Preslia 85: 483–503, 2013 483 Oreojuncus, a new genus in the Juncaceae Oreojuncus, nový rod čeledi Juncaceae Lenka Z á v e s k á D r á b k o v á & Jan K i r s c h n e r Department of Taxonomy and Biosystematics, Institute of Botany, Academy of Sciences, Zámek 1, CZ-252 43 Průhonice, Czech Republic, e-mail: drabkova@ibot.cas.cz, jan.kirschner @ibot.cas.cz Záveská Drábková L. & Kirschner J. (2013): Oreojuncus, a new genus in the Juncaceae. – Preslia 85: 483–503. Juncus trifidus and J. monanthos, two species traditionally included in section Steirochloa of the Juncaceae, are shown to differ substantially in their morphology from other members of this section. Their relationships with the other groups in this family based on DNA data of selected regions of plastome, chondriome and nrDNA, were examined using phylogenetic analyses of the combined data set of all these DNA regions. The resultant cladograms place J. trifidus and J. monanthos at a very basal position in the Juncaceae. Both the morphology and the phylogenetic analysis support the exclusion of these two species from Juncus and placing them in a separate genus. This genus is described under the name Oreojuncus Záveská Drábková et Kirschner. A detailed comparison with the other genera of the Juncaceae and a key for identifying these genera are provided. K e y w o r d s: chondriome, Juncaceae, J. monanthos, Juncus trifidus, nrDNA, phylogeny, plastome, taxonomy Introduction Since the early studies on Juncaceae, summarized in Buchenau (1890), the supraspecific system of the genus Juncus has not undergone substantial changes (if we disregard the newly described taxa and changes in the taxonomic rank of individual groups). It is primarily based on the morphology of the inflorescence and leaf anatomy and morphology. The system is very practical for the elementary classification into species groups, which is one of the reasons for its stability. There are several recent studies that elucidate the generic system of the family (Balslev 1996, 1998) and the supraspecific division of Juncus and make necessary nomenclatural adjustments (Novikov 1990, Kirschner et al. 1999, 2002a, b, c). The latter of these studies also provides a complete taxonomic and nomenclatural revision of the Juncaceae (see also Table 1). The widely used system of the family can be outlined as given in Table 1; in the literature individual groups may appear as subgenera but their circumscription, in principle, does not vary. There is, however, a group that represents an exception to the generally accepted system of Juncus, which consists of Juncus trifidus and J. monanthos. The evaluation of the aberrant morphological attributes of this group and results of comprehensive molecular analyses of the Juncaceae (Drábková et al. 2003, 2004, 2006, Záveská Drábková 2010, Záveská Drábková & Vlček 2009, 2010) showing an unexpected position of J. trifidus and J. monanthos make it possible to draw the conclusions presented below. 484 Preslia 85: 483–503, 2013 Previous molecular studies of the Juncaceae and their limitations Even the first molecular studies of samples of Juncaceae cast doubts upon this well established system of genera and sections. Based on plastome rbcL sequences several authors (Chase et al. 1993, Duval et al. 1993, Plunkett et al. 1995, Muasya et al. 1998, 2000) thought that the genus Oxychloë did not belong to the same group as the similar Juncaceous genera, Distichia and Patosia. This, however, was soon reliably disproven by Kristiansen et al. (2005) who analysed in detail both the voucher specimens and the sequences used by the previous authors and show that the previous results were based on contaminated and chimaeric sequences. Starr et al. (2007) arrived at the same conclusions, probably independently, and Oxychloë was returned to its original position. In order to summarize why so much effort was spent on the Oxychloë problem, we have to emphasize the difficulties associated with removing erroneous data from databases, mostly due to the lack of taxonomic expertise and a non-functional system of data validation or verification. Later on, the molecular phylogeny of the Juncaceae attracted attention of several teams. In the majority of the studies, two main problems emerged: the existence of a so called Southern Hemisphere Clade (SHC) and the position of the group of Juncus trifidus L. Drábková et al. (2003) published a rbcL phylogeny characterized by substantially more representative sampling and another analysis of plastome (trnL-trnF intergenic spacer and the trnF intron, Drábková et al. 2004) corroborated the general phylogenetic pattern outlined in the previous work, but not the aberrant position of J. trifidus. However, in a comprehensive phylogeny of cyperids based on the rbcL of more than 320 taxa, Juncus trifidus is a sister taxon to Luzula (Záveská Drábková et al., unpubl.). These studies were followed by that of Roalson (2005), a nrDNA study of a relatively limited sample of species and sections, which did not include three genera; this study again showed the above two problematic groups and in addition highlighted the unstable position of Juncus capitatus. It should be pointed out that Roalson (2005) identified Juncus trifidus as a sister taxon of Luzula. Drábková et al. (2006) summarized the plastome results for an even wider selection of species and corroborated the previous tree topologies that indicated a separate position for the J. trifidus group and elucidated the composition of SHC species (the Juncus sections Juncus, Graminifolii and Caespitosi were also suspected of belonging to it). Unstable position of J. capitatus and J. filiformis was accounted for in terms of a LongBranch Attraction (LBA) phenomenon, an artefact of the MP analyses. Jones et al. (2007) extended the plastome study by including sequences of the rps16 intron combined with the trnL-trnF data, which placed Juncus trifidus in a basal position in the subg. Agathryon clade (however, they only present a single tree of the 206 equally parsimonious trees obtained). The low representativeness of the sampling (less than 40 species of Juncaceae) and the great gaps in the rps16 sequences (five of 10 sections of Juncus are missing and two of three subgenera and three of seven sections in the subg. Luzula not covered) make the results rather unreliable. The most recent studies (Záveská Drábková & Vlček 2009, 2010 and Záveská Drábková 2010) are characterized by extensive taxon sampling, with Luzula almost complete, and the inclusion of sequences of additional DNA regions. A comparison and interpretation of the information content of plastome and chondriome data (Záveská Drábková & Vlček 2009) confirm two remarkable results, i.e. the strange position of the J. trifidus group and the existence of the SHC. The last study (Záveská Drábková 2010) repeated Záveská Drábková & Kirschner: A new genus in the Juncaceae 485 some of the analyses on a wider and corrected selection of taxa and compared the results of the chondriome, plastome and nrDNA studies. Phylogenetic position of the Juncus trifidus group on the basis of previous molecular analyses In the previous studies on the phylogeny and evolution of the Juncaceae, the problematic position of Juncus trifidus was first revealed by the analysis of the rbcL sequence data (Drábková et al. 2003). The rbcL analysis was repeated by Záveská Drábková (2010) using a corrected data set that did not include the wrongly determined GenBank Oxychloë sample (see also Kristiansen et al. 2005, Drábková & Vlček 2007). The analysis of the rbcL data showed that J. trifidus occupies a basal split in the tree for the whole genus Juncus and the Southern Hemisphere Clade (Záveská Drábková 2010; simplified tree Fig. 1A). Later on, further DNA regions were analysed by the present authors to ascertain whether there is a support for the unexpected position of the Juncus trifidus group. When more variable non-coding regions of plastome are used (trnL intron and trnL-F intergenic spacer; Drábková et al. 2006), the Juncus trifidus group, together with J. filiformis of the sect. Juncotypus, form a sister group to the subg. Juncus (Fig. 1B). The close relationship between J. trifidus and J. filiformis seem to be the result of the long-branch attraction (LBA) phenomenon and varied within the topology depending on outgroup selection, as elucidated by Drábková et al. (2006). The maximum likelihood analysis of this data set revealed various positions for Juncus trifidus, J. monanthos, J. capitatus (sect. Caespitosi) and J. filiformis (for details, see Drábková et al. 2006). Furthermore, we examined atp1 and ITS regions in order to determine whether it is possible that the evolution of the plastome, chondriome and nuclear genomes followed divergent pathways. Juncus trifidus and J. monanthos form an early derived branch within the Juncaceae with very high support in the atp1 tree (Záveská Drábková & Vlček 2009; Fig. 1C). In the nrDNA ITS tree, J. trifidus and J. monanthos are in an unresolved position with J. capensis, J. fontanesii and J. capitatus as a sister group to Luzula (Záveská Drábková 2010; Fig. 1D). Results of the above analyses are summarized in Fig. 1, which is a slightly modified version of the original sources. Supraspecific division of Juncus: an introductory overview In principle, the current morphological system of Juncus includes two groups, subg. Juncus (or “eprophyllati” of Buchenau 1906) and subg. Agathryon (or “prophyllati” of Buchenau, l.c.), each with a series of sections (subgenera in the works of Buchenau). Within the latter subgenus, four sections are recognized (Table 1). The sect. Tenageia is comprised of tiny annuals sharing many features of general habit, the sect. Juncotypus is easily characterized by leafless stems and the lowest bract forming an apparent prolongation of the stem (the inflorescence therefore is pseudolateral), the monotypic sect. Forskalina has a thick creeping rhizome and leafy stem with terete leaf blades; all three sections are relatively homogeneous. On the other hand, the section Steirochloa in the traditional system consists of taxa of a very varied appearance, having stem leaves or only 486 Preslia 85: 483–503, 2013 A B C D Fig. 1. – Summary of previous studies showing relationships among major clades within the Juncaceae with the position of Juncus trifidus and J. monanthos marked by a shaded box. A simplified strict consensus maximum parsimony (MP) tree based on (A) rbcL sequence data. Length 659, CI = 0.55, RI = 0.84, total number of characters TNC = 1,428, total number of parsimony informative characters PIC = 209, modified from Záveská Drábková (2010), (B) trnL-F sequence data. Length 2,664, CI = 59, RI = 79, TNC = 2,377, PIC = 612, modified from Drábková et al. (2006), (C) atp1 sequence data. Length 374, CI = 86, RI = 90, TNC = 1,264, PIC = 169, modified from Záveská Drábková and Vlček (2009), (D) ITS1-5.8S-ITS2 sequence data. Length 1,941, CI = 0.55, RI = 0.90, TNC = 743, PIC = 404, modified from Záveská Drábková (2010). Numbers above branches indicate jackknife support. basal ones, flowers solitary on pedicells or sessile or in distinct clusters, leaves flat or almost round. The most aberrant group usually included in the sect. Steirochloa is comprised of Juncus trifidus L. and J. monanthos Jacq. It is characterized by a number of attributes otherwise not found in this section or even absent from, or rare in, the genus Juncus. These differences (Table 2) are conspicuous; two authors who accorded this group the rank of section, sect. Trifidi Rouy (Rouy 1912, Novikov 1990) used some of them, i. e. general habit characters and the lacerate auricles, to support the exclusion of the J. trifidus group from the sect. Steirochloa. 487 Záveská Drábková & Kirschner: A new genus in the Juncaceae Table 1. – The established traditional system of the Juncaceae (Kirschner et al. 2002a, b, c). Genus Subgenus Section Juncus L. [type: J. acutus L.] Juncus Juncus Graminifolii Engelm. Caespitosi Cout. Stygiopsis Gand. ex Kuntze Iridifolii Snogerup et Kirschner Ozophyllum Dumort. Tenageia Dumort. Steirochloa Griseb. Juncotypus Dumort. Forskalina Kuntze Agathryon Raf. Luzula DC. [type: L. campestris (L.) DC.] Marlenia Ebinger Luzula Anthelaea Griseb. Atlanticae Kirschner Nodulosae Chrtek et Křísa Diprophyllatae Satake Alpinae Chrtek et Křísa Thyrsanochlamydeae Satake Luzula Pterodes (Griseb.) Buchenau Marsippospermum Desv. Oxychloë Philippi Patosia Buchenau Rostkovia Desv. Distichia Nees et Meyen Table 2. – Diagnostic morphological features distinguishing Juncus trifidus, J. monanthos and the sect. Steirochloa (data modified from Kirschner et al. 2002c). Feature Juncus trifidus L. J. monanthos Jacq. sect. Steirochloa Auricles Leaf margin Leaf margin sclerenchyma Enlarged cells in adaxial leaf epidermis Inflorescence lacerate-fimbriate serrulate not or weakly developed weakly developed lacerate-fimbriate serrulate not or weakly developed weakly developed entire smooth, entire well developed usually well developed (1–)2–3(–4) flowered 1–(2–3) flowered Tepals Cataphyll blade Seeds 2.0–4.2 mm up to 1 cm seed coat loose, appendages 0.2–0.3 mm long 4.0–5.2 mm 6–10 cm seed coat loose, appendages 0.4–0.5 mm long more than 3-flowered, usually more than 10-fl. 2–6 mm 1.5–6.0 cm seed coat tight, appendages ± absent1 1 In J. vaseyi appendages are present, 0.2–0.5 mm long but have a different nature from those of J. trifidus and J. monanthos: they are smooth, thin, tail-like in the former while in the latter two taxa, they are thick and wrinkled. 488 Preslia 85: 483–503, 2013 Material and methods Plant samples A list of taxa and specimens used in the phylogenetic analyses, including the GenBank sequence accessions, are given in Drábková et al. (2003, 2004, 2006) and Záveská Drábková & Vlček (2009, 2010), see also Appendix 1. The plant material was collected with permission in AAU, BM, C, K, NY, PRA, RSA-POM, and the specimens newly collected by the authors are deposited in PRA (all the abbreviations follow the Index Herbariorum, see http://sciweb.nybg.org/science2/IndexHerbariorum.asp). Data for the morphological characters were compiled from the literature (Kirschner et al. 2002a, b, c), supplemented and specified by herbarium studies, preferably in PRA, but also in AAU, B, BM, BP, C, E, GOET, K, KRA, KRAM, L, LI, NY, RSA-POM, etc. Names of taxa and authors follow the monographic treatment of the family by Kirschner et al. (2002a, b, c). Electron microscopy SEM examinations of seeds were made on at least five seeds per taxon under a scanning electron microscope FEI Quanta 200 in the high vacuum mode (HV), low vacuum mode (LV) and Environmental Scanning Electron Microscope mode (ESEM). Upon comparison, the HV mode proved to be the best. Samples were used without any treatment as it was not necessary to remove the outer membranous layer of the seeds (Ertter 1983). Molecular data analysis DNA sequences were assembled in GeneSkipper (EMBL Heidelberg). Alignment of sequences was done manually using the sequence alignment editor BioEdit (Hall 1999) and insertions or deletions were detected. The nucleotide sequences used in this paper are deposited in GenBank (Appendix 1), and sequence alignments stored in TreeBase. Phylogenetic analyses were performed on the combined data sets and separately on subsets of trnL intron plus trnL-F intergenic spacer, rbcL, atpA and ITS. In total, the matrix comprised ca 5.7 kb per taxon. Phylogenetic trees were constructed using Prionium serratum (Prioniaceae), Thurnia polycephala and T. sphaerocephala (Thurniaceae) as outgroups based on previous analyses (e.g. Plunkett et al. 1995, Drábková et al. 2003, 2006, Záveská Drábková & Vlček 2009, Záveská Drábková 2010). In order to reduce the potential effect of distant outgroups (from the Cyperaceae) on inference of relationships within the Juncaceae, these outgroups were excluded from the first analyses, although the inferred topologies from these different analyses are very similar. The ingroup taxa contained representatives of all currently recognized subgenera and sections of Juncus and Luzula. The gaps in the data matrix were treated as missing data. We included all codon positions based on presupposition that the third position in plastid data contains most of the phylogenetic information (Rydin et al. 2002). Two methods, maximum parsimony (MP) and maximum likelihood (ML) were used to avoid errors where high support values are found for groups not supported by the original data (Goloboff et al. 2003) and also to test the effect of long and extremely short branches on the tree topology. Initially, phylogenetic analyses were performed using PAUP* version 4.0b10 (Swofford 2002). Due to the size of the matrix only heuristic search could be used with Záveská Drábková & Kirschner: A new genus in the Juncaceae 489 1000 random input orders saving 100 trees per replicate, tree bisection and reconnection branch swapping (TBR), holding five trees per step and using steepest descent. A strict consensus tree was constructed. In the second analysis, WinClada version 1.00.08 (Nixon 2002) running NONA version 2.0 (Goloboff 1999) as a daughter process, was used for the parsimony ratchet procedure to search treespace by re-weighting alternating iterations of a search (Nixon 1999). We performed the ratchet procedure (Nixon 1999) running 1000 replicates holding 25 trees at each replicate and sampling 45 characters. A strict consensus tree was constructed. The resultant cladograms were then submitted to commands “hard collapse unsupported nodes in all trees” and “keep best trees only”. Nodal support was determined by jackknifing where proportions were calculated in NONA using 1000 replicates, TBR branch swapping, holding five trees per replication. Maximum likelihood analysis of the complete Juncaceae matrix was performed in GARLI version 0.96b8 (Zwickl 2006) to identify and avoid influence of the long-branch attraction to the tree. The best fitting model of DNA GTR+Γ+I evolution was chosen using AIC criterion as implemented in Modeltest 3.7 (Posada & Crandall 1998). Multiple runs were performed to ensure that results are consistent as the algorithm is stochastic (Zwickl 2006). The log likelihood values of each run were retained in order to compare the individual runs. Branch support was determined by 1000 ML bootstrap iterations in GARLI. Results Separate status of Juncus trifidus and J. monanthos: evidence from morphology With the exception of the lacerate-fimbriate auricles, the peculiar characters of these two species remained unnoticed. Buchenau (1890, 1906) mentioned only the unusual character of auricles, and from his drawing (Buchenau 1906: 110, fig. 60) it is obvious that he also recorded the absence of marginal sclerenchyma strands in transverse sections of its leaves, a character mentioned by S. Snogerup (Nilsson & Snogerup 1972). The latter author also summarized the other important characters of J. trifidus: mucronate connective, serrulate leaf margins, etc., which had not attracted the attention of Buchenau (opera varia). Not even the authors who accepted a separate section for the two species (Rouy 1912, Novikov 1990) mention any of the important distinguishing features summarized in Table 2 (except for lacerate auricles). The lacerate-fimbriate auricles of the group of J. trifidus and J. monanthos are unique in this family but the other remarkable characters, when considered separately, are not so diagnostic or conspicuous. The marginal sclerenchyma strands ± missing from the leaves of this group and consistently present in the leaves of the rest of the sect. Steirochloa, for instance, are not developed in the closely related sect. Tenageia Dumort. The mucronate connective can be found also in two species of the subg. Juncus, J. (Ozophyllum) scheuchzerioides Gaudich. and J. (Stygiopsis) castaneus Sm. Also the loose seed coat is very occasionally found in other species of Juncus (Figs 2 and 3). Only when all these characters are simultaneously compared with those of Juncus and other genera of Juncaceae, is it obvious that, morphologically speaking, the J. trifidus group is quite distinct (Table 3). 490 Preslia 85: 483–503, 2013 Fig. 2. – Seeds (left) and seed surface (right) of Juncus trifidus (A, B), J. monanthos (C, D). SEM pictures, scale bars on the photographs. Separate status of the Juncus trifidus and J. monathos group: evidence from karyology Karyology traditionally is a source of important data for inferring evolutionary trends in the Juncaceae. As a consequence of chromosome fragmentation (agmatoploidy), it should be emphasized that chromosome number cannot be used for determining ploidy level, at least in Luzula. Most of the Juncaceae proved to be a rather difficult material for karyological studies and many chromosome counts are thought to be inaccurate (although often close to the correct number). There was a recent attempt to summarize chromosome evolution in the Juncaceae (Roalson 2005) in which conclusions were drawn about its phylogenetic importance. We shall try to rectify some of the most serious mistakes in the latter work. The chromosome counts cited below are referred to in the revision of this family by Kirschner et al. (2002a, b, c) and in Záveská Drábková (2013). Fig. 3. – Seeds (left) and seed surface (right) of Juncus gerardii (section Steirochloa, A, B) and J. squarrosus (sect. Steirochloa (C, D), J. filiformis (sect. Juncotypus, E, F) and Luzula confusa (Luzula sect. Thyrsanochlamydeae, G, H). SEM pictures, scale bars on the photographs. ¤ Záveská Drábková & Kirschner: A new genus in the Juncaceae 491 492 Table 3. – A morphological comparison of the Juncus trifidus group with the other genera of the Juncaceae. Juncus trifidus group the rest of Juncus Luzula Distichia Patosia Oxychloë Rostkovia Marsippospermum Auricles lacerate-fimbriate absent or entire absent entire entire entire entire entire Anther connectives mucronate without mucro1 without mucro mucronate mucronate mucronate mucronate mucronate2 Leaf margin serrulate smooth serrulate smooth serrulate smooth smooth smooth 6 Seed coat loose tight or loose tight loose loose loose tight tight Leaf indumentum absent absent present absent absent absent absent absent seeds in capsule many many three many many many many many Leaf sheath open open or closed3 closed open open open open open Cushion growth absent absent absent4 present present present absent absent 5 Leaves arrangement spirally spirally spirally distichous spirally spirally spirally spirally Gynophore absent absent absent developed absent absent absent absent 1 In J. castaneus Sm. and J. scheuchzerioides Gaudich., connectives may bear a short mucro; its presence is variable within species 2 In M. grandiflorum (L. fil.) Hook., the connective is shorter than the pollen sacs and the tips of pollen sacs are horn-like. 3 In J. engleri Buchenau, J. lomatophyllus Spreng., J. dregeanus Kunth and J. capensis Thunb., all belonging to the so called Southern Hemisphere Clade, leaf sheaths are initially closed and split later. 4 A group of New Zealand species, L. pumila Hook. fil., L. crenulata Buchenau and L. colensoi Hook. fil. form small dense cushions. 5 6 See also Figs 2 and 3. Preslia 85: 483–503, 2013 An aberrant form of the group of J. capensis, originally described as J. singularis Steud., is also characterized by distichously arranged leaves. This form requires further study (it was found only once in the early 19th century). Záveská Drábková & Kirschner: A new genus in the Juncaceae 493 The genus Luzula has been studied extensively by several authors, most so by Nordenskiöld (opera varia, for summary see Kirschner et al. 2002a), and its karyotype evolution may be summarized as follows: In Luzula, for which there are counts for about 70% of the species, invariably x = 6 and four basic karyotype evolution phenomena are known, i.e. (i) true polyploidy (up to octoploid level), (ii) agmatoploidy (in Luzula, it is simultaneous chromosome fragmentation reaching and in the subg. Pterodes exceeding the number 2n = 48 of twice fragmented chromosomes), (iii) chromosome fusion (resulting in 2n = 6 in L. elegans, subg. Marlenia) and (iv) mixed alloploidy with euploid and agmatoploid genome donors (or resulting from the combination of fragmented and unfragmented chromosomes, e.g. in some hybrids, or variously fragmented chromosomes, mainly subg. Pterodes). These phenomena are not randomly distributed in the genus and the following list shows the differences among subgenera and sections: subg. Marlenia (2n = 6, chromosome fusion) subg. Luzula sect. Anthelaea (only eudiploids, 2n = 12) sect. Diprophyllatae (eudiploids and agmatoploids, 2n = 12 AL 1, 24 BL) sect. Alpinae (eudiploids and agmatoploids, some agmatoploids combining AL and BL chromosomes, e.g. L. spicata subsp. italica in the Caucasus with 2n = 16 = 8AL+8BL) sect. Thyrsanochlamydeae (polyploid group of hybrid origin) sect. Luzula (diploids, polyploids, agmatoploids, mixed polyploids, rarely also with partially fragmented sets of chromosomes) subg. Pterodes (only agmatoploids, sometimes mixed agmatoploids, with 2n = 24BL to 2n = 66 CL or CL+DL) As regards the genera confined to the southern hemisphere, we conclude that virtually nothing is known about their chromosome evolution. There is a single reliable chromosome count for these five genera of 2n = 56 for Rostkovia magellanica, which is either a highly derived polyploid or agmatoploid number (disregarded by Roalson 2005). The repeatedly cited chromosome number (n = 8) for Oxychloë andina (Sasaki 1937, see also Kirschner et al. 2002a) was not critically examined, also because of the relative unavailability of Sasaki’s paper. It was not until we studied the electronic version of the paper that we realized that the number was probably false. Sasaki (1937) published not only the count of n = 8 for O. andina but also the chromosome numbers of Juncus bufonius, J. compressus, J. filiformis, J. lamprocarpus and J. squarrosus, all invariably n = 8–10, J. maritimus (n = 20) and L. campestris, L. multiflora and L. nivea, all n = 9. In view of the obviously wrong numbers given for nine species, the Oxychloë andina count is at least very doubtful. The speculation of Roalson (2005) that “the low chromosome number (2n = 16) of Oxychloë andina appears likely to be an agmatoploid reduction from a higher chromosome number, as all of the related Juncus species that have been counted range from 2n = 36 to 48”, is therefore superfluous. 1 The terminology of agmatoploid chromosomes in Luzula recognizes AL chromosomes (euploid, unfragmented), BL chromosomes (simultaneously fragmented once) and CL chromosomes (fragmented twice). 494 Preslia 85: 483–503, 2013 There are relatively reliable chromosome counts for 93 species of Juncus, less than one third of its species (the counts are cited in Kirschner et al. 2002a, b, c). However, the counts cover all sections of the genus and the coverage is reasonably representative. The following conclusions can be made when the chromosome numbers are generalized: (i) All the species counted, with one notable exception, are highly derived polyploids. (ii) Agmatoploidy has not been documented in Juncus; the most probable polyploid series group to be considered as agmatoploids, the J. bufonius agg., was recently studied in detail and found to be eupolyploid (Rooks et al. 2011). (iii) The basic chromosome numbers that can be derived from the known counts vary between x = 9 and x = 12 in most cases, although the majority of existing gametic numbers vary between n = 19 and n = 24 (the most frequent being n = 20) and reflect the probable paleopolyploid history of most of the genus. (iv) The only diploid with a gametic number corresponding to the above conclusion is J. capitatus with 2n = 18 (x = 9). The idea of Roalson (2005) that “diploid chromosome numbers based on multiples of 6 or 12 appear to provide a synapomorphy for the Luzula + J. capitatus clade”, is false. The following account includes the gametic numbers recorded for the different sections of the genus (the less common numbers are in brackets): subg. Juncus sect. Juncus sect. Graminifolii sect. Caespitosi Juncus caespitosus sect. Stygiopsis sect. Iridifolii sect. Ozophyllum subg. Agathryon sect. Tenageia sect. Steirochloa Juncus trifidus Juncus monanthos sect. Juncotypus sect. Forskalina n = (23) 24 n = (18, 19) 20 n = 16–19 [excluding J. caespitosus] n=9 very high polyploids with multiples of 20 (often 2n = 60 or ca. 120) n = 20 n = (22) 20 (and polyploids with 2n = 80) n = 13–18 (and high polyploids with 2n around 72 and 108) n = 20–22 (and corresponding polyploids) n = 15 n = 15 n = 20, 21 (and polyploids) n = 21 As regards the chromosome number 2n = 30, recorded for both J. trifidus and J. monanthos, it clearly deviates from the numbers recorded for the rest of the sect. Steirochloa. A comparison with the other sections shows that similar chromosome numbers are found in the sections Tenageia and Caespitosi (for details, see discussion). Evidence from the phylogenetic analysis of combined molecular data One of the key new results of the previous analyses is the separate position of Juncus trifidus and J. monanthos outside the section Steirochloa (Fig. 1). This result is apparent from all the phylogenetic analyses of Juncaceae based on all available molecular data 495 Záveská Drábková & Kirschner: A new genus in the Juncaceae A B Fig. 4. – Molecular phylogenetic tree of the Juncaceae based on data from five regions of all three genomes, rbcL, trnL, trnL-F, atp1 and ITS1-5.8S-ITS2. (A) Strict consensus maximum parsimony (MP) tree from 1292 most parsimonious trees. Length 5279, CI = 0.59, RI = 0.80, TNC = 5674, PIC = 1310. (B) Best tree from maximum likelihood (ML) analysis under the evolution model GTR+Γ+I (-lnL = 31471.6336). Numbers above branches indicate jackknife support values above 50%. Representatives of Juncus sect. Steirochloa are demarcated by a box. 1 Thurnia, 2 Oreojuncus, 3 Juncus subg. Juncus I., 4 Southern Hemisphere Clade + sect. Graminifolii, 5 Luzula, 6 Juncus subg. Agathryon, 7 Juncus subg. Juncus II. sources. However, Juncus trifidus (and J. monanthos) are either a sister clade to the rest of the genus Juncus (rbcL, Fig. 1A) or majority of Juncus subg. Juncus (trnL-F; Fig. 1B), or a separate clade close to the root of the Juncaceae tree (atp1, Fig. 1C) or a sister group to the genus Luzula (ITS1-5.8S-ITS2, Fig. 1D). As each of the DNA regions used place the Juncus trifidus group in a different (though remarkable) position, each potentially leading to a different interpretation, the four individual data partitions were combined (Fig. 4A) using the maximum parsimony criterion in terms of maximizing the explanatory power and informativeness of all the data. 496 Preslia 85: 483–503, 2013 Simultaneous molecular analyses of combined chloroplast, mitochondrial and nuclear regions (rbcL, trnL-F, atp1 and ITS1-5.8S-ITS2, 5684 characters in total) under the maximum parsimony criterion generated 1,292 most parsimonious trees (MPT) of length 5,279, CI = 0.59 and RI = 0.80 (Fig. 4A). Total number of parsimony informative characters (PICs) was 1310. We then used maximum likelihood analysis to determine the occurrence of long branches within the tree (Fig. 4B). The ML analysis using the evolution model GTR+G+I (-lnL = 31471.6336) yielded the same topology for relationships among the major clades that are depicted in Fig. 4B. One of the longest branches showed Juncus trifidus–J. monanthos supported by 100% JS. These analyses show seven main groups within the Juncaceae: Juncus subg. Agathryon I. & II., Juncus subg. Juncus I. & II., the genus Luzula, the so called Southern Hemisphere Clade (SHC) and J. trifidus and J. monanthos. The majority of members of subgenera Agathryon and Juncus form well-defined clades generally corresponding to the accepted taxonomic idea based on Buchenau (1890). They are characterized by a pair of floral bracteoles and a cymose inflorescence or absence of bracteoles and racemose inflorescence, respectively. However, two traditional representatives of the sect. Steirochloa, Juncus trifidus and J. monanthos, occupy the first derived lineage in the Juncaceae tree as a sister clade to Juncus, Luzula and the Southern Hemisphere Clade. Atp1 was revealed as the most diagnostic region within the whole matrix, where Juncus trifidus and J. monanthos appeared exactly in the same position in the basal-most split (Fig. 1C). When the molecular evidence gained from the combined data sets (Fig. 4) is compared with the established traditional system (Table 1), a few conspicuous features can be emphasized: (i) The monophyletic status of the genus Luzula and paraphyletic status of Juncus and the other genera. (ii) Subgenus Agathryon partially retained in the traditional circumscription (but see the case of J. trifidus) with little support for sectional subdivision. (iii) The southern hemisphere genera (Distichia, Marsippospermum, Oxychloë, Patosia and Rostkovia), together with a few S. African members of the sect. Graminifolii (and with some species of other Graminifolii and with all the members of the sect. Juncus studied), form a clade embedded in the Juncus clade, which is supported also by all separate data sets. (iv) The rest of subg. Juncus form a separate clade (but there is an unclear, variable and not very well supported position of J. capitatus of the sect. Caespitosi). (v) Juncus trifidus and J. monanthos form a sister clade to the rest of the family. Evidence from the host-pathogen compatibility in the Juncaceae as an indicator for generic taxonomy There is an interesting link in the field of compatibility between plants and pathogenic fungi. The ascomycetous fungus, Septoria chanousiana Ferraris, which was previously thought to infect only species of Luzula, was recently found also in leaves of Juncus trifidus (Suková & Chlebicki 2004). Another similar example is the ascomycetous filamentous fungus, Arthrinium juncoideum (J. G. Hall) Sacc. that occurs in the leaves of Juncus spp., but not those of Juncus trifidus, which are infected with Arthrinium luzulae M. B. Ellis, a typical pathogen of the genus Luzula (Suková & Chlebicki 2004). These findings support the hypothesis that Záveská Drábková & Kirschner: A new genus in the Juncaceae 497 Juncus trifidus (fungi in J. monanthos have not been studied) occupies a unique position. The above facts are even more significant when we take into account the specificity of fungi found infesting representatives of the Juncaceae. For instance, Ustilago vuijckii Oudem. et Beijer is known to be widely distributed and to infest various species of Luzula in North America, Australia and Eurasia (Hämet-Ahti 1972, 1979, 1982) while in the genus Juncus it is Ustilago abstrusa Malen., with the exception of South Africa, where it is Ustilago capensis Reess (Hämet-Ahti & Piispala 1971). Discussion In the present study we analysed the evidence from multiple sources in order to elucidate the character divergence and phylogenetic position of Juncus trifidus and J. monanthos. These two species form a sister clade to the rest of the Juncaceae, and evidence from morphology, chromosome numbers and host-pathogen compatibility strongly support the separate status of this group. We therefore treat these two species as members of a separate genus described here, Oreojuncus. Molecular studies based on different molecular markers (rbcL, trnL, trnL-F, atp1, ITS and rps16) place J. trifidus and J. monanthos in different separate positions within the Juncaceae tree topology (Drábková et al. 2003, 2004, 2006, Drábková & Vlček 2007, Záveská Drábková & Vlček 2009, 2010, Záveská Drábková 2010) even though a limited number of taxa from crucial groups were used in several of the studies (Roalson 2005, Jones et al. 2007). All the principal morphological characters usually used in the delimitation of genera in the Juncaceae are displayed in Table 3. It is evident, that the traditional Juncaceae system using natural and evolutionary principles and taking character divergence into account is primarily based on autapomorphies or combinations of rare character states. As regards Juncus trifidus and J. monanthos, our attention to this species pair was attracted by the results of molecular analyses showing their position outside the traditional system (sect. Steirochloa or even outside Juncus). When the same morphological criteria as those applied to the genera of Juncaceae are used for the J. trifidus group, the most appropriate solution is its classification as a separate genus. The Juncus trifidus group is easily distinguished from the rest of Juncus on the basis of distinct characters – the lacerate-fimbriate auricles, the serrulate leaf margin and the mucronate connectives, and the differences between the J. trifidus group and the other Juncaceae genera are of equal magnitude (see Table 3 and the identification key to genera below). However, the position of the J. trifidus group is not the only fundamentally new phenomenon revealed in the Juncaceae. Within Juncus, a distinct and well supported clade exists including all the genera confined to the southern hemisphere (Marsippospermum, Rostkovia, Distichia, Patosia and Oxychloë, the latter having been safely returned to the Juncaceae after some adventures caused by sample confusion, see the chapter dealing with previous molecular studies of this family) and probably some or perhaps all the species of several sections of Juncus (sect. Juncus, Graminifolii and Caespitosi). The paraphyly of Juncus (as a result of the existence of the Southern Hemisphere Clade, SHC) is not dealt with in detail in the present paper. This is because of incomplete sampling of the SHC, particularly the section Graminifolii (in the two main regions of its occurrence, 498 Preslia 85: 483–503, 2013 southern and south-eastern Africa and western North America, six of 22 species sampled). The African and western North American members of the sect. Caespitosi might also belong to SHC but they have been sampled inadequately for drawing such conclusion (three out of 17 species, and only two from the U.S., none of the at least six African taxa). The same is to be concluded about the section Juncus, with three out of nine species covered by the present study. We therefore refrain from making conclusions about the generic delimitation in Juncus (excluding Oreojuncus, see below), and a new, more detailed study of the SHC is needed. On the other hand, we have to point out that we would anyway refrain from sinking the SHC genera into the broad Juncus merely on the basis of departures from holophyly (cf. Hörandl & Stuessy 2010) when it is in conflict with the character evolution in the family. The other problematic taxa are J. capitatus, perhaps in a sister position to the rest of Juncus and the SHC, and J. potaninii, a slender, morphologically reduced putative member of the sect. Stygiopsis, otherwise a distinct, monophyletic group. J. potaninii appears in a position sister to the clade formed by the majority of subg. Juncus (sections Ozophyllum, Iridifolii and Stygiopsis). This problem cannot be resolved without a new study. It was Roalson (2005) who first discussed the unexpected phylogenetic position of J. capitatus and suggested a tentative solution, i.e. that it is a consequence of the LongBranch Attraction (LBA) phenomenon, an artefact of the MP analyses. The summary of the karyological data raises a few questions. First, the highly derived chromosome numbers in Juncus indicate a high probability of reticulation and horizontal gene transfer in Juncus, a factor potentially complicating the phylogenetic inferences using cladistic methods. Another fact to be discussed is the chromosome number in the Juncus trifidus group, n = 15. There are another two groups where gametic numbers of n = 13 to n = 18 are known, the sect. Tenageia and the Californian members of the sect. Caespitosi (nothing is known about chromosome numbers of the South African members). It is hypothesized that these numbers can be derived from n = 20, a common number in closely related perennial taxa, and that in both groups, these gametic numbers were reached by gradual reduction of the paleopolyploid genome as is the case in annual, predominantly selfing taxa in other groups (Albach & Greilhuber 2004, Wright et al. 2008). The chromosome number of the J. trifidus group might have undergone a similar evolution but there is a hypothetical possibility that the number is derived from a separate basic number of x = 8. A cytogenetic study is needed to solve this problem. Taxonomic treatment Oreojuncus Záveská Drábková et Kirschner, genus novum Plantae perennes estolonosae, laminis foliorum canaliculatis marginibus minute serrulatis, auriculis laceratofimbriatis, inflorescentiis paucifloris, floribus diprophyllatis, hexandris, connectivis mucronatis, capsulis trilocularibus rostratis, seminibus numerosis bicaudatis testa laxa. Plants perennial, without stolons, leaf blade canaliculate with minutely serrulate margin, auricules lacerate-fimbriate, inflorescence few-flowered, flowers with two bracteoles, six anthers, anthers with mucronate connective, capsule trilocular, rostrate, seeds numerous, with loose coat and two appendages. Záveská Drábková & Kirschner: A new genus in the Juncaceae 499 Type of the generic name, designated here: Oreojuncus monanthos (Jacq.) Záveská Drábková et Kirschner [≡ Juncus monanthos Jacq.] Members of the genus: Oreojuncus trifidus (L.) Záveská Drábková et Kirschner, comb. nova Basionym: Juncus trifidus L., Sp. Pl. 326 (1753) Oreojuncus monanthos (Jacq.) Záveská Drábková et Kirschner, comb. nova Basionym: J. monanthos Jacq., Enum. Stirp. Vindob. 61, 236 (1762) Key to the genera of Juncaceae 1a 1b 2a 2b 3a 3b 4a 4b 5a 5b 6a 6b 7a 7b Auricles lacerate ........................................................................................................................... Oreojuncus Auricles absent or entire ................................................................................................................................. 2 Leaf margin hairy, at least sparsely so near sheath opening; seeds 3 ..................................................... Luzula Leaf margin not developed (leaves round, glabrous), or glabrous; seeds many ............................................... 3 Leaf margin minutely serrulate ........................................................................................................... Patosia Leaf margin not developed (leaves round) or smooth ..................................................................................... 4 Flowers solitary and anthers mucronate (if anthers not mucronate then outer tepals at least 15 mm long) ...... 5 Flowers in inflorescences; anther connective not mucronate (if flowers occasionally solitary then tepals shorter than 10 mm) .............................................................................................................................. Juncus Plants cushion-forming, upper part of stem densely covered with leaves, flower lateral (subterminal, axillary) ....................................................................................................................................................................... 6 Plants not cushion-forming, upper part of stem leafless, flower terminal ....................................................... 7 Leaves regularly distichous; gynophore developed, elongating during capsule ripening .................. Distichia Leaves ± spirally arranged; gynophore absent................................................................................... Oxychloë Flower bracts 2, the lower one herbaceous, conspicuously longer than perianth, upper bract ± equalling perianth, capsule suborbicular to obovoid, obtuse, to c. 5 mm long, seeds without conspicuous appendages .......................................................................................................................................................... Rostkovia Flower bracts 1 or 2, membranous, much shorter than perianth, capsule oblong to ellipsoidal, trigonous, acuminate, at least 7 mm long, seeds with two distinct appendages .................................. Marsippospermum Acknowledgements This study was done in the laboratories of Center for Integrated Genomics – Institute of Molecular Genetics ASCR, Institute of Botany ASCR and was supported in part by SYNTHESIS DK-TAF 1295, SYNTHESIS GBTAF 2052 and GAČR 206/07/P147, and by a research centre grant no. LC06073 (Ministry of Education). It was also supported by institutional research plan AV0Z60050516 and a long-term research development project no. RVO 67985939. L. Z. D. is grateful to Christian Köbele (University of Munich) for useful discussions on the morphology and anatomy of seeds, Oldřich Benada (Institute of Microbiology ASCR) for help with the electron microscopy of seeds and Jiří Machač (Optical laboratory of the Institute of Botany ASCR) for help with electron microscopy of the seed surface, and to Jiří Zázvorka for fruitful discussions. Souhrn Dvojice druhů rodu Juncus, J. trifidus L. a J. monanthos Jacq., byla tradičně řazena do sekce Steirochloa, ačkoliv někteří autoři je na základě pozoruhodných morfologických znaků řadili do samostatné sekce Trifidi Rouy. Předchozí molekulárně-fylogenetické studie čeledi Juncaceae ukázaly, že tyto dva druhy do sekce Steirochloa nepatří. Předložená práce shrnuje dostupné údaje morfologické, karyologické, a zejména kombinované výsledky molekulárních analýz s využitím sekvencí kódujících i nekódujících úseků chloroplastové DNA, nukleární ribosomální DNA a mitochondriální DNA. Ukázalo se, že J. trifidus a J. monanthos z fylogenetického hlediska představují raně odštěpenou vývojovou větev čeledi Juncaceae, sesterskou rodu Luzula. Rovněž soubor morfologických znaků, zejména třásnitá ouška, prašníky s nasazenou špičkou, semena s volným vnějším osemením a s přívěsky a papilnatě pilovitý okraj listů, ale také rodová specifičnost houbových patogenů, tyto dva druhy staví mimo sekci Steirochloa, ba dokonce mimo rod Juncus. Proto jsou tyto dva druhy odděleny do samostatného nového rodu Oreojuncus jako O. trifidus a O. monanthos. Je též uveden klíč k určování rodů čeledi Juncaceae. 500 Preslia 85: 483–503, 2013 References Albach D. C. & Greilhuber J. (2004): Genome size variation and evolution in Veronica. – Ann. Bot. 94: 897–911. Balslev H. (1996): Juncaceae. – In: Flora neotropica monographs 68: 1–167, NYBG Press, New York. Balslev H. (1998): Juncaceae. – In: Kubitzki K. (ed.), The families and genera of vascular plants, Vol. 4. Flowering plants: monocotyledons – Alismatanae and Commelinanae (except Gramineae), p. 252–260, SpringerVerlag, Berlin & Heidelberg. Buchenau F. (1890): Monographia Juncacearum. – Bot. Jahrb. Syst. 12: 1–495. Buchenau F. (1906): Juncaceae. – In: Engler A. (ed.), Das Pflanzenreich (Regni vegetabilis conspectus) 25: IV, 36: 1–284, W. Engelman Verlag, Leipzig. Chase M. W. et al. (1993): Phylogenetics of seed plants: an analysis of nucleotide sequences from the plastid gene rbcL. – Ann. Missouri Bot. Gard. 80: 528-580. Drábková L. & Vlček Č. (2007): The phylogenetic position of Oxychloë (Juncaceae): evidence from one nuclear, three plastid regions and morphology. – Taxon 56: 95–102. Drábková L., Kirschner J., Seberg O., Petersen G. & Vlček Č. (2003): Phylogeny of the Juncaceae based on rbcL sequences, with special emphasis on Luzula DC. and Juncus L. – Pl. Syst. Evol. 240: 133–147. Drábková L., Kirschner J. & Vlček Č. (2006): Phylogenetic relationships within Luzula DC. and Juncus L. (Juncaceae): a comparison of phylogenetic signals of trnL-trnF intergenic spacer, trnL intron and rbcL plastome sequence data. – Cladistics 22: 132–143. Drábková L., Kirschner J., Vlček Č. & Pačes V. (2004): TrnL-trnF intergenic spacer and trnL intron define clades within Luzula and Juncus (Juncaceae). – J. Mol. Evol. 59: 1–10. Duvall M. R., Clegg M. T., Chase M. W., Clark W. D., Kress W. J., Hills H. G., Eguiarte L. E., Smith J. F., Gaut B. S., Zimmer E. A. & Learn G.H. (1993): Phylogenetic hypotheses for the monocotyledons constructed from rbcL sequences. – Ann. Missouri Bot. Gard. 80: 607–619. Ertter B. (1983): A re-evaluation of the Juncus triformis complex (Juncaceae). – PhD thesis, City Univ. New York, New York. Goloboff P. A. (1999): NONA, version 2.0. – URL: http:\\www.cladistics.com/aboutNONA.htm. Goloboff P. A., Farris J., Källersjö M., Oxelmann B., Ramirez M. & Szumik C. (2003): Improvements to resampling measures of group support. – Cladistics 19: 324–332. 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. Hämet-Ahti L. (1972): Notes on Ustilago vuijckii Oudem. and Beijer. on some Luzula species in North America. – Syesis 5: 83–85. Hämet-Ahti L. (1979): Ustilago vuijckii Oud. et Beijer. na Luzula multiflora subsp. sibirica i L. pseudosudetica iz aziatskoi chasti SSSR. [Ustilago vuijckii Oud. et Beijer. on Luzula multiflora subsp. sibirica and L. pseudosudetica from the Asiatic part of the USSR.] – Nov. Sist. Nizsh. Rast. 16: 113–114. Hämet-Ahti L. (1982): Ustilago vuijckii found in Australia. – Karstenia 22: 17–19. Hämet-Ahti L. & Piispala E. (1971): Ustilago abstrusa found on Juncus gerardii in Finland and Sweden. – Aquilo, ser. bot., 11: 1–3. Hörandl E. & Stuessy T. F. (2010): Paraphyletic groups as natural units of biological classification. – Taxon 59: 1641–1653. Jones E., Simpson D. A., Hodkinson T. R., Chase M. W. & Parnell J. A. N. (2007): The Juncaceae-Cyperaceae interface: a combined plastid sequence analysis. – Aliso 23: 55–61. Kirschner J. et al. (2002a): Juncaceae 1: Rostkovia to Luzula. – In: Orchard A. E. (ed.), Species plantarum: flora of the world. Part 6: 1–237. – ABRS, Canberra, Australia. Kirschner J. et al. (2002b): Juncaceae 2: Juncus subg. Juncus. – In: Orchard A. E. (ed.), Species plantarum: flora of the world. Part 7: 1–336. – ABRS, Canberra, Australia. Kirschner J. et al. (2002c): Juncaceae 3: Juncus subg. Agathryon. – In: Orchard A. E. (ed.), Species plantarum: flora of the world. Part 8: 1–192. – ABRS, Canberra, Australia. Kirschner J., Novara L. J., Novikov V. S., Snogerup S. & Kaplan Z. (1999): Supraspecific division of the genus Juncus (Juncaceae). – Folia Geobot. 34: 377–390. Kristiansen K., Cilieborg M., Drábková L., Jørgensen T., Petersen G. & Seberg O. (2005): DNA taxonomy: the riddle of Oxychloë. – Syst. Bot. 30: 284–289. Muasya A. M. , Simpson D. A., Chase M. W. & Culham A. (1998): An assessment of suprageneric phylogeny in Cyperaceae using rbcL DNA sequences. – Pl. Syst. Evol. 211: 257–271. Záveská Drábková & Kirschner: A new genus in the Juncaceae 501 Muasya A. M., Bruhl J. J., Simpson D. A., Culham A. & Chase M. W. (2000): Suprageneric phylogeny of Cyperaceae: a combined analysis. – In: Wilson K. L. & Morrison D. A. (eds.), Monocots: systematics and evolution, p. 593–601, CSIRO, Melbourne. Nilsson Ö. & Snogerup S. (1972): Drawings of Scandinavian plants 65–68. Juncus L. – Bot. Not. 125: 2–8. Nixon K. C. (1999): The parsimony ratchet, a new method for rapid phylogenetic analysis. – Cladistics 15: 407–414. Nixon K. C. (2002): WinClada, version 1.00.08. – URL: http://www.cladistics.com/aboutWinc.htm. Novikov V. S. (1990): Konspect sistemy roda Juncus L. (Juncaceae) [Conspect of the system of the genus Juncus L. (Juncaceae)]. – Byull. Moskov. Obshch. Ispyt. Prir., sect. biol. 95: 111–125. Plunkett G. M., Soltis D. E., Soltis P. S. & Brooks R. E. (1995): Phylogenetic relationships between Juncaceae and Cyperaceae: insights from rbcL sequence data. – Am. J. Bot. 82: 520–525. Posada D. & Crandall K. A. (1998): MODELTEST: testing the model of DNA substitution. – Bioinformatics 14: 817–818. Roalson E. H. (2005): Phylogenetic relationships in the Juncaceae inferred from nuclear ribosomal DNA internal transcribed spacer sequence data. – Int. J. Pl. Sci. 166: 397–413. Rooks F., Jarolímová V., Záveská Drábková L. & Kirschner J. (2011): The elusive Juncus minutulus: a failure to separate tetra- and hexaploid individuals of the Juncus bufonius complex in a morphometric comparison of cytometrically defined groups. – Preslia 83: 565–589. Rouy G. (1912): Juncaceae. – In : Flore de France, Vol. 13, p. 222–223, Les Fils d’Émile Deyrolle, Paris. Rydin C., Källersjö M. & Friis E. M. (2002): Seed plant relationships and systematic position of Gnetales based on nuclear and chloroplast DNA: conflicting data, rooting problems, and the monophyly of conifers. – Int. J. Pl. Sci. 163: 197–214. Sasaki M. (1937): Chromosome numbers of Miscanthus and juncaceous plants (A preliminary note). – Japan. J. Gen. 13: 260. Starr J. R., Gravel G., Bruneau A. & Muasya A. M. (2007): Phylogenetic implications of a unique 5.8S nrDNA insertion in Cyperaceae. – Aliso 23: 84–98. Suková M. & Chlebicki A. (2004): Fungi on Juncus trifidus in the Czech Republic (II) with taxonomical notes to some species. – Czech Mycol. 56: 203–222. Swofford D. L. (2002): PAUP*. Phylogenetic analysis using parsimony (* and other methods). Version 4.0b10. – Sinauer Associates, Sunderland, Massachusets. Wright S. I., Ness R. W., Foxe J. P. & Barrett S. C. H. (2008): Genomic consequences of outcrossing and selfing in plants. – Int. J. Pl. Sci. 169: 105–118. Záveská Drábková L. (2010): Phylogenetic relationships within Juncaceae: evidence from all three genomic compartments with notes to the morphology. – In: Seberg O., Petersen G., Barford A. S. & Davis J. I. (eds), Diversity, phylogeny, and evolution in the monocotyledons, p. 389–416, Aarhus University Press, Aarhus. Záveská Drábková L. & Vlček Č. (2009): DNA variation within Juncaceae: comparison of impact of organelle regions on phylogeny. – Pl. Syst. Evol. 278: 169–186. Záveská Drábková L. (2013): A survey of karyological phenomena in the Juncaceae with emphasis on chromosome number variation and evolution. – Bot. Rev. 79: 401–446. Záveská Drábková L. & Vlček Č. (2010): Molecular phylogeny of the genus Luzula DC. (Juncaceae, Monocotyledones) based on plastome and nuclear ribosomal regions: a case of incongruence, incomplete lineage sorting and hybridisation. – Mol. Phylogen. Evol. 57: 536–551. Zwickl D. J. (2006): Genetic algorithm approaches for the phylogenetic analysis of large biological sequence datasets under the maximum likelihood criterion. – PhD thesis, University of Texas at Austin. Received 5 December 2012 Revision received 28 May 2013 Accepted 10 June 2013 502 Preslia 85: 483–503, 2013 Appendix 1. – Species included in the study with GenBank accession numbers. Family, genus, subgenus, section, species: accession numbers of atp1, trnL-F, rbcL, ITS1-5.8S-ITS2. JUNCACEAE: Luzula subg. Marlenia, L. elegans Lowe: EU523158 D, AY437928 B, AY216648 A, FJ213845 E, AB261692. Luzula subg. Luzula, section Anthelaea, L. canariensis Poir.: EU523159 D, AY437929 B, AY216655 A, AY973498 C. L. luzuloides (Lam.) Dandy & E. Eillm.: EU523160 D, −, −, FJ213848 E. L. nivea (Nath.) DC.: EU523161 D, AY437930 B, AY216650 A, −. L. purpureosplendens Seub.: EU523162 D, AY437931 B, AY216654 A, FJ213850 E. L. sylvatica (Huds.) Gaudin: EU523163 D, AY437932 B, AY216637 A, FJ213851 E. Luzula subg. Luzula, section Atlanticeae, L. atlantica Braun-Blanq.: −, DQ099455 B, AY216639 A, FJ213852 E . Luzula subg. Luzula, section Nodulosae, L. nodulosa (Bory & Chaub.) E. Mey.: EU523164 D, AY437933 B, AY216636 A, AY973499 C. Luzula subg. Luzula, section Diprophyllatae, L. divaricata S. Watson: −, AY437934 B , −, FJ213854 E, AY727771. L. glabrata (Hoppe) Desv.: EU523165 D, AY437935 B, AY216644 A, FJ213856 E. L. parviflora (Ehrh.) Desv.: −, −, U49228, −. L. wahlenbergii Rupr.: EU523166 D, AY437936 B, AY216649 A, AY973500 C. Luzula subg. Luzula, section Alpinae, L. peruviana Desv.: −, AY437937 B, −, FJ213863 E. L. racemosa Desv.: −, AY437938 B, −, FJ213864 E. L. spicata (L.) DC. s. str.: EU523167 D, AY437939 B, AY216645 A, FJ213865 E, AY727775. L. mendocina Barros: −, DQ099456 B, −, FJ213866 E. L. traversii (Buchenau) Cheesem. s str.: −, AY437940 B, AY216635 A, FJ213867 E. Luzula subg. Luzula, section Thyrsanochlamydeae, L. arcuata (Wahlenb.) Sw.: −, AY437941 B, AY216651 A, FJ213869 E. L. kjellmaniana Miyabe & Kudo: −, AY437942 B, AY216633 A, FJ213871 E. L. subcongesta (S. Watson) Jeps.: −, AY216657 A, −, FJ213873 E. Luzula subg. Luzula, section Luzula, L. campestris (L.) DC. s. str.: EU523168 D, AY437943 B, AY216652 A, FJ213882 E. L. comosa E. Mey.: −, DQ099457 B, FJ213887 E, FJ213888 E, AY727776. L. divulgata Kirschner: EU523169 D, AY437944 B, AY216646 A, FJ213893 E. L. meridionalis H. Nordensk.: −, −, AY216641 A , −. L. rufa Edgar: EU523170 D, AY437945 B, AY216642 A, FJ213903 E. L. sudetica (Willd.) Schult.: EU523171 D, AY437946 B¸ AY216647 A, AY973501 C. Luzula subg. Pterodes, L. acuminata Raf.: AY1245212, AY4379471, B, AY2166561, A, FJ213905 E. L. pilosa (L.) Willd.: −, AY437948 B, AY216653 A, −. Juncus subg. Juncus, section Juncus, J. cooperi Engelm.: EU523172 D, DQ099458 B, −, AY727781. J. kraussii Hochst. in C. Krauss s. str.: −, −, −, AY216609 A, −. J. maritimus Lam.: EU523173 D, AY437949 B, AY216629 A, −. Juncus subg. Juncus, section Caespitosi, J. capitatus Weigel.: EU523174 D, DQ099459 B, −, AY727769. J. tiehmii Ertter: −, DQ099460 B, −, −. J. uncialis Greene: −, DQ099461 B, −, −. Juncus subg. Juncus, section Graminifolii, J. capensis Thunb.: EU523175 D, AY437950 B, AY216616 A, AY973502 C, AY277825. J. covillei Piper s. str. : EU523176 D, AY437951 B, AY216606 A, −. J. lomatophyllus Spreng.: EU523177 D, AY437952 B, AY216617 A, AY973503. J. longistylis Torr.: EU523186 D, −, −, AY727786. J. prominens (Buchenau) Miyabe & Kudo: EU523178 D, DQ099465 B, −, −. J. regelii Buchenau: EU523179 D, DQ099466 B, −, −. J. repens Michx.: −, AY216627 A, −, ITS: AY727785. Juncus subg. Juncus, section Stygiopsis, J. allioides Franch.: EU523180 D, −, −, AY727809. J. amplifolius A. Camus: EU523181 D, −, −, AY727807. J. bengalensis Kunth: EU523182 D, −, −. J. biglumis L.: EU523183 D, AY437953 B, −, −. J. castaneus Sm.: EU523184 D, AY437954 B, AY216623 A, AY727908. J. gracilicaulis A. Camus: EU523185 D, −, −, −. J. himalensis Klotzsch: −, −, AY216628 A, −. J. membranaceus Royle: EU523187 D, −, −, −. J. minimus Buchenau: EU523188 D, −, −, −. J. sikkimensis Hook. f.: EU523189 D, −, −, −. J. potaninii Buchenau: EU523190 D, −, −, −. J. przewalskii Buchenau: EU523191 D, −, −, −. J. stygius L.: EU523192 D, AY437955 B, AY216610 A, AY973504 C. J. thomsonii Buchenau: EU523193 D, −, −, −. J. triglumis L.: EU523194 D, AY437956 B, AY216605 A, −. Juncus subg. Juncus, section Iridifolii, J. ensifolius Wikstr.: EU523195 D, AY437957 B, AY216611 A, −. J. oxymeris Engelm.: EU523196 D, AY437958 B, AY216621 A, AY973505 C. J. polycephalus Michx.: −, −, AY216626 A, ITS: AY727813, J. xiphioides E. Mey.: EU523197 D, AY437959 B, AY216624 A, −. Juncus subg. Juncus, section Ozophyllum, J. articulatus L. s. str.: −, AY437961 B, AY216614 A, AY727819. J. bulbosus L.: EU523199 D, AY437962 B, AY216622 A, AY973506 C. J. nevadensis S. Watson: −, AY437963 B, AY216601 A, AY727815. J. subnodulosus Schrank: EU523200 D, AY437964 B, AY216630 A, −. J. oxycarpus E. Mey. ex Kunth: −, −, AY216631 A, −. Juncus subg. Agathryon, section Tenageia, J. bufonius L.: EU523201 D, AY437965 B, AY216615 A, AY727889. J. turkestanicus V. Krecz. & Gontsch.: EU523202 D, AY437966 B, −. Juncus subg. Agathryon, section Steirochloa, J. capillaceus Lam.: −, AY437967 B, AY216604 A, AY727796. J. compressus Jacq.: EU523203 D, AY437968 B, AY216625 A, AY973507 C. J. dichotomus Elliott: −, DQ099467 B, AY216607 A, AY727800. J. gerardii Loisel.: −, AY437969 B, AY216613 A, −. J. imbricatus Laharpe: EU523204 D, −, AY216602 A, −. J. monanthos Jacq.: EU523205 D, DQ099464 B, −, −. J. squarrosus L.: EU523206 D, AY437970 B, AY216619 A, −. J. trifidus L.: EU523207 D, AY437971 B, AY216618 A, AY973508 C, AY727770. Juncus subg. Agathryon, section Juncotypus, J. arcticus Willd.: EU523208 D, AY437972 B, −, −. J. balticus Willd. s. str.: EU523209 D, AY437973 B, AY216620 A, −. J. conglomeratus L.: EU523210 D, AY437974 B, −, −. J. drummondii E. Mey.: −, AY437975 B, −, −. J. effusus L. s. str.: 1EU523211 D, 1AY437976 B, 1AY216612 A, 1AY973509 C, Záveská Drábková & Kirschner: A new genus in the Juncaceae 2–5 503 AY727791-AY727794. J. filiformis L.: EU523212 D, AY437977 B, −, AY727790. J. hallii Engelm.: EU523213 D, DQ099462 B, AY216603 A, −. J. jacquinii L.: EU523214 D−, −, −. J. mexicanus Willd ex Schult. & Schult. f.: EU523215 D, DQ099463 B −, −. J. parryi Engelm.: −, AY437978 B, AY216600 A, −. J. vaginatus R. Br.: −, −, AY216608 A, −. Juncus subg. Agathryon, section Forskalina, J. subulatus Forssk.: EU523216 D, − , −, −. Marsippospermum, M. grandiflorum (L. f.) Hook. f.: EU523217 D, AY973535, U49226, AY973515 C. Distichia, D. muscoides Nees et Meyen: EU523218 D, AY973537, U49227, AY727784. D. acicularis Balslev & Laegaard: −, AY973538, AJ419944, AY973513 C. Patosia, P. clandestina (Phil.) Buchenau: EU523219 D, AY973536, U49225, AY973514 C. Oxychloë, O. andina Phil.: −, AY437980 B, AY660587, AY727783 C. O. bisexualis Kuntze: EU523220 D, AY437981 B, AY660584, AY973510 C. O. castellanosii Barros: −, −, AY660585, AY973512 C. O. haumaniana (Barros) Barros: −, −, AY660586, AY973511 C. PRIONIACEAE: Prionium, P. serratum (L. f.) Drège ex E. Mey.: AY124527, AY344155, U49223, −. THURNIACEAE: Thurnia, T. polycephala Schnee.: AY124532, −, AY123239, −. T. sphaerocephala Hook f.: −, −, AF03688, −. CYPERACEAE: Cyperus, C. involucratus Rottb.: −,−, Y12967.1, AY242052.1. Gahnia, G. deusta Benth: −,−, U49231, −. Eleocharis, E. pauciflora Link: −,−, U49232.1, −. Kobresia, K. simpliciuscula Mack.: −,−, U49232.1, AY241971.1. Rhynchospora, R. fascicularis (Michx. ) Vah.: −,−, U49223, −. Published sequences: Superscript codes for site of first publication: A) Drábková et al. (2003); B) Drábková et al. (2004 and 2006); C) Drábková & Vlček (2007); D) Záveská Drábková & Vlček (2009); and E) Záveská Drábková & Vlček (2010); accesssions without letter codes are from Genbank (without additional voucher information, for most, see Roalson 2005).