Molecular evidence for genetic heterogeneity and the hybrid origin of Acrorumohra subreflexipinna from Taiwan.

ABSTRACT.--Acrorumohra subreflexipinna, an endemic fern of Taiwan, has been suspected to be a hybrid species. The aims of this study were to detect possible multiple origins of this species, determine the genetic variation in different populations, and clarify their lineages. One nuclear and three organellar DNA fragments were sequenced to determine parentage of this putative hybrid and to examine genetic differentiation among populations. Sequence data support the conclusion that A. subreflexipinna arose from the hybridization of A. hasseltii and A. diffracta, and the hybridization was uni-directional, i.e., based on the assumption of maternal inheritance in organellar DNA, the former was its maternal species while the latter was its paternal source. A convincing interpretation is that the female gametes of A. hasseltii gametophyte could be fertilized by the male gametes from apogamous A. diffracta. Unique nuclear alleles present in different populations of A. subreflexipinna and A. hasseltii demonstrated that hybridization occurred many times independently. The nuclear haplotypes present in A. subreflexipinna were subsets of those found in the parental species, and A. subreflexipinna always had lower haplotype diversity than A. hasseltii at sympatric sites. Our results show that any genetic variation of A. subreflexipinna came from its parents and that it maintains this significant genetic variability because of recurrent hybridization.

KEY WORDS.--Acrorumohra diffracta, Acrorumohra hasseltii, Acrorumohra subreflexipinna, hybridization, monilophytes, multiple origins

**********

Hybridization followed by polyploidization is an important mechanism driving the formation of new lineages of ferns and other plants (Paun et al., 2007). By means of diploidization processes, such as chromosomal rearrangements, intergenome recombination, and gene silencing, the genomic constitution of many extant taxa might be the outcome of ancient hybridization and polyploidy (Bowers et al., 2003; De Bodt et al., 2005; Haufler, 1987; Paun et al., 2007). Hybridization events often begin these cycles and high chromosomal base number in ferns was achieved as the result of repeated cycles of polyploidization (Haufler, 1987; Klekowski and Baker, 1966; Nakazato et al., 2006).

Accessing the parentage of hybrids or allopolyploids is essential for understanding relationships within taxonomically complex groups. Although allozyme studies could provide tenable evidence to indicate the possible origin of hybrid-originated taxa, they have rarely been utilized to distinguish maternal lines from paternal ones. However, direct DNA evidence, such as nucleotide sequences and DNA fingerprints, can provide more informative insights into these evolutionary processes than enzymes. In most plants, organelle genomes are maternally inherited via female gametes while nuclear DNA is biparentally inherited (Soltis et al., 1992). Comparing organellar DNA of hybrid taxa and their possible parents therefore could reveal the maternal origin (Gastony and Yatskievich, 1992; Vogel et al., 1998) while comparing nuclear DNA of those taxa could show both putative parentages (Small et al., 2004). In addition, any evolutionary trace, theoretically, would be deposited in nucleotide sequences and could be detected by DNA-based molecular technology. Unique local variation would be detectable if applicable DNA markers were chosen (Soltis et al., 1992). DNA markers containing non-coding regions have been shown to be the best choice to reconstruct genealogies of hybrid and parental populations (Small et al., 2004; Xiang et al., 2000).

Studies of north temperate ferns have clearly indicated the contribution of hybridization and polyploidization to fern evolution (Barrington et al,. 1989; Bennert et al., 2005; Pinter et al., 2002; Wagner, 1973; Werth et al., 1985). Out of the 420 species of lycophytes and ferns that grow in North America, nearly 20% are of hybrid origin (Flora of North America Editorial Committee, 1993), and reticulate networks and ploidy levels of most taxonomically complex groups have been well studied (Barrington, 1986; Stein and Barrington, 1990; Wagner, 1954, 1962, 1973; Xiang et al., 2000). However, only a few ferns from other regions have received taxonomic attention like those in Europe and North America (e.g., Barrington, 1990; Ebihara et al., 2005; Takamiya et al., 2001; Terada and Takamiya, 2006). Some hybrid ferns have been recorded from Taiwan (Holttum and Edwards, 1986; Kuo, 1988, 1990; Miyamoto and Nakamura, 1983), but until now, no direct evidence has been reported to test and verify their parentage.

Acrorumohra is a small genus with about seven species distributed in Eastern and Southeastern Asia. This genus has an intermediate morphology between Dryopteris and Arachniodes; therefore, species of Acrorumohra were once treated in these genera. However, Acrorumohra was treated as an independent genus in the Flora of Taiwan (Shieh et al., 1994) and Flora Reipublicae Popularis Sinicae (Hsieh, 2000) based on the presence of the zigzag rachis and anadromous pinnules of pinnae. Acrorumohra subreflexipinna (M. Ogata) H. Ito, an endemic species of Taiwan, produces shriveled and abortive spores and has an intermediate morphology between A. hasseltii (Blume) Ching and A. diffracta (Baker) H. Ito. Given its morphological characteristics, A. subreflexipinna has been suspected as a hybrid of these two species (Moore, 2000). Moreover, the fact that A. subreflexipinna always grows sympatrically with the later two species reinforces the reasonable hypothesis of its hybrid origin. The narrowly defined genus 'Acrorumohra' was followed and the scientific name 'A. subreflexipinna' is used throughout the study, although palynological and unpublished breeding data indicates it a sterile F1 hybrid. In this study, chloroplast, mitochondria and nucleus DNA markers were used to identify the parentage of this suspected hybrid. Furthermore, the hypothesis that hybrid populations in Taiwan each originated independently was tested. In addition to haplotype comparison, genetic variation in different populations was determined to clarify lineage relationships.

[FIGURE 1 OMITTED]

MATERIALS AND METHODS

Plants of Acrorumohra subreflexipinna were sampled from three sites in Taiwan: Mt. Howeishan, Lake Chunglingchih and Mt. Kentuerhshan (Fig. 1). Leaf tissue of four to 11 individuals per population was collected for molecular analyses. Ten individuals of the two putative parent species, A. hasseltii and A. diffracta, were also sampled in each sympatric site (Table 1). Two plants of Dryopteris polita Rosenst. were also sampled to detect any possible parental relationship because based on phylogenetic analysis of a chloroplast trnS-rps4 data set, D. polita and A. hasseltii are sister species (Li and Lu, 2006).

Two chloroplast intergenic spacers (trnL-trnF and trnS-rps4 IGS) and one mitochondrial intron (had5 intron 2), which have been frequently used for phylogenetic analysis at lower taxonomic levels, were employed to reveal maternal history, while introns of the single-copy nuclear gene pgiC (including introns 14 and 15, and exon 15) were used to observe bi-parental inheritance. These sequences were chosen because of their significant phylogenetic information relative to other fragments and the availability of usable primers (Ishikawa et al., 2002; Nadot et al., 1995; Smith and Cranfill, 2002; Vangerow et al., 1999).

Dry or fresh tissues of young leaves were homogenized with liquid nitrogen. Genomic DNA was extracted from ca. 100 mg of leaf tissue by using a Plant Genomic DNA Mini Kit (Viogene, USA). The PCR amplification of all segments was performed in an ABI thermocycler (9700). Primers for trnL-trnF IGS, trnLF-11 5'-GCG CAA GTT GCG GTA GAA CGA-3' and trnLF-12 5'-CTG CTC TAC CGA CTG AG CTA-3', were modifications of those utilized by Taberlet et al. (1991). The primers tsr4-f/tsr4-r 5'-CCC GCA AAG CTT AGT GAT CA-3'/ 5'-CCG AGG GTT CGA ATC CCT C-3', nadh2-f/nadh2-r 5'-GGG GCT ATA TCG CCA TCC-3'/5'-CCG CAC GTG CAA GTT TCC-3', and pgiC-14fA/pgiC-16rA 5'GTG CTT CTG GGT CTT TTG AG-3'/5'-GTT GTC CAT TAG TTC CAG GT-3' were developed for this study referring to Smith and Cranfill (2002), Vangerow et al. (1999) and Ishikawa et al. (2002), respectively. PCR reactions were carried out in 20 [micro]L reactions containing 2 [micro]L unstandardized template DNA, 0.2 mmol/L of each dNTP, 0.8 units of Taq polymerase (ABgene, USA) and 6.25 pmol each of the forward and reverse primers, and programmed for 5 min at 95[degrees]C, 35 cycles of i min at 95[degrees]C, 1 min at annealing temperature and 2 min at 72[degrees]C, followed by a 8 min extension at 72[degrees]C. The annealing temperature was 59[degrees]C in amplifying the chloroplast trnL-trnF IGS and the mitochondrial nad5 intron 2, and 52[degrees]C in amplifying the chloroplast trnS-rps4 fragment. When amplifying nuclear pgiC intron 14-15 segment, annealing was performed at 57[degrees]C for the first 3 cycles, at 55[degrees]C for the next 3 and at 54[degrees]C for the final 29. PCR products were directly sequenced, using one amplification primer, on an ABI 373A automated sequencer (Applied Biosystems, USA) with the Taq Dye Dideoxy Terminator Cycle Sequencing Kit (Applied Biosystems). For the electrophoresed bands with lengths greater than 500 bp, sequences were determined in both directions. Additionally, pgiC intron 14-15 segments of all A. subreflexipinna samples and 3-5 samples in each population of A. hasseltiiand A. diffracta were cloned. The PCR products of the nuclear segment were purified by electrophoresis using 1x TAE buffer on a 1.2% agarose gel. Electrophoresed bands were cut and eluted using the Gel-M gel extraction system (Viogene). Purified nuclear DNA was cloned with the yT&A cloning kit (Yeastern Biotech, Taiwan) following the manufacturer's protocol. Five to eight colonies were chosen to perform colony PCR using TA-F forward and TA-R reverse primers (Yeastern Biotech). Purified nuclear DNA was sequenced with M13 universal and reverse primers which are located on the DH5[alpha] vector termination site. When any different haplotype was detected, repeated PCR reactions using a different Taq polymerase (Genomics, Taiwan) or using DNA from another three colonies were chosen to check whether it was a real variant or not. All sequences were deposited in the GenBank nucleotide sequence database, and accession numbers and their corresponding DNA regions are listed in Table 1.

The sequences were aligned by BioEdit 7.0 and manual correction, and compared with nucleotide sequences available through GenBank to determine their boundaries of coding region. Haplotypes were named after the first letter of the specific epithet, and followed by a lowercase letter and number to designate different, minor haplotypes (those differing from their corresponding major haplotype at only one base) of A. hasseltii. Genetic diversity at population and species levels was estimated with the software package DNA Sequence Polymorphism (DnaSP 4.20.2, Rozas et al., 2003). The haplotype diversity (h) and nucleotide diversity ([pi]) of these three populations were calculated separately and totally. Genetic differentiation ([[gamma].sub.ST], Nei, 1982) among these three populations and between pairs of populations was also calculated by this package. [[gamma].sub.st], but not [F.sub.ST] or [N.sub.ST], was used because the three sampled populations were the only ones of interest (Lynch and Crease, 1990). Because no variation was detected in the nuclear sequences of A. diffracta, only the A. hasseltii haplotypes cloned from A. subrefiexipinna were used when analyzing genetic diversity and differentiation among the populations of A. subreflexipinna. Haplotypes of A. hasseltii and A. subreflexipinna were identified and coded by direct sequence comparison, and unrooted haplotype networks were constructed with the program TCS 1.21 (Clement et al., 2000).

RESULTS

Total aligned length and GC content of the sequences of nuclear pgiC intron 14-15, chloroplast trnL-trnF IGS and trnS-rps4 IGS, and mitochondrial nad5 intron 2 were 725 bp/37.8%, 268 bp/34.9%, 374 bp/36.5%, and 728 bp/ 52.6%, respectively. Low GC content of chloroplast segments agreed with the AT-rich property of most non-coding spacers (Graur and Li, 2000).

In the chloroplast and mitochondria segments, all 50 individuals of A. subreflexipinna and A. hasseltii had the same nucleotide sequences but were different from those of A. diffracta and D. polita (Tables 2-4). In the nuclear pgiC intron 14-15 sequences, A. subreflexipinna possessed both the A. hasseltii and the A. diffracta haplotypes (Table 5) but not that of D. polita (data not shown), pgiC intron 14-15 sequences of A. diffracta from the three populations were all the same (haplotype 'D'), but those of A. hasseltii and A. subreflexipinna in each population had two to three haplotypes (Tables 5 and 6). There were two major (Ha and Hb) and another three minor haplotypes (Hal, Hbl and Hb2) found in A. hasseltii (Table 5). These minor haplotypes differ from their corresponding major haplotypes at only one base, and were found in the three populations respectively. In total, five and four haplotypes were found in A. hasseltii and A. subrefiexipinna, respectively.

When calculating haplotype diversity (h) and nucleotide diversity ([pi]) (Table 6), the haplotype "D" was, a priori, removed from the genetic pool of A. subreflexipinna to avoid interference in comparison with that of A. hasseltii. In A. hasseltii, haplotype diversity (h) among these three populations ranged from 0.533 to 0.689, and it was 0.724 at the species level. In A. subreflexipinna, haplotype diversity among populations ranged from 0.400 to 0.667, and it was 0.674 at the species level. Nucleotide diversity ([pi]] among the three populations of A. hasseltii ranged from 0.00074 to 0.00446, and it was 0.00485 at the species level. In A. subreflexipinna, nucleotide diversity among populations ranged from 0.00055 to 0.00553, and it was 0.00466 at the species level.

For the nuclear pgiC segment of A. hasseltii and A. subreflexipinna, the Ha haplotype could be clearly distinguished from Hb by six sites with different base pairs and two indel sites (Table 5; Fig. 2). The Ha haplotype has a minor type (Hal) with a single base difference. This minor haplotype is found only in the Mt. Kentuerhshan population of A. hasseltii and A. subreflexipinna (Fig. 2). On the other hand, the Hb haplotype has two single base change minors (Hbl and Hb2) occurring respectively in Lake Chunglingchih and Mt. Howeishan populations of A. hasseltii (Fig. 2(i)). In A. subreflexipinna, genetic variation among different individuals and/or populations directly came from different haplotypes of A. hasseltii. For example, in A. subreflexipinna of Mt. Kentuerhshan, except for the haplotype that was identical to A. diffracta, there were two haplotypes (Ha and Hal) that were also found in A. hasseltii of the sympatric site. Nuclear haplotypes of A. hasseltii in Mt. Howeishan and Lake Chunglingchih were identical except for the two minors (Hb1 and Hb2). However, only one major haplotype (Ha) was found in A. hasseltii and A. subreflexipinna of Mt. Kentuerhshan. The Hb and derivatively minor haplotypes were found neither in A. hasseltii nor A. subreflexipinna of Mt. Kentuerhshan.

The level of divergence among the three populations could not be revealed by the organellar fragments because only one haplotype was detected in each species (Tables 2-4). For nuclear pgiC intron 14-15 sequences, however, DnaSP analysis revealed high levels of genetic differentiation among three populations of A. hasseltii and A. subreflexipinna ([[gamma].sub.ST] = 0.44377 and [[gamma].sub.ST] = 0.26399; Table 7). Additionally, higher levels of genetic differentiation were also detected between northern and southeastern populations (A-B and A-C; Table 7) of these two species while little differentiation was found between those northern two (B-C; Table 7). For this same fragment, on the other hand, A. diffracta had only one haplotype and indicated no pattern of population structure.

DISCUSSION

Hybridization and parentage.--Similar to the traditional circumscription of species, hybrid species and hybrid parentage are usually postulated initially based on morphological characters and degree of spore/pollen abortion (Barrington, 1989, 1990). Acrorumohra subreflexipinna is suspected as a natural hybrid between A. hasseltii and A. diffracta (Moore, 2000) because A. subreflexipinna has abortive spores and intermediate morphology between A. hasseltii and A. diffracta, and occurs sympatrically with these two species. In addition, A. subreflexipinna's spores show no germination, but those of A, hasseltii and A. diffracta germinate at a rate of more than 80% (unpublished data). Therefore, A. subreflexipinna appears to be a sterile F1 hybrid. Acrorumohra subreflexipinna with its perennial habit, however, could occupy an original habitat for a long time despite of all spores being sterile. Repeated hybridization where the putative parents sympatrically exist might also replenish the stock of this hybrid.

Organelle genomes are generally maternally inherited in monilophytes (Gastony and Yatskievich, 1992; Vogel et al., 1998). The assumption that chloroplast and mitochondria are maternally inherited is adopted through this study. All organellar sequence data indicated that A. hasseltii was the maternal parent of A. subreflexipinna. Nuclear pgiC sequences indicated that A. diffracta was the other genome donor of this hybrid.

In addition to the three taxa of Acrorumohra discussed here, another species, A. yoroii (Seriz.) Shieh, was reported in the second edition of Flora of Taiwan (Shieh et al., 1994). In Taiwan, it grows in high montane regions and never sympatrically with other three taxa of Acrorumohra. Samples of that species were also collected from Taiwan and sequenced. It has organellar and nuclear sequences different from those of A. subreflexipinna, and phylogenetic analysis indicates a distant relationship between them (data not shown).

There are three other species of this narrowly defined genus. Acrorumohra dissecta Ching ex Hsieh is distributed in a few locations of southwestern China, and A. obtusissima (Mett. ex Kuhn) Ching and A. undulata (Bedd.) Ching are distributed throughout Sri Lanka. Though we cannot reject the hypothesis, the possibility of these species contributing to the formation of this hybrid is extremely low because of their restricted habitats and disjunct distribution from this hybrid.

Although the phylogenetic analysis based on chloroplast trnS-rps4 IGS sequence show a sister-group relationship between D. polita and A. hasseltii (Li and Lu, 2006; our unpublished data), this study reveals that A. subreflexipinna has sequences different from those of D. polita in both organellar (Tables 2-4) and nuclear (data not shown) genomes. Therefore, Dryopteris polita did not contribute to the formation of this hybrid. These molecular data plus morphological and ecological information explicitly suggest that A. subreflexipinna arose through hybridization of A. hasseltii and A. diffracta, and that the former was its putative maternal parent while the latter was its paternal source.

Acrorumohra hasseltii is distributed in tropical Asia, including Java, Borneo, Thailand, Nepal, East Himalayas, Vietnam, Hainan, Taiwan and southern Japan (Fig. 1). The range of A. diffracta overlaps with A. hasseltii in the northern portion of the range of A. hasseltii, i.e., East Himalayas, northern Thailand, Vietnam, Hainan and Taiwan (Fig. 1). Except for southwestern China, the geographic range of A. diffracta almost completely overlaps with that of A. hasseltii. However, A. subreflexipinna has only been reported from Taiwan. It is suspected that A. subreflexipinna might be established at some sites across this widely overlapping range of these two putative parents but misidentified as A. diffracta because of their similar zigzag rachis. Careful recognition is needed to identify this hybrid in future field investigation where the range of these two species overlaps.

Gender bias in hybridization events has been demonstrated many times in plants (Emms et al., 1996; Vogel et al., 1998; Weiblen and Brehm, 1996; Xiang et al., 2000). For reasons not entirely clear, the hybridization of A. hasseltii and A. diffracta in our study was absolutely biased, i.e., A. hasseltii always was the supplier of egg while A. diffracta was that of sperm. This phenomenon has also been found in other hybrid species (e.g., Arnold and Bennett, 1993; Peng and Chiang, 2000; Smith and Sytsma, 1990; Wendel et al., 1991). In ferns, mating systems usually correlate with ploidy levels and could be a decisive factor in the nuclear-organellar combination pattern of parental genotypes in hybridization. In fact, A. diffracta was reported as a tetraploid (Tsai and Shieh, 1975) and A. hasseltii a diploid (Iwatsuki, 1995), and our observation showed that A. hasseltii has 64 spores per sporangium but A. diffracta has 32 spores (unpublished data). These agree with the general rule that a sexual species usually has 64 spores per sporangium whereas there are typically 32 spores in an apogamous species. According to the Dopp-Manton Scheme and Braithwaite modes of reproduction (Raghavan, 1989), Acrorumohra diffracta might be an obligate apogamous species with functional antheridia. These observations and molecular evidence lead to a convincing hypothesis for unidirectional hybridization that sexual A. hasseltii could adopt sperm from the apogamous tetraploid, A. diffracta. Additionally, a lack of heterozygosity of the A. hasseltii genotypes in this hybrid further supports that A. hasseltii was the haploid female gamete donor in these hybridization events. In this case, because fertile sperm were liberated from an apogamous, tetraploid gametophyte, A. subreflexipinna should be a pentaploid. C-value results based on flow cytometry show that A. subreflexipinna has the highest ploidy level among these taxa (unpublished data), which confirms this hypothesis. Further observations on chromosome counts and the mating system of A. subreflexipinna and both parents in the future would provide more direct evidence.

[FIGURE 2 OMITTED]

Multiple independent origins.--Although A. subreflexipinna has a larger body size than both putative parental species, it produces no fertile spores, and its small population size and rare occurrence strongly suggests it is a product of occasional hybridization event(s). Both cpDNA and mtDNA fragments of populations of A. hasseltii and A. subreflexipinna were identical in sequence. Single direction hybridization and allopolyploidization may produce cytoplasmically uniform hybrids or allopolyploids despite distinct origins (Soltis et al., 1992). The variation in cpDNA and mtDNA from the taxa of this study was uninformative regarding multiple origins of A. subreflexipinna. Nuclear pgiC sequences, however, revealed different haplotypes and genetic differentiation among populations of A. hasseltii and A. subreflexipinna. Southern and northern populations of A. hasseltii had distinct and unique genotypes, and the genetic uniqueness was transmitted to their hybrid offspring (Table 5; Fig. 2). In both A. hasseltii and A. subreflexipinna, there were nine variable sites that had different bases in southern and northern populations (Table 5). Haplotype Hal was found in A. hasseltii of Mt. Kentuerhshan and was present in A. subreflexipinna in the same location. On the other hand, major haplotype Hb was only found in the northern populations of both species but not in the southern ones (Fig. 2). These are direct indicators of multiple independent origins of A. subreflexipinna.

There was no variation of organellar DNA fragments in populations of A. hasseltii and A. subreflexipinna, but high nucleotide diversity was found in the nuclear pgiC intron 14-15 of the same species. The fact that, usually, evolutionary rates of nuclear DNA are faster than those of chloroplast and mitochondria in plants (Graur and Li, 2000) may provide a reasonable interpretation of this significant difference. This also indicates that nuclear DNA markers may be a better choice when analyzing population variation due to relatively short evolutionary time. In this case, geographical barriers might effectively hinder gene flow, causing geographical subdivision in nuclear DNA, but the time of isolation might not have been long enough to accumulate new mutations in organellar DNA.

Several studies indicate that multiple origins of hybrid and polyploid species are common, if not the rule, in plants (see Soltis et al., 1992 and references therein). Genetic variation of the parental species could be incorporated into and preserved in hybrids and their derivative taxa by these processes (Arft and Ranker, 1998; Paun et al., 2007; Peng and Chiang, 2000). This phenomenon was also found in this study. Unlike a small population usually having fixed alleles for most loci, populations of A. subreflexipinna, even when only four individuals were found in a population, have high haplotype diversity. Moreover, the results agree with our anticipation that in each collecting site there were fewer haplotypes and lower haplotype diversity of A. subreflexipinna than those of A. hasseltii. Because A. subreflexipinna is a sterile hybrid, any genetic variation should come from its parents. It is the recurrent hybridization of this hybrid that might maintain its significant genetic variability and provide operative materials for future evolution, i.e., polyploidization and subsequent fertilization.

ACKNOWLEDGMENTS

The authors thank James R. Shevock for advice and revision of English. We are thankful to Mrs. Lu, Pi-Fong and Dr. Liu, Yea-Chen for the information of distribution sites and to Miss Chao, YiShan for help in experiment. We also extend appreciation to Dr. Geiger and two anonymous reviewers for their critical reviews of the manuscript. This study was supported by Council of Agriculture (grant number 96AS-11.2.3-EI-W2), Taiwan.

LITERATURE CITED

ARFT, A. M. and T. A. RANKER. 1998. Allopolyploid origin and population genetics of the rare orchid Spiranthes diluvialis. Am. J. Bet. 85:110-122.

ARNOLD, M. L. and B. D. BENNETT. 1993. Natural hybridization in Louisiana irises: genetic variation and ecological determinants. In R. G. HARRISON (ed.), Hybrid zones and the evolutionary process, pp. 115-139. Oxford University Press, New York.

BARRINGTON, D. S. 1986. The morphology and cytology of Polystichum x potteri hybr. nov. (=P. acrostichoides x P. braunii). Rhodora 88:297-313.

BARRINGTON, D. S. 1989. New species and combination in tropical American Polystichum (Dryopteridaceae). Ann. Miss. Bot. Gard. 76:365-373.

BARRINGTON, D. S. 1990. Hybridization and allopolyploidy in central American polystichum: cytological and isozyme documentation. Ann. Miss. Bot. Gard. 77:297-305.

BARRINGTON, D. S., C. H. HAUFLER and C. R. WERTH. 1989. Hybridization, reticulation, and species concepts in the ferns. Am. Fern J. 79(2):55-64.

BENNERT, W., M. LUBIENSKI, S. KORNER and M. STEINBERG. 2005. Triploidy in Equisetum subgenus Hippochaete (Equisetace, Pteridophyta). Ann. Bot. 95:807-815.

BOWERS, J. E., B. A. CHAPMAN, J. RONG and A. H. PATERSON. 2003. Unravelling angiosperm genome evolution by phylogenetic analysis of chromosome duplication events. Nature 422:433-438.

CLEMENT, M., D. POSADA and K. A. CRANDALL. 2000. TCS: a computer program to estimate gene genealogies. Mol. Ecol. 9:1657-1660.

DE BODT, S., S. MAERE and Y. VAN DE PEER. 2005. Genome duplication and the origin of angiosperms. Trends Ecol. Evol. 20:591-597.

EBIHARA, A., H. ISHIKAWA, S. MATSUMOTO, S. LIN, K. IWATSUKI, M. TAKAMIYA, Y. WATANO and M. ITO. 2005. Nuclear DNA, chloroplast DNA, and ploidy analysis clarified biological complexity of the Vandenboschia radicans complex (Hymenophyllaceae) in Japan and adjacent areas. Am. J. Bot. 92:1535-1547.

EMMS, S. K., S. A. HODGES and M. L. ARNOLD. 1996. Pollen-tube competition, siring success, and consistent asymmetric hybridization in Louisiana irises. Evolution 50:2201-2206.

FLORA OF NORTH AMERICA EDITORIAL COMMITTEE. 1993. Flora of North America. Vol. 2, Pteridophytes and Gymnosperms. Oxford University Press, New York.

GASTONY, G. J. and G. YATSKIEVICH. 1992. Maternal inheritance of the chloroplast and mitochondrial genomes in cheilanthoid ferns. Am. J. Bot. 79:716-722.

GRAUR, D. and W.-H. LI. 2000. Fundamentals of molecular evolution, 2nd ed. Sinaner Associates, Sunderland.

HAUFLER, C. H. 1987. Electrophoresis is modifying our concepts of evolution in homosporous pteridophytes. Am. J. Bot. 74:953-966.

HOLTTUM, R. E. and P. J. EDWARBS. 1986. Studies in the fern genera allied to Tectaria II. Dryopsis, a new genus. Kew Bull. 41:171-204.

HSIEH, Y.-T. 2000. Acrorumohra (H. Ito) H. Ito. In S.-G. Wu (ed.), Flora Reipublicae Popularis Sinicae 5(1):8-11. Science Press, Beijing. (in Chinese)

ISHIKAWA, H., Y. WATANO, K. KANO, M. ITO and S. KURITA. 2002. Development of primer sets for PCR amplification of the PgiC gene in ferns. J. Plant Res. 115:65-70.

IWATSUKI, K. 1995. Dryopteridaceae. In K. IWATSUKI, et al. (eds.), Flora of Japan I, Pteridophyta and Gymnospermae, pp. 120-173. Kodansha, Tokyo.

KLEKOWSKI, E. J. and H. G. BAKER. 1966. Evolutionary significance of polyploidy in the Pteridophyta. Science 153:305-307.

KUO, C.-M. 1988. A new Asplenium hybrid from Taiwan. Bot. Bull. Acad. Sin. 29:109-111.

KUO, C.-M. 1990. Materials for the Lomariopsidaceae of Taiwan. Bot. Bull. Acad. Sin. 31:305-314.

LI, C.-X. and S.-G. Lu. 2006. Phylogenetics of Chinese Dryopteris (Dryopteridaceae) based on the chloroplast rps4-trnS sequence data. J. Plant Res. 119:589-598.

LYNCH, M. and T. J. CREASE. 1990. The analysis of population survey data on DNA sequence variation. Mol. Biol. Evol. 7:377-394.

MIYAMOTO, F. and T. NAKAMURA. 1983. Three new hybrids of Polystichum from Taiwan. J. Jap. Bot. 58:146-150.

MOORE, S.-J. 2000. Acrorumohra subreflexipinna. In S.-Y. Lu, et al. (eds.), Rare and endangered plants in Taiwan, pp. 1-2. Council of Agriculture Executive Yuan, Taipei. (in Chinese)

NADOT, S., G. BITTAR, L. CARTER, R. LACROIX and B. LEIEUNE. 1995. A phylogenetic analysis of monocotyledons based on the chloroplast gene rps4, using parsimony and a new numerical phenetics method. Mol. Phylogen. Evol. 4:257-282.

NAKAZATO, T., M.-K. JUNG, E. A. HOUSWORTH, L. H. RIESEBERG and G. J. GASTONY. 2006. Genetic map-based analysis of genome structure in the homosporous fern Ceratopteris richardii. Genetics 173:1585-1597.

NEI, M. 1982. Evolution of human races at the gene level. In B. BONNE-TAMIR (ed.), Human Genetics, Part A: The Unfolding Genome, pp. 167-181. Alan R. Liss, New York.

PAUN, O., M. F. FAY, D. E. SOLTIS and M. W. CHASE. 2007. Genetic and epigenetic alterations after hybridization and genome doubling. Taxon 56:649-656.

PENG, C.-I. and T.-Y. CHIANG. 2000. Molecular confirmation of unidirectional hybridization in Begonia x taipeiensis Peng (Begoniaceae) from Taiwan. Ann. Miss. Bot. Gard. 87:273-285.

PINTER, I., F. BAKKER, J. BARRETT, C. Cox, M. GIBBY, S. HENDERSON, M. MORGAN-RICHARDS, F. RUMSEY, S. RUSSELL, S. TREWICK, H. SCHNEIDER and J. VOGEL. 2002. Phylogenetic and biosystematic relationships in four highly disjunct polyploid complexes in the subgenera Ceterach and Phyllitis in Asplenium (Aspleniaceae). Org. Divers. Evol. 2:299-311.

RAGHAVAN, V. 1989. Developmental biology of fern gametophytes. Cambridge University Press, Cambridge.

ROZAS, J., J. C. SANCHEZ-DELBARRIO, X. MESSEGYER and R. ROZAS. 2003. DnaSP, DNA polymorphism analyses by the coalescent and other methods. Bioinformatics 19:2496-2497.

SHIEH, W.-C., C. E. DEVOL and C.-M. Kuo. 1994. Dryopteridaceae. In T.-C. HUANG, et aL (eds.), Flora of Taiwan 2nd ed, 1:303-351. Editorial Committee of the Flora of Talwan, Second Edition, Taipei.

SMALL, R. L., R. C. CRONN and J. F. WENDEL. 2004. L. A. S. Johnson Review No. 2, Use of nuclear genes for phylogeny reconstruction in plants. Aust. Syst. Bet. 17:145-170.

SMITH, A. R. and R. B. CRANFILL. 2002. Intrafamilial relationships of the thelypteroid ferns (Thelypteridaceae). Am. Fern J. 92(2):131-149.

SMITH, R. L. and K. J. SYTSMA. 1990. Evolution of Populus nigra (sect. Aigeiros): introgressive hybridization and the chloroplast contribution of Populus elba (sect. Populus). Am. J. Bet. 77:1176-1187.

SOLTIS, P. S., J. J. DOYLE and D. E. SOLONS. 1992. Molecular data and polyploidy evolution in plants. In P. S. SOLTIS, et al. (eds.), Molecular systematics of plants, pp. 177-201. Chapman & Hall, New York.

STEIN, D. B. and D. S. BARRINGTON. 1990. Recurring hybrid formation in a population of Polystichum x potteri: evidence from chloroplast DNA comparisons. Ann. Miss. Bet. Gard. 77:334-339.

TABERLET, P., L. GIELLY, G. PAUTOU and J. BOUVET. 1991. Universal primers for amplification of three non=coding regions of chloroplast DNA. Plant Mol. Biol. 17:1105-1109.

TAKAMIYA, M., N. OHTA, Y. YATABE and N. MURAKAML 2001. Cytological, morphological, genetic, and molecular phylogenetic studies on intraspecific differentiations within Diplazium doederleinii (Woodsiaceae: Pteridophyta). Int. J. Plant Sci. 162:625-636.

TERADA, Y. and M. TAKAMIYA. 2006. Cytological and genetic study of two putative hybrids and their parents of Athyrium (Woodsiaceae; Pteridophyta) in Yakushima Island, southwestern Japan. Acta Phytotax. Geobot. 57(1):95-106.

TASI, J.-L. and W.-C. SHIEH. 1975. Chromosome numbers of the fern family Aspidiaceae (sensu Copeland) in Taiwan. J. Sci. Engin. 12:321-334.

VANGEROW, S., T. TEERKORN and V. KNOOP. 1999. Phylogenetic information in the mitochondrial ned5 gene of pteridophytes: RNA editing and intron sequences. Plant Biol. 1:235-243.

VOGEL, J. C., S. J. RUSSELL, F. J. RUMSEY, J. A. BARRETT and M. GIBBY. 1998. Evidence for maternal transmission of chloroplast DNA in the genus Asplenium (Aspleniaceae, Pteridophyta). Bet. Acta 111:247-249.

WAGNER, W. H. 1954. Reticulate evolution in the Appalachian aspleniums. Evolution 8:103-118.

WAGNER, W. H. 1962. Irregular morphological development in hybrid ferns. Phytomorphology 12:87-100.

WAGNER, W. H. 1973. Reticulation of holly ferns (Polystichum) in the western United States and adjacent Canada. Am. Fern J. 63(3):99-115.

WEIBLEN, G. D. and B. G. BREHM. 1996. Reproductive strategies and. barriers to hybridization between Tellima grandiflora and Tolmeia menziesii (Saxifragaceae). Am. J. Bet. 83:910-918.

WENDEL, J. F., J. M. STEWART and J. H. RETTIG. 1991. Molecular evidence for homoploid reticulate evolution among Australian species of Gossypium. Evolution 45:694-711.

WERTH, C. R., S. I. GUTTMAN and W. H. ESHBAUGH. 1985. Electrophoretic evidence of reticulate evolution in the Appalachian Asplenium complex. Syst. Bet. 10:184-192.

XIANG, L., C. R. WERTH, S. N. EMERY and D. E. McCAULEY. 2000. Population-specific gender-biased hybridization between Dryopteris intermedia and D. carthusiana: evidence from chloroplast DNA. Am. J. Bet. 87:1175-1180.

Ho-MING CHANG

Department of Life Science, National Taiwan Normal University, No.88, Ting-Chow Rd., Sec 4, Taipei 116, Taiwan

Division of Botany, Endemic Species Research Institute, No.l, Ming-Sheng E. Rd., Chi-Chi, Nantou 552, Taiwan

WEN-LIANG CHIOU

Division of Forest Biology, Taiwan Forestry Research Institute No. 53, Nanhai Rd., Taipei 100,

Taiwan

JENN-CHE WANG *

Department of Life Science, National Taiwan Normal University No.88, Ting-Chow Rd., Sec 4,

Taipei 116, Taiwan

* corresponding author: e-mail: [e-mail]

TABLE 1. Voucher information, quantity of sample and GenBank
accession numbers for taxa used in molecular analysis of this study.

                Locality (population             Cloned sample no./no.
Taxon           code)/sample no.                       of clones

Acrorumohra     1. Mt. Kentuerhshan (A)/10               3/15
  diffracta
                2. Lake Chunglingchih (B) /10            3/15

                3. Mt. Howeishan (C) /10                 3/15

Acrorumohra     1. Mt. Kentuerhshan (A) /10              3/15
  hasseltii
                2. Lake Chunglingchih (B) /10            5/29

                3. Mt. Howeishan (C) /10                 4/23

Acrorumohra     1. Mt. Kentuerhshan (A)/5                5/52
  subre-
  flexipinna    2. Lake Chunglingchih (B)/4              4/37

                3. Mt. Howeishan (C)/11                  11/91

Dryopteris      Lienhuachih/2                            2/10
  polita

                Voucher/deposited   DNA region/GenBank
Taxon           herbarium           accession no.

Acrorumohra     Chang 6316/TNU      trnL-trnF/EU797681
  diffracta                         trnS-rps4/EU797685
                                    nad5 intron 2/EU797695
                                    pgiC intron 14-15/EU797705

                Chang 6538/TAIE     trnL-trnF/EU797682
                                    trnS-rps4/EU797686
                                    nad5 intron 2/EU797696
                                    pgiC intron 14-15/EU797706

                Chang 6667/TAIE     trnL-trnF/EU797683
                                    trnS-rps4/EU797687
                                    nad5 intron 2/EU797697
                                    pgiC intron 14-15/EU797707

Acrorumohra     Chang 6319/TNU      trnL-trnF/EU797679
  hasseltii                         trnS-rps4/EU797691
                                    nad5 intron 2/EU797701

                Chang 6320/TNU      pgiC intron 14-15/EU797708
                                    pgiC intron 14-15/EU797709

                Chang 6539/TAIE     trnL-trnF/EU797680
                                    trnS-rps4/EU797692
                                    nad5 intron 2/EU797702

                Chang 6544/TAIE     pgiC intron 14-15/EU797710

                Chang 6540/TAIE     pgiC intron 14-15/EU797711
                                    pgiC intron 14-15/EU797712

                Chang 6673/TAIE     trnL-trnF/EU797677
                                    trnS-rps4/EU797693
                                    nad5 intron 2/EU797703

                Chang 6674/TAIE     pgiC intron 14-15/EU797713

                Chang 6675/TAIE     pgiC intron 14-15/EU797714
                                    pgiC intron 14-15/EU797715

Acrorumohra     Chang 6317/TNU      trnL-trnF/EU797678
  subre-                            trnS-rps4/EU797688
  flexipinna                        nad5 intron 2/EU797698

                Chang 6322/TNU      pgiC intron 14-15/EU797720
                                    pgiC intron 14-15/EU797716
                                    pgiC intron 14-15/EU797719

                Chang 6541/TAIE     trnL-trnF/EU797675
                                    trnS-rps4/EU797689
                                    nad5 intron 2/EU797699

                Chang 6542/TAIE     pgiC intron 14-15/EU797717
                                    pgiC intron 14-15/EU797721
                                    pgiC intron 14-15/EU797723

                Chang 6671/TAIE     trnL-trnF/EU797676
                                    trnS-rps4/EU797690
                                    nad5 intron 2/EU797700
                                    pgiC intron 14-15/EU797718

                Chang 6672/TAIE     pgiC intron 14-15/EU797722
                                    pgiC intron 14-15/EU797724

Dryopteris      Chang 6903/TAIE     trnL-trnF/EU797684
  polita                            trnS-rps4/EU797694
                                    nad5 intron 2/EU797704
                                    pgiC intron 14-15/EU797725

Table 2. The variable nucleotide sites (indel & base substitution)
of chloroplast trnL-trnF intergenic spacer sequences. Columns
shaded are sites identical to the hybrid sequences.

                                      Variable sites

Species              0     0     0     0     0     0    0    0    1
                     0     1     2     2     2     2    4    6    0
                     7     9     0     1     3     3    9    8    5

A. diffracta         G    --#   --#   --#   --#   --#   G#   A    C
A. subreflexipinna   A#   --#   --#   --#   --#   --#   A#   G#   T#
A. hasseltii         A#   --#   --#   --#   --#   --#   A#   G#   T#
D. polita            G     T     A     G     T     T    A#   G#   T#

                         Variable sites

Species                 1     1   1    1    1    1    1     2    2
                        1     5   5    6    8    9    9     1    3
                        2     1   6    6    8    8    9     6    2

A. diffracta            T    C#   A#   T    C    G    C    --#   C
A. subreflexipinna      A#   C#   A#   C#   A#   A#   T#   --#   T#
A. hasseltii            A#   C#   A#   C#   A#   A#   T#   --#   T#
D. polita               A#   T    G    C#   G    A#   T#    A    C

Note: Columns shaded are sites identical to the hybrid sequences is
indicated with #.

TABLE 3. The variable nucleotide sites (indel & base substitution) of
chloroplast trnS-rps4 intergenic spacer sequences. Columns shaded are
sites identical to the hybrid sequences.

                                    Variables sites

Species              0    0    0    0    0     1     1    1    1    2
                     3    7    7    8    8     4     4    4    9    0
                     9    2    6    2    1     0     1    2    3    8

A. diffracta         A    T    G    G#   G    --#   --#   T    A    T
A. subreflexipinna   G#   G#   A#   G#   A#   --#   --#   C#   G#   C#
A. hasseltii         G#   G#   A#   G#   A#   --#   --#   C#   G#   C#
D. polita            A    G#   G    T    G     T     T    T    G#   T

Species              2    3    2    2    2     2    2     2    3    3
                     1    1    1    3    4     5    5     6    2    4
                     2    4    7    0    1     2    8     8    7    7

A. diffracta         A    G#   C#   C    C     T    A     C#   C    G#
A. subreflexipinna   C#   G#   C#   T#   T#    C#   C#    C#   T#   G#
A. hasseltii         C#   G#   C#   T#   T#    C#   C#    C#   T#   G#
D. polita            C#   C    T    C    C     C#   C#    T    T#   A

Species              3    3
                     1    7
                     8    3

A. diffracta         A    T
A. subreflexipinna   G#   C#
A. hasseltii         G#   C#
D. polita            G#   T

Note: Columns shaded are sites identical to the hybrid sequences is
indicated with #.

TABLE 4. The variable nucleotide sites (indel & base substitution)
of mitochondrion nods intron 2 sequences. Columns shaded are sites
identical to the hybrid sequences.

                                     Variables sites

Species                2     2     2     2     2    3     3    3    3
                       1     1     1     1     1    7     7    8    8
                       3     4     5     6     7    8     9    0    1

A. diffracta           T     G     G     T     T    T    --    T    T
A. subreflexipinna    --#   --#   --#   --#   --#   C#   A#   C#   C#
A. hasseltii          --#   --#   --#   --#   --#   C#   A#   C#   C#
D. polita              T     G     G     T     T    T    --    T    T

                                Variables sites

Species               4     7     7     7     7     7
                      1     0     0     0     0     1
                      1     6     7     8     9     0

A. diffracta          T#    C     C     G     C     T
A. subreflexipinna    T#   --#   --#   --#   --#   --#
A. hasseltii          T#   --#   --#   --#   --#   --#
D. polita            --    --#   --#   --#   --#   --#

Note: Columns shaded are sites identical to the hybrid sequences is
indicated with #.

TABLE 5. The haplotype and variable nucleotide sites (indel & base
substitution) of nuclear pgiC intron 14-15 sequences. Columns shaded
are sites different from those of other populations.

                                                        Variable sites
                                Haplotype
Site (a)   Species              code (b)    Frequency   0   0    0   1
                                                        4   6    9   2
                                                        8   3    1   2

A          A. diffracta         D           10          C   G    A   G
                                D           5           C   G    A   G
           A. subreflexipinna   Ha          1           T   G    T   C
                                Ha1         4           T   G    T   C
           A. hasseltii         Ha          6           T   G    T   C
                                Ha1         4           T   G    T   C

B          A. diffracta         D           10          C   G    A   G
                                D           4           C   G    A   G
           A. subreflexipinna   Ha          2           T   G    T   C
                                Hb          2           T   A#   T   C
           A. hasseltii         Ha          2           T   G    T   C
                                Hb          5           T   A#   T   C
                                Hb1         3           T   A#   T   C

C          A. diffracta         D           10          C   G    A   G
                                D           11          C   G    A   G
           A. subreflexipinna   Ha          5           T   G    T   C
                                Hb          6           T   A#   T   C
           A. hasseltii         Ha          4           T   G    T   C
                                Hb          5           T   A#   T   C
                                Hb2         1           T   G    T   C

                                     Variable sites

Site (a)   Species              1    1   2   2   2   2   2    3   3   3
                                4    6   2   4   5   5   6    0   4   5
                                4    9   6   7   3   4   6    8   4   0

A          A. diffracta         C    C   T   A   T   G   C    T   A   G
                                C    C   T   A   T   G   C    T   A   G
           A. subreflexipinna   T    T   C   G   C   T   C    C   A   A
                                T    T   C   G   C   T   T#   C   A   A
           A. hasseltii         T    T   C   G   C   T   C    C   A   A
                                T    T   C   G   C   T   T#   C   A   A

B          A. diffracta         C    C   T   A   T   G   C    T   A   G
                                C    C   T   A   C   T   C    C   A   G
           A. subreflexipinna   T    T   C   G   C   T   C    C   A   A
                                C#   T   C   G   C   T   C    C   A   A
           A. hasseltii         T    T   C   G   C   T   C    C   A   A
                                C#   T   C   G   C   T   C    C   A   A
                                C#   T   C   G   C   T   C    C   G#  A

C          A. diffracta         C    C   T   A   T   G   C    T   A   G
                                C    C   C   A   T   G   C    T   A   G
           A. subreflexipinna   T    T   T   G   C   T   C    C   A   A
                                C#   T   T   G   C   T   C    C   A   A
           A. hasseltii         T    T   T   G   C   T   C    C   A   A
                                C#   T   T   G   C   T   C    C   A   A
                                C#   T   T   G   C   T   C    C   A   A

                                             Variable sites

Site (a)   Species              4     4     4    4    5    5    5   5
                                8     5     5    6    7    2    2   3
                                1     0     1    5    8    2    3   4

A          A. diffracta         A    C     A     C    T   --   --   A
                                A    C     A     C    T   --   --   A
           A. subreflexipinna   G    --    --    C    C   C    T    G
                                G    --    --    C    C   C    T    G
           A. hasseltii         G    --    --    C    C   C    T    G
                                G    --    --    C    C   C    T    G

B          A. diffracta         A    C     A     C    T   --   --   A
                                A    C     A     C    T   --   --   A
           A. subreflexipinna   G    --    --    C    C   C    T    G
                                A#   C #   A #   G#   C   C    T    G
           A. hasseltii         G    --    --    C    C   C    T    G
                                A#   C #   A #   G#   C   C    T    G
                                A#   C #   A #   G#   C   C    T    G

C          A. diffracta         A    C     A     C    T   --   --   A
                                A    C     A     C    T   --   --   A
           A. subreflexipinna   G    --    --    C    C   C    T    G
                                A#   C #   A #   G#   C   C    T    G
           A. hasseltii         G    --    --    C    C   C    T    G
                                A#   C #   A #   G#   C   C    T    G
                                A#   C #   A #   G#   C   C    T    G

                                  Variable sites

Site (a)   Species              5   6    6   6   6
                                7   0    1   6   7
                                3   9    0   4   0

A          A. diffracta         G   C    A   T   T
                                G   C    A   T   T
           A. subreflexipinna   A   T    G   C   G
                                A   T    G   C   G
           A. hasseltii         A   T    G   C   G
                                A   T    G   C   G

B          A. diffracta         G   C    A   T   T
                                G   C    G   C   T
           A. subreflexipinna   A   T    G   C   G
                                A   C#   G   C   T#
           A. hasseltii         A   T    G   C   G
                                A   C#   G   C   T#
                                A   C#   G   C   T#

C          A. diffracta         G   C    A   T   T
                                G   C    A   T   T
           A. subreflexipinna   A   T    G   C   G
                                A   C#   G   C   T#
           A. hasseltii         A   T    G   C   G
                                A   C#   G   C   T#
                                A   C#   G   C   T#

Note: Columns shaded are sites different from those of other
populations is indicated with #.

Table 6. Number of haplotypes, estimates of haplotype diversity (h)
and nucleotide diversity ([pi]) of A. hasseltii and A. subreflexipinna.
The A. diffracta haplotype 'D' cloned froth A. subreflexipinna
were excluded prior to this analysis.

                                                No. of        No. of
Populations                Species            individuals   haplotypes

Total (A+B+C)            A. hasseltii             30            5
                         A. subreflexipinna       20            3
Mt. Kentuerhshan (A)     A. hasseltii             10            2
                         A. subreflexipinna        5            2
Lake Chunglingchih (B)   A. hasseltii             10            3
                         A. subreflexipinna        4            2
Mt. Howeishan (C)        A. hasseltii             10            3
                         A. sabreflexipinna       11            2

                           Haplotype      Nucleotide
Populations              diversity (h)   diversity (n)

Total (A+B+C)                0.724          0.00485
                             0.674          0.00466
Mt. Kentuerhshan (A)         0.533          0.00074
                             0.400          0.00055
Lake Chunglingchih (B)       0.689          0.00360
                             0.667          0.00553
Mt. Howeishan (C)            0.644          0.00446
                             0.545          0.00453

TABLE 7. Gene, tic differentiation between pairwise and among all
populations of A. hasseltii and A. subreflexipinna.

                       Genetic
                    differentiation
                 among all populations
Species           ([[gamma].sub.ST])       A-B       A-C       B-C

A. hasseltii            0.44377          0.60815   0.40303   0.06301
A. subreflexi-          0.26399          0.40000   0.31460   0.60162
   pinna

(a) A: Mt. Kentuerhshan, B: Lake Chunglingchih, C: Mt. Howeishan.