Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Molecular Species Identification with Rich Floristic Sampling: DNA Barcoding the Pteridophyte Flora of Japan

  • Atsushi Ebihara ,

    Affiliation Department of Botany, National Museum of Nature and Science, Tsukuba-shi, Ibaraki, Japan

  • Joel H. Nitta,

    Affiliation Department of Integrative Biology, University of California, Berkeley, California, United States of America

  • Motomi Ito

    Affiliation Departments of System Sciences, Graduate School of Arts and Sciences, University of Tokyo, Meguro-ku, Tokyo, Japan



DNA barcoding is expected to be an effective identification tool for organisms with heteromorphic generations such as pteridophytes, which possess a morphologically simple gametophyte generation. Although a reference data set including complete coverage of the target local flora/fauna is necessary for accurate identification, DNA barcode studies including such rich taxonomic sampling on a countrywide scale are lacking.

Methodology/Principal Findings

The Japanese pteridophyte flora (733 taxa including subspecies and varieties) was used to test the utility of two plastid DNA barcode regions (rbcL and trnH-psbA) with the intention of developing an identification system for native gametophytes. DNA sequences were obtained from each of 689 (94.0%) taxa for rbcL and 617 (84.2%) taxa for trnH-psbA. Mean interspecific divergence values across all taxon pairs (K2P genetic distances) did not reveal a significant difference in rate between trnH-psbA and rbcL, but mean K2P distances of each genus showed significant heterogeneity according to systematic position. The minimum fail rate of taxon discrimination in an identification test using BLAST (12.52%) was obtained when rbcL and trnH-psbA were combined, and became lower in datasets excluding infraspecific taxa or apogamous taxa, or including sexual diploids only.


This study demonstrates the overall effectiveness of DNA barcodes for species identification in the Japanese pteridophyte flora. Although this flora is characterized by a high occurrence of apogamous taxa that pose a serious challenge to identification using DNA barcodes, such taxa are limited to a small number of genera, and only minimally detract from the overall success rate. In the case that a query sequence is matched to a known apogamous genus, routine species identification may not be possible. Otherwise, DNA barcoding is a practical tool for identification of most Japanese pteridophytes, and is especially anticipated to be helpful for identification of non-hybridizing gametophytes.


Pteridophytes (ferns and lycophytes) are the only land plants possessing distinct, free-living sporophyte (2n) and gametophyte (1n) generations. As the site of fertilization and recruitment, the gametophyte is highly ecologically significant; however, previous taxonomic and ecological studies on pteridophytes have largely been confined to the sporophyte generation because of the difficulties associated with finding and identifying tiny (<1 cm), morphologically simple gametophytes [1]. The relatively few studies on pteridophyte gametophyte ecology have shown that in some lineages this generation occupies different (often broader) ecological niches than the sporophyte generation [2][5], and in extreme cases may persist without producing sporophytes at all as ‘independent gametophytes’ [6]. Pteridophyte gametophytes are likely to be extremely important in determining the distribution of sporophytes, the success of populations, and ultimately, the evolution of the fern and lycophyte lineages.

DNA barcoding [7] is a tool that can potentially be used to rapidly identify pteridophyte gametophytes to the species level [8][10], thus enabling detailed field studies linking the distribution and ecology of pteridophyte gametophytes and sporophytes. However, a standardized protocol for DNA barcoding in plants is still being developed, and pteridophytes as a group have not yet been the focus of a DNA barcode study. Previous plant DNA barcoding studies have typically taken one of three sampling approaches: broad studies including many species pairs across diverse genera to identify universal markers with high species resolution (e.g. [11][16]), floristic studies testing a few markers for all the plants in a given geographic area (e.g. [11], [17][19]), and densely sampled taxonomic studies testing the applicability of proposed markers on a chosen group (e.g. [20][25]). The current study combines floristic and taxonomic approaches, and tests the applicability of two proposed plastid barcode markers, rbcL and trnH-psbA, on the pteridophytes of Japan. Although the Consortium for the Barcode of Life (CBoL) proposed two DNA barcode regions for plants (rbcL and matK)[26], the current universal primer set for matK fails for most pteridophytes; thus, trnH-psbA, one of the regions designated as a supplementary locus by CBoL [26], was used instead.

In addition to the ecological insights that can be gained from species-level identification of the gametophyte generation, pteridophytes represent an important group to test methods of identification using DNA barcodes because of their high rate of polyploids [27]. Because the chloroplast has been shown to be maternally inherited in all pteridophytes studied so far (e.g. [28][30]), plastid markers alone cannot distinguish allopolyploid taxa that have recovered fertility by chromosome doubling after hybridization from sexual, non-hybrid taxa; however, most proposed plant barcode protocols do not include nuclear markers due to difficulty in sequencing and complex molecular evolution [31]. One of the goals of the current study is to determine the capabilities of a DNA barcode system using only plastid markers to distinguish species in a polyploid-rich taxonomic group such as pteridophytes.

Apogamous reproduction, the formation of a sporophyte from a gametophyte without fertilization, is not a rare reproductive mode in ferns [32]. Apogamous ‘species’ are often morphologically intermediate between two sexual species and/or other apogamous species, probably as a result of overcoming hybrid sterility by acquisition of apogamy [33]. Therefore, apogamous taxa are expected to pose similar problems to DNA barcoding as sexual allopolyploids.

For practical application of DNA barcoding, a complete set of reference data that includes all the taxa that may occur in the study area is necessary. To our knowledge, however, there is no example of a DNA barcode reference data set that covers a whole country for any taxonomic group above the family level. The Japanese pteridophyte flora is an ideal target for a DNA barcode model study with complete flora sampling because 1) the flora is well-known not only at species-level but also for ploidy and reproductive mode, 2) the number of native species (approximately 650, although this number varies depending on breadth of species concept [34], [35]) which represent 32 of all 40 extant pteridophyte families sensu Smith et al. 2006 [36], is relatively rich considering its land area (378,000 km2), and 3) since the country is completely bounded by the sea, it stands not only as a political region but also a biogeographically significant area (notwithstanding the low rate of endemism of Japanese pteridophytes compared to flowering plants; [37]). Most DNA barcoding studies have neglected infraspecific taxa, but our sampling includes subspecific and varietal ranks. Considering that rates of pteridophyte species diversity are frequently underestimated as demonstrated by the discovery of a number of cryptic species [38], present infraspecific taxa may represent independent units (biological species or different polyploids). Sampling for the current study is based on our updated checklist of Japanese pteridophytes reflecting the results of the latest research, which enumerates 733 taxa including subspecies, varieties and a few naturalized species (A. Ebihara, unpublished data). The Japanese pteridophyte flora is also characterized by the presence of a large number of putative interspecific hybrids; Nakaike (2004) listed more than 300 combinations [39]. However, putatively F1 sterile hybrid taxa are not included in the present sampling because of the difficulty in complete collection.

Candidate DNA barcode regions are usually tested with multiple individuals per taxon, even if this does not include all taxa from a target flora or taxonomic group. However, we concentrated the present sampling on high coverage for the flora because a sampling strategy that includes multiple collections of some taxa while leaving many completely unsampled makes little sense for the discussion of marker discrimination ability in a local context. On the other hand, the results obtained from analysis of a single sample per taxon with almost complete taxon coverage may demonstrate a “maximum success rate” of species identification which will never increase but will probably decrease by adding multiple samples per taxon.


Success rates for PCR and sequencing

rbcL sequences were obtained from all 689 DNA samples, while trnH-psbA sequences could only be obtained from 617 of the 652 samples attempted (94.6%). Polymerase chain reaction (PCR) of trnH-psbA was almost always successful (failed in only four samples [0.6%] after multiple trials using various PCR conditions), but problems frequently occurred during sequencing due to polyG repeats. In Selaginellaceae, contamination by algal sequences was observed in a significant number of samples (7/16 samples for trnH-psbA); contaminated samples were removed from the analysis.

Divergence values

Figure 1 shows the distribution of interspecific K2P distances among all combinations of taxa whose sequences were successfully obtained for trnH-psbA, rbcL, and rbcL-a (the upper ca. 700 bp of rbcL). A comparison of average K2P distances within genera (θ' values) between rbcL and trnH-psbA is summarized in Figure 2. For the 66 genera that had θ' values for both rbcL and trnH-psbA, distances for trnH-psbA were significantly larger than those of rbcL (Wilcoxon test p = 1.98×10−5) and rbcL-a (p = 4.241×10−6).

Figure 1. Distribution of interspecific K2P distances across all taxon pairs of Japanese pteridophytes.

Infraspecific taxa are treated as distinct species. (A) rbcL, (B) rbcL-a, (C) trnH-psbA.

Figure 2. Average interspecific K2P distances for each genus containing multiple taxa (θ' values).

Infraspecific taxa are treated as distinct species. θ'values are placed upon the latest family-level classification [36]. Gray: rbcL, black: trnH-psbA.

BLAST test

Summaries of species discrimination fail rates based on the three regions and their combinations are shown in Table 1 (among the all taxa) and Table 2 (within major families). The single region with the lowest fail rate is rbcL (22.08%). The combination of regions showing the lowest fail rate is rbcL + trnH-psbA (16.25%), and fail rates decreased if apogamous taxa or infraspecific taxa were excluded. Ploidies and reproductive modes are known for 78.3% of Japanese taxa (Supporting Information Table S1). When sampling was restricted to 254 taxa reported only as sexual diploids, fail rates decreased substantially: 7.91% in rbcL and 3.56% in rbcL + trnH-psbA.

Table 2. Results of BLAST species identification tests by major family (>20 taxa per family).


The molecular divergence rate of trnH-psbA has been reported to be faster than rbcL in previous plant DNA barcode studies (e.g. [11][13], [16]). This trend was not observed in the distribution of interspecific K2P divergence values across all taxon combinations (Figure 1); however, comparison of mean divergence values within each genus (θ' values) by genus pair did statistically support an overall faster rate of divergence in trnH-psbA relative to rbcL (see Results). These seemingly conflicting results are best explained by a few species-rich genera (e.g. Dryopteris, Thelypteris, Polystichum) with low rates of divergence in trnH-psbA that skew the overall distribution of interspecific values. Furthermore, inspection of θ' values (Figure 2) reveals that the divergence rate of trnH-psbA seems to be correlated with phylogenetic position, apparently evolving more rapidly in early diverging groups and more slowly in recently diverging groups. Lycophytes and “basal” ferns showed relatively frequent sequence variation in trnH-psbA (e.g. θ' = 0.15 in Lycopodium, θ' = 0.48 in Crepidomanes) which is also reflected in Nitta's (2008) successful use of trnH-psbA for species identification in the early diverging fern family Hymenophyllaceae [40]. On the other hand, molecular divergence rates for trnH-psbA are much lower, and nearly equal to those of rbcL in eupolypods [41], the recently derived lineage including the bulk of extant fern diversity [42].

In the BLAST test, the minimum fail rate for discrimination (16.25%) was observed when rbcL and trnH-psbA were used in combination as a two-locus barcode. When rbcL was shortened to the upper 700 bp region (rbcL-a) as suggested in other DNA barcode studies [13], [14], [16], the fail rate increased to 20.27%. Species discrimination fail rates were also dependent on taxonomy: if we ignored infraspecific taxa, the fail rate decreased to 12.57%. Of course, we should bear in mind that these values are derived from a data set comprising only one sample per taxon, and that future addition of reference data will undoubtedly increase the fail rate. That said, it is an appropriate interpretation of the present results that species identification for Japanese pteridophyte flora succeeds in nearly 85% of the trials using a combination of full-length rbcL and trnH-psbA. trnH-psbA apparently works well as the second DNA barcode for pteridophytes instead of matK; nevertheless, trnH-psbA had fatal problems during sequencing caused by frequent polymers. In our case, only 75% of the samples could be successfully sequenced on the first trial. It is probably not worthwhile to invest in laborious sequencing protocols such as cloning [43] to obtain barcodes for a region that can only reduce the fail rate of the BLAST test by 7–15%, and in contrast, full-length rbcL as a single DNA barcode locus seems the more practical choice.

Since there are few examples of DNA barcoding studies with dense taxonomic sampling, it is difficult to compare our fail rates directly with those of previous studies. But our minimum fail rate values seem acceptable when compared with the results of recent DNA barcoding studies (Table 3, [13], [14], [44]. Debates over the relative performance of each candidate DNA barcode region aside, a certain degree of discrimination failure can be attributed to uncertainty of species circumscription. The Japanese pteridophyte flora is known to have a high occurrence of apogamous taxa, approximately 15% [45]. The apogamous habit shows a close relationship to phylogenetic position: 87% of apogamous species of Japan belong to only five genera in four families, namely Cyrtomium (Dryopteridaceae), Dryopteris (Dryopteridaceae), Diplazium (Woodsiaceae), Pteris (Pteridaceae) and Hymenasplenum (Aspleniaceae). Although the apogamous taxa of these genera have scientific names, many such ‘species’ in fact comprise morphologically ill-defined species complexes. The borders between multiple apogamous species are often ambiguous; in other words, the morphological variation within one apogamous species is often larger than closely related sexually reproducing species [46], [47]. These characteristics presumably reflect the origins of apogamous taxa – hybridization, recurrent origins and reticulate relationships [48][51] – and they lead to recognition of non-monophyletic species on the gene-tree. In fact, the observed fail rates of the BLAST test in families with a large proportion of apogamous taxa tended to be high (Table 2), and the BLAST test excluding apogamous taxa resulted in improved species resolution (Table 1). Discrimination failure attributed to non-monophyletic history will not be avoidable even if a codominant nuclear DNA marker is adopted as a DNA barcode. At present, there is no solution to the discrimination difficulty of apogamous taxa, and identification results based on molecular data need to be reviewed by specialists if the best-match taxon belongs to an apogamous-rich genus. Even if apogamous taxa are removed, allopolyploids, common in pteridophytes as noted above, are also expected to increase the fail rate. In the current study, the greatest reduction in fail rates was observed when sampling of the BLAST test was limited to only sexual, diploid taxa (Table 1).

Table 3. Comparison of minimum species discrimination fail rates in selected DNA barcoding studies.

Overall, DNA barcoding is a highly effective tool for identifying species among the Japanese pteridophyte flora, despite its high rate of apogamous taxa. Although the present study only includes a single sample per taxon, additional data will better characterize molecular variation and may reveal cryptic species, which are not uncommon in pteridophytes [52]. Furthermore, DNA barcoding using the present markers is expected to be particularly useful for identifying gametophytes, in which interspecific sterile F1 hybrids are usually not a concern. One exciting possible future result of our strategy of ‘complete regional taxonomic sampling’ may be the discovery of independently growing gametophytes that lack accompanying counterpart sporophytes within the same country.

Materials and Methods


DNA of all but 42 taxa listed in our checklist has been newly extracted for this study (DNA/sequences used in our previous studies were reused for some Hymenophylloid taxa) from their native localities or cultivated stocks in botanical gardens, reaching a final representation of 689/733 taxa ( = 94.0%) (see Supporting Information Table S1). Materials used were primarily sporophytic leaf tissue samples/cultivated stocks collected from localities in Japan, but conspecific material from outside the country was substituted for 17 taxa that are difficult to obtain from Japanese localities. All extracted DNA is deposited in the Center for Molecular Biodiversity Research, National Museum of Nature and Science.

PCR and sequencing

DNA sequence data were obtained for two chloroplast regions, rbcL and trnH-psbA. For the rbcL gene, ca. 1400 bp was amplified using the primers F1 and 1379R [53], and sequencing reactions were performed using combinations of four primers: F1, 645F, 888R and 1379R [54], [55]. The trnH-psbA intergenic spacer was amplified using the primers psbA-F and trnH-R [56]; sequencing reactions were performed with PCR primers, but the following taxon-specific primers were also used for some samples: trnH-Rn1 (GGACGTGAACRAGATCTATC) mostly for Pteridaceae, psbA-Fn1 (CGTCTGGTTATGCAGCACAA) and trnH-Rn2 (CCTTGATCCACTTGGCTACG) both mainly for Dryopteridaceae.

Measures of resolution

Three DNA barcode regions (rbcL-a [upper 700 bp of rbcL], full length rbcL [1205bp], and the trnH-psbA intergenic spacer [lengths ranged from 130 bp to 514 bp after deletion of primer regions]) were analyzed for their suitability as markers for species identification by characterizing interspecific divergence values and conducting a local BLAST search to test for rates of successful species identification. The rbcL-a region was tested for its utility as a DNA barcode marker by using trimmed rbcL sequences; hence, PCR and sequencing success rates of rbcL-a were not observed.

Intragenic, interspecific divergence values (K2P genetic distance) were calculated for each pair of species using PAUP* 4.0b10 [57] and averaged over all species pairs for each barcode region (rbcL, rbcL-a, and trnH-psbA) and combination of regions (rbcL + trnH-psbA and rbcL-a + trnH-psbA). Infraspecific taxa were treated as distinct species. trnH-psbA sequences could not be obtained for some taxa, so no statistical test was attempted to compare total average interspecific divergence values between barcode regions. Average interspecific divergence values were also calculated for each genus with more than one species (e.g. theta prime [θ'] values [58]) for each barcode region and combination of regions. The Wilcoxon Matched-Pairs Signed-Ranks Test was used to test for significant differences in θ' values between all pairs of genera for which both rbcL and trnH-psbA data were available.

A local BLAST search was used as a test of species identification ability. First, a reference library was constructed using the "makeblastDB" command in BLAST+ [59] for each region (rbcL, rbcL-a, and trnH-psbA) and combination of regions (rbcL + trnH-psbA and rbcL-a + trnH-psbA). The barcode sequence of each species was then queried against the reference library with the "blastn" command. Infraspecific taxa were treated as distinct species. Species matching 100% over the entire sequence length with their own reference sequence only were counted as successful identifications; those that also matched 100% over the entire sequence length with the reference sequence of one or more different species were counted as failures. Since a large portion of trnH-psbA sequences was missing data at one or both ends, the alignment was trimmed and 11 taxa with missing trnH-psbA data were removed from the BLAST analysis; the rbcL alignment did not require trimming, but three taxa with missing rbcL data were also removed (Supporting Information Table S2). Additional BLAST searches were conducted on the following datasets: apogamous (80 taxa) excluded, infraspecific (47 taxa) excluded, sexual diploids (253 taxa) only. Mean BLAST results were also calculated by family for each major family (>20 taxa per family).

Supporting Information

Table S1.

List of plant material (voucher) and GenBank accession numbers.


Table S2.

List of samples removed from the BLAST test.



We thank the following persons for their assistance with collection of plant material: S. Fujimoto, T. Fujita, K. Fuse, M. Higuchi, Y. Horii, E. Ichikawa, R. Imaichi, Y. Ishizaka, T. Ishiwatari, T. Iwata, Y. Kakugawa, F. Kasetani, K. Kawahara, Y. Kimura, K. Kokubo, G. Kokubugata, T. Kuramata, S. Matsumoto, T. Minamitani, M. Mishima, H. Miyazaki, K. Mizunashi, A. Murata, T. Nakamura, M. Nakazawa, A. Narita, K. Ohora, Ta. Oka, To. Oka, O. Okegawa, Y. Okuyama, N. Sahashi, S. Serizawa, Y. Sugisaki, M. Suzuki, T. Suzuki, S. Tagane, M. Takamiya, T. Takara, K. Takayama, Y. Tsujita, C. Tsutsumi, A. Uchida, K. Ueno, M. Uzawa, T. Yahara, T. Yamada, K. Yamaoka, H. Yamashita, Y. Yamashita, K. Yasuda and M. Yokota. We also thank Y. Hirayama for her assistance with DNA analysis, M. Takamiya for providing unpublished data of ploidies and reproductive modes, and M. Kato and two anonymous reviewers for providing valuable comments on the manuscript.

Author Contributions

Conceived and designed the experiments: AE MI. Performed the experiments: AE. Analyzed the data: JHN AE. Contributed reagents/materials/analysis tools: AE MI. Wrote the paper: AE JHN.


  1. 1. Momose S (1967) Gametophytes of Japanese Pteridophytes. Tokyo: University of Tokyo Press. 627 p.
  2. 2. Rumsey FJ, Sheffield E (1996) Inter-generational ecological niche separation and the ‘independent gametophyte’ phenomenon. In: Camus JM, Gibby M, Johns RJ, editors. Pteridology in Perspective. London: Royal Botanic Gardens Kew. pp. 563–570.
  3. 3. Sato T, Sakai A (1981) Cold tolerance of gametophytes and sporophytes of some cool temperature ferns native to Hokkaido. Canad J Bot 59: 604–608.
  4. 4. Watkins JE, Cardelús C (2009) Habitat differentiation of ferns in a lowland tropical rainforest. Am Fern J 99: 162–175.
  5. 5. Nitta JH, Epps MJ (2009) Hemi-epiphytism in Vandenboschia collariata (Hymenophyllaceae). Brittonia 61: 392–397.
  6. 6. Farrar DR (1967) Gametophytes of four tropical fern genera reproducing independently of their sporophytes in the southern Appalachians. Science 155: 1266–1267.
  7. 7. Hebert PDN, Cywinska A, Ball SL, de Waard JR (2003) Biological identifications through DNA barcodes. Proc Roy Soc London B Biol Sci 270: 313–321.
  8. 8. Schneider H, Schuettpelz E (2006) Identifying fern gametophytes using DNA sequences. Mol Ecol Notes 6: 989–991.
  9. 9. Ebihara A, Matsumoto S, Ito M (2009) Hybridization involving independent gametophytes in the Vandenboschia radicans complex (Hymenophyllaceae): a new perspective on the distribution of fern hybrids. Mol Ecol 18: 4904–4911.
  10. 10. Li F-W, Tan BC, Buchbender V, Moran RC, Rouhan G, et al. (2009) Identifying a mysterious aquatic fern gametophyte. Plant Syst Evol 281: 77–86.
  11. 11. Kress WJ, Wurdack KJ, Zimmer EA, Weigt LA, Janzen DH (2005) Use of DNA barcodes to identify flowering plants. Proc Nat Acad Sci U S A 102: 8369–8374.
  12. 12. Newmaster SG, Fazekas AJ, Ragupathy S (2006) DNA barcoding in land plants: evaluation of rbcL in a multigene tiered approach. Canad J Bot 84: 335–341.
  13. 13. Kress WJ, Erickson DL (2007) A two-locus global DNA barcode for land plants: the coding rbcL gene complements the non-coding trnH-psbA spacer region. PLoS ONE 2: e508.
  14. 14. Fazekas AJ, Burgess KS, Kesanakurti PR, Graham SW, Newmaster SG, et al. (2008) Multiple multilocus DNA barcodes from the plastid genome discriminate plant species equally well. PLoS ONE 3: e2802.
  15. 15. Ford CS, Ayres KL, Toomey N, Haider N, van Alphen Stahl J, et al. (2009) Selection of candidate coding DNA barcoding regions for use on land plants. Bot J Linn Soc 159: 1–11.
  16. 16. Hollingsworth ML, Clark A, Forrest LL, Richardson J, Pennington RT, et al. (2009) Selecting barcoding loci for plants: evaluation of seven candidate loci with species-level sampling in three divergent groups of land plants. Mol Ecol Resour 9: 439–457.
  17. 17. Lahaye R, van der Bank M, Bogarin D, Warner J, Pupulin F, et al. (2008) DNA barcoding the floras of biodiversity hotspots. Proc Natl Acad Sci U S A 105: 2923–2928.
  18. 18. Gonzalez MA, Baraloto C, Engel J, Mori SA, Ptronelli P, et al. (2009) Identification of Amazonian trees with DNA barcodes. PLoS ONE 4: e7483.
  19. 19. Kress WJ, Erickson DL, Jones FA, Swenson NG, Perez R, et al. (2009) Plant DNA barcodes and a community phylogeny of a tropical forest dynamics plot in Panama. Proc Natl Acad Sci U S A 106: 18621–18626.
  20. 20. Newmaster SG, Fazekas AJ, Steeves RAD, Janovec J (2007) Testing candidate plant barcode regions in the Myristicaceae. Mol Ecol Notes 8: 480–490.
  21. 21. Edwards D, Horn A, Taylor D, Savolainen V, Hawkins JA (2008) DNA barcoding of a large genus, Aspalathus L. (Fabaceae). Taxon 57: 1317–1327.
  22. 22. Newmaster SG, Ragupathy S (2009) Testing plant barcoding in a sister species complex of pantropical Acacia (Mimosoideae, Fabaceae). Mol Ecol Resour 9(Suppl. 1): 172–180.
  23. 23. Ren B, Xiang X, Chen Z (2009) Species identification of Alnus (Betulaceae) using nrDNA and cpDNA genetic markers. Mol Ecol Resour 10: 594–605.
  24. 24. Seberg O, Petersen G (2009) How many loci does it take to DNA barcode a crocus? PLoS ONE 4: e4598.
  25. 25. Starr JR, Naczi RFC, Chouinard BN (2009) Plant DNA barcodes and species resolution in sedges (Carex, Cyperaceae). Mol Ecol Res 9(Suppl. 1): 151–163.
  26. 26. CBOL Plant Working Group (2009) A DNA barcode for land plants. Proc Natl Acad Sci U S A 106: 12794–12797.
  27. 27. Soltis PS, Soltis DE (2000) The role of genetic and genomic attributes in the success of polyploids. Proc Natl Acad Sci U S A 97: 7051–7057.
  28. 28. Stein DB, Barrington DS (1990) Recurring hybrid formation in a population of Polystichum x potteri: evidence from chloroplast DNA comparisons. Ann Missouri Bot Gard 77: 334–339.
  29. 29. Gastony GJ, Yatskievych G (1992) Maternal inheritance of the chloroplast and mitochondrial genomes in cheilanthoid ferns. Am J Bot 79: 716–722.
  30. 30. Vogel JC, Russell SJ, Rumsey FJ, Barrett JA, Gibby M (1998) Evidence for maternal transmission of chloroplast DNA in the genus Asplenium (Aspleniaceae, Pteridophyta). Bot Acta 111: 247–249.
  31. 31. Fazekas AJ, Kesanakurti PR, Burgess KS, Percy DM, Graham SW, et al. (2009) Are plant species inherently harder to discriminate than animal species using DNA barcoding markers? Mol Ecol Resour 9: 130–139.
  32. 32. Sheffield E (2008) Alternation of generations. In: Ranker TM, Haufler CH, editors. Biology and Evolution of Ferns and Lycophytes. New York: Cambridge University Press. pp. 49–74.
  33. 33. Park C-H, Kato M (2003) Apomixis in the interspecific triploid hybrid fern Cornopteris christenseniana (Woodsiaceae). J Pl Res 116: 93–103.
  34. 34. Iwatsuki K, Yamazaki T, Boufford DE, Ohba H (1995) Flora of Japan Vol. I Pteridophyta and Gymnospermae. Tokyo: Kodansha.
  35. 35. Nakaike T (1992) New Flora of Japan, Pteridophyta. Revised and Enlarged. Tokyo: Shibundo.
  36. 36. Smith AR, Pryer KM, Schettpelz E, Korall P, Schneider H, et al. (2006) A classification for extant ferns. Taxon 55: 705–731.
  37. 37. Guo Q, Kato M, Ricklefs RE (2003) Life history, diversity and distribution: a study of Japanese pteridophytes. Ecography 26: 129–138.
  38. 38. Haufler CH (2008) Species and speciation. In: Ranker TM, Haufler CH, editors. Biology and Evolution of Ferns and Lycophytes. New York: Cambridge University Press. pp. 303–331.
  39. 39. Nakaike T (2004) Index to Scientific Names of Japanese Pteridophytes. J Nippon Fernist Club vol. 3,: suppl. 2
  40. 40. Nitta JH (2008) Exploring the utility of three plastid loci for biocoding the filmy ferns (Hymenophyllaceae) of Moorea. Taxon 57: 725–736.
  41. 41. Schuettpelz E, Pryer KM (2007) Fern phylogeny inferred from 400 leptosporangiate species and three plastid genes. Taxon 56: 1037–1050.
  42. 42. Schneider H, Schuettpelz E, Pryer KM, Cranfill R, Magallon S, et al. (2004) Fern diversified in the shadow of angiosperms. Nature 428: 553–557.
  43. 43. Spooner DM (2009) DNA barcoding will frequently fail in complicated groups: An example in wild potatoes. Am J Bot 96: 1177–1189.
  44. 44. Liu Y, Yan H-F, Cao T, Ge X-J (2010) Evaluation of 10 plant barcodes in Bryophyta (mosses). J Syst Evol 48: 36–46.
  45. 45. Takamiya M (1996) Index to Chromosomes of Japanese Pteridophyta (1910-1996). Tokyo: Japan Pteridological Society.
  46. 46. Holman JE, Playford J (2000) Molecular and morphological variation in the Senna artemisoides complex. Austr J Bot 48: 569–579.
  47. 47. Andrew RL, Miller JT, Peakall R, Crisp MD, Bayer RJ (2003) Genetic, cytogenetic and morphological patterns in a mixed mulga population: evidence for apomixis. Austr Syst Bot 16: 69–80.
  48. 48. Grant V (1981) Plant Speciation. 2nd ed. New York: Columbia University Press.
  49. 49. Hörandl E (2004) Comparative analysis of genetic divergence among sexual ancestors of apomictic complexes using isozyme data. Int J Plant Sci 165: 615–622.
  50. 50. Ishikawa H, Ito M, Watano Y, Kurita S (2003) Extensive electrophoretic variation in the apogamous fern species, Dryopteris nipponensis (Dryopteridaceae). Acta Phytotax Geobot 54: 59–68.
  51. 51. Robertson A Newton AC, Ennos RA (2004) Multiple hybrid origins, genetic diversity and population genetic structure of two endemic Sorbus taxa on the Isle of Arran, Scotland. Mol Ecol 13: 123–134.
  52. 52. Yatabe Y, Masuyama S, Darnaedi D, Murakami N (2001) Molecular systematics of the Asplenium nidus complex from Mt. Halimun National Park, Indonesia: evidence for reproductive isolation among three sympatric rbcL sequence types. Am J Bot 88: 1517–1522.
  53. 53. Pryer KM, Smith AR, Hunt JS, Dubuisson J-Y (2001) rbcL data reveal two monophyletic groups of filmy ferns (Filicopsida: Hymenophyllaceae). Am J Bot 88: 1118–1130.
  54. 54. Ebihara A, Iwatsuki K, Ohsawa TA, Ito M (2003) Hymenophyllum paniense (Hymenophyllaceae), a new species of filmy fern from New Caledonia. Syst Bot 28: 228–235.
  55. 55. Ebihara A, Iwatsuki K, Ito M, Hennequin S, Dubuisson J-Y (2007) A global molecular phylogeny of the fern genus Trichomanes (Hymenophyllaceae) with special reference to stem anatomy. Bot J Linn Soc 155: 1–27.
  56. 56. Tate JA, Simpson BB (2003) Paraphyly of Tarasa (Malvaceae) and diverse origins of the polyploid species. Syst Bot 28: 723–737.
  57. 57. Swofford DL (2002) PAUP*: Phylogenetic Analysis Using Parsimony (* and Other Methods), version 4.0b10 [computer program]. Sunderland, Mass.: Sinauer Associates.
  58. 58. Chen S, Yao H, Han J, Liu C, Song J, et al. (2010) Validation of the ITS2 region as a novel DNA barcode for identifying medicinal plant species. PLoS ONE 5: e8613.
  59. 59. Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, et al. (2009) BLAST+: architecture and applications. BMC Bioinf 10: 421.