Libor Ekrt 2,5 , Renata Holubov Á 3 , Pavel Tr Á Vn Í Ček 3,4 , and Jan Suda

American Journal of Botany 97(7): 1208–1219. 2010.

S PECIES BOUNDARIES AND FREQUENCY OF HYBRIDIZATION IN THE D RYOPTERIS CARTHUSIANA (DRYOPTERIDACEAE) COMPLEX: A TAXONOMIC PUZZLE RESOLVED USING GENOME SIZE DATA 1

Libor Ekrt 2,5 , Renata Holubov á 3 , Pavel Tr á vn íý ek 3,4 , and Jan Suda 3,4

2 Department of Botany, Faculty of Science, University of South Bohemia, Brani š ovsk á 31, CZ-370 05 ý esk é Budě jovice, Czech Republic; 3 Department of Botany, Faculty of Science, Charles University in Prague, Ben á tsk á 2, CZ-128 01 Prague, Czech Republic; and 4 Institute of Botany, Academy of Sciences of the Czech Republic, Prů honice 1, CZ-252 43 Prů honice, Czech Republic

• Premise of the study : Genome duplication and interspecifi c hybridization are important evolutionary processes that signifi - cantly infl uence phenotypic variation, ecological behavior, and reproductive biology of land plants. These processes played a major role in the evolution of the Dryopteris carthusiana complex. This taxonomically intricate group composed of one diploid ( D. expansa ) and two allotetraploid (D. carthusiana a n d D. dilatata ) species in Central Europe. Overall phenotypic similarity, great plasticity, and the incidence of interspecifi c hybrids have led to a continuous dispute concerning species circumscription and delimitation. • Methods : We used fl ow cytometry and multivariate morphometrics to assess the level of phenotypic variation and the frequency of hybridization in a representative set covering all recognized species and hybrids. • Key results : Flow cytometric measurements revealed unique genome sizes in all species and hybrids, allowing their easy and reliable identifi cation for subsequent morphometric analyses. Different species often formed mixed populations, providing the opportunity for interspecifi c hybridization. Different frequencies of particular hybrid combinations depended primarily on evolutionary relationships, reproductive biology, and co-occurrence of progenitors. • Conclusions : Our study shows that genome size is a powerful marker for taxonomic decisions about the D. carthusiana complex and that genome size data may help to resolve taxonomic complexities in this important component of the temperate fern fl ora.

Key words: Central Europe; Dryopteris carthusiana ; ferns; ﬂ ow cytometry; interspeciﬁ c hybridization; mixed populations; multivariate morphometrics; nuclear DNA content; polyploidy; Pteridophyta; taxonomy.

Hybridization has long been recognized as a prominent force varies by taxon and location ( Ellstrand et al., 1996 ). From a in plant speciation, with up to 70% of extant plants being taxonomic point, species prone to interspecifi c hybridization descendants of interspecifi c hybrids ( Whitham et al., 1991 ). often pose serious problems because gene fl ow may blur spe- Hybridization can have several important evolutionary cies boundaries and affect levels of variability in natural popu- consequences, including increased genetic diversity, the origin lations. Considering the ploidy level of putative parents, of new ecotypes and taxa, and the reinforcement or breakdown homoploid and heteroploid hybridization can be distinguished of reproductive barriers ( Rieseberg, 1997 ). However, interspe- ( Grant, 1971 ; Rieseberg, 1997 ). While heteroploid hybridiza- cifi c crossing may not always be a source of genetic and func- tion is usually quite easy to detect because of the intermediate tional novelty, because resulting hybrids can suffer from number of chromosomes (e.g., using conventional karyological reduced fi tness and thus merely represent evolutionary dead techniques), recognition of homoploid crosses is a more chal- ends ( Seehausen, 2004 ). Similarly, the detrimental effect of hy- lenging task ( Rieseberg and Carney, 1998 ; Kron et al., 2007 ). bridization on the genetic integrity of rare species has been re- Nevertheless, the last decade has seen signifi cant advances in peatedly documented ( Ellstrand, 1992 ; Levin et al., 1996 ). In the understanding of patterns and dynamics of hybrid specia- general, interspecifi c hybridization plays a central role in the tion, catalyzed mainly by the advent of novel molecular ap- evolutionary history of many plant species, although its impact proaches ( Hegarty and Hiscock, 2005 ; Uyeda and Kephart, 2006 ). One of the taxonomically intricate plant groups that have 1 Manuscript received 11 July 2009; revision accepted 17 May 2010. been signifi cantly affected by interspecifi c hybridization (often The authors thank S. Jessen for several plant samples (including D. coupled with genome duplication) is the woodfern (Dryopteris u sarvelae ) and V. Jarol í mov á for karyological analyses of D. expansa . The Adans., Dryopteridaceae, Pteridophyta) ( Manton, 1950 ). In authors are also grateful to K. Kub á t, E. Ekrtov á , A. H á jek, M. Lep š í , and temperate regions, Dryopteris belongs among the most hybrid- J. Ku þ era for help during fi eldwork and to J. Rauchov á for assistance with prone genera, with almost every combination of species pairs FCM analyses. This study was supported by the Ministry of Education, Youth and Sports of the Czech Republic (project no. MSM 0021620828) and recorded ( Wagner, 1971 ; Fraser-Jenkins, 1982 ; Dost á l et al., the Academy of Sciences of the Czech Republic (no. AV0Z60050516). 1984 ; Frey et al., 1995 ; Krause, 1998 ) as well as one interge- 5 Author for correspondence (e-mail: [email protected]) neric hybrid ( Wagner et al., 1992 ). A paradigmatic example in the Central European fl ora is the Dryopteris carthusiana (for- doi:10.3732/ajb.0900206 merly D. spinulosa ) complex. This conspicuous component of

temperate woodlands consists of one diploid (2 n = 2 x = 82) spe- heteroploid hybridization ( Manton, 1950 ), while studies of cies D. expansa (C. Presl) Fraser-Jenk. et Jermy and two tetra- chromosome pairing during meiosis in both natural and artifi - ploids (2 n = 4x = 164), D. carthusiana (Vill.) H. P. Fuchs and cial hybrids shed some light on species relationships ( Manton D. dilatata (Hoffm.) A. Gray. The latter (allo)tetraploid has and Walker, 1953 ; Walker, 1955 , 1961 ). In recent years, an- been hypothesized to be derived from a cross between D. ex- other cytogenetic character, genome size, has become accessi- pansa var. alpina (Moore) Viane and D. intermedia subsp. ma- ble ( Leitch and Bennett, 2007 ). The knowledge that genome derensis (Alston) Fraser-Jenk. ( Gibby and Walker, 1977 ; Viane, size is usually constant within the same taxonomic entity 1986 ; Krause, 1998 ), while D. carthusiana is thought to com- (Greil huber, 2005 ) but may vary tremendously even among bine genomes of D. intermedia (Muhlenberg ex Willdenow) A. closely related taxa ( Bennett and Leitch, 2005 ) provides a foun- Gray subsp. intermedia from North America and an undiscov- dation for employing genome size as an important taxonomic ered diploid that has been variously identifi ed as D. “ semicris- marker. Indeed, this character has proven successful in resolv- tata ” ( Wagner, 1971 ; Werth and Lellinger, 1992 ) or D. ing complex low-level taxonomies, delimiting species boundar- “ stanley-walkeri ” ( Fraser-Jenkins, 2001 ). The unknown diploid ies, and revealing cryptic taxa ( Dimitrova et al., 1999 ; Mahelka has been characterized using isozymes ( Werth, 1989 ) and DNA et al., 2005 ; Kron et al., 2007 ; Suda et al., 2007 ). Despite its ( Hutton, 1992 ), and its morphology has been extrapolated by potential taxonomic value, genome size has largely been ne- comparing its extant allotetraploid derivatives ( Werth and glected in the biosystematics of ferns in general (but see Bure š Kuhn, 1989 ; Fraser-Jenkins, 2001 ). The species are classifi ed in et al., 2003 ; Ebihara et al., 2005 ; Ekrt and Š tech, 2008 ; Ekrt the subgenus Dryopteris , section Lophodium (Newman) C. et al., 2009 ), and in the D. carthusiana group in particular. The Chr. ex H. It ô ( Fraser-Jenkins, 1986 ), which includes ferns with Plant DNA C-values database ( Bennett and Leitch, 2005 ) con- usually thick, glossy, and bicolor stipe-base scales, narrow tains only four estimates for the whole genus Dryopteris , in- lobes of ultimate segments, terminating in long, hair-tipped cluding the value for D. dilatata (1C = 8.05 pg), as determined aristate teeth, and unique spore morphology with minutely by Feulgen microdensitometry. spinulose surface on the perispore. Formerly, D. cristata (L.) A. In this study, we used relative genome size data together with Gray was also a part of the D. carthusiana group (e.g., Walker, analysis of multivariate morphometric data to obtain new in- 1955 , 1961 ; Wid é n et al., 1967 ). However, more recent treat- sights into phenotypic variation and the frequency of interspe- ments have placed D. cristata into the separate section Pandae cifi c hybridization in the D. carthusiana group in Central Fraser-Jenk. ( Fraser-Jenkins, 1986 ). Europe. Our investigation was inspired by the demonstration The characteristics of petiole scales and pinnule margins, that genome size is usually stable within the same taxonomic shape and color of the frond, and the presence and density of entity ( Greilhuber, 2005 ), while it often varies between differ- glandular hairs are some of the diagnostic characters used for ent taxa ( Bennett and Leitch, 2005 ). Consequently, genome species recognition ( Dost á l et al., 1984 ; Fraser-Jenkins, 1993 ; size can be employed as a useful marker for resolving taxo- Frey et al., 1995 ). However, overall phenotypic similarity, a nomic complexities, circumscribing species, and unveiling high degree of phenotypic plasticity, and interspecifi c hybrid- cryptic diversity ( Kron et al., 2007 ). An added benefi t of ge- ization often make unambiguous species determination diffi - nome size data is that they can also be used to differentiate be- cult. An important fi nding for the taxonomy of the group was tween parental species and their hybrids, provided there are the discovery that interspecifi c hybrids are sterile, producing suffi cient differences in the amount of nuclear DNA. This ap- only aborted, irregularly shaped spores ( Manton, 1950 ; Wagner proach can be successfully applied even in groups with the and Chen, 1965 ; Wid é n et al., 1967 ; Fraser-Jenkins, 1977 ; Jessen same number of chromosomes, when traditional karyological and Rasbach, 1987 ; Leonhards et al., 1990 ). All three possible treatments would be in vain ( Loureiro et al., 2010 ). hybrid combinations within the D. carthusiana group have been Using fl ow cytometry and multivariate morphometrics, we collected in the fi eld in the past, though with markedly different addressed three main questions: (1) What is the level of genome abundances. Records exist for D. u ambroseae Fraser-Jenk. size variation in the group? Can genome size be used as an in- et Jermy (= D. dilatata u D. expansa ) and D. u deweveri (Jansen) formative marker for taxonomic decisions? (2) Which pheno- Jansen et Wachter (= D. carthusiana u D. dilatata ) from west- typic traits are correlated with different genome size categories? ern and northern Europe ( D ö pp and G ä tzi, 1964 ; Wid é n et al., What are the species- and hybrid-specifi c morphological char- 1967; Benl and Eschelmü ller, 1970; Pi ę k o ś , 1974 ; Fraser- acters? (3) What is the abundance and distribution of particular Jenkins, 1977 ; Leonhards et al., 1990 ); more recently, these species and hybrids in the area studied? hybrids have also been repeatedly confi rmed in Central and Eastern Europe ( Holubov á , 2006 ; Ivanova, 2006 ; Ekrt and P ů bal, 2008 ). Conversely, Dryopteris u sarvelae Fraser-Jenk. et MATERIALS AND METHODS Jermy (= D. carthusiana u D. expansa ) is much rarer; occasional reports of its occurrence come only from Finland ( Wid é n Plant material — Plants were collected in the fi eld between 2003 and 2007 et al., 1967 ; Sorsa and Wid é n, 1968 ), Scotland ( Corley and from 85 localities ( Fig. 1 ; see Appendix 1 for details), with 76 of them in the Czech Republic. Additional samples originated from Germany (fi ve locali- Gibby, 1981 ), and the island of R ü gen in Germany ( Jessen and ties), Slovakia (three localities), and Austria (one locality). Special attention Rasbach, 1987 ). This hybrid was synthesized experimentally was paid to localities where more species grew in sympatry (to assess the ( Walker, 1955 ) before it was documented from the wild. frequency of interspecifi c hybridization) and localities where D. expansa was In addition to the evaluation of morphological variation, the expected (i.e., a comparatively rare member of the complex). Plants were col- group has also been subjected to several cytological and chemot- lected randomly in an area of approximately 20 u 20 m. In localities with axonomic analyses ( Tryon and Britton, 1966 ; Wid é n et al., multiple samples, we attempted to collect the whole morphological variation present at the locality. The number of samples per locality varied from one to 1967, 1970 ; Sorsa and Wid é n, 1968 ; Wid é n and Sorsa, 1968 ; 81, refl ecting both the population structure (uniform vs. mixed populations) Britton, 1972 ; Gibby and Walker, 1977 ; Pię ko ś -Mirkowa, 1979 , and locality size; at least fi ve plants were collected at 34 localities. Collec- 1987 , 1993 ; Benl and Eschelm ü ller, 1983 ; Gibby, 1983 ). Con- tively, 858 plants were available, involving all species and hybrid combina- ventional chromosome counts provided robust evidence for tions (D. carthusiana : 2 3 7 s a m p l e s ; D. dilatata : 2 4 4 s a m p l e s ; D. expansa : 1210 American Journal of Botany [Vol. 97

Fig. 1. Geographic distribution of 83 studied populations from the Dryopteris carthusiana group in Central Europe. AT = Austria, CZ = Czech Repub- lic, DE = Germany, SK = Slovakia. Two additional localities from R ü gen Island, Germany are not shown.

277 samples; D. u a m b r o s e a e : 8 2 s a m p l e s ; D. u d e w e v e r i : – 1 6 s a m p l e s ; D. 39 quantitative characters (21 primary characters and 18 ratios; Appendix S1, see u s a r v e l a e : 2 samples). Herbarium vouchers are kept in PR, PRC, HR and CB Supplemental Data at http://www.amjbot.org/cgi/content/full/ajb.0900206/DC1) (see Appendix 1). Plants from Germany were deposited in the private her- were measured and scored on dry herbarium vouchers. In addition, spore color barium of S. Jessen (Chemnitz). was also recorded. All characters previously used for taxa recognition (e.g., Pi ę ko ś -Mirkowa, 1979 ; Chrtek, 1988 ; Ekrt, 2000 ; Kub á t et al., 2002 ) were in- Flow cytometry — Relative genome sizes were estimated by fl ow cytometry cluded, as well as additional potentially informative traits chosen on the basis (FCM) using the two-step methodology originally developed by Otto (1990) . of our own observations. The protocol generally followed Dole ž el et al. (2007) . Briefl y, approximately Morphometric data were analyzed with the SAS 8.1 statistical package 4.5 cm 2 of fresh pinna tissue, devoid of sori, from the analyzed sample was (SAS Institute, Cary, NC, USA) using the procedures CANDISC, CORR, DIS- chopped with leaf tissue of the internal standard, using a sharp razor blade, in a CRIM, PRINCOMP and UNIVARIATE (see Rosenbaumov á et al., 2004 for Petri dish containing 0.5 ml of ice-cold Otto I buffer (0.1 M citric acid, 0.5% details). In discriminant analyses, individual plants were selected as operational Tween 20). The suspension was fi ltered through a nylon mesh (42 P m) and in- taxonomic units (OTUs), and genome size with spore fertility (v21) defi ned cubated for 20 min at room temperature (20q C), with occasional shaking. Sam- particular taxonomic groups (species and hybrids). Because the distribution within groups was not multivariate normal, the nonparametric k -nearest-neigh- ples were stained with 1 ml of Otto II buffer (0.4 M Na2 HPO4 · 12 H2 O) supplemented with AT-selective fl uorochrome DAPI (at fi nal concentration of bor method was employed ( Klecka 1980 ). Discriminant power was determined 4 P g/mL) and E -mercaptoethanol (fi nal concentration of 2 P L/mL). The stain- by cross-validation. Various modifi cations of discriminant analyses (all species ing lasted 1 – 2 min at room temperature, after which the relative fl uorescence and hybrids together / parental species only / hybrids only / parental species intensity of 3000 – 5000 particles was recorded on a Partec PA II fl ow cytometer with a corresponding hybrid) were performed. (Partec GmbH, M ü nster, Germany) equipped with a mercury-arc lamp for UV Differences in genome size values (relative fl uorescence intensities) were excitation. Vicia faba ‘ Inovec ’ (2C = 26.90 pg; Dole ž el et al., 1992 ) was se- tested with a general linear model (procedure GLM) due to unbalanced data lected as an appropriate primary internal reference standard (with genome size design (i.e., different sample size in different OTUs). close to but not overlapping that of most samples). A secondary reference standard (Pisum sativum ‘ Ctirad ’ ; 2C = 9.09 pg, Dole ž el et al., 1998 ) was used in analyses of D. u sarvelae because genome size of this hybrid nearly overlapped with that of the primary standard. Resulting values were recalculated to the RESULTS fl uorescence intensity of Vicia faba (using a mean Vicia to Pisum ratio of 3.14, based on eight replications on different days). A karyologically confi rmed plant Genome size variation — Flow cytometric analyses of 858 of D. expansa ( n = 41 II ; counted by V. Jarol í mov á , Prů honice, Czech Republic) Dryopteris plants yielded high-resolution histograms with dis- from the Czech Republic, Kosteln í Myslov á , Velk ý Huli š ť sk ý pond (loc. 40) tinct peaks of the sample and the internal reference standard, was used to interpret FCM results. and with little background signal ( Fig. 2 ). The average coeffi - cients of variation of the G0/G1 peaks of the analyzed sample Multivariate morphometrics — A subset of 609 plants, representing all spe- and internal reference standard were 2.67% (range 1.37 – 5.89) cies and hybrids except D. u sarvelae , was subjected to morphometric analyses ( D. carthusiana : 209; D. dilatata : 188; D. expansa : 163; D. u ambroseae : 37; D. and 2.36% (range 1.18 – 4.44), respectively. Table 1 summarizes u deweveri : 12; see Appendix 1). All the plants were classifi ed to a particular relative fl uorescence intensities for all species and hybrids in species or hybrid on the basis of relative genome size as determined by fl ow the D. carthusiana group. The estimated values differed signifi - cytometry. Only individuals with well-developed sori were considered. A total of cantly (GLM procedure; F 5,852 = 15914, P < 0.0001) and were July 2010] Ekrt et al. — Taxonomy of DRYOPTERIS CARTHUSIANA complex 1211

Fig. 2. Representative fl ow cytometric histograms for the estimation of relative genome sizes in species and hybrids from the Dryopteris carthusiana group. Analyses of (A) D. carthusiana , (B) D. dilatata , (C) D. expansa , (D) D. u deweveri , and (E) D. u ambroseae together with Vicia faba as an internal reference standard (marked by an asterisk). (F) Simultaneous analysis of species and hybrids with the most similar genome sizes: D. u ambroseae and D. carthusiana (labeled as ExD and C, respectively). Simultaneous analyses of parental taxa and their putative hybrids: (G) D. carthusiana , D. u deweveri and D. dilatata (labeled as C, H and D, respectively) and (H) D. expansa , D. u ambroseae and D. dilatata (labeled as E, H and D, respectively). Nuclei from all samples were simultaneously isolated, stained with DAPI, and analyzed. specifi c for species and hybrids. In fact, relative fl uorescence sured by FCM) and theoretical (= calculated from the data of intensities for particular species and hybrids were mostly non- putative parents) relative genome size values of hybrid plants; overlapping; the only exception was D. carthusiana and D. the difference never exceeded 2.5%. u ambroseae , with an average difference of about 4.7%. Never- The knowledge of genome composition of individual species theless, the overlap among their variation intervals was not sig- and hybrids ( Table 1 ) allowed us to compare the relative size of nifi cant (Mann – Whitney U test; Z = 9.64, U = 145, df = 1, P < parental genomes. Setting the genome of D. expansa (genome 0.0001). There was a good congruence between actual (= mea- E) to a unit value, genomes of D. intermedia (genome I) and 1212 American Journal of Botany [Vol. 97

Table 1. Relative ﬂ uorescence intensities of 858 individuals corresponding to three parental species and their hybrids from the Dryopteris carthusiana group. Isolated nuclei were stained with DAPI, and Vicia faba ‘ Inovec ’ (2C = 26.90 pg) was set as the unit value.

Relative ﬂ uorescence intensity Taxon Genome composition a DNA ploidy level (mean r SD) No. of samples

D. expansa EE 2x 0.815 r 0.020 277 D. carthusiana IISS 4x 1.163 r 0.014 237 D. dilatata IIEE 4x 1.408 r 0.019 244 D. u sarvelae (= D. carthusiana u D. expansa ) ISE 3x 0.980 r 0.015 2 D. u ambroseae (= D. dilatata u D. expansa ) IEE 3x 1.111 r 0.022 82 D. u deweveri (= D. carthusiana u D. dilatata ) IIES 4x 1.254 r 0.028 16 a Genome composition: E = D. expansa , I = D. intermedia , S = D. “ stanley-walkeri ” / “semicristata ”

D. “ stanley-walkeri ” / “ semicristata ” (genome S) equal to 0.728 parental species and two hybrids) revealed three distinct and 0.699, respectively. clusters ( Fig. 4 ): (1) plants corresponding to D. carthusiana , (2) plants corresponding to D. expansa + D. dilatata , and (3) Species abundance and hybrid frequency — All three species hybrid individuals: D. u ambroseae + D. u deweveri . Despite a from the D. carthusiana complex together with all putative hy- high number of analyzed groups (fi ve different OTUs), a very brid combinations were recorded from various locations within high rate of correctly classifi ed objects was achieved (nearly the areas studied. Dryopteris carthusiana a n d D. dilatata were 96%; Table 2 ). Because hybrid plants were unambiguously de- the most common components of deciduous broad-leaved, mixed, limited based on aborted spores, separate analyses of parental and coniferous forests from lowlands to mountain regions (see species and hybrids were performed in the next step. Appendix 1). While D. dilatata has no special requirement for Discriminant analysis of three parental species corresponded moisture, D. carthusiana occurs more frequently in moist habi- well with the results of the whole data set. Once again, D. carthusi- tats and swamps. Diploid D. expansa was also quite common (ca. ana formed an isolated cluster (all individuals correctly classifi ed), 66% of localities), mainly in higher altitudes and/or at various while about 5% of individuals corresponding to D. dilatata a n d D. humid stands such as bottoms of deep valleys or alder forests. expansa were not assigned to the correct group. The characters Hybrid plants were collected from 39 of the 85 localities most tightly correlated with the fi rst canonical axis (presented ac- (46%) and accounted for 11.7% of all samples (100 individu- cording to a decreasing discrimination power) and thus separating als). The frequency of particular hybrid combinations, however, D. carthusiana from the other two species were v33 (length of the differed markedly. The most common hybrid was D. u ambro- dark central stripe/total length of the petiole scale), v18 (length of seae (= D. dilatata u D. expansa ), only distantly followed by the dark central stripe on the scale), v16 (length of the petiole scale), D. u deweveri (= D. carthusiana u D. dilatata ) and the rarest, v19 (glandularity of the rachis), v2 (width of the leaf lamina), v17 D. u sarvelae (= D. carthusiana u D. expansa ). Disregarding (scale width), and v6 (length of the lowermost pinna). localities with fewer than fi ve sampled individuals, D. u ambro- The second canonical axis discriminated between the pheno- seae was recorded in 72% of localities with the presence of both typically similar D. dilatata and D. expansa . The following parents (see Fig. 3 ). The homoploid hybrid D. u deweveri occurred at about 5% of localities inhabited by both parents ( Fig. 3 ), while D. u sarvelae was not found in 14 mixed D. carthusiana – D. expansa localities (this rare hybrid was only confi rmed in two previously known sites; Jessen and Rasbach, 1987 ). Sympatric occurrence of more species and/or hybrids was a very common phenomenon. Considering localities with multiple fern samples ( t 5 sampled plants), only two of 34 localities contained a single species (see Appendix 1). The taxonomic composition in other localities was as follows: two species/ hybrids, 12 localities; three species/hybrids, 9 localities; four species/hybrids, 10 localities; fi ve species/hybrids, 1 locality. The most salient example was population no. 40 (Czech Repub- lic, ý eskomoravsk á vrchovina, Kosteln í Myslov á ), where all taxa, except for one hybrid (D. u sarvelae ), were recorded.

Morphometric analyses — Phenetic relationships among individual Dryopteris plants were visualized by principal component analysis (PCA). Particular species or hybrids usually showed high variation and formed loose, more or less overlapping clusters (results not shown). Discriminant analyses were employed to select a set of characters that allowed the best separation of groups of objects (i.e., Fig. 3. The occurrence of hybrid individuals in localities with sympat- species and hybrids characterized by relative genome sizes) de- ric growth of parental species. Only localities with at least fi ve sampled fi ned a priori and to determine the proportion of correctly clas- plants were considered. C = Dryopteris carthusiana , D = D. dilatata , E = sifi ed plants. Discriminant analysis of all OTUs (i.e., three D. expansa . July 2010] Ekrt et al. — Taxonomy of DRYOPTERIS CARTHUSIANA complex 1213

Fig. 4. Canonical discriminant analysis of 609 individuals representing three parental species and two hybrids from the Dryopteris carthusiana group. Cube = D. carthusiana , heart = D. dilatata , cross = D. expansa , pyramid = D. u deweveri , club = D. u ambroseae .

characters contributed most to this division: v17 (scale width), biology because of their impacts on the maintenance of species v28 (length of the basal basiscopic pinnula of the lowermost distinctions, enhancements of genetic diversity, and diversifi ca- pinna/length of the lowermost pinna), v32 (length/width of the tion ( Hersch-Green and Cronn, 2009 ). The present work aimed petiole scale), v33 (length of the dark central stripe/total length to evaluate phenotypic variation and assess the frequency of of the petiole scale), v31 (length of the basal basiscopic pinnula interspecifi c hybridization in a taxonomically challenging poly- of the lowermost pinna/length of the basal acroscopic pinnula ploid fern group, the Dryopteris carthusiana complex. Flow of the lowermost pinna), v27 (length of the basal basiscopic cytometric analyses of a representative set of samples from pinnula of the lowermost pinna/lamina length), and v7 (length Central Europe revealed large differences in relative genome of the basal basiscopic pinnula of the lowermost pinna). size among the three species constituting this group. Average Hybrid plants ( D. u ambroseae and D. u deweveri ) can be suc- fl uorescence intensities of two tetraploid species (D. carthusi- cessfully discriminated by scale characters (v18 [length of the ana and D. dilatata ) differed by 21%, while diploid D. expansa dark central stripe], v33 [ratio between the length of the dark had a 43% smaller genome than D. carthusiana (the tetraploid central stripe and the total length of the scale], and v16 [length with the smaller amount of nuclear DNA). Such differences in of the scale]), glandularity of the rachis (v19), and length/width genome size not only allowed unambiguous determination of of the leaf lamina (v22). the species, but also reliable detection of interspecifi c hybrids Appendix S2 (see online Supplemental Data) summarizes of any parental combination. The closest fl uorescence values descriptive statistics of 13 taxonomically most informative were found between D. carthusiana and D. u ambroseae (aver- characters for the three parental species and two hybrids from age difference 4.7%) and between D. carthusiana and D. u dew- the D. carthusiana group. everi (average difference 7.8%). Nevertheless, such plants can be easily discriminated by observing spore quality (well developed in D. carthusiana but aborted in hybrids). Collectively, DISCUSSION the data obtained on genome size variation show that fl ow cytometry provides a powerful tool for reliable identifi cation of Genome size variation and its value for taxonomic deci- species and hybrids in the D. carthusiana complex. In this fern sions — Interspecifi c hybridization and polyploidization are group, neither conventional phenotype-based taxonomy nor processes of central importance in population and evolutionary conventional karyology can be easily used to recognize the

Table 2. Results of classiﬁ catory discriminant analysis of 609 individuals corresponding to three parental species and two hybrids from the Dryopteris carthusiana group.

Predicted group membership (number of observations / percentage classiﬁ ed into groups) Actual group membership D E C D u E C u D

D 180 / 95.7 8 / 4.3 — — — E 15 / 9.2 148 / 90.8 — — — C — — 209 / 100.0 — — D u E — — — 37 / 100.0 — C u D — — — 1 / 8.3 11 / 91.7 Notes: C = D. carthusiana , D = D. dilatata , E = D. expansa , D u E = D. u ambroseae , C u D = D. u deweveri 1214 American Journal of Botany [Vol. 97 species and hybrids because of the considerable phenotypic polyploids are predominantly selfi ng ( Soltis and Soltis, 1987 , plasticity and the diffi culties inherent in counting high numbers 1990 ; Soltis et al., 1988 ; Masayuma and Watano, 1990 ; Xiang of chromosomes, respectively. et al., 2000 ; Flinn, 2006 ). Therefore, the probably selfi ng breeding system in tetraploid D. carthusiana and D. dilatata ( Xiang Frequency of interspecifi c hybridization — Signifi cant dif- et al., 2000 ; Flinn, 2006 ) likely reduces opportunities for inter- ferences in relative genome size between different species al- specifi c hybridization, which is consistent with the lower num- lowed a reliable assessment of the frequency of interspecifi c ber of crosses (D. u deweveri ) discovered in the current study. hybridization. Dryopteris is considered one of the most hybrid- prone fern genera in temperate regions, as evidenced by 21 dif- Species and hybrid delimitation — A clear delimitation of ferent hybrid combinations reported from Europe ( Dost á l et al., species and hybrids in our study opened the way for a critical 1984 ) and 25 from North America ( Montgomery and Wagner, assessment of phenotypic variation and allowed us to select a 1993 ). However, the putative hybrids were almost exclusively set of species- and hybrid-specifi c morphological characters. identifi ed based on morphological characters while more robust Unlike previous studies, which either exclusively used subjective data (e.g., karyological and/or molecular) supporting their exis- morphological criteria for species recognition (e.g., Nannfeldt, tence were often lacking. In our study, we detected hybrid indi- 1966 ; Wid é n et al., 1967 ; Benl and Eschelm ü ller, 1983 ) or viduals in nearly half of the localities (39 of 85) subjected to karyologically examined only a few individuals (e.g., Wid é n FCM investigation. This is certainly a much higher incidence of et al., 1970 ; Nardi, 1976 ; Pi ę ko ś -Mirkowa, 1979 ; Leonhards interspecifi c hybridization than would be expected based on et al., 1990 ), our approach involved unambiguous assignment previous works. For example, some studies of the D. carthusi- of all samples to a particular species or hybrid based on the rela- ana group performed in different European countries have not tive genome size. Therefore, we consider the present morpho- reported any crosses (e.g., Wid é n and Sorsa, 1968 ; Pię ko ś - metric results and taxonomic conclusions to be highly robust Mirkowa, 1979 ; Seifert and Holderegger, 1995 ), while others and reliable. The results of discriminant analyses showed that found only a single hybrid individual (e.g., D ö pp and G ä tzi, D. carthusiana is phenotypically a well-defi ned species that 1964 ; Benl and Eschelm ü ller, 1970 ; Fraser-Jenkins, 1977 ; possesses several unique characters and whose determination is Leonhards et al., 1990 ). Triploid Dryopteris u ambroseae ( = D. usually clear. On the other hand, differentiation between D. ex- dilatata u D. expansa ) was the most common hybrid combina- pansa and D. dilatata is much more challenging, and some in- tion, detected in 32 localities (82 individuals). Despite the iden- dividuals will likely be misclassifi ed even if a set of characters tical number of chromosomes in putative parental species, with the highest discrimination power is used. The taxonomi- D. u deweveri (= D. carthusiana u D. dilatata ) occurred much cally most informative characters in the D. carthusiana group less frequently, recorded from only nine localities (16 individ- include scale characteristics (length and width of the scale and uals altogether). Nevertheless, D. u deweveri is likely distrib- the length of the dark central stripe), size of the leaf lamina to- uted throughout Europe as indicated by reports of karyologically gether with the size and relative proportions of certain lamina parts verifi ed plants ( Walker, 1955 ; Sorsa and Wid é n, 1968 ; Wid é n (e.g., characteristics of the lowermost pinna, the basal basiscopic et al., 1970 ; Leonhards et al., 1990 ). The last hybrid combina- pinnula of the lowermost pinna), and the glandularity of the rachis. tion D. u sarvelae (= D. carthusiana u D. expansa ) is extremely Despite their high degree of overall morphological similarity with rare. Only two individuals from two previously known locali- putative parental species, hybrid ferns can be easily recognized ties ( Jessen and Rasbach, 1987) were encountered. by their aborted spores. Therefore, only fertile specimens allow Marked differences in the frequency of particular hybrid reliable determination. All Central European members of the combinations can be explained by the presence or absence of D. carthusiana groups (except the very rare hybrid D. u s a r v e l a e ) common genomic elements by the putative parental species. can be determined according to the following key: Based on evidence of genome homology between D. expansa and D. dilatata ( D. expansa is hypothesized as one of the pro- 1a Spores abortive 2 genitors of the allotetraploid D. dilatata ; Gibby and Walker, 1b Spores well-developed 3 1977 ; Viane 1986 ; Krause 1998 ; see also Table 1 ), extensive 2a Rachis sparsely glandular; petiole scales (8 – )11 – 13( – 16) interspecifi c hybridization appears reasonable. An analogous mm long, with distinct, (6.5 – )8.5 – 11( – 13.5) mm long dark cen- situation is known in North America, where allotetraploid D. tral stripe, usually extending to the upper half of the scale (oc- carthusiana hybridizes frequently with one of its diploid ances- casionally up to the apex); lamina (0.9 – )1.5 – 1.8( – 1.9) mes tors, D. intermedia , resulting in the common and widespread longer than wide hybrid D. u triploidea Wherry ( Tryon and Britton, 1966 ; Xiang D . u ambroseae Fraser-Jenk. et Jermy et al., 2000 ). Dryopteris carthusiana and D. dilatata also share (= D . dilatata u D . expansa ) the D. intermedia genome ( Table 1 ), which may facilitate their interspecifi c hybridization. On the contrary, the lack of genome 2b Rachis densely glandular; petiole scales (6 – )8 – 10.5( – 13.5) identity between D. carthusiana and D. expansa may reduce mm long, concolorous or with (2 – )3 – 4( – 6.5) mm long, dark, the likelihood of successful hybridization between them. A central stripe, typically terminating in the lower half of the congruence between the frequency of interspecifi c hybridiza- scale; lamina (1.6 – )1.7 – 2.1( – 2.4) times longer than wide tion and species relationship has been documented in several D . u deweveri (Jansen) Jansen et Wachter plant groups, both fl owering ( Mosseler, 1990 ) and spore-bear- (= D . carthusiana u D . dilatata ) ing ( Yatabe et al., 2009 ). Another, nonexclusive explanation of differences in the fre- 3a Petiole scales (4 – )5 – 6( – 8.5) mm long, concolorous; rachis quency of interspecifi c hybrids, takes into account breeding glabrous or very scarcely glandular (1 – 6 glands per rachis); systems and associated sex expression in gametophytes of dip- lamina (10 – )15 – 20( – 35) cm wide, the lowermost pinna (4.5 – ) loid vs. polyploid Dryopteris species. While the majority of 7 – 10( – 17.5) cm long diploids possess an outcrossing (or mixed) mating system, D. carthusiana (Vill.) H. P. Fuchs July 2010] Ekrt et al. — Taxonomy of DRYOPTERIS CARTHUSIANA complex 1215

3b Petiole scales (6.5 – )11 – 14( – 17) mm long, with distinct Ekrt , L. 2000 . Komplex Dryopteris carthusiana agg. na Š umav ě a v dark central stripe; rachis densely glandular; lamina (14 – )22 – P ř ed š umav í . Bc. thesis, ý esk é Budě jovice, University of South 30( – 41) cm wide, the lowermost pinna (7 – )10.5 – 15( – 21) cm Bohemia, Czech Republic. long 4 Ekrt , L. , and D. P Ů bal . 2008 . Novinky v kv ě ten ě c é vnat ý ch rostlin þ esk é 4a Petiole scales (1.5 – )3 – 4( – 12) mm wide, (1.0 – )3.2 – 4.2( – Š umavy a p ř il é haj í c í ho P ř ed š umav í . I. Silva Gabreta 14 : 19 – 38 . Ekrt , L. , and M. Š tech . 2008 . A morphometric study and revision of 9.3) times longer than wide, with dark, central stripe often ex- the Asplenium trichomanes group in the Czech Republic. Preslia 80 : tending up to the apex, covering (65 – )90 – 100% of the scale 325 – 347 . length; basal basiscopic pinnula of the lowermost pinna (2 –)3.5 – Ekrt , L. , P. Tr á vn í ý ek , V. Jarol í mov á , P. V í t , and T. Urfus . 2009 . 5.5( – 9) cm long, its ratio to the lowermost pinna (0.15 – )0.35 – Genome size and morphology of the Dryopteris affi nis group in 0.45( – 0.70); spores dark brown Central Europe. Preslia 81 : 261 – 280 . D. dilatata (Hoffm.) A. Gray Ellstrand , N. C. 1992 . Gene fl ow by pollen — Implications for plant conservation genetics. Oikos 63 : 77 – 86 . 4 b P e t i o l e s c a l e s ( 2 . 5 – ) 4 . 5 – 5 . 5 ( – 7 . 5 ) m m w i d e , ( 1 . 9 – ) 2 . 4 – Ellstrand , N. C. , R. Whitkus , and L. H. Rieseberg . 1996 . Distribution 2.8( – 4.7) times longer than wide, the dark central stripe usually of spontaneous hybrids. Proceedings of the National Academy of not extending up to the apex, covering (40 – )70 – 90( – 100)% of Sciences, USA 93 : 5090 – 5093 . the scale length; basal basiscopic pinnula of the lowermost pinna Flinn , K. M. 2 0 0 6. R e p r o d u c t i v e b i o l o g y o f t h r e e f e r n s p e c i e s m a y c o n - tribute to differential colonization success in post-agricultural forests. (2.5– )5.5 – 7.5( – 10.5) cm long, its ratio to the lowermost pinna American Journal of Botany 93 : 1289 – 1294 . (0.25 – )0.45 – 0.55( – 0.65); spores yellowish to light brown Fraser-Jenkins , C. R. 1977 . Dryopteris u brathaica Fraser-Jenk. et D. expansa (C. Presl) Fraser-Jenk. et Jermy Reichstein hybr. nov., the putative hybrid of D. carthusiana u D. fi lix- mas. Fern Gazette 11 : 337 . Fraser-Jenkins , C . R . 1 9 8 2 . Dryopteris in Spain, Portugal and Macaronesia. LITERATURE CITED Boletim da Sociedade Broteriana S é ries 2 , 55: 175 – 336. Fraser-Jenkins , C. R. 1986 . A classifi cation of the genus Dryopteris Benl , G. , and A. Eschelm ü ller . 1970 . Dryopteris dilatata u assimi- (Pteridophyta: Dryopteridaceae). Bulletin of the British Museum lis in Bayern. Berichte der Bayerischen Botanischen Gesellschaft 42 : (Natural History) . Botany 14 : 183 – 218 . 185 – 188 . Fraser-Jenkins , C. R. 1993 . Dryopteris. In T. G. Tutin, N. A. Burges, Benl , G. , and A. Eschelm ü ller . 1983 . Zum Vorkommen weniger beka- A. O. Chater, et al., [eds.], Flora europaea, vol. 1, 2nd ed., 27 – 30. nnter Dryopteris -Sippen im bayerischen Alpen- und Voralpenraum. Cambridge University Press, Cambridge, UK. Berichte der Bayerischen Botanischen Gesellschaft 54 : 77 – 102 . Fraser-Jenkins , C. R. 2 0 0 1 . Dryopteris stanley-walkeri Fras.-Jenk., the Bennett , M. D. , and I. J. Leitch . 2005 . Nuclear DNA amounts in an- missing diploid common ancestor of D. carthusiana a n d D. cristata giosperms: Progress, problems and prospects. Annals of Botany 95 : (Pteridophyta: Dryopteridaceae). In P. C. Panda and S. S. Samant [eds.], 45 – 90 . Plant diversity of the Himalaya: Prof. Y. P. S. Pangtey commemoration Britton , D. M. 1972 . Spore ornamentation in the Dryopteris spinulosa volume, 119 – 152. Gyanodaya Prakashan, Nainital, India. complex. Canadian Journal of Botany 50 : 1617 – 1621 . Frey , W. , J. P. Frahm , E. Fischer , and W. Lobin . 1995 . Kleine Bure š , P. , L. Tich ý , Y. Wang , and J. Barto š . 2003 . Occurrence of Kryptogamenfl ora Band IV, Die Moss- und Farnpfl anzen Europas. Polypodium u mantoniae and new localities for P. interjectum in Gustav Fischer Verlag, Stuttgart. the Czech Republic confi rmed using fl ow cytometry. Preslia 75 : Gibby , M. 1983 . The Dryopteris dilatata complex in Macaronesia and the 293 – 310 . Iberian Peninsula. Acta Botanica Malacitana 8 : 59 – 72 . Chrtek , J. 1988 . Dryopteris Adanson, kapraď . In S. Hejn ý and B. Slav í k Gibby , M. , and S. Walker . 1977 . Further cytogenetic studies and a reap- [eds.], Kvě tena ý SR 1, 262 – 272. Academia, Praha, Czech Republic. praisal of the diploid ancestry in the Dryopteris carthusiana complex. Corley , H. V. , and M. Gibby . 1981 . Dryopteris u sarvelae in Scotland: A The Fern Gazette 11 : 315 – 324 . new hybrid for the British Isles. Fern Gazette 12 : 178 . Grant , V. 1971 . Plant speciation. Columbia University Press, New York, Dimitrova , D. , I. Ebert , J. Greilhuber , and S. Kozhuharov . 1999 . New York, USA. Karyotype constancy and genome size variation in Bulgarian Crepis Greilhuber , J. 2005 . Intraspecifi c variation in genome size in angio- foetida s. l. (Asteraceae). Plant Systematics and Evolution 217 : sperms: Identifying its existence. Annals of Botany 95 : 91 – 98 . 245 – 257 . Hegarty , M. J. , and S. J. Hiscock . 2005 . Hybrid speciation in Dole ž el , J. , J. Greilhuber , S. Lucretti , A. Meister , M. A. Lys á k , L. plants: New insights from molecular studies. New Phytologist 165 : Nardi , and R. Obermayer . 1 9 9 8 . P l a n t g e n o m e s i z e e s t i m a t i o n b y fl ow 411 – 423 . cytometry: Inter-laboratory comparison. Annals of Botany 82 : 17 – 26 . Hersch-Green , E. I. , and R. Cronn . 2009 . Tangled trios?: Characterizing Dole ž el , J. , J. Greilhuber , and J. Suda . 2007 . Estimation of nuclear a hybrid zone in Castilleja (Orobanchaceae). American Journal of DNA content in plants using fl ow cytometry. Nature Protocols 2 : Botany 96 : 1519 – 1531 . 2233 – 2244 . Holmgren , P. K. , N. H. Holmgren , and L. C. Barnett . 1990 . Index Dole ž el , J. , S. Sgorbati , and S. Lucretti . 1992 . Comparison of three herbariorum, part I: The herbaria of the world, 8th ed. New York DNA fl uorochromes for fl ow cytometric estimation of nuclear DNA Botanical Garden, Bronx, New York, USA. content in plants. Physiologia Plantarum 85 : 625 – 631 . Holubov á , R. 2006 . Cytotaxonomick á a morfologick á studie okruhu D ö pp , W. , and W. G ä tzi . 1964 . Der Bastard zwischen tetraploider und dip- Dryopteris dilatata v ý esk é republice s dů razem na ý esk é stř edohoř í. loider Dryopteris dilatata. Bericht der Schweizerischen Botanischen M.S. thesis, Charles University, Prague, Czech Republic. Gesellschaft 74 : 45 – 53 . Hutton , C. 1992 . The common ancestor of the allotetraploid ferns Dost á l , J. , C. R. Fraser-Jenkins , and T. Reichstein . 1984 . Dryopteris. Dryopteris carthusiana and Dryopteris cristata : A chloroplast In K. U. Kramer [ed.], Illustrierte Flora von Mitteleuropa. Band I, DNA analysis. M.S. thesis, Mount Holyoke College, South Hadley, Teil 1. Pteridophyta , 3. Aufl age, 136 – 169. Verlag Paul Parey, Berlin, Massachussetts, USA. Germany. Ivanova , D. 2006 . Dryopteris u ambroseae (Dryopteridaceae: Ebihara , A. , H. Ishikawa , S. Matsumoto , S. Lin , K. Iwatsuki , M. Pteridophyta) a hybrid new to Bulgaria. Phytologica Balcanica 12 : Takamiya , Y. Watano , and M. Ito . 2005 . Nuclear DNA, chloro- 351 – 356 . plast DNA, and ploidy analysis clarifi ed biological complexity of the Jessen , S. , and H. Rasbach . 1987 . Dryopteris u sarvelae , erster Vandenboschia radicans complex (Hymenophyllaceae) in Japan and Nachweis f ü r Mitteleuropa. Abstammung, verwandte Arten und adjacent areas. American Journal of Botany 92 : 1535 – 1547 . Hybriden (Pteridophyta). Gleditschia 15 : 57 – 72 . 1216 American Journal of Botany [Vol. 97

Klecka , W. R. 1980 . Discriminant analysis. Sage University paper se- Seehausen , O. 2004 . Hybridization and adaptive radiation. Trends in ries: Quantitative applications in the social sciences series no. 07-019. Ecology & Evolution 19 : 198 – 207 . Sage Publications, Beverly Hills, California, USA. Seifert , M. , and R. Holderegger . 1 9 9 5 . M o r p h o l o g i s c h e U n t e r s u - Krause , S. 1998 . Dryopteris Adans. In R. Wisskirchen and H. Haeupler chungen innerhalb der Dryopteris carthusiana Gruppe. Farnbl ä tter [eds.], Standardliste der Farn- und Bl ü tenpfl anzen Deutschlands, 182 – 26/27 : 58 – 77 . 190. Ulmer, Stuttgart, Germany. Soltis , P. S. , and D. E. Soltis . 1987 . Population structure and esti- Kron , P. , J. Suda , and B. C. Husband . 2007 . Applications of fl ow cy- mates of gene fl ow in the homosporous fern Polystichum munitum. tometry to evolutionary and population biology. Annual Review of Evolution 41 : 620 – 629 . Ecology, Evolution, and Systematics 38 : 847 – 876 . Soltis , P. S. , and D. E. Soltis . 1990 . Evolution of inbreeding and out- Kub á t , K. , L. Hrouda , J. Chrtek Jr ., Z. Kaplan , J. Kirschner , and J. crossing in ferns and fern-allies. Plant Species Biology 5 : 1 – 11 . Š tĚ p á nek [eds.]. 2002 . Kl íþ ke kv ě ten ě ý esk é republiky. Academia, Soltis , P. S. , D. E. Soltis , and K. E. Holsinger . 1988 . Estimates of Praha, Czech Republic. intragametophytic selfi ng and interpopulational gene fl ow in homo- Leitch , I. J. , and M. D. Bennett . 2007 . Genome size and its uses: The impact sporous ferns. American Journal of Botany 75 : 1765 – 1770 . of fl ow cytometry. In J. Dole ž el, J. Greilhuber, and J. Suda, [eds.], Flow Sorsa , V. , and C. J. Wid é n . 1968 . The Dryopteris spinulosa complex in cytometry with plant cells, 153 – 176. Wiley-VCH, Weinheim, Germany. Finland, a cytological and chromatographic study of some hybrids. Leonhards , W. , H. Rasbach , W. J ä ger , and H. W. Bennert . 1990 . Hereditas 60 : 273 – 293 . Vorkommen und Cytologie von Dryopteris u deweveri (= Dryopteris Suda , J. , A. Krahulcov á , P. Tr á vn í ý e k , R. Rosenbaumov á , T. Peckert , carthusiana u dilatata , Dryopteridaceae, Pteridophyta) in Nordrhein- and F. Krahulec . 2007 . Genome size variation and species rela- Westfalen. Tuexenia 10 : 17 – 24 . tionships in Hieracium sub-genus Pilosella (Asteraceae) as inferred Levin , D. A. , J. Francisco Ortega , and R. K. Jansen . 1996 . Hybridization by fl ow cytometry. Annals of Botany 100 : 1323 – 1335 . and the extinction of rare plant species. Conservation Biology 1 0 : 1 0 – 1 6 . Tryon , R. , and M. Britton . 1966 . A study of variation in the cytotypes of Loureiro , J. , P. Tr á vn íý ek , J. Rauchov á , T. Urfus , P. V í t , M. Š tech , S. Dryopteris spinulosa. Rhodora 68 : 59 – 92 . Castro , and J. Suda . 2010 . The use of fl ow cytometry in the biosys- Uyeda , J. C. , and S. R. Kephart . 2006 . Detecting species boundaries and tematics, ecology and population biology of homoploid plant. Preslia hybridization in Camassia quamash and C. leichtlinii (Agavaceae) us- 82 : 3 – 21 . ing allozymes. Systematic Botany 31 : 642 – 655 . Mahelka , V. , J. Suda , V. Jarol í mov á , P. Tr á vn í ý ek , and F. Krahulec . Viane , R. L. 1986 . Taxonomical signifi cance of the leaf indument in 2005 . Genome size discriminates between closely related taxa Dryopteris (Pteridophyta). I. Some North American, Macronesian and Elytrigia repens and E. intermedia (Poaceae: Triticeae) and their hy- European taxa. Plant Systematics and Evolution 153 : 77 – 105 . brid. Folia Geobotanica 40 : 367 – 384 . Wagner , W. H. Jr . 1971 . Evolution of Dryopteris in relation to the Manton , I. 1950 . Problems of cytology and evolution of the Pteridophyta. Appalachians. In P. C. Holt [ed.], The distributional history of the [Publisher unknown], Cambridge, UK. biota of the southern Appalachians. Research Division Monograph Manton , I. , and S. Walker . 1953 . Cytology of the Dryopteris spinulosa 2, 147 – 192. Virginia Polytechnic Institute and State University, complex in eastern North America. Nature 171 : 1116 – 1117 . Blacksburg, Virginia, USA. Masuyama , S. , and Y. Watano . 1990 . Trend for inbreeding in polyploid Wagner, W. H. Jr., and L. K. Chen. 1965. Abortion of spores and sporan- pteridophytes. Plant Species Biology 5 : 13 – 17 . gia as a tool in the detection of Dryopteris hybrids. American Fern Montgomery , J. D. , and W. H. Wagner Jr . 1993 . Dryopteris. In Flora of Journal 55: 9 – 29. North America Editorial Committee [eds.], Flora of North America Wagner , W. H. Jr ., F. S. Wagner , A. A. Reznicek , and C. R. Werth . north of Mexico, 280 – 289. Oxford University Press, New York, New 1 9 9 2 . u Dryostichum singulare (Dryopteridaceae), a new fern notho- York, USA. genus from Ontario. Canadian Journal of Botany 70 : 245 – 253 . Mosseler , A. 1990 . Hybrid performance and species crossability re- Walker , S. 1955 . Cytogenetic studies in the Dryopteris spinulosa com- lationships in willows (Salix ). Canadian Journal of Botany 68 : plex I. Watsonia 3 : 193 – 209 . 2329 – 2338 . Walker , S. 1961 . Cytogenetic studies in the Dryopteris spinulosa com- Nannfeldt , J. A. 1966 . Dryopteris dilatata och Dr. assimilis i Sverige. plex II. American Journal of Botany 48 : 607 – 614 . Botaniska Notiser 119 : 136 – 152 . Werth , C. R. 1989 . Isozyme evidence on the origin of Dryopteris cristata Nardi , E. 1976 . Dryopteris assimilis S.Walker in Italia. Webbia 30 : and D. carthusiana. American Journal of Botany 76 ( supplement ): 457 – 478 . 208 [abstract]. Otto , F. 1990 . DAPI staining of fi xed cells for high-resolution fl ow cy- Werth , C. R. , and C. Kuhn . 1989 . A study of the morphological attributes tometry of nuclear DNA. In H. A. Crissman and Z. Darzynkiewicz of the hypothetical diploid fern Dryopteris “semicristata ” . American [eds.], Methods in cell biology, vol. 33, 105 – 110. Academic Press, Journal of Botany 76 ( supplement ): 208 [abstract]. New York, New York, USA. Werth , C. R. , and D. B. Lellinger . 1992 . Genomically preserved plants Pi Ę ko Ś , H. 1974 . Mieszaniec Dryopteris assimilis u dilatata w Polsce. – and their nomenclature. Taxon 41 : 513 – 521 . The hybrid Dryopteris assimilis u dilatata in Poland. Fragmenta Wid é n , C. J. , J. Sarvela , and T. Ahti . 1967 . The Dryopteris spinulosa Floristica et Geobotanica 20 : 333 – 338 . complex in Finland. Acta Botanica Fennica 77 : 1 – 25 . PiĘ ko Ś -Mirkowa , H. 1979 . Paprocie z grupy Dryopteris dilatata w Wid é n , C. J. , and V. Sorsa . 1968 . On the intraspecifi c variability of Polsce. Monographiae Botanicae 59 : 1 – 75 . Dryopteris assimilis S. Walker and Dryopteris spinulosa Watt, a chro- PiĘ ko Ś -Mirkowa , H. 1987 . The Dryopteris dilatata complex in the matographic and cytological study. Hereditas 62 : 1 – 13 . Soviet Far East. Botanica Helvetica 97 : 167 – 177 . Wid é n , C. J. , V. Sorsa , and J. Sarvela . 1970 . Dryopteris dilatata s. lat. in Pi Ę k o Ś - M i r k o w a , H. 1 9 9 3 . A l t i t u d i n a l d i s t r i b u t i o n p a t t e r n o f Dryopteris Europe and the Island of Madeira, a chromatographic and cytological dilatata and D. expansa (Dryopteridaceae) in the Alps, the Carpathians study. Acta Botanica Fennica 91 : 1 – 30 . and the Caucasus. Floristica et Geobotanica ( supplementum 2 ): Whitham , T. G. , P. A. Morrow , and B. M. Potts . 1991 . Conservation of 307 – 312 . hybrid plants. Science 254 : 779 – 780 . Rieseberg , L. H. 1997 . Hybrid origins of plant species. Annual Review of Xiang , L. , C. R. Werth , S. N. Emery , and D. E. McCauley . 2000 . Ecology, Evolution, and Systematics 28 : 359 – 389 . Population-specifi c gender-biased hybridization between Dryopteris Rieseberg , L. H. , and S. E. Carney . 1998 . Plant hybridization. New intermedia and D. carthusiana : Evidence from chloroplast DNA. Phytologist 140 : 599 – 624 . American Journal of Botany 87 : 1175 – 1180 . Rosenbaumov á , R. , I. Pla ý kov á , and J. Suda . 2004 . Variation Yatabe , Y. , W. Shinohara , S. Matsumoto , and N. Murakami . 2009 . in Lamium subg. Galeobdolon (Lamiaceae): Insights from ploidy Patterns of hybrid formation among cryptic species of bird-nest fern, levels, morphology and isozymes. Plant Systematics and Evolution Asplenium nidus complex (Aspleniaceae), in West Malesia. Botanical 244 : 219 – 244 . Journal of the Linnean Society 160 : 42 – 63 . July 2010] Ekrt et al. — Taxonomy of DRYOPTERIS CARTHUSIANA complex 1217

Appendix 1. Locality details, including geographic coordinates (WGS-84 system), altitude (m a.s.l.), collector name(s), date of collection and herbarium collections where vouchers are stored (abbreviated according to Holmgren et al., 1990 ) for 85 populations of the Dryopteris carthusiana group. Total number of analyzed plants together with taxonomic composition and number of particular species and hybrids is also provided. Samples used for morphometric analyses are marked by an asterisk. Country abbreviations: AT = Austria, CZ = Czech Republic, DE = Germany, SK = Slovakia. Sample abbreviations: car = D. carthusiana , dil = D. dilatata , exp = D. expansa , u amb = D. u ambroseae (= D. dilatata u D. expansa ), u dew = D. u deweveri (= D. carthusiana u D. dilatata ), u sar = D. u sarvelae (= D. carthusiana u D. expansa ).

Species and hybrids Total number of detected No. Locality details samples (number of individuals)

1 CZ: Labsk é st ř edoho ř í — Tluþ ensk é ú dol í valley, 50q 34 ƍ 42.7 Ǝ N, 14q 05 ƍ 13.8 Ǝ E, 310 m, leg. R. Holubov á , 17. 4 dil(4)* VIII. 2004, PRC. 2 CZ: Labsk é st ř edoho ř í — Op á rensk é ú dol í valley, 50q 32 ƍ 19.8 Ǝ N, 13q 59 ƍ 22.8 Ǝ E, 275 m, leg. R. Holubov á , 12. 26 car(21)*, dil(5)* VI. 2005, PRC. 3 CZ: D ž b á n — Bil í chov, 50q 15 ƍ 59.3 ƍ ’ N, 13q 54 ƍ 50.7 Ǝ E, 395 m, leg. R. Holubov á , 25. VI. 2005, PRC. 48 car(15)*, dil(30)*, exp(3)* 4 CZ: Pardubick é Polab í — Krakovany u Podě brad, 50q 04 ƍ 14.3 ƍ ’ N, 15q 24 ƍ 32.2 Ǝ E, 220 m, leg. R. Holubov á , 10.X. 20 car(14)*, dil(6)* 2004, PRC. 5 CZ: Kru š nohorsk é podhů ř í vlastn í — M í sto, 50q 26 ƍ 17 Ǝ N, 13q 15 ƍ 44 Ǝ E, 405 m, leg. L. Ekrt, 2. IX. 2007, PR. 2 u dew(2)* 6 CZ: ý esk ý les — Pivo ň , ruins of Star ý Her š tejn, 49q 28 ƍ 19 Ǝ N, 12q 42 ƍ 48 Ǝ E, 850 m, leg. L. Ekrt, 4. IX. 2007, PR. 5 dil(1), u amb(1), u dew(3)* 7 CZ: ý esk ý les — Broumov near Zadn í Chodov, 49q 53 ƍ 33 Ǝ N, 12q 34 ƍ 38 Ǝ E, 595 m, leg. L. Ekrt, 3. IX. 2007, PR. 3 dil(1), exp(2) 8 CZ: ý esk ý les — P ř imda, 49q 40 ƍ 54 Ǝ N, 12q 39 ƍ 07 Ǝ E, 595 m, leg. L. Ekrt, 3. IX. 2007, PR. 1 exp(1) 9 CZ: ý esk ý les — Ostrů vek near Lesn á , 49q 44 ƍ 48 Ǝ N, 12q 27 ƍ 08 Ǝ E, 680 m, leg. L. Ekrt, 3. IX. 2007, PR. 2 exp(2) 10 CZ: Holoubkovsk é Podbrdsko — Stra š ice, 49q 43 ƍ 57 Ǝ N, 13q 44 ƍ 01 Ǝ E, 470 m, leg. L. Ekrt, 2. IX. 2007, PR. 2 dil(2) 11 CZ: Ho ř ovick á kotlina — Jivina near Hoř ovice, 49q 47 ƍ 19 Ǝ N, 13q 49 ƍ 38 Ǝ E, 600 m, leg. L. Ekrt, 2. IX. 2007, PR. 3 dil(3) 12 CZ: Prachatick é Př ed š umav í — Kahov, Dehetn í k, 49q 00 ƍ 49 Ǝ N, 13q 58 ƍ 22 Ǝ E, 755 m, leg. L. Ekrt, 19. X. 2004, 4 car(2)*+(1), dil(1)* PR. 13 CZ: Kaplick é mezihoř í — Horn í Dvo ř i š t ě , 48q 37 ƍ 44 Ǝ N, 14q 23 ƍ 02 Ǝ E, 550 m, leg. L. Ekrt, 5. IX. 2007, PR. 1 dil(1) 14 CZ: T ř ebo ň sk á p á nev — Borkovick á blata peatbog, 49q 14 ƍ 13 Ǝ N, 14q 37 ƍ 56.2 Ǝ E, 430 m, leg. R. Holubov á , 27. 28 car(26)*, dil(2)* IX. 2004, PRC. 15 CZ: T ř ebo ň sk á p á nev — Ma ž ick á blata peatbog, 49q 14 ƍ 14 Ǝ N, 14q 37 ƍ 00.8 Ǝ E, 425 m, leg. R. Holubov á , 28. IX. 23 car(16)*+(7) 2004, PRC. 16 CZ: Mile š ovsk é stř edoho ř í — Š t ě p á novsk á hora hill, 50q 32 ƍ 18.8 Ǝ N, 13q 52 ƍ 18.7 Ǝ E, 550 m, leg. K. Kub á t & R. 12 dil(3)*, exp(9)* Holubov á , 14. VIII. 2004, PRC. 17 CZ: Mile š ovsk é stř edoho ř í — Mile š ovka hill, 50q 33 ƍ 21.3 ƍ ’ N, 13q 55 ƍ 45.5 ƍ ’ E, 770 m, leg. R. Holubov á , 26. VIII. 43 car(7)*, dil(24)*+(6), 2004 & 11. VI. 2005, PRC. exp(2)*, u amb(4)* 18 CZ: Love þ kovsk é stř edoho ř í — Kalich hill, 50q 36 ƍ 30.2 Ǝ N, 14q 12 ƍ 27.4 Ǝ E, 435 m, leg. R. Holubov á , 15. VIII. 3 dil(3)* 2004, PRC. 19 CZ: Love þ kovsk é stř edoho ř í — Panna hill, 50q 36 ƍ 31 Ǝ N, 14q 10 ƍ 49.5 ƍ ’ E, 475 m, leg. R. Holubov á , 15. VIII. 9 car(1)*, dil(8)* 2004, PRC. 20 CZ: Love þ kovsk é stř edoho ř í — ý ertova jizba hill, 50q 36 ƍ 59.4 Ǝ N, 14q 05 ƍ 08.1 Ǝ E, 305 m, leg. R. Holubov á , 16. 9 exp(9)* VIII. 2004, PRC. 21 CZ: Love þ kovsk é stř edoho ř í — Bob ř í soutě ska valley, 50q 39 ƍ 26.3 ƍ ’ N, 14q 21 ƍ 29.3 ƍ ’ E, 435 m, leg. R. Holubov á , 54 car(14)*, dil(19)*, 4. IX. 2004 & 26. VI. 2005, PRC. exp(13)*, u amb(8)* 22 CZ: Love þ kovsk é stř edoho ř í — Kamenec hill, 50q 42 ƍ 25.1Ǝ N, 14q 21 ƍ 20.4 Ǝ E, 315 m, leg. R. Holubov á , 6. IX. 81 car(49)*, dil(19)*, 2004 & 4. X. 2004, PRC. exp(13)* 23 CZ: Love þ kovsk é stř edoho ř í — Binov hill, 50q 38 ƍ 44.6 Ǝ N, 14q 21 ƍ 52.5 ƍ ’ E, 505 m, leg. R. Holubov á , 7. IX. 2004, 11 car(1)*, dil(10)* PRC. 24 CZ: Love þ kovsk é stř edoho ř í — Prů þ elsk á rokle gorge, 50q 37 ƍ 03.8 Ǝ N, 14q 06 ƍ 28.5 ƍ ’ E, 450 m, leg. R. Holubov á , 21 car(4)*, dil(16)*, exp(1)* 20. IX. 2004, PRC. 25 CZ: Love þ kovsk é stř edoho ř í — Star ý Š achov, 50q 43 ƍ 12.7 Ǝ N, 14q 22 ƍ 53.7 Ǝ E, 230 m, leg. R. Holubov á , 21. IX. 7 car(6)*, dil(1)* 2004, PRC. 26 CZ: R ů ž ovsk á tabule — Dě þ í n, R ů ž ovsk ý vrch hill, 50q 50 ƍ 17.7 Ǝ N, 14q 19 ƍ 50.2 Ǝ E, 470 m, leg. R. Holubov á , 19. 57 car(19)*, dil(11)*, IX. 2004, PRC. exp(22)*, u amb(5)* 27 CZ: Lu ž ick é hory — Kl íþ hill, 50q 47 ƍ 18.4 Ǝ N, 14q 34 ƍ 14.4 Ǝ E, 555 m, leg. R. Holubov á , 5. IX. 2004, PRC. 24 car(3)*, dil(5)*, exp(13)*, u amb(3) 28 CZ: Polomen é hory — Polomen é Hory, 50q 36 ƍ 11.3 ƍ ’ N, 14q 40 ƍ 35 Ǝ E, 365 m, leg. R. Holubov á , 3. X. 2004, PRC. 1 car(1)* 29 CZ: Ralsko – bezd ě zsk á tabule — Ralsko u Mimon ě , 50q 40 ƍ 31.7 Ǝ N, 14q 45 ƍ 54.8 Ǝ E, 600 m, leg. R. Holubov á , 18. 36 car(3)*, dil(18)*, IX. 2004, PRC. exp(8)*, u amb(7)* 30 CZ: Ralsko – bezdě zsk á tabule — Star é Splavy, 50q 35 ƍ 57.3 ƍ ’ N, 14q 38 ƍ 46.1 Ǝ E, 260 m, leg. R. Holubov á , 3. X. 1 car(1)* 2004, PRC. 31 CZ: Troseck á pahorkatina — Př í hrazy, Krtola gorge, 50q 31 ƍ 48 Ǝ N, 15q 03 ƍ 53 Ǝ E, 340 m, leg. L. Ekrt, 25. VI. 1 u amb(1) 2004, PR. 32 CZ: Polick á kotlina — Horn í Adr š pach, 50q 38 ƍ 23 Ǝ N, 16 q 05 ƍ 20 Ǝ E, 620 m, leg. L. Ekrt, 19. X. 2004, PR. 2u amb(2)* 33 CZ: Ž altman — Zbeþ n í k, Maternice, 50 q 29 ƍ 31 Ǝ N, 16q 08 ƍ 44 Ǝ E, 435 m, leg. L. Ekrt, 9. IX. 2005, PR. 19 car(3), dil(10), exp(4), u amb(2) 34 CZ: Broumovsk é stě n y — H l a v ň ov, Koruna rock, 50q 3 1 ƍ 0 9 Ǝ N , 1 6 q 1 8 ƍ 4 5 Ǝ E, 720 m, leg. L. Ekrt, 10. X. 2004, PR. 3 dil(1)*, exp(1)*, u amb(1) 35 CZ: Broumovsk é stě n y — K o v á ř ova rokle gorge, 50 q 3 3 ƍ 3 5Ǝ N , 1 6 q 1 6 ƍ 0 7 Ǝ E, 605 m, leg. L. Ekrt, 8. IX. 2005, PR. 17 exp(11), u amb(6) 36 CZ: Broumovsk é st ě ny — Zajeþ í rokle gorge, 50q 32 ƍ 16 Ǝ N, 16q 17 ƍ 57 Ǝ E, 595 m, leg. L. Ekrt, 8. IX. 2005, PR. 17 car(3), dil(3), exp(6), u amb(5) 37 CZ: Broumovsk é st ě ny — Bo ž anov, 50q 29 ƍ 59 Ǝ N, 16q 20 ƍ 43 Ǝ E, 630 m, leg. A. H á jek, 27. VI. 2005, PR. 1 u dew(1) 1218 American Journal of Botany [Vol. 97

Appendix 1. Continued.

Species and hybrids Total number of detected No. Locality details samples (number of individuals)

38 CZ: ý eskomoravsk á vrchovina — Ol š any u Telþ e, 49q 10 ƍ 46 Ǝ N, 15q 33 ƍ 36 Ǝ E, 600 m, leg. L. Ekrt, 28. VII. 2003, 3 dil(1), exp(1), u amb(1)* PR. 39 CZ: ý eskomoravsk á vrchovina — Mysliboř , Obecn í rybn í k pond, 49q 13 ƍ 23 Ǝ N, 15q 28 ƍ 04 Ǝ E, 550 m, leg. L. Ekrt, 2 car(1), dil(1) 30. X. 2004, PR. 40 CZ: ý eskomoravsk á vrchovina — Kosteln í Myslov á , Velk ý Huli šť sk ý rybn í k pond, 49q 09 ƍ 09 Ǝ N, 15q 24 ƍ 25 Ǝ E, 22 car(4), dil(9), 520 m, leg. L. Ekrt, 8. VIII. 2005, PR. exp(1)*+(4), u amb(2), u dew(2)* 41 CZ: ý eskomoravsk á vrchovina — Tř e š ť , Velk ý Š pi þ á k hill, 49q 18 ƍ 45 Ǝ N, 15q 30 ƍ 40 Ǝ E, 695 m, leg. L. Ekrt, 16. 18 car(4), dil(8), exp(6) VII. 2005, PR. 42 CZ: ý eskomoravsk á vrchovina — Stonař ov, 49 q 16 ƍ 57 Ǝ N, 15 q 33 ƍ 30 Ǝ E, 630 m, leg. L. Ekrt, 19. IX. 2006, PR. 2 u dew(1)+(1)* 43 CZ: ý eskomoravsk á vrchovina — Poles í near Poþ á tky, 49q 17 ƍ 16 Ǝ N, 15q 15 ƍ 24 Ǝ E, 630 m, leg. L. Ekrt, 11. VII. 1 u dew(1) 2006, PR. 44 CZ: ý eskomoravsk á vrchovina — Jezdovice, 49q 18 ƍ 28 Ǝ N, 15q 29 ƍ 07 Ǝ E, 545 m, leg. L. Ekrt, 15. IX. 2007, PR. 1 exp(1) 45 CZ: ý eskomoravsk á vrchovina — Tř e š tice, 49 q 14 ƍ 40 Ǝ N, 15q 28 ƍ 20 Ǝ E, 605 m, leg. L. Ekrt, 9. IX. 2007, PR. 2 exp(2) 46 CZ: ý eskomoravsk á vrchovina — Panensk á Rozs íþ ka, 49q 14 ƍ 43 Ǝ N, 15q 30 ƍ 42 Ǝ E, 585 m, leg. L. Ekrt, 8. IX. 1 exp(1) 2007, PR. 47 CZ: ý eskomoravsk á vrchovina — Bezdě kov, 49q 15 ƍ 06 Ǝ N, 15q 31 ƍ 16 Ǝ E, 565 m, leg. L. Ekrt, 7. IX. 2007, PR. 1 exp(1) 48 CZ: Ceskomoravsk á vrchovina — Sta š ov, 49q 40 ƍ 32 Ǝ N, 16q 22 ƍ 00 Ǝ E, 650 m, leg. J. Ko š nar, 31. VIII. 2007, PR. 1 dil(1) 49 CZ: Jesenick é podhů ř í — Krasov, 50q 05 ƍ 04 Ǝ N, 17q 30 ƍ 02 Ǝ E, 565 m, leg. L. Ekrt, 17. IX. 2007, PR. 2 exp(2) 50 CZ: B í l é Karpaty lesn í — V á penky, Velk á Javoř ina hill, 48q 51 ƍ 58 Ǝ N, 17q 39 ƍ 40 Ǝ E, 620 m, leg. L. Ekrt, 18. IX. 3 dil(1), exp(2) 2007, PR. 51 CZ: Zl í nsk é vrchy — Lideþ ko, 49q 13 ƍ 13 Ǝ N, 18q 02 ƍ 48 Ǝ E, 595 m, leg. L. Ekrt, 18. IX. 2007, PR. 1 exp(1) 52 CZ: Beskydsk é podh ů ř í — Jablunkov, 49q 35 ƍ 30 Ǝ N, 18q 47 ƍ 07 Ǝ E, 430 m, leg. L. Ekrt, 17. IX. 2007, PR. 4 exp(2), u amb(2) 53 CZ: Brdy — Leletice, 49q 31 ƍ 58 Ǝ N, 13q 52 ƍ 06 Ǝ E, 535 m, leg. L. Ekrt, 2. IX. 2007, PR. 5 exp(1), u amb(2), u dew(2)* 54 CZ: Kr á lovsk ý hvozd — cirque of the ý ern é jezero lake, 49q 10 ƍ 29 Ǝ N, 14q 10 ƍ 54 Ǝ E, 1050 m, leg. L. Ekrt, 14. 12 dil(1), exp(9), u amb(2) VII. 2005, PR. 55 CZ: Kr á lovsk ý hvozd — Ž elezn á Ruda, Debrnick é ú dol í valley, 49q 06 ƍ 57 Ǝ N, 13q 14 ƍ 20 Ǝ E, 770 m, leg. L. Ekrt & 1 exp(1) J. Hadinec, 29. VIII. 2007, PR. 56 CZ: Boub í nsko – sto ž eck á hornatina —ý esk é Ž leby, Sp á leni š tě hill, 48q 52 ƍ 30 Ǝ N, 13q 48 ƍ 01 Ǝ E, 870 m, leg. L. 13 car(5)*+(1), dil(1), Ekrt, 19. X. 2004, PR. exp(2)*, u amb(4)* 57 CZ: Boub í nsko – sto ž eck á hornatina — Boub í n hill, 48q 58 ƍ 31 Ǝ N, 13q 48 ƍ 36 Ǝ E, 1100 m, leg. L. Ekrt, 15. XI. 2 exp(1), u amb(1) 2003, PR. 58 CZ: Trojmezensk á hornatina — cirque of the Ple š n é jezero lake, 48q 46 ƍ 38.2 Ǝ N, 13q 51 ƍ 51.5 ƍ ’ E, 1120 m, leg. L. 13 dil(1), exp(6)+(2)*, Ekrt & E. Hofhanzlov á , 7. VIII. 2003 & 15. VII. 2005, PR. u amb(4) 59 CZ: 88e. Trojmezensk á hornatina — Nov á Pec, Smrþ ina hill, 48q 45 ƍ 27 Ǝ N, 13q 56 ƍ 27 Ǝ E, 880 m, leg. L. Ekrt, 25. 2 exp(1), u amb(1) IX. 2007, PR. 60 CZ: Trojmezensk á hornatina — Nov á Pec, Bul í k hill, 48q 45 ƍ 07 Ǝ N, 13q 55 ƍ 32 Ǝ E, 1150 m, leg. L. Ekrt, 25. IX. 2 exp(2) 2007, PR. 61 CZ: Ž elnavsk á hornatina — Pě kn á , ý ern ý les hill, 48q 50 ƍ 26 Ǝ N, 13q 58 ƍ 38 Ǝ E, 920 m, leg. L. Ekrt & E. Ekrtov á , 5 car(1)*, dil(1)*, exp(2)*, 19. X. 2004, PR. u amb(1)* 62 CZ: Novohradsk é hory — Š ejby, Hojn á voda, 48q 42 ƍ 23 Ǝ N, 14q 45 ƍ 09 Ǝ E, 850 m, leg. L. Ekrt & M. Lep š í , 5. IX. 2 exp(1), u amb(1) 2007, PR. 63 CZ: Novohradsk é hory — Pohorsk á Ves, Ž of í nsk ý prales, 48q 40 ƍ 03.3 ƍ ’ N, 14q 42 ƍ 04.5 ƍ ’ E, 810 m, leg. M. Lep š í , 20 car(3), dil(4), exp(10), 10. VIII. 2005, CB. u amb(3) 64 CZ: Novohradsk é hory — Ž of í n, 48° 40 ƍ 30 Ǝ N, 14° 40 ƍ 26 Ǝ E, 765 m, leg. L. Ekrt & M. Lep š í , 5. IX. 2007, PR. 1 exp(1) 65 CZ: Krkono š e lesn í — Medvě d í potok, 50° 44 ƍ 44.9 Ǝ N 15° 36 ƍ 01.7 Ǝ E, 820 m, leg. R. Holubov á , 8. VII. 2005, 13 exp(12)*, × amb(1)* PRC. 66 CZ: Krkono š e lesn í — Medvě d í boudy, 50° 45 ƍ 44.36 Ǝ N, 15 ° 35 ƍ 33.1 Ǝ E, 1025 m, leg. R. Holubov á , 9. VII. 2005, 29 exp(27)*, × amb(2)* PRC. 67 CZ: Krkono š e lesn í — Labsk ý dů l valley, 50 ° 46 ƍ 09.4 Ǝ N, 15 ° 32 ƍ 59.8 Ǝ E, 1200 m, leg. R. Holubov á , 10. VII. 23 exp(21)*, × amb(2)* 2005, PRC. 68 CZ: Teplicko – adr š pa š sk é sk á ly — Sibiř , 50 ° 35 ƍ 33 Ǝ N, 16 ° 07 ƍ 32Ǝ E, 660 m, leg. L. Ekrt, 8. IX. 2005, PR. 19 exp(14), × amb(5), 69 CZ: Teplicko – adr š pa š sk é sk á ly — Homole cukru rock, 50° 35 ƍ 12.2 Ǝ N, 16 ° 7 ’ 10.4 Ǝ E, 660 m, leg. A. H á jek, 2005, 1 dil(1) HR. 70 CZ: Teplicko – adr š pa š sk é sk á ly — Divok á rokle gorge, 50° 35 ƍ 32.7 Ǝ N, 16 ° 08 ƍ 25.1 Ǝ E, leg. A. H á jek, 2005, HR. 1 exp(1) 71 CZ: ý esk ý h ř eben — Bukaþ ka, 50 ° 20 ƍ 20 Ǝ N, 16° 22 ƍ 36 Ǝ E, 1000 m, leg. L. Ekrt & J. Kuþ era, 17. IX. 2007, PR. 3 exp(2), × amb(1) 72 CZ: ý esk ý h ř eben — Pansk é Pole, ý ern ý d ů l valley, 50 ° 12 ƍ 06 Ǝ N, 16 ° 31 ƍ 05 Ǝ E, 860 m, leg. L. Ekrt & J. Kuþ era, 2 exp(2) 17. IX. 2007, PR. 73 CZ: Hrub ý Jesen í k — Vidly, 50° 06 ƍ 13 Ǝ N, 17 ° 15 ƍ 17 Ǝ E, 845 m, leg. L. Ekrt, 17. IX. 2007, PR. 2 exp(2) 74 CZ: Radho šť sk é Beskydy — Pustevny, 49° 29 ƍ 34 Ǝ N, 18 ° 15 ƍ 57 Ǝ E, 1090 m, leg. L. Ekrt, 18. IX. 2007, PR. 5 exp(2), × amb(1), × dew(2)* 75 CZ: Radho šť sk é Beskydy — Doln í Lomn á , Mion š í reserve, 49° 31 ƍ 52 Ǝ N, 18° 40 ƍ 08 Ǝ E, 625 m, leg. L. Ekrt, 18. 2 exp(2) IX. 2007, PR. 76 CZ: Slezsk é Beskydy — N ý dek, 49° 40 ƍ 09 Ǝ N, 18° 46 ƍ 24 Ǝ E, 485 m, leg. L. Ekrt, 17. IX. 2007, PR. 3 exp(2), × amb(1) 77 SK: Kriv á nska Mal á Fatra — Krasň any, 49 ° 11 ƍ 34 Ǝ N, 18° 56 ƍ 01 Ǝ E, 605 m, leg. L. Ekrt, 30. IX. 2004, PR. 2 dil(1)*, exp(1)* 78 SK: Mal á Fatra — Terchov á , 49° 14 ƍ 39 Ǝ N, 19 ° 05 ƍ 24 Ǝ E, 945 m, leg. L. Ekrt, 30. IX. 2004, PR. 1 exp(1)* July 2010] Ekrt et al. — Taxonomy of DRYOPTERIS CARTHUSIANA complex 1219

Appendix 1. Continued.

Species and hybrids Total number of detected No. Locality details samples (number of individuals) 79 SK: Z á padn í Tatry — Liptovsk ý Mikul á š , Parichvost valley, 49° 11 ƍ 17 Ǝ N, 19° 43 ƍ 51 Ǝ E, 1250 m, leg. L. Ekrt & 1 exp(1) E. Hofhanzlov á , 24. VI. 2003, PR. 80 AT: Rottenmanner Tauren — Rottenmann, 47° 28 ƍ 21 Ǝ N, 14 ° 24 ƍ 56 Ǝ E, 1600 m, leg. L. Ekrt & E. Hofhanzlov á , 3 exp(2), × amb(1) 24. VIII. 2003, PR. 81 DE: Harz — B ä renklippe near Schirke, 51° 45 ƍ N, 10 ° 40 ƍ E, 860 m, leg. S. Jessen, 8. VI. 1981, cult. in Arktisch – 1 exp(1) Alpine Garten, Chemnitz, SJ359. 82 DE: Th ü ringen — Stadtroda, Borntal valley near Ruttersdorf, 50° 52 ƍ N, 11 ° 44 ƍ E, leg. S. Jessen & J. Riethausen, 1 car(1) 1992, cult. in Arktisch – Alpine Garten, Chemnitz, SJ2821. 83 DE: Th ü ringen — Fuchsgraben near Wolfsgef ä rth near Gera, 50° 48 ƍ N, 12 ° 03 ƍ E, leg. S. Jessen, 17. VII. 1982, 1 × dew(1) cult. in Arktisch – Alpine Garten, Chemnitz, SJ147/1. 84 DE: R ü gen — Kranichbruch near Neu Mukran, 54° 28 ƍ N, 13 ° 34 ƍ E, 3 m, leg. S. Jessen, 28. VI. 1983, cult. in 1 × sar(1) Arktisch – Alpine Garten, Chemnitz, SJ337. 85 DE: R ü gen — Stubbnitz, Herthasee, 54° 32 ’ N, 13 ° 38 ƍ E, 115 m, leg. S. Jessen, 12. VI. 1984, cult. in Arktisch – 1 × sar(1) Alpine Garten, Chemnitz, SJ374.