepted species, in 30 genera (CrandallStotler et al., 2009; Müller, 2007). With a cosmopolitan distribution, the family has its greatest species diversity in the southern hemisphere leading some to postulate a Gondwanan origin (Schuster, 2000), which is consistent with an early to mid-cretaceous origin of the family (Heinrichs et al., 2007). Unrivalled in the breadth of gametophyte morphological diversity, the Lepidoziaceae are united by conserved sporophyte morphology and isophyllous gynoecial branches (Schuster, 2000). Nevertheless, the morphological contrast between the robust leafy shoots of Bazzania and Lepidozia and the lax-celled, pseudo-thalli of Zoopsis and Zoopsidella, was sufficient for some authors to divide the taxa amongst six families (Evans, 1939; Fulford, 1963a, 1966, 1968; Nakai, 1943). Furthermore, the degree of gametophyte polymorphism within genera of the Lepidoziaceae is at times greater than that exhibited by entire families of Jungermanniales (Schuster, 2000). This morphological diversity is accompanied by an equally broad range of ecological ⇑ Corresponding author. Address: Rm 409 Heydon-Laurence Building (A08), School of Biological Sciences, University of Sydney, NSW 2006, Australia. Fax: +61 2 9351 4119. E-mail address: endymion.cooper@sydney.edu.au (E.D. Cooper). 1055-7903/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.ympev.2011.02.006 differentiation, with plants occurring in habitats ranging from sandy alpine soils to lowland tree trunks. Recent treatments of the Marchantiophyta (Crandall-Stotler and Stotler, 2000; Crandall-Stotler et al., 2009) provide a conservative precis of the Lepidoziaceae and do not propose a structure below family level. Crandall-Stotler et al. (2009) noted in their review of liverwort classification, based on a decade of phylogenetic analyses, that broader sampling within the Lepidoziaceae would be necessary before a revision of its internal classification could be presented. The most recent and complete subfamilial classification is that of Schuster (2000), an account that largely follows Schuster (1969), but places Drucella integristipula, Neogrollea notabilis and Protocephalozia ephemeroides in monotypic subfamilies. In the absence of robust modern phylogenetic hypotheses, extensive taxonomic study of the family has not provided stable subfamily, generic or subgeneric delimitations. For example, the monotypic genus Neogrollea was recently removed from the Lepidoziaceae to its own monotypic suborder Neogrollineae (Engel and Braggins, 2001) and Megalembidium insulanum was transferred from Lepidozioideae to a monotypic Megalembidioideae (Engel and Braggins, 2005). The delimitation of genera has followed often geographically restricted traditional morphological treatments and as a result generic limits remain fluid (Engel and Merrill, 1994, 2004; Engel and Schuster, 2001; Evans, 1934; Grolle, 1964; Hodgson, 1955; Schuster, 1980). 490 E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 Despite their significance in biological systematics, modern phylogenetic methods have only been applied to the Lepidoziaceae in the last decade (Engel and Merrill, 2004; Heslewood and Brown, 2007; Renner et al., 2006). A cladistic study of Telaranea, based on 32 morphological characters, led to a reinterpretation of the genus and its synonymisation with Arachniopsis (Engel and Merrill, 2002, 2004). A second study also indicated the paraphyly of Zoopsis and Zoopsidella, but the authors suggested that limitations of morphological cladistics might have hampered their attempt to recover an accurate phylogeny (Renner et al., 2006). Several molecular phylogenetic studies of liverworts have included representatives of the Lepidoziaceae (Davis, 2004; Forrest et al., 2006; He-Nygrén et al., 2006; Heinrichs et al., 2007; Hendry et al., 2007) but the few taxa included therein have been sufficient only to indicate monophyly of the family. Heslewood and Brown (2007) provided the first explicit test of Schuster’s (2000) classification. Their analyses of Australasian representatives of the family using two chloroplast loci (trnL-trnF and rbcL) and one mitochondrial gene (nad5) revealed extensive paraphyly of subfamilies and genera, suggested possible polyphyly of the family, and contradicted the works of Engel and Braggins (2001, 2005). They emphasised the need for further taxonomic sampling, including greater representation of the Lembidioideae and morphologically reduced Telaranea species in the molecular phylogeny (Heslewood and Brown, 2007). In this paper, we use a large molecular dataset (10 loci) and a broad taxonomic sample (93 species from 20 genera) to investigate phylogenetic relationships between species belonging to the Lepidoziaceae. We aim to test the monophyly of the family and evaluate the phylogenetic basis for removing Neogrollea to the Neogrollineae. By testing the monophyly of subfamilies and genera we aim to provide a phylogenetic foundation for future taxonomic revisions. 2. Materials and methods 2.1. Taxonomic sample Species were sampled haphazardly from available material in order to encompass as much taxonomic and geographic diversity as possible. We sampled 93 species from 20 genera, in seven subfamilies. Zoopsis leitgebbiana, Lembidium nutans, Arachniopsis major and Hygrolembidium rigidum were represented by multiple accessions because preliminary data (E.D. Cooper, unpublished) suggested that these species are not monophyletic. We included nine outgroup taxa based on the phylogenies of Forrest et al. (2006) and He-Nygrén et al. (2006) (see Appendix A for complete accession information). 2.2. Molecular methods DNA extraction was performed using a modification of the CTAB protocol (Doyle and Doyle, 1987) or using a DNeasy Plant Minikit as previously described (Heslewood and Brown, 2007). One nuclear, two mitochondrial and seven chloroplast markers were sequenced (Table 1). All PCR reactions were completed under standard conditions. Each 10 ll reaction contained 1 PCR Buffer, 0.16 mM dNTPs, 2 mM MgCl2, 0.5 lm each primer, 0.75 mg/ml BSA and 0.2 units Taq polymerase. Genomic DNA was added as 0.5 ll to each reaction without standardising concentration. A standard temperature profile was used: 95 °C for 1 min, then 30 cycles of 95 °C for 1 min, 50 °C for 1 min and 72 °C for 1 min, followed by a final extension step of 72 °C for 10 min. The same primers were used for sequencing (Table 1). 2.3. Sequence alignment and phylogenetic analyses Initial sequence alignments were produced using MUSCLE (Edgar, 2004) on the European Bioinformatics Institute web server (www.ebi.ac.uk) and then manually edited in Mesquite v2.72 (Maddison and Maddison, 2009). Coding sequences were aligned to Marchantia polymorpha sequences from Genbank (chloroplast genome X04465 and mitochondrial genome M68929) and translated to check for premature stop codons. Maximum parsimony (MP) analyses were run using PAUP⁄ v4.0 (Swofford, 2002) as implemented on the University of Oslo bioportal (www.bioportal.uio.no). Branch swapping was performed using TBR and 500 random sequence addition replicates, and the strict Table 1 Details of markers used in this study. a b c Marker Genome Primer [sequence 50 –30 ] References nr26S Nuclear nad1 Mitochondrial rps3 Mitochondrial atpB Chloroplast psbA Chloroplast psbT-psbH Chloroplast rbcL Chloroplast rps4 Chloroplast trnG Chloroplast trnL-trnF Chloroplast LS0F [ACCCGCTGTTTAAGCATAT] LS12R [ATCGCCAGTTCTGCTTACCA] Mp-nad1-1 [ATGAGAATTTATCTAATTGG] Mp-nad1-966 [TTAAGGRAGCCATTCAAAGGCT] rps3F1 [TCGTAGTTCAGATTCCAGTTG] rps3R2 [CCAAAACGTACAAAATTTCG] atpB672F [TTGATACGGGAGCYCCTCTWAGTGT] atpE384R [GAATTCCAAACTATTCGATTAGG] trnK2F [GACGAGTTCCGGGTTCGA] psbA576R [TGGAATGGGTGCATAAGG] psbA501F [TTTCTCAGACGGTATGCC] trnHR [GAACGACGGGAATTGAAC] psbT [ATGGAAGCWTTAGTWTATACWTT] psbH [GTHCCCCARCCDGGDRVHACTTTWCC] mtrnRR [GCTCTAATCCACTGAGCTAC] nm34 [GTTGTTGGATTTAAAGCTGGTGTT] rbcL1 [GGGATTTATGTCACCACAAACAGA] rbcL2 [GATCTCCTTCCATACTTCACAAGC] rps5 [ATGTCCCGTTATCGAGGACCT] trnas [TACCGAGGGTTCGAATC] trnGF [ACCCGCATCGTTAGCTTG] trnGR [GCGGGTATAGTTTAGTGG] trnF [ATTTGAACTGGTGACACGAG] trnC [CGAAATCGGTAGACGCTACG] Shaw (2000) Shaw (2000) Dombrovska and Qiu (2004) Dombrovska and Qiu (2004) This studya This studya Forrest and Crandall-Stotler (2004)b Forrest and Crandall-Stotler (2004)b Forrest and Crandall-Stotler (2004)c Forrest and Crandall-Stotler (2004)c Forrest and Crandall-Stotler (2004)c Forrest and Crandall-Stotler (2004)c Krellwitz et al. (2001) Krellwitz et al. (2001) Cox et al. (2000) Cox et al. (2000) Gadek and Quinn (1993) Gadek and Quinn (1993) Nadot et al. (1994) Souza-Chies et al. (1997) Pacak and Szweykowska-Kuliñska (2000) Pacak and Szweykowska-Kuliñska (2000) Taberlet et al. (1991) Taberlet et al. (1991) Designed for the Liverwort Tree of Life (LiToL: www.biology.duke.edu/bryology/LiToL). Source given as P. Wolf (http://bioweb.usu.edu.wolf/atpB%20primer%20map.htm). Source given as C. Cox, pers. comm. E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 consensus of most parsimonious trees was computed. Autapomorphic and constant characters were excluded prior to the analysis. Clade support was assessed by conducting 10,000 bootstrap replicates with five random sequence addition replicates per bootstrap replicate. Clades are considered supported if found in P70% of the bootstrap replicates and strongly supported if found in P90% of replicates. Rapid maximum likelihood (ML) analyses were run using RAxML v7.2.6 (Stamatakis, 2006) provided on the CIPRES portal (Miller et al., 2010). Analyses used the GTRGAMMA model and 10,000 bootstrap replicates were run. Clade support is assessed using the same thresholds as for maximum parsimony bootstrapping. Prior to Bayesian analysis, model selection was completed using jModeltest (Posada, 2008). The highest ranked model according to the Akaike information criterion (AIC, Akaike, 1973; Burnham and Anderson, 2004) was used for Bayesian analyses. Where the highest ranked model was not the GTR + C + I model, analyses were also performed with the GTR + C + I model and their performance compared using Bayes Factors (Kass and Raftery, 1995). Bayesian inference (BI) was run using a hybrid version (Pratas et al., 2009) of Mr. Bayes (Huelsenbeck and Ronquist, 2001) as implemented on the CIPRES Portal (Miller et al., 2010). Two independent metropolis coupled MCMC runs were run for five million generations, using one heated and three cold chains, and sampling every 1000 generations. After plotting log likelihood values and calculating the standard deviation of split frequencies to check convergence, the first 1001 samples were discarded as burnin. The 10 loci were analysed separately and then concatenated by genome. A concatenated 10 locus dataset was also analysed. For BI analyses of concatenated datasets, a homogeneous analysis (single partition) using the GTR + C + I model and a heterogeneous analysis (partitioned by marker, using the best model per partition) were run. Clades with a posterior probability P0.9 are considered supported and those with posterior probability P0.95 are considered strongly supported. 3. Results 3.1. Sampling We sampled 93 species, representing seven subfamilies and 20 genera, for an average of eight out of ten markers each. The number of sequences per marker ranged from 66 (atpB) to 89 (psbA). A total of 784 new sequences were generated for this study (Appendix A). 3.2. Model selection The AIC universally preferred the general time reversible model (GTR) with gamma distributed rates and, in all but two cases, a proportion of invariant sites (Table 2). For rps3 the GTR + C model was preferred, but the DAIC between it and the next best GTR + C + I model was within the interval of strongly supported models (Burnham and Anderson, 2004). For trnG, the GTR + C + I model had considerably less support than the GTR + G model (DAIC 7.98) but was still within the approximate cut off for supported models of 10 AIC units (Burnham and Anderson, 2004). Despite this, analyses under the GTR + C + I model resulted in better marginal likelihoods for these markers than the GTR + C model. For the combined analyses, partitioning by marker always improved model fit. 3.3. Phylogenetic analyses Support values from individual BI, ML and MP analyses of the 10 loci are summarised in Table 2. Maximum likelihood and MP boot- 491 strap percentages for analyses of concatenated datasets are shown on the Bayesian topologies (Figs. 1–3 and 5). All trees are available as online Supplementary material (Appendix B). 3.3.1. Nuclear data The single nuclear marker (26S) was the least variable of the ten regions used. Of 1009 aligned base pairs, 77 (8%) were parsimony informative (Table 2). Sequences were available for 76 of the 98 ingroup accessions. Little incongruence is observed between the three methods of analysis, which yield largely unresolved phylogenetic estimates with few supported nodes (Fig. 1). The family is strongly supported in MP, ML and BI analyses of 26S (pp 1, ML-BS 100, MP-BS 100) and is separated by a long branch from the outgroup (phylogram not shown). Additionally, the family is characterised by a four base pair deletion at position 662 and a three base pair insertion at position 917. Four species of Zoopsidoideae are resolved as sister to the rest of the family at node B, which is supported in BI and ML analyses (pp 0.96, ML-BS 77). An unsupported clade comprised of Lepidozia and some Telaranea species is recovered in all analyses. Eight out of nine Bazzania species are consistently resolved as a clade with strong support in BI analysis (pp 0.95). The two Paracromastigum species are sisters (pp 0.99), as are the two Pseudocephalozia species (pp 1, ML-BS 99, MP-BS 99). 3.3.2. Mitochondrial data The concatenated mitochondrial dataset had 89 ingroup accessions of which 70 had data for both nad5 and rps3. We were unable to obtain nad1 sequences for Telaranea meridiana and T. pulcherrima var. mooreana. The concatenated alignment had 2119 positions of which 246 (12%) were parsimony informative. The resolution, support values and topology are consistent across BI, ML and MP analyses, and are largely congruent with topologies from the individual marker analyses (Fig. 2, Table 2). Monophyly of the family is supported (pp 0.92, ML-BS 94, MP-BS 90). Node B, which places Zoopsids I and II sister to the remainder of the family, has weak support (pp 0.9). Bazzania (pp 0.99, ML-BS 89, MP-BS 80) and Acromastigum (pp 0.96) are always recovered as sister clades, but are separated by relatively long branches (phylogram not shown). The Lembidioid/Kurzia clade is consistently recovered without support, and Pseudocephalozia is separated from the remainder of this clade (node iv) by a long branch. The Lepidozia clade is also recovered without support but has minimal internal resolution. Unlike the individual analyses, the combined mitochondrial topologies always recover an unsupported monophyletic Zoopsids III. Relationships amongst these clades are unresolved, as are the relationships of Paracromastigum, Hyalolepidozia, Drucella and Neogrollea. 3.3.3. Chloroplast data The combined chloroplast alignment had 6569 positions of which 1951 (30%) were parsimony informative. A total of 552 sequences for 98 ingroup accessions represents a 80% complete sequence set. The Lepidoziaceae are strongly supported (pp 0.99, ML-BS 100, MP-BS 99) and several early divergent nodes are resolved and supported in Bayesian analyses (nodes A–E, Fig. 3). Six clades corresponding approximately to current generic concepts are recovered, Hygrolembidium, Pseudocephalozia, Bazzania, Acromastigum, Kurzia, and Lepidozia with varying support (Table 2). Zoopsis and Telaranea are dispersed throughout the tree, and three clades containing these genera are labelled in Fig. 3 (Zoopsids I–III). Three clades of Telaranea (i.e. Telaranea I–III) form consecutive sister relationships with Lepidozia. The monotypic genera Neogrollea, Psiloclada, Sprucella, Megalembidium and Isolembidium have supported relationships with other taxa. Drucella, however, occupies a phylogenetically isolated position. 492 Table 2 Summarised support values for major clades of interest from all individual and combined analyses (pp/ML-BS/MP-BS). Clade names follow Figs. 1–5. The aligned length, proportion of parsimony informative (PI) sites and best fit model (the number of partitions are indicated) are shown for each dataset. Where insufficient representation is available for a particular clade, the support values are labelled not available (n/a). The following abbreviations are used when nodes are unsupported: ns = recovered but not supported, nr = not recovered, cn = topological incongruence not supported, and cs = topological incongruence with support. nr26S nad1 rps3 atpB psbA psbT-psbH rbcL rps4 trnG trnL-trnF 770 33% GTR + C + I Concatenated mitochondrial 2119 12% 2  GTR + C + I Concatenated chloroplast 6569 30% 7  GTR + C + I Total evidence (all markers) 9697 23% 10  GTR + C + I Aligned length PI sites Best model 1009 8% GTR + C + I 943 12% GTR + C + I 1176 12% GTR + C 1143 27% GTR + C + I 1151 26% GTR + C + I 541 26% GTR + C + I 1386 26% GTR + C + I 612 41% GTR + C + I 966 42% GTR + C Lepidoziaceae Node A Node B Node C Node D Node E Node F Bazzanioideae Lembidioid/Kurzia clade Lepidozia clade Zoopsids I Zoopsids II Zoopsids III Paracromastigum clade Node i Acromastigum Bazzania Pseudocephalozia Lembidioids Node ii. Node iii. Kurzia Node iv. Node v. Telaranea II Node vi. Lepidozia Telaranea III Node vii. 1/100/100 cn/cn/cn 0.96/77/ns cn/cn/nr cn/cn/cn cn/cn/cn cn/cn/cn cn/cn/cn nr/ns/nr ns/ns/ns n/a nr/nr/cn cs/cn/cn n/a n/a cs/cn/cn nr/cn/nr 1/99/99 nr/cn/nr nr/cn/nr nr/cn/nr nr/cn/nr nr/cn/nr nr/cn/nr cn/cn/nr cn/cn/nr cn/cn/cn cn/cn/cn cn/cn/cn 1/97/89 nr/cn/cn nr/ns/cn nr/cn/cn cn/cn/cn cn/cn/cn cn/cn/cn nr/cn/cn 1/ns/ns ns/ns/ns 1/97/98 nr/cn/nr nr/ns/cn cn/cn/cn 0.93/76/70 ns/ns/nr ns/ns/nr 1/100/ns nr/ns/nr ns/80/nr 0.99/ns/ns nr/cn/cn 0.96/ns/ns n/a cn/cn/nr nr/ns/nr cn/ns/nr cn/cn/nr cn/ns/ns 1/100/99 cn/cn/nr 0.94/ns/ns cn/cn/nr cn/cn/nr cn/cn/nr cn/cn/nr 0.93/ns/ns 1/94/96 ns/ns/nr 1/100/100 1/100/100 cn/cn/nr cn/cn/nr 1/99/99 1/86/76 0.91/81/ns 1/100/100 ns/ns/ns 0.98/98/ns 1/92/89 ns/ns/ns 1/100/100 n/a ns/ns/nr 0.95/ns/ns ns/ns/ns n/a 1/87/81 1/98/94 n/a n/a ns/ns/nr cn/cn/nr cn/cn/nr nr/cn/nr ns/ns/nr 1/100/100 1/99/100 n/a n/a cn/cn/nr 1/75/ns 1/100/100 1/96/89 1/100/99 1/100/100 1/85/76 1/94/97 1/99/98 0.96/ns/nr 0.99/79/ns 0.92/85/nr 0.99/92/77 1/98/95 ns/ns/nr n/a 1/99/96 1/89/ns cn/ns/ns cn/cn/cn cn/cn/cn cn/cn/nr cs/cn/nr cn/cn/nr ns/ns/nr 1/87/ns 1/96/82 0.93/ns/ns 0.99/ns/ns cs/cn/nr cn/cn/nr 1/98/89 1/98/89 1/91/94 1/100/100 0.99/90/87 1/92/92 cn/ns/nr cs/cn/cn 0.99/78/nr nr/ns/nr 0.98/74/ns 0.9/ns/nr cn/cn/cn cn/cn/nr ns/ns/ns 0.99/74/ns nr/ns/cn ns/ns/cn nr/ns/cn nr/ns/cn nr/ns/cn nr/cn/cn nr/ns/nr 1/92/87 1/76/nr 0.99/75/ns 1/97/93 nr/ns/ns nr/cn/cn 1/93/93 1/76/ns 0.94/ns/nr 1/100/100 ns/cn/nr 1/99/98 nr/cn/nr nr/ns/nr nr/ns/nr 0.94/ns/nr 0.96/ns/ns nr/ns/nr ns/ns/nr n/a 1/78/75 1/96/75 cs/cn/ns cs/cn/nr cs/cn/cn cs/cn/nr cs/cn/nr nr/cn/cn ns/cn/nr 1/98/86 1/98/91 1/100/100 n/a 1/71/ns nr/cn/cn 1/100/100 1/96/84 1/100/100 1/100/100 0.97/ns/ns 1/81/83 1/93/90 nr/cn/ns 0.97/cn/nr 0.98/83/75 nr/cn/cn ns/ns/ns 0.99/78/79 ns/ns/ns 1/90/80 1/99/98 nr/cn/nr 0.91/ns/ns cn/cn/cn/ nr/cn/nr nr/cn/nr cn/cn/cn cn/cn/cn 1/88/76 1/78/85 n/a 1/100/100 nr/ns/nr 0.93/ns/nr 1/100/100 0.98/ns/nr 1/99/100 1/100/100 1/92/95 1/99/99 1/91/91 nr/ns/nr 0.99/76/81 n/a 1/89/ns nr/cn/cn nr/cn/cn n/a 1/99/97 1/91/84 cn/cn/cn cn/cn/cn cn/cn/cn cn/cn/cn cn/cn/cn cn/cn/cn 0.91/ns/nr 1/97/98 1/100/98 1/100/97 1/98/96 0.97/ns/nr cn/cn/cn 1/100/99 1/99/96 1/99/100 1/100/100 0.99/75/nr 1/98/97 0.99/74/70 nr/cn/cn ns/cn/ns 1/99/99 1/99/96 1/100/97 ns/ns/ns cn/cs/cn 1/98/95 1/99/97 nr/cn/ns/ 0.95/ns/cn cn/cn/cn cn/cn/cn cn/cn/cn cn/cn/cn nr/cn/cn 1/95/86 1/94/93 1/96/86 0.99/86/87 0.99/71/ns cn/cn/cn 1/99/99 1/95/82 1/100/100 1/100/100 1/85/80 1/89/93 1/99/94 cn/cn/nr ns/72/ns cn/cn/nr cs/cn/cn cn/cn/nr cn/cn/nr cn/cn/nr 1/98/95 0.92/94/90 nr/cn/nr 0.9/ns/ns nr/cn/ns nr/cn/cn nr/cn/cn nr/cn/cn ns/ns/ns ns/ns/ns ns/ns/ns 99/96/99 nr/cn/nr ns/ns/ns nr/cn/cn 1/98/98 0.96/ns/ns 0.99/89/80 1/99/100 0.95/75/72 0.99/93/82 1/91/88 nr/cn/nr ns/ns/ns na cn/cn/nr cn/cn/nr cn/cn/nr cn/cn/nr ns/ns/nr 0.99/100/99 ns/ns/ns 0.99/ns/ns 0.91/ns/ns 0.97/ns/cn 0.97/ns/cn nr/nr/ns 0.99/92/ns 0.99/100/100 0.99/100/100 0.99/100/100 0.99/100/100 0.99/92/ns 0.99/86/ns 1/100/100 1/100/100 1/100/100 1/100/100 1/100/100 1/100/100 1/100/100 cn/cn/cn 0.99/100/100 0.99/100/100 0.99/100/97 1/100/99 0.99/84/86 ns/ns/nr 1/100/100 0.99/100/100 nr/cn/75 0.99/96/76 ns/ns/ns ns/ns/ns ns/ns/cn ns/cn/ns 0.99/95/ns 0.99/100/100 0.99/100/100 1/100/100 1/100/100 0.98/93/70 0.99/83/ns 1/100/100 1/100/100 0.99/100/100 1/100/100 1/100/100 1/100/100 1/100/100 nr/ns/nr 0.99/100/100 0.99/100/99 0.99/99/95 0.99/100/100 0.99/85/86 ns/ns/nr 1/100/100 E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 Alignment E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 493 Fig. 1. Majority rule consensus tree from Bayesian analysis of the nuclear 26S dataset. Numbers above branches are posterior probabilities >0.9, maximum likelihood bootstraps >70%, and maximum parsimony bootstraps >70% (pp/ML/MP). Node B is discussed in the text and support values for all analyses given in Table 2. 494 E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 Fig. 2. Majority rule consensus tree from Bayesian analysis of the concatenated mitochondrial dataset partitioned by marker. Numbers above branches are posterior probabilities >0.9, maximum likelihood bootstraps >70%, and maximum parsimony bootstraps >70% (pp/ML/MP). Nodes B and (i)–(iv) are discussed in the text and support values for all analyses presented in Table 2. E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 495 Fig. 3. Majority rule consensus tree from Bayesian analysis of the concatenated chloroplast dataset partitioned by marker. Numbers above branches are posterior probabilities >0.9, maximum likelihood bootstraps >70%, and maximum parsimony bootstraps >70% (pp/ML/MP). Nodes A–E and (i)–(vii) are discussed in the text and support values are given in Table 2. 496 E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 Fig. 4. Majority rule consensus tree from Bayesian analysis of the concatenated chloroplast dataset partitioned by marker. Branch lengths are drawn proportional to the average number of substitutions per site. The topology is identical to that shown in Fig. 3. E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 497 Fig. 5. Majority rule consensus tree from Bayesian analysis of the concatenated 10 locus (total evidence) dataset partitioned by marker. Numbers above branches are posterior probabilities >0.9, maximum likelihood bootstraps >70%, and maximum parsimony bootstraps >70% (pp/ML/MP). Nodes B–F and (i)–(vii) are discussed in the text. 498 E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 Three large clades frequently recovered in individual analyses, and strongly supported in the combined analyses, are illustrated in Fig. 3. These are, the clade containing Bazzania and Acromastigum (Bazzanioideae; pp 0.99, ML-BS 92), the clade containing the lembidioid species, Kurzia, and Pseudocephalozia (Lembidioid/Kurzia clade; pp 0.99, ML-BS 100, MP-BS 100), and the clade containing Lepidozia and some Telaranea species (Lepidozia clade; pp 0.99, ML-BS 100, MP-BS 100). Branch lengths corresponding to early divergences (i.e. subtending nodes A–E) are relatively short, as is the branch that subtends the Bazzanioideae and those that separate Telaranea I–III and Lepidozia (Fig. 4). 3.3.4. Total evidence The 10 locus, three genome, concatenated dataset had 778 sequences for 98 ingroup accessions (79%) and a total of 9697 alignment positions of which 2274 (23%) were parsimony informative. Of the 2274 parsimony informative characters, 1951 (86%) were contributed by the seven chloroplast loci. The Lepidoziaceae are strongly supported (pp 0.99, ML-BS 100, MP-BS 100; Fig. 5). Maximum parsimony analysis provides support for node A, which places Zoopsids I sister to the remainder of the Lepidoziaceae, but this node is not recovered in BI analysis (MP-BS 75). Node A, however, is contradicted by an unsupported sister relationship between Zoopsids I and II in the ML analysis. The Paracromastigum clade is an unsupported sister to the Bazzanioideae in BI and MP analyses (node F) but this is incongruent with the ML topology, in which the Paracromastigum clade is sister to the other clades resolved at node C. With the exception of nodes A and F, the topology and support values are largely the same as the chloroplast tree. 4. Discussion Biological classification aims to facilitate communication. In order to serve this aim well, classification should be maximally predictive. A widely held belief, frequently applied in practise, is that maximally predictive classifications are those based on the theory of descent with modification. In addition, communication is best served by stability of classification. We take the view that predictive content and stability are promoted by naming monophyletic groups such that the least nomenclatural change is required. By monophyletic we mean groups that contain all, and only, descendants of a common ancestor (monocladistic, sensu Podani, 2010). 4.1. Incongruence between the molecular phylogeny and the subfamilial and generic classification of the Lepidoziaceae The currently accepted broad definition of the Lepidoziaceae (Schuster, 2000) is corroborated by the monophyly of the family in individual and concatenated analyses of ten molecular markers across three genomic compartments. The strong evidence for a monophyletic Lepidoziaceae, gives some support to Schuster’s (1969) choice of morphological characters as reliable predictors of phylogeny, i.e. a wide array of branching modes including general retention of lateral-terminal and ventral-intercalary branches; distinct, usually large, underleaves with rhizoids restricted to the basal sector; strictly isophyllous gynoecial branches of determinate length; long, usually slender, trigonous perianths; flagelliform/stoloniferous branches; sporophytes with the seta having a fixed number of cortical cell rows (usually 8 or 16 ± 1) and numerous rows of much smaller, medullary cells; and a capsule wall that undergoes a two-phase development. The molecular phylogeny strongly supports placement of N. notabilis within the Lepidoziaceae. Engel and Braggins (2001), pro- viding the first description of its sporophyte, considered Neogrollea morphologically distinct from the remainder of the Lepidoziaceae and erected a monotypic family Neogrolleaceae. They listed nine morphological characters which they consider preclude placing the new family in the Lepidoziineae. Only one of these characters, baculate spore ornamentation, strictly distinguish Neogrollea from the Lepidoziaceae. Of the remaining characters, Neogrollea is simply at the extreme end of the range or variation seen in the Lepidoziaceae. Engel and Braggins (2001) are incorrect in their statement that reddish pigments are not produced in the Lepidoziaceae. Several species of Paracromastigum and Hyalolepidozia microphylla produce reddish to brown pigments. Indeed, the morphology of Neogrollea fits within the diagnostic criteria given by Schuster (1969), suggesting that focussing attention on perceived differences in typical expressions of character states, rather than similarity in general morphological trends, might obscure phylogenetic relatedness. Given that Neogrollea shares character similarity with several distantly related members of the Lepidoziaceae, we hypothesise that a pleisiomorphic morphology is retained in this species. In contrast to the strong support for a monophyletic Lepidoziaceae, is the lack of support, and often evidence against, the monophyly of several subfamilies and genera. Of the five polytypic subfamilies, only Bazzanioideae are monophyletic. Lembidioideae are paraphyletic by the exclusion of Megalembidium and Kurzia dendroides. Lepidozioideae and Zoopsidoideae are polyphyletic. Neither Micropterygium nor Mytilopsis were sampled and monophyly of Micropterygioideae remains untested. Furthermore, the subfamily status of two of the three sampled monotypic genera, Neogrollea and Megalembidium, is brought into question. Seven of the 20 genera sampled are not monophyletic. The monophyly of five genera, Sprucella, Hygrolembidium, Amazoopsis, Hyalolepidozia and Zoopsidella, remains untested as only one representative of each was sampled. Monophyly is supported for only three genera, Acromastigum, Bazzania and Pseudocephalozia. The five remaining genera, Psiloclada, Drucella, Neogrollea, Isolembidium and Megalembidium, are monotypic (Engel and Glenny, 2008; Schuster, 2000). 4.2. Unresolved early divergence Despite a dataset containing over 2000 parsimony informative sites, only the Bayesian analysis of the concatenated chloroplast data provides support for early divergent relationships. It is possible that resolving the sequence of early divergence is hampered by missing lineages. However, phylograms show a clustering of short branches around these basal nodes and in analyses of individual markers alternative, unsupported arrangements were recovered (Appendix B Supplementary figures). Thus, an alternative explanation is that this pattern of incongruent but unsupported gene trees, with relatively short basal branch lengths is the result of rapid divergence of main lineages. Unresolved basal relationships are also found in ferns (Schneider et al., 2006), mosses (Buck et al., 2000) and other liverworts (Hentschel et al., 2009; Wilson et al., 2007a). Few attempts have been made to date liverwort divergence but rapid establishment of main lineages of Lejeuneaceae in the cretaceous (Wilson et al., 2007b) is consistent with the hypothesis that spore producing plants diversified following, or concurrently with, the establishment of angiosperm forests (Newton et al., 2007; Schneider et al., 2004). The chronogram of Heinrichs et al. (2007) is consistent with rapid basal diversification in Lepidoziaceae at this time. The extent of this pattern needs to be empirically tested and the conceptual basis for reflecting such information in classification schemes needs consideration. E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 4.3. Zoopsids I and II The entire suite of gametophyte morphologies observed in the Jungermanniales can be found within the Zoopsidioideae (Schuster, 2000). Given this degree of polymorphism, it is not surprising that this subfamily is polyphyletic. Zoopsids I and Zoopsids II, are resolved sister to the remainder of the family (node B), whereas Zoopsids III, is sister to the Lembidioid/Kurzia clade. Zoopsids I contains Telaranea major, Zoopsidella caledonica and Zoopsis ceratophylla. Little can be concluded about the relationships of Zoopsidella because only Z. caledonica was included in this analysis and morphological data indicate a closer relationship to Z. ceratophylla than to other Zoopsidella species (Renner et al., 2006). In resolving T. major with species of Zoopsis, Heslewood and Brown (2007) took this as evidence supporting Schuster’s (2000) concept of Arachniopsis and Telaranea. In doing so they emphasised the need for further representation of morphologically reduced Telaranea species in the molecular phylogeny. Our results place these species in three different clades, namely Zoopsids I, the Lembidioid/Kurzia clade, and the Zoopsids III, lending support to the suggestion that morphological convergence is at play (Heslewood and Brown, 2007). These genera are discussed further under the relevant clades. The Zoopsids II clade is comprised of Zoopsis argentea, Z. leitgebiana, and Z. macrophylla. The later two species are similar in their morphology, sharing succubous leaf insertion and characteristically asymmetric bilobed leaves. However, their morphological connection with Z. argentea, which is more similar to Z. liukiuensis (Schuster, 2000), is unclear and it is separated from the rest of the clade by a long branch. In our analyses, and the earlier analyses of Heslewood and Brown (2007), Z. liukiuensis is resolved sister to Z. setulosa (Zoopsid III clade in this study). The unsupported early divergent node A separates the Zoopsids I and Zoopsids II clades. However, in several individual marker analyses these clades are either resolved as a polytomy or as sister taxa. Consequently, a sister relationship between these two clades cannot be ruled out. Although Z. argentea, the nomenclatural type of Zoopsis, is placed in Zoopsids II, further molecular sampling is required before clade membership can be determined and an appropriate taxonomic treatment provided. 4.4. Drucella integristipula The monotypic Drucella is placed in an isolated position amongst the early diverging nodes of the family. Only in BI analysis of rbcL was a supported placement sister to Zoopsids I recovered (pp 0.98), but a conflicting relationship with Neogrollea is weakly supported in BI analysis of psbA (pp 0.9). Neither placement reflects previous taxonomic treatments of D. integristipula, which was first described as a Lepidozia (Stephani, 1909) and was until recently retained in the Lepidozioideae. The phylogenetic isolation of Drucella was anticipated by Schuster (1980) on the basis of gametophyte morphology and this notion gained further support when sporophytes were discovered (Boesen, 1982). In contrast to the molecular analysis of Heslewood and Brown (2007), we consistently resolved Drucella within the family. Our phylogeny supports retention of a monotypic subfamily for this taxon. 4.5. Paracromastigum clade Paracromastigum, currently assigned to the Zoopsidoideae, has a complex taxonomic history (Fulford, 1968; Fulford and Taylor, 1961; Schuster, 2000; Schuster and Engel, 1996). We recover two species of Paracromastigum in a clade with H. microphylla, Chloranthelia denticulata and N. notabilis. This clade is not recovered in the analyses of nuclear and concatenated mitochondrial datasets, but 499 it is supported in the analysis of chloroplast and total evidence datasets. A subclade containing Paracromastigum microstipum, H. microphylla and C. denticulata is always supported, confirming the conclusion of Heslewood and Brown (2007) that it is unlikely that C. denticulata belongs with the lembidioids. Paracromastigum, Hyalolepidozia and Chloranthelia bergrennii were not included in their study, and they resolved a sister relationship with Z. leitgebbiana. Our analyses do not support an association between C. denticulata and Z. leitgebbiana, and place Chloranthelia bergrenii within the lembidioid clade. Several authors have speculated that C. bergrenii, with its strongly lembidioid appearance, would show different phylogenetic affinities to those of C. denticulata, the type of the diatypic genus (Engel and Glenny, 2008; Heslewood and Brown, 2007; Schuster, 2000; Schuster and Engel, 1987). The placement of H. microphylla sister to P. microstipum supports Schuster’s (2000) suggestion that Hyalolepidozia should be reduced to a subgenus of Paracromastigum. However, the type species of the diatypic genus, Hyalolepidozia bicuspidata, was not sampled. Morphologically these species are distinguished from Paracromastigum by their bifid leaves and their highly reduced stems, which consist of only six cortical cell rows (Schuster, 2000). Similar reduction in leaf and stem size occurs in unrelated lineages of Lepidoziaceae and does not always indicate phylogenetic relationships. Although the phylogeny supports transfer of H. microphylla to Paracromastigum, the generic status of Hyalolepidozia remains untested. The monotypic Neogrollea was recently removed from the Lepidoziaceae and placed in a monotypic suborder, Neogrollineae (Engel and Braggins, 2001), but phylogenies based on molecular data (Hendry et al., 2007; Heslewood and Brown, 2007) have consistently placed it within the Lepidoziaceae. Schuster (1972) placed it tentatively in the Lembidioideae but recently moved it to its own subfamily and suggested possible links to the Bazzanioideae (Schuster, 2000). In the analyses of the concatenated total evidence dataset Neogrollea is sister to the rest of the Paracromastigum clade, a position that better reflects its morphological uniqueness than placement within Paracromastigum as suggested by the chloroplast analyses. The two Paracromastigum species and H. microphylla are all from New Zealand and inclusion of south American representatives would be desirable. The phylogeny supports transfer of C. denticulata, and H. microphylla to Paracromastigum and suggests subfamily status for the clade comprised of Paracromastigum and Neogrollea would be appropriate. 4.6. Bazzanioideae Morphologically the best defined subfamily, Bazzanioideae are monophyletic but with little support. A close relationship between Bazzania and Acromastigum has long been hypothesised based on shared pseudo-dichotomous terminal branching and formation of frequent geotropic, flagelliform ventral branches (Evans, 1934, 1939; Fulford, 1963b; Schuster, 2000). Bazzania is a genus of between 100 (Schuster, 2000) and 450 species (Stephani, 1909) and its internal classification is widely viewed as unsatisfactory (Engel and Glenny, 2008; Meagher, 2008; Schuster, 1969, 2000). The placement of Bazzania nitida sister to the rest of the genus is in agreement with Heslewood and Brown (2007) who resolve two other members of sect. Vittatae, B. monilinervis and B. tayloriana, in this position. Whilst this indicates that there might be support for some morphologically defined subgeneric groupings, sampling is currently insufficient to evaluate the competing hypotheses (Fulford, 1963a; Schuster, 2000). Acromastigum exile from the phylogenies of Davis (2004) and Forrest et al. (2006) was included in the analyses. It is consistently nested in the Bazzania clade (Bazzania sp. Mb28). Following examination of the voucher we are of the opinion that the specimen, which has 500 E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 shallowly tridentate falcate leaves, ventral intercalary stolons and lacks ventral terminal (Acromastigum-type) branches, is a Bazzania. Acromastigum is a smaller genus and our resolution of three clades agrees with the results of Heslewood and Brown (2007) and has some similarities with the current subgeneric classification. We resolve the elobate nearly isomorphic A. stellare and A. caevifolium from subg. Acromastigum as sister to the remainder of the genus. A clade of small, wiry, pigmented species, A. mooreanum, A. anisostomum and A. furcatifolium, from subg. Inaequilaterae, and the small wiry A. filum from subg. Subcomplicatae, form an unsupported clade. The third clade consists of Bazzania-like species from subgenus Inaequilaterae. The phylogeny indicates that morphologically circumscribed subgeneric groups exist within this genus. 4.7. Zoopsids III The Zoopsids III clade is comprised of Psiloclada clandestina, Amazoopsis diplopoda, two species of Zoopsis, and four species of Telaranea. These small plants share reduced stems and leaves, often without a disk present, and a preference for sheltered microhabitats (Schuster, 2000). Of the four Telaranea species, T. diacantha and T. herzogii were previously assigned to Arachniopsis, which is currently considered a synonym of Telaranea (Engel and Merrill, 2002, 2004). Telaranea maorensis was recently described (Pócs, 2006) and closely resembles T. pectin, also previously an Arachniopsis. The fourth, T. pseudozoopsis, has not been previously assigned to the genus Arachniopsis. It is possible that the nomenclatural types of Telaranea and Arachniopsis belong within the Zoopsids III clade and not with the bulk of Telaranea species, here placed with Lepidozia. The type of Telaranea, Telaranea chaetophylla, has been the subject of persistent nomenclatural confusion; sometimes considered a synonym of Telaranea nematodes (e.g. Howe, 1902; Schuster, 1969, 2000) or a synonym of Telaranea (Arachniopsis) sejuncta (Fulford, 1963b), but on recent revision it was recognised as a distinct species (Engel and Merrill, 2004). It is morphologically similar to T. herzogii, which is placed in the Zoopsids III. Telaranea coactilis, the type of Arachniopsis, was considered a synonym of Telaranea (Arachniopsis) diacantha (Pócs, 1984). Although, it is morphologically similar to T. diacantha, Engel and Merrill (2004) recognise T. coactilis as a separate species. We were unable to obtain sequence data for either T. chaetophylla or T. coactilis and cannot confirm their phylogenetic placement. Nevertheless, it is clear from the current phylogeny that Telaranea species with reduced stem and leaf anatomy exhibit a convergent morphology and more complete taxonomic sampling will be necessary before taxonomic changes are made. 4.8. Lembidioid/Kurzia clade The Lembidioid/Kurzia clade is comprised of Kurzia, some Telaranea species, the Lembidioideae, Megalembidium and Pseudocephalozia. A subclade at node iv, containing taxa of the Lembidioideae and Kurzia was recovered by Heslewood and Brown (2007), and is strongly supported in our analyses. Within this clade a further subclade, which contains species mostly referable to Lembidioideae, is strongly supported. However, two additional species are resolved within this clade, M. insulanum (Megalembidioideae) and K. dendroides. The placement of M. insulanum with the lembidioids is in contradiction to its placement in the Lepidozioideae (Schuster, 2000) and its placement in a monotypic subfamily (Engel and Braggins, 2005). The molecular data presented here provide very strong support for a sister relationship with Isolembidium. The association of these elements, namely Megalembidium, the lembidioids, and Kurzia, is hinted at in the controversy surrounding K. dendroides which has been variously placed in Lembidium, Kurzia or Dendrolembidium (Heslewood and Brown, 2007; Schuster, 2000). The results presented here clearly show that this species is not a Kurzia and provide support for recognition of a separate genus. However, the paraphyly of Lembidium indicates that generic limits need to be revisited in this clade. The placement of C. bergrenii within the Lembidioid/Kurzia clade demonstrates that it has no phylogenetic affinity with the type, C. denticulata, of the diatypic genus and perhaps should be returned to Lembidium (Herzog, 1952; Schuster and Engel, 1987). However, in analyses of concatenated mitochondrial and total evidence datasets, C. bergrenii is sister to Hygrolembidium with which it shares a lack of terminal branching. Furthermore, Lembidium is paraphyletic to the exclusion of Hygrolembidium and Chloranthelia, and re-evaluation of the morphological characters separating these genera is necessary. It is possible that a broader generic concept for Lembidium should be applied to encapsulate these various elements. The lack of support for a monophyletic Kurzia suggests that the delimitation of this genus also needs revision. Nevertheless, strong support for the clade containing Kurzia and the lembidioids demonstrates that the recent transfer of Kurzia species with reduced leaf and stem anatomy to Telaranea (Engel and Merrill, 2004) is not compatible with the phylogeny. 4.9. Lepidozia clade Lepidozia is paraphyletic to the exclusion of Sprucella. However, Sprucella has often been considered a synonym of Lepidozia (Pócs, 1994; Vanden Berghen, 1983) and the phylogeny indicates that it should not be retained as a separate genus. Lepidozia is placed in a clade with several subclades of Telaranea (i.e. Telaranea I–III). These three clades contain the Lepidozia-like members of Telaranea, which were included in Lepidozia until Fulford and Taylor (1959) erected the genus Neolepidozia to accommodate them. Common to these species are the deeply lobed leaves, except in Sprucella, with a distinct disc present. Telaranea centipes, which is resolved sister to the remainder of the clade, is placed in Telaranea sect. Ceraceae (Engel and Merrill, 2004) with other glaucous Telaranea species. This monophyletic section (E. D. Cooper, unpublished data) is not phylogenetically associated with Lepidozia subg. Glaucophylla as represented by L. bisbifida, implying multiple origins of waxy cuticle. Telaranea II contains taxa with a reduced leaf disc and elongated to ±filiform lobes. The form of the lobes and short disc are similar to the reduced leaves seen in Telaranea species placed in the Zoopsids I, Zoopsids III and Lembidioid/Kurzia clades, and are in some species, accompanied by a similar preference for damp niches. Telaranea III is resolved as sister to Lepidozia (node vi) in many analyses, often with strong support. Given that the clade contains the most Lepidozia-like of the Telaranea species, this association is not surprising. Taxa placed in this clade (i.e., of Lepidozia and Telaranea III) have 3–4 lobed leaves and a distinct disc. In Lepidozia these leaves are asymmetric. The association between the four components of the Lepidozia clade recovered in all but two analyses (Table 2) indicates strong support for the clade. The morphological similarity of these four clades suggests that returning to a broader concept of this genus might be appropriate. Our analysis included only 33 species from this clade and internal relationships are not consistently resolved. A study focussing on this clade is underway and aims to provide the phylogenetic background for a taxonomic revision. 4.10. Perspectives and conclusions Our data from 93 species, representing 20 genera in seven subfamilies, significantly improves upon previous phylogenetic studies of the Lepidoziaceae (Engel and Merrill, 2004; Heslewood and E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 Brown, 2007; Renner et al., 2006). The extent of incongruence between the current classification and the phylogeny indicates that reassessment of morphological homologies and, in some groups, intensive taxonomic sampling are necessary. Consequently, a fully revised classification is not presented here but several preliminary taxonomic conclusions can be made: 1. Neogrollea notabilis belongs within Lepidoziaceae and subfamily status is not warranted. This change is formalised below. 2. Megalembidium insulanum should not be retained in a monotypic subfamily and this change is formalised below. 3. Drucella integristipula is phylogenetically isolated, and retention of a monotypic Drucelloideae reflects this. 4. Despite weak support, Bazzanioideae are likely to be monophyletic and no changes to the classification are required at this time. 5. Expanding the definition of Lembidioideae to encompass all members of the Lembidioid/Kurzia clade is appropriate. 6. Delimitation of Lembidium and Kurzia need to be revised, but provisional transfer of C. bergrenii back to Lembidium and resurrection of Dendrolembidium for K. dendroides are justified. 7. The definition of Lepidozioideae should be restricted to include only Telaranea I, II and III and Lepidozia. Sprucella is not an autonomous genus and should be accommodated within Lepidozia. 8. The Zoopsidioideae are a polyphyletic assemblage of morphologically convergent taxa. Until adequate taxonomic sampling is completed, the Zoopsids I–III, and the Paracromastigum clade should be considered unaligned. The phylogeny suggests that recognition of three or four subfamilies and significant revision of generic boundaries are required. 9. Transfer of H. microphylla and C. denticulata to Paracromastigum is warranted and this change will be made elsewhere. Several phylogenetically important taxa remain unsampled or under-sampled. The Micropterygioideae, for example, are yet to be included in phylogenetic studies of the family. The phylogenetic incongruence in Zoopsis and Telaranea indicate that a more thorough taxonomic sampling within these groups is desirable before a stable, unequivocal classification is possible. In particular, inclusion of South and Central American species is necessary to improve the geographically biased sampling within Pseudocephalozia, Zoopsidella, Hygrolembidium, and Paracromastigum. Nevertheless, the phylogeny presented here provides a solid foundation for evaluating and revising the classification. 4.11. Taxonomic changes Lepidoziaceae Limpr., in Cohn, Krypt.-Fl. Schlesien 1: 310 (1877). 501 Lepidozieae Limpr., in Cohn, Krypt.-Fl. Schlesien 1: 310 (1877). Zoopsidaceae Nakai, in Ogura, Ord. Fam. Trib. Gen. Sect. . . . nov. ed. 199. 1943. Bazzaniaceae Nakai, in Ogura, Ord. Fam. Trib. Gen. Sect. . . . nov. ed. 200. 1943. Hyalolepidoziaceae Fulford, Mem. New York Bot. Gard. 11: 376. 1968. Regredicaulaceae Fulford, Mem. New York Bot. Gard. 11: 358. 1968. Paracromastigaceae Fulford, Mem. New York Bot. Gard. 11: 384. 1968. Neogrolleaceae (R.M.Schust.) J.J.Engel & Braggins, J. Hattori Bot. Lab. 91: 195. 2001. syn. nov. Type: Lepidozia (Dumort.) Dumort., Recueil Observ. Jungerm. 19, 1835. Lepidoziaceae subfamily Lembidioideae R.M.Schust., Hepat. Anthocerotae N. Amer. 2: 11, 1969. Lepidoziaceae subfamily Megalembidioideae J.J.Engel, in J.J.Engel and Braggins, J. Hattori Bot. Lab. 97: 88. 2005. syn. nov. Type: Lembidium Mitt., in Hook.f., Handb. N. Zeal. Fl. 2: 751, 754. 1867. Acknowledgments We would like to thank the New Zealand Department of Conservation/Te Papa Atawhai, and people of Te Roroa, Piki te Aroha Marae and Te Rünanga for permission to collect specimens from their land. We thank John Braggins for his invaluable assistance in the field. Matt von Konrat and John Engel (F), and Jochen Heinrichs (GOET) provided specimens. Sandy Boles and Carolyn Porter provided E.D.C. with technical assistance for laboratory work. We thank Matthew Renner for useful comments on an earlier draft of this manuscript. An Australian Biological Resources Study, student travel bursary (to E.D.C.) partially funded field work in New Zealand. The Hermon Slade Foundation supported DNA extraction at NSW (Grant HSF 03-14 to E.A.B.). This research was supported by US National Science Foundation Grant EF-0531730-002 (to A.J.S.). Appendix A. List of taxa, voucher information and accession numbers. Ingroup taxa are listed alphabetically before outgroup taxa (also alphabetically). Accession numbers in bold indicate sequences generated for this study. Superscripts following accession numbers indicate place of publication as follows: aForrest et al. (2006); b Heslewood and Brown (2007). Source nr26S nad1 Acromastigum adaptatum Hürl. Acromastigum anisostomum (Lehm. et Lindenb.) A.Evans Acromastigum cavifolium R.M.Schust. Acromastigum cunninghamii (Steph.) A.Evans Acromastigum divaricatum (Nees) A.Evans Acromastigum filum (Steph.) A.Evans Acromastigum furcatifolium (Steph.) E.A.Br. ms Acromastigum mooreanum (Steph.) E.A.Hodgs. Acromastigum stellare N.Kitag. Amazoopsis diplopoda (Pócs) J.J.Engel et G.L.Merr. Bazzania sp. Mb28 E.A.Brown, 06/ 137, NSW E.C.Davis, NZ40, DUKE New Caledonia New Zealand JF315899 JF316100 JF316378 JF315966 JF316167 JF315905 JF316097 - E.A.Brown, 08/ 278, NSW New Zealand - JF316102 JF316373 - J.J.Engel 23662, F New Zealand JF315888 JF316096 - E.A.Brown 04/ 84e, NSW Papua New Guinea New Caledonia Australia - JF316098 - Bazzania affinis (Lindenb. et Gottsche) Trevis. Bazzania intermedia (Lindenb. et Gottsche) Trevis. Bazzania involuta (Mont.) Trevis. Mastigobryum [Bazzania] lenormandii Steph. Mastigobryum [Bazzania] limbatum Steph. Bazzania nitida (F.Weber) Grolle E.A.Brown 03/ 139, NSW E.A.Brown 02/ 100, NSW rps3 rps4 trnG JF316259 JF316320 - JF316541 JF316589 JF316263 JF316321 JF316457 JF316545 JF316588 JF316143 JF316264 JF316302 JF316437 JF316540 JF316585 - JF316168 JF316268 - JF316459 JF316542 JF316590 - JF316169 - - JF316460 JF316543 JF316591 JF315901 JF316101 JF316376 JF316001 JF316166 - EF1009752 - JF316544 EF1010822 - - - JF316165 - EF1009772 JF316456 JF316547 EF1010842 JF316164 - JF316318 JF316455 JF316546 JF316587 - atpB psbA JF316002 JF316163 psbTpsbH rbcL trnL- trnF B.Shaw 6407, DUKE Australia JF315904 JF316103 JF316377 - E.A.Brown 03/ 184, NSW A.Szabo 9607/ CB, New Caledonia Reunion Island - JF316099 JF316375 JF315972 JF316148 JF316198 EF1009782 - JF316528 EF1010852 - - - - - - - - JF316499 JF316582 Chua-Petiot Mb28, NY E.C.Davis, 146, Duke Kenya AY6081921 - - - AY6079191 - - AY6080411 JF316535 JF316636 Jamaica AY6082001 JF316093 JF316386 JF316008 AY6079271 JF316272 DQ4396801 AY6080481 JF316497 JF316633 - JF316089 JF316388 JF316006 JF316153 JF316275 - JF316445 JF316537 - JF315915 JF316095 JF316389 JF316012 JF316155 JF316277 JF316323 JF316448 JF316493 JF316631 JF315916 JF316092 JF316390 JF316003 JF316156 JF316278 EF1009892 JF316449 JF316534 EF1010962 JF315908 JF316091 JF316392 JF316005 JF316152 JF316274 EF1009902 JF316446 JF316536 EF1010972 JF315919 JF316088 JF316385 JF316004 JF316151 JF316276 JF316322 JF316444 JF316492 JF316630 E.A.Brown, 04/ 67f, NSW Papua New Guinea J.J.Engel, 37 New Tarn 1 (2006), F Zealand E.A.Brown, 03/ New 140c, NSW Caledonia D.M.Crayn, 829, NSW Australia J.J.Engel, Bearley New 36 (2006), F Zealand E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 Voucher [Collector, date, Herbarium] 502 Taxon Kurzia brevicalycina (Steph.) Grolle Kurzia calcarata (Steph.) Grolle Kurzia compacta (Steph.) Grolle Kurzia dendroides (Carrington et Pearson) Grolle Kurzia gonyotricha (Sande Lac.) Grolle Kurzia helophila R.M.Schust. Kurzia hippurioides G.P.Rothero, 11022, E D.Long, 34721, E D.Kosma, SIU culture IV T.Yamaguchi, 29079, F Scotland - JF316080 - JF316010 JF316154 JF316269 JF316282 JF316447 JF316494 JF316632 China JF315877 JF316094 JF316381 JF316009 JF316157 JF316271 JF316310 JF316451 JF316498 JF316635 U.S.A. JF315876 JF316090 JF316387 JF316007 JF316113 JF316273 - JF316453 JF316496 JF316634 Japan JF315878 JF316081 JF316391 JF316011 JF316158 JF316270 - JF316452 JF316495 JF316637 E.A.Brown, 08/ 282, NSW E.A.Brown, 03/ 147, NSW New Zealand New Caledonia JF315931 JF316035 JF316354 JF315960 JF316126 JF316248 JF316296 JF316423 JF316520 JF316601 - JF316085 JF316382 - JF316266 EF1009912 - - E.A.Brown, 08/ 219, NSW E.A.Brown, 08/ 301, NSW New Zealand New Zealand JF315897 JF316075 JF316380 JF315999 JF316147 JF316197 JF316287 JF316436 JF316502 JF316576 - JF316087 JF316384 JF315968 JF316149 JF316267 JF316304 JF316435 JF316527 JF316629 E.A.Brown, 08/ 310 New Zealand JF315932 JF316033 JF316355 - JF316131 JF316252 JF316299 JF316422 JF316522 JF316602 J.J.Engel, 28134, F New Zealand JF315928 JF316032 JF316358 JF315957 JF316129 JF316250 JF316297 JF316420 - E.A.Brown, 08/ 283, NSW New Zealand JF315930 JF316031 JF316357 JF315958 JF316130 JF316251 JF316298 JF316421 JF316521 JF316603 E.A.Brown, 08286, NSW New Zealand JF315926 JF316041 JF316345 JF315962 JF316123 JF316244 JF316294 JF316416 JF316517 JF316600 E.A.Brown, 03/ 140a, NSW E.A.Brown, EAB 03/131, NSW E.D.Cooper, 330, NSW M.A.M.Renner, 02102, AK E.A.Brown, 04/ 02, NSW New Caledonia New Caledonia New Zealand New Zealand Australia JF315890 JF316029 JF316342 - JF316237 JF316292 JF316415 JF316508 JF316595 JF315918 JF316047 JF316349 JF315963 JF316112 JF316239 EF1009932 JF316427 JF316511 EF1011022 JF315921 JF316043 JF316347 JF315965 JF316121 JF316240 JF316301 JF316428 JF316513 JF316596 - - - - JF316512 EF1011042 JF315920 JF316039 JF316341 JF315956 JF316124 JF316246 EF1009962 JF316418 JF316518 EF1011052 M.von Konrat, 6/8-8b, F E.D.Cooper, 383, NSW E.A.Brown, 04/ Fiji - JF316046 JF316348 - - JF316426 JF316510 JF316604 New Zealand Australia JF315923 JF316042 JF316350 JF315964 JF316133 JF316241 JF316300 JF316431 JF316515 JF316597 JF315924 JF316044 JF316346 JF315951 JF316132 JF316243 EF1009972 - JF316514 - - - JF316109 JF316117 JF316120 - EF1009952 - EF1011002 - 503 (continued on next page) E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 Bazzania pearsonii Steph. Bazzania tricrenata (Wahlenb.) Lindenb. Bazzania trilobata (L.) Gray Bazzania yoshinagana (Steph.) Steph. ex Yoshin. Chloranthelia berggrenii (Herzog) R.M.Schust. Chloranthelia denticulata (Steph.) R.M.Schust. Drucella integristipula (Steph.) E.A.Hodgs. Hyalolepidozia microphylla R.M.Schust. ex J.J.Engel Hygrolembidium rigidum R.M.Schust. et J.J.Engel Hygrolembidium rigidum R.M.Schust. et J.J.Engel Hygrolembidium rigidum R.M.Schust. et J.J.Engel Isolembidium anomalum (Rodway) Grolle Kurzia sp. 504 Appendix A (continued) Voucher [Collector, date, Herbarium] (Hook.f. et Taylor) Grolle Kurzia trichoclados (Müll.Frib.) Grolle Lembidium longifolium R.M.Schust. Lembidium nutans (Hook.f. et Taylor) Mitt. ex A.Evans Lembidium nutans (Hook.f. et Taylor) Mitt. ex A.Evans Lepidozia sp. 1 42, NSW Lepidozia sp. 2 Source nr26S nad1 rps3 atpB psbA psbTpsbH rbcL rps4 trnG trnL- trnF JF316319 JF316430 - - G.P.Rothero, 11014, E M.A.M.Renner, 2530, F E.A.Brown, 02/ 322, NSW UK JF315875 JF316045 JF316351 JF315955 - - New Zealand New Zealand JF315927 JF316037 JF316352 - JF316128 JF316253 - JF316425 - JF315922 JF316034 JF316353 - JF316125 JF316247 EF1009922 JF316419 JF316519 EF1010462 J.J.Engel, 28415, F New Zealand JF315929 JF316036 JF316356 JF315959 JF316127 JF316249 JF316295 JF316424 - - M.von Konrat, 6/20-4, F B.Shaw, 5789, DUKE E.D.Cooper, 430, NSW D.G.Long, 29184, E T.Yamaguchi, 29076, F E.A.Brown, 04/ 08, NSW Fiji JF315909 JF316068 JF316408 JF315996 JF316180 JF316219 JF316325 JF316478 - - China JF315934 JF316062 JF316405 JF315991 JF316189 JF316218 JF316307 JF316477 - - JF315940 JF316060 JF316410 JF315995 JF316193 JF316223 JF316305 JF316480 JF316560 JF316627 - JF316063 JF316403 JF315992 JF316191 JF316220 - JF316474 JF316556 JF316623 JF315933 JF316066 JF316406 JF315988 JF316187 JF316214 - JF316476 JF316551 JF316620 JF315906 JF316067 JF316409 JF315994 JF316192 JF316224 EF1010002 JF316472 JF316559 EF1011092 JF315871 JF316055 - - - JF316210 - - JF316555 JF316622 JF315938 JF316048 - - JF316185 JF316215 - JF316475 - - AY6082291 JF316064 JF316404 JF315990 AY6079621 JF316217 JF316327 AY6080831 - - JF315907 JF316059 JF316402 JF315998 JF316110 JF316221 JF316326 JF316471 JF316552 JF316625 JF315935 JF316061 JF316407 JF315989 JF316188 JF316216 EF1010022 - JF316554 EF1011112 JF315898 JF316058 - - JF316479 JF316558 JF316624 Lepidozia bisbifida New Steph. Zealand UK Lepidozia cupressina (Sw.) Lindenb. Lepidozia fauriana Japan Steph. Australia Lepidozia laevifolia (Hook.f. et Taylor) Gottsche, Lindenb. et Nees N.J.M.Gremmen, Gough Lepidozia laevifolia 2000-0385, F Island (Hook.f. et Taylor) Gottsche, Lindenb. et Nees Lepidozia microphylla J.J.Engel, 23699, New (Hook.) Lindenb. F Zealand Lepidozia reptans (L.) Sargent living Dumort. culture collection, Berkeley, UC Lepidozia setigera Steph. J.J.Engel, 41 Bog New 20, F Zealand Lepidozia spinosissima M.A.M.Renner, New (Hook.f. et Taylor) MR810, NSW Zealand Mitt. Papua Lepidozia trichodes E.A.Brown, 04/ New (Reinw. ex Blume et 67g, NSW JF315997 JF316179 JF316324 E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 Taxon Telaranea sp. 2 Telaranea centipes (Taylor ex Gottsche, Lindenb. et Nees) R.M.Schust. Telaranea chaetocarpa (Pearson) Grolle Telaranea diacantha (Mont.) J.J.Engel et G.L.Merr. Telaranea herzogii (E.A.Hodgs.) E.A.Hodgs. Telaranea heterotexta (Steph.) J.J.Engel et G.L.Merr. Telaranea lawesii (Steph.) Grolle E.A.Brown, 08/ 273, NSW M.A.M.Renner, 02100a, NSW Guinea New Zealand New Zealand E.A.Brown, 08/ 312, NSW E.A.Brown, 08/ 315, NSW New Zealand New Zealand E.A.Brown, 08/ 218, NSW JF315939 JF316065 JF316411 JF315993 JF316190 JF316222 JF316306 JF316473 JF316557 JF316626 JF315925 JF316040 JF316344 JF315961 JF316122 JF316245 EF1010052 JF316417 JF316516 EF1011152 JF315891 JF316104 JF316394 JF315948 JF316114 JF316254 JF316288 JF316443 JF316523 JF316605 JF315887 JF316079 JF316393 JF316000 JF316170 JF316234 JF316283 JF316461 JF316532 JF316586 New Zealand JF315874 JF316086 JF316383 JF315967 JF316150 JF316265 JF316303 JF316434 JF316526 JF316628 E.A.Brown, 08/ 216, NSW New Zealand JF315893 JF316027 JF316365 JF315949 JF316144 JF316235 JF316289 JF316413 JF316505 JF316593 E.A.Brown, 08/ 293, NSW New Zealand JF315894 JF316028 JF316366 JF315950 JF316145 JF316236 JF316290 JF316414 JF316506 JF316594 E.A.Brown, 03/ 131c, NSW T.Pocs, 97108/D, F J.J.Engel, 18Apr08, F M.von Konrat, 6/7-9, F D.A.Meagher, 4237, NSW New Caledonia Uganda JF315868 JF316084 JF316379 JF315971 JF316159 JF316262 EF1017002 JF316458 JF316533 EF1011162 - - - - JF316173 JF316211 - JF316470 JF316550 JF316617 Fiji - - - - JF316111 JF316232 - - JF316563 JF316614 Fiji JF315896 JF316073 - JF315980 JF316177 JF316231 - - JF316562 JF316613 Australia JF315895 - JF315974 JF316172 JF316205 JF316311 - JF316549 JF316616 - E.A.Brown, 03/ 146a, NSW T.Pocs, 03156/N, F New JF315886 Caledonia Dominican JF315902 Republic JF316049 JF316367 JF315973 JF316171 JF316206 EF1010112 JF316462 JF316548 EF1011202 JF316106 JF316363 - JF316160 - - JF316454 - J.J.Engel, 23927, F New Zealand JF315903 JF316108 JF316364 - JF316162 - JF316309 JF316440 JF316539 JF316639 E.A.Brown, 03/ 129, NSW New Caledonia JF315910 JF316071 JF316398 JF315979 JF316175 JF316229 EF1010122 JF316464 JF316564 EF1011212 E.A.Brown, 04/ 84h, NSW Papua New Guinea JF315885 JF316053 JF316374 JF315977 JF316195 JF316207 JF316332 JF316469 - - E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 Nees) Gottsche Lepidozia ulothrix (Schwägr.) Lindenb. Megalembidium insulanum (W.Martin et E.A.Hodgs.) R.M.Schust. Neogrollea notabilis E.A.Hodgs. Paracromastigum drucei (R.M.Schust.) R.M.Schust. Paracromastigum macrostipum (Steph.) R.M.Schust. Pseudocephalozia lepidozioides R.M.Schust. Pseudocephalozia paludicola R.M.Schust. Psiloclada clandestina Mitt. Sprucella succida (Mitt.) Steph. Telaranea sp. 1 JF316606 (continued on next page) 505 506 Appendix A (continued) Taxon Voucher [Collector, date, Herbarium] nr26S nad1 New Zealand JF315884 JF316054 - Fiji - JF316023 JF316335 - New Caledonia Comoros JF315873 JF316024 JF316336 - rps3 atpB psbA rps4 trnG JF316209 JF316316 JF316482 JF316570 JF316608 - - - JF316500 - JF316116 JF316204 JF316333 - JF316501 JF316583 - JF316025 JF316339 JF315947 JF316115 JF316203 JF316284 - - New Zealand JF315936 - - - JF316484 JF316571 JF316615 New Zealand - - JF316399 - JF316226 - JF316466 JF316566 JF316610 New Zealand JF315872 JF316030 - - - - New Zealand Australia JF315914 JF316072 JF316400 JF315987 JF316184 JF316228 JF316317 JF316468 JF316568 JF316612 JF315867 JF316069 JF316401 JF315985 JF316181 JF316225 EF1010162 JF316465 JF316565 EF1011252 Chile JF315889 JF316056 JF316337 - JF316176 - JF316312 JF316483 - Chile JF315900 JF316107 - JF316161 - JF316308 - JF316538 JF316638 New Zealand JF315883 - JF316485 JF316573 JF316619 Australia AY6082561 JF316050 JF316396 JF315983 AY6079931 JF316212 DQ4397041 AY6081141 JF316572 JF316618 Australia - JF316052 - JF315978 - - JF316331 - - Chile JF315937 JF316057 - - - JF316315 - JF316553 JF316621 Australia JF315879 JF316026 JF316340 JF315952 JF316146 JF316238 JF316285 - JF316507 JF316592 JF315976 JF316194 JF315982 - JF316182 JF315953 JF316118 - JF316397 JF315984 JF316174 JF316186 psbTpsbH rbcL - JF316330 - JF316213 - trnL- trnF - JF316598 - - E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 J.J.Engel, 23605, Telaranea lindenbergii NSW (Gottsche) J.J.Engel et G.L.Merr. Telaranea major Herzog M.von Konrat, 6/26-15, F Telaranea major Herzog E.A.Brown, 03/ 180, NSW Telaranea maorensis T.Pocs, 05090/B, F Pócs Telaranea martinii J.J.Engel, 24903, (E.A.Hodgs.) F R.M.Schust. J.J.Engel, 24863, Telaranea meridiana F (E.A.Hodgs.) E.A.Hodgs. J.J.Engel, 23623, Telaranea pallescens F (Grolle) J.J.Engel et G.L.Merr. Telaranea palmata M.A.M.Renner, J.J.Engel et G.L.Merr. 2522, NSW Telaranea patentissima E.A.Brown, 03/ (Hook.f. et Taylor) 61, NSW E.A.Hodgs. J.J.Engel, 26422, Telaranea plumulosa F (Lehm. et Lindenb.) Fulford Telaranea pseudozoopsis J.J.Engel, 25743, (Herzog) Fulford F M.A.M.Renner, Telaranea pulcherrima 01-100, F var. mooreana (Steph.) J.J.Engel et G.L.Merr. Telaranea pulcherrima H.Streimann, var. pulcherrima 59554, NY (Steph.) R.M.Schust. D.A.Meagher, Telaranea quadriseta 1064, F (Steph.) J.J.Engel et G.L.Merr. Telaranea seriatitexta J.J.Engel, 25953, (Steph.) J.J.Engel F M.A.M.Renner, Telaranea tasmanica 1954, NSW (Steph.) J.J.Engel et G.L.Merr. Source Outgroup species Calypogeia muelleriana (Schiffn.) Müll.Frib. Chiloscyphus semiteres (Lehm.) Lehm. & Lindenb. Herbertus sakurai S.Hatt. Lepicolea attenuata (Miit.) Steph. Lepicolea rara (Steph.) Grolle Mastigophora woodsii (Hook.) Nees D.G.Long, 29502, E UK - JF316051 - JF315975 - JF316208 - JF316481 JF316569 JF316607 J.J.Engel, 24292, F New Zealand JF315913 JF316074 - JF315986 JF316183 JF316227 JF316329 JF316467 JF316567 JF316611 E.A.Brown, 08/ 250, NSW New Zealand JF315917 JF316038 JF316343 JF315954 JF316119 JF316242 JF316293 JF316429 JF316509 JF316599 E.A.Brown, 04/ 829, NSW Papua New Guinea New Caledonia New Zealand JF315892 JF316070 JF316395 JF315981 JF316178 JF316230 JF316328 JF316463 JF316561 JF316609 - - - - - JF316233 EF1010222 - - JF315880 JF316105 - - JF316141 JF316255 - - JF316504 JF316577 E.A.Brown, 03/ 179, NSW J.J.Engel, 23962, F EF1011312 E.A.Brown, 08/ 320, NSW R.Coveny, sn, NSW [614795] New Zealand Australia - JF316022 - - JF316142 JF316199 JF316314 JF316412 JF316503 JF316584 - JF316078 JF316368 - JF316134 JF316256 EF1010232 JF316432 JF316529 EF1011322 B.Shaw, 6319, DUKE Australia JF315881 JF316077 JF316369 - JF316135 JF316258 - - JF316530 - E.A.Brown, 03/ 132, NSW E.A.Brown, 08/ 212, NSW E.A.Brown, 04/ 39, NSW New Caledonia New Zealand Australia JF315911 JF316082 JF316371 JF315969 JF316137 JF316260 EF1010242 JF316441 JF316524 EF1011342 JF315882 JF316076 JF316370 - JF316257 JF316286 JF316433 JF316531 JF316574 JF315912 JF316083 JF316372 JF315970 JF316138 JF316261 EF1010252 JF316442 JF316525 EF1011352 E.C.Davis, 130, DUKE J.J.Engel, 24780, F USA AY6082031 JF316013 - New Zealand JF315869 JF316014 JF316338 JF315942 JF316139 Steere, 74(2)882, NY R.Stotler, 4586, ABSH D.Norris, 66575, DUKE Alaska AY6082161 JF316017 JF316361 JF315945 AY6079491 JF316201 AY6080311 New Zealand Papua New Guinea China DQ2688881 JF316020 JF316360 - AY5074941 JF316280 AY5074101 AY6082271 JF316021 - AY6079601 JF316281 DQ4396911 AY6080811 JF316488 JF316579 JF315870 JF316019 JF316362 JF315946 JF316140 D.Long, 33696, E JF316136 JF315941 AY6079311 JF316196 JF316291 - - JF316334 JF316200 JF316313 AY6080521 JF316438 - JF316487 - AY6080761 - E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 Telaranea tetradactyla (Hook.f. et Taylor) E.A.Hodgs. Telaranea tetrapila (Hook.f. et Taylor) J.J.Engel et G.L.Merr. Telaranea trilobata (R.M.Schust.) J.J.Engel et G.L.Merr. Telaranea wallichiana (Gottsche) R.M.Schust. Zoopsidella caledonica (Steph.) R.M.Schust. Zoopsis argentea (Hook.f. et Taylor) Gottsche, Lindenb. et Nees Zoopsis ceratophylla (Spruce) Hamlin Zoopsis leitgebiana (Carrington et Pearson) Bastow Zoopsis leitgebiana (Carrington et Pearson) Bastow Zoopsis liukiuensis Horik. Zoopsis macrophylla R.M.Schust. Zoopsis setulosa Leitg. - AY5074521 JF316491 JF316578 JF316439 JF316486 JF316581 (continued on next page) 507 508 E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 AY6081181 JF316016 JF316359 JF315944 AY6079971 JF316279 AY6080401 AY6082611 Ecuador E.C.Davis, 368, DUKE JF315943 AY6079961 JF316202 DQ4397061 AY6081171 JF316490 JF316580 JF316018 AY6082591 Venezuela Ricardi, 9730/T, F - AY6079951 JF316015 AY608257 New Zealand J.E.Braggins, 92/ 87, F Temnoma pulchellum (Hook.) Mitt. ex Bastow Triandrophyllum subtrifidum (Hook. & Taylor) Fulford & Hatcher Trichocolea tomentosa (Sw.) Gottsche rps3 atpB psbA psbTpsbH rbcL nad1 nr26S Source DQ4397051 AY6081151 JF316489 JF316575 References Voucher [Collector, date, Herbarium] Appendix A (continued) Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ympev.2011.02.006. Taxon rps4 trnG trnL- trnF Appendix B. Supplementary material Akaike, H., 1973. Information theory as an extension of the maximum likelihood principle. In: Petrov, B.N., Csaki, F. (Eds.), Second International Symposium on Information Theory. Akademiai Kiado., Budapest, pp. 267–281. Boesen, D.F., 1982. The taxonomy of Drucella Hodgs. and Druceleae, trib. nov. and their position within Lepidozioideae Limpr. Lindbergia 8, 77–88. Buck, W.R., Goffinet, B., Shaw, A.J., 2000. Testing morphological concepts of orders of pleurocarpous mosses (Bryophyta) using phylogenetic reconstructions based on trnL–trnF and rps4 sequences. Mol. Phylogenet. Evol. 16, 180–198. Burnham, K.P., Anderson, D.R., 2004. Multimodel inference. Understanding AIC and BIC in model selection. Social. Methods Res. 33, 261–304. Cox, C.J., Goffinet, B., Newton, A.E., Shaw, A.J., Hedderson, T.A., 2000. Phylogenetic relationships among the Diplolepideous-alternate mosses (Bryidae) inferred from nuclear and chloroplast DNA sequences. Bryologist 103, 224–241. Crandall-Stotler, B., Stotler, R.E., 2000. Morphology and classification of the Marchantiophyta. In: Shaw, A.J., Goffinet, B. (Eds.), Bryophyte Biology. Cambridge University Press, Cambridge, pp. 21–70. Crandall-Stotler, B., Stotler, R.E., Long, D.G., 2009. Phylogeny and classification of the Marchantiophyta. edinb. J. Bot. 66, 155–198. Davis, E.C., 2004. A molecular phylogeny of leafy liverworts (Jungermaniidae: Marchantiophyta). In: Goffinet, B., Hollowell, V., Magill, R. (Eds.), Molecular Systematics of Bryophytes. Missouri Botanic Garden Press, St. Louis, Missouri, pp. 61–86. Dombrovska, O., Qiu, Y.-L., 2004. Distribution of introns in the mitochondrial gene nad1 in land plants phylogenetic and molecular evolutionary implications. Mol. Phylogenet. Evol. 32, 246–263. Doyle, J.J., Doyle, J.L., 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochem. Bull. 19, 11–15. Edgar, R.C., 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucl. Acids Res. 32, 1792–1797. Engel, J.J., Braggins, J.E., 2001. Austral Hepaticae 34. The sporophyte of Neogrollea Hodgs. and the taxonomic position of Neogrolleaceae (Schust.) Engel & Braggins Comb, & Stat. Nov. J. Hattori Bot. Lab. 91, 337–421. Engel, J.J., Braggins, J.E., 2005. Austral Hepaticae 37. The sporophyte of Megalembidium Schust., together with a re-evaluation of the taxonomic position of the genus. J. Hattori Bot. Lab. 97, 81–96. Engel, J.J., Glenny, D., 2008. A Flora of the Liverworts and Hornworts of New Zealand Volume 1. Monographs in Systematic Botany from the Missouri Botanical Garden 110, pp. 1–897. Engel, J.J., Merrill, G.L.S., 1994. Studies of the New Zealand Hepaticae. 8–13. Bazzania and Acromastigum. Bryologist 97, 313–320. Engel, J.J., Merrill, G.L.S., 2002. Proposal to conserve the name Telaranea against Arachniopsis (Hepaticae). Taxon 51, 571–572. Engel, J.J., Merrill, G.L.S., 2004. Austral hepaticae. 35. A taxonomic and phylogenetic study of Telaranea (Lepidoziaceae), with a monograph of the genus in temperate Australasia and commentary on extra-Australasian taxa. Fieldiana Bot. 44, 1– 265. Engel, J.J., Schuster, R.M., 2001. Austral Hepaticae. 32. A revision of the genus Lepidozia (Hepaticae) for New Zealand. Fieldiana Bot. 42, 1–107. Evans, A.W., 1934. A revision of the genus Acromastigum. Ann. Bryol. Suppl. 3, 1– 178. Evans, A.W., 1939. The classification of the Hepaticae. Bot. Rev. 5, 49–96. Forrest, L.L., Crandall-Stotler, B.J., 2004. A molecular phylogeny of the simple thalloid liverworts (Jungermanniopsida, Metzgeriidae) inferred from five chloroplast genes. In: Goffinet, B., Hollowell, V., Magill, R. (Eds.), Molecular Systematics of Bryophytes. Missouri Botanic Garden Press, St. Louis, Missouri, pp. 61–86. 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. Fulford, M., 1963a. Manual of the leafy Hepaticae in Latin America. Part I. Mem. N. Y. Bot. Garden 11, 1–172. Fulford, M., 1963b. Segregate genera of the Lepidozia complex (Hepaticae). Part 4. Telaranea and a review of the Lepidoziaceae. Brittonia 15, 65–86. Fulford, M., 1966. Manual of the leafy Hepaticae in Latin America. Part II. Mem. N. Y. Bot. Garden 11, 1–172. Fulford, M., 1968. Manual of the leafy Hepaticae in Latin America. Part III. Mem. N. Y. Bot. Garden 11, 1–172. Fulford, M., Taylor, J., 1959. The segregate genera of the Lepidozia complex (Hepaticae). Part 1. Sprucella Steph. and Neolepidozia gen. nov. Brittonia 11, 77–85. Fulford, M., Taylor, J., 1961. Segregate genera of the Lepidozia complex (Hepaticae). Part 2. Two new genera, Bonneria and Paracromastigum. Brittonia 13, 334–339. Gadek, P.A., Quinn, C.J., 1993. An analysis of relationships within Cupressaceae sensu stricto based on rbcL sequences. Ann. Missouri Bot. Garden 80, 581–586. Grolle, R., 1964. Notulae hepaticologicae XV. Neue Notizen über Kurzia v. Mart. und Verwandte. J. Jpn. Botany 39, 79–81. E.D. Cooper et al. / Molecular Phylogenetics and Evolution 59 (2011) 489–509 Heinrichs, J., Hentschel, J., Wilson, R., Feldberg, K., Schneider, H., 2007. Evolution of leafy liverworts (Jungermanniidae, Marchantiophyta): estimating divergence times from chloroplast DNA sequences using penalized likelihood with integrated fossil evidence. Taxon 56, 31–44. Hendry, T.A., Wang, B., Yang, Y., Davis, E.C., Braggins, J.E., Schuster, R.M., Qui, Y.-L., 2007. Evaluating phylogenetic positions of four liverworts from New Zealand, Neogrollea notabilis, Jackiella curcata, Goebelobryum unguinculatum, and Herzogianthus vaginatus using three chloroplast genes. Bryologist 110, 738– 751. Hentschel, J., von Konrat, M.J., Pócs, T., Schäfer-Verwimp, A., Shaw, A.J., Schneider, H., Heinrichs, J., 2009. Molecular insights into the phylogeny and subgeneric classification of Frullania Raddi (Frullaniaceae, Porellales). Mol. Phylogenet. Evol. 52, 142–156. He-Nygrén, X., Juslén, A., Ahonen, I., Glenny, D., Piippo, S., 2006. Illuminating the evolutionary history of liverworts (Marchantiophyta)—towards a natural classification. Cladistics 22, 1–31. Herzog, T., 1952. Revision der Lebermoosgattung Lembidium Mitt. Arkiv Bot. N.S. 1, 471–503. Heslewood, M.M., Brown, E.A., 2007. A molecular phylogeny of the leafy liverwort family Lepidoziaceae Limpr. in Australasia. Plant Syst. Evol. 265, 193–219. Hodgson, E.A., 1955. New Zealand Hepaticae (Liverworts)—IX a review of the New Zealand species of the genus Lepidozia. Trans. R. Soc. N. Z. 83, 589–620. Howe, M.A., 1902. Notes on American Hepaticae. Bull. Torrey Bot. Club 29, 281– 289. Huelsenbeck, J.P., Ronquist, F., 2001. MRBAYES: Bayesian inference of phylogeny. Bioinformatics 17, 754–755. Kass, R.E., Raftery, A.E., 1995. Bayes factors. J. Am. Stat. Assoc. 90, 773–795. Krellwitz, E.C., Kowallik, K.V., Manos, P.S., 2001. Molecular and morphological analyses of Bryopsis (Bryopsidales, Chlorophyta) from the western North Atlantic and Caribbean. Phycologia 40, 330–339. Maddison, W.P., Maddison, D.R., 2009. Mesquite: a modular system for evolutionary analysis. Version 2.72. <http://www.mesquiteproject.org>. Meagher, D., 2008. Studies on Bazzania 1. Some new and little known species from Australia. Nova Hedwig. 86, 477–495. Miller, M.A., Holder, M.T., Vos, R., Midford, P.E., Liebowitz, T., Chan, L., Hoover, P., Warnow, T., 2010. The CIPRES Portals. <http://www.phylo.org>. Müller, F., 2007. Meinungeria mouensis (Lepidoziaceae), a new genus and species from New Caledonia. Bryologist 110, 494–499. Nadot, S., Bajon, R., Lejeune, B., 1994. The chloroplast gene rps4 as a tool for the study of Poaceae phylogeny. Plant Syst. Evol. 191, 27–38. Nakai, T., 1943. A list of Prof. Nakai’s Papers with Indices to Names of Plants and Plant-Groups Published as New to Science by Him, Tokyo. Newton, A.E., Wiksrtöm, N., Bell, N., Forrest, L.L., Ignatov, M.S., 2007. Dating the diversification of the pleurocarpous mosses. In: Newton, A.E., Tangney, R.S. (Eds.), Pleurocarpous Mosses: Systematics and Evolution. CRC Press, Boca Raton, pp. 337–366. Pacak, A., Szweykowska-Kuliñska, Z., 2000. Molecular data concerning alloploid character and the origin of chloroplast and mitochondrial genomes in the liverwort species Pellia borealis. J. Plant Biotech. 2, 101–108. Pócs, T., 1984. Synopsis of the African Lepidozioideae K. Müll. In: Vána, J. (Ed.), Proceedings of the Third Meeting of Bryologists from Central and Eastern Europe. Univerzita Karlova, Praha. Pócs, T., 1994. Taxonomic results of the BRYOTROP expedition to Zaïre and Rwanda. Trop. Bryol. 9, 123–130. 509 Pócs, T., 2006. East African Bryophytes, XXI. Two new species of Telaranea, sect. Tenuifoliae and records on Amazoopsis (Lepidoziaceae) from the Indian Ocean Islands. Acta Bot. Hung. 48, 119–138. Podani, J., 2010. Monophyly and paraphyly: a discourse without end? Taxon 59, 1011–1015. Posada, D., 2008. JModelTest: phylogenetic model averaging. Mol. Biol. Evol. 25, 1253–1256. Pratas, F., Trancoso, P., Stamatakis, A., Sousa, L., 2009. Fine-grained parallelism using Multi-core, Cell/BE, and GPU systems: Accelerating the Phylogenetic Likelihood Function. In: Proceedings of the ICPP, Vienna, Austria. Renner, M.A.M., Brown, E.A., Glenny, D., 2006. Two new Zoopsis species and their relationships to other zoopsids (Jungermanniopsida: Lepidoziaceae). J. Bryol. 28, 331–334. Schneider, H., Schuettpelz, E., Pryer, K.M., Cranfill, R., Magallón, S., Lupia, R., 2004. Ferns diversified in the shadow of angiosperms. Nature 428, 553–557. Schneider, H., Kreier, H.-P., Wilson, R., Smith, A.R., 2006. The Synammia enigma: evidence for a temperate lineage of polygrammoid ferns (Polypodiaceae, Polypodiidae) in southern South America. Syst. Bot. 31, 31–41. Schuster, R.M., 1969. The Hepaticae and Anthocerotae of North America. Columbia University Press, New York. Schuster, R.M., 1972. Phylogenetic and taxonomic studies on Jungermaniidae. J. Hattori Bot. Lab. 36, 321–405. Schuster, R.M., 1980. Studies on Hepaticae, LIV–LVII. Kurzia v.Mart [Microlepidozia (Spr.) Joerg.], Megalembidium Schust., Psiloclada Mitt., Drucella Hodgs. and Isolembidium Schust. J. Hattori Bot. Lab. 48, 337–421. Schuster, R.M., 2000. Austral Hepaticae Part 1. Nova Hedwig. 118, 1–542. Schuster, R.M., Engel, J.J., 1987. A monograph of Lepidoziaceae subfam. Lembidioideae (Hepaticae). J. Hattori Bot. Lab. 63, 247–350. Schuster, R.M., Engel, J.J., 1996. Austral Hepaticae. XXI. Paracromastigum fiordlandiae (sp. nov.) and the Delimitation of Paracromastigum and Hyalolepidozia (Lepidoziaceae). Brittonia 48, 165–173. Shaw, A.J., 2000. Phylogeny of the Sphagnopsida based on chloroplast and nuclear DNA sequences. Bryologist 103, 277–306. Souza-Chies, T.T., Bittar, G., Nadot, S., Carter, L., Besin, E., Lejeune, B., 1997. Phylogenetic analysis of Iridaceae with parsimony and distance methods using the plastid gene rps4. Plant Syst. Evol. 204, 109–123. Stamatakis, A., 2006. RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics 22, 2688– 2690. Stephani, F., 1909. Species Hepaticarum, Geneva. Swofford, D.L., 2002. PAUP⁄. Phylogenetic Analysis Using Parsimony (⁄and Other Methods). Sinauer Associates, Sunderland, Massachusetts. Taberlet, P., Gielly, L., Pautou, G., Bouvet, J., 1991. Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Mol. Biol. 17, 1105–1109. Vanden Berghen, C., 1983. Lepidozia Dum. emend. Joerg. subgen Sprucella (Steph.) Vanden Berghen comb. et stat. nov. (Hepaticae). Bull. Jard. Bot. Nat. Belg. 53, 321–330. Wilson, R., Gradstein, S.R., Schneider, H., Heinrichs, J., 2007a. Unravelling the phylogeny of Lejeuneaceae (Jungermanniopsida): Evidence for four main lineages. Mol. Phylogenet. Evol. 43, 270–282. Wilson, R., Heinrichs, J., Hentschel, J., Gradstein, S.R., Schneider, H., 2007b. Steady diversification of derived liverworts under Tertiary climatic fluctuations. Biol. Lett. 3, 566–569.