HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Supplementary Data Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Weiss-Schneeweiss, H. Articles by Schneeweiss, G. M. Search for Related Content PubMed Articles by Weiss-Schneeweiss, H. Articles by Schneeweiss, G. M. Agricola Articles by Weiss-Schneeweiss, H. Articles by Schneeweiss, G. M.

Genome size evolution in holoparasitic Orobanche (Orobanchaceae) and related genera¹

²Department of Systematic and Evolutionary Botany, Institute of Botany, University of Vienna, Rennweg 14, A-1030 Vienna, Austria; ³Department of Biogeography, Institute of Botany, University of Vienna, Rennweg 14, A-1030 Vienna, Austria

Received for publication June 6, 2005. Accepted for publication September 21, 2005.

ABSTRACT

Genome size was estimated using Feulgen densitometry for 76 accessions of 40 taxa of Orobanche and two taxa each of the related genera Phelypaea and Cistanche, providing the first data set for any group of nonphotosynthetic angiosperms. The 2C-values were 16.8–19.9 pg in Cistanche, 2.9–11.6 pg in Orobanche sect. Orobanche, 6.8–10.8 pg in sect. Trionychon, 4.3–5.1 pg in sect. Myzorrhiza, and 4.9–5.8 pg and 10.5 pg in the two diploid species Phelypaea coccinea and P. tournefortii, respectively. Distribution of genome size is congruent with phylogenetic lineages identified by analyses of nuclear ITS sequence data, in particular regarding the distinctness of O. anatolica from the rest of sect. Orobanche. With the exception of tetraploid O. transcaucasica, polyploid taxa of sect. Orobanche are among those with the smallest C-values, suggesting substantial genome downsizing after polyploidization (as in other angiosperm groups). In O. sect. Orobanche, genome size evolved more rapidly in earlier stages of its evolution than in later stages. This might be indicative of adaptive radiation, but this hypothesis requires corroboration in the form of genome size estimates on more taxa and accessions.

Key Words: Cistanche • genome size evolution • holoparasitic plants • Orobanche • Phelypaea • polyploids

The amount of DNA in an unreplicated gametic nuclear genome varies extensively across flowering plants. This is the 1C-value, which is not to be confused with the Cx-value ("genome size" in the definition of Bennett et al., 1998 ), which describes the amount of DNA in an unreplicated, basic, monoploid chromosome set (Greilhuber et al., 2005 ). Excluding low estimates with little reliability (see Bennett and Leitch, 2005 ), such as the 0.06 pg reported for Cardamine amara (Brassicaceae; Bennett and Smith, 1991 ), the 1C value ranges from 0.16 pg in Arabidopsis thaliana (Brassicacaeae) to 127.40 pg in Fritillaria assyriaca (Liliaceae; Bennett and Smith, 1976 ). The difference is thus about 800-fold, although the basic complement of genes required for normal growth and development appears to be essentially the same. This leads to what is referred to as the "C-value paradox" (Thomas, 1971 ; reviewed in Gregory, 2001 , 2005 ). Several mechanisms have been proposed for this large variation in DNA content. One of them is the existence of repeated cycles of polyploidy (e.g., Leitch and Bennett, 1997 ; Otto and Whitton, 2000 ; Soltis and Soltis, 2000 ; Wendel, 2000 ), a hypothesis recently substantiated by genomic evidence for ancient polyploidy, even in species with small genomes such as Arabidopsis thaliana (Arabidopsis Genome Initiative, 2000 ; Bowers et al., 2003 ). Transposable elements have also been suggested to contribute strongly to an increase in genome size throughout eukaryotes (e.g., Bennetzen, 2000 ; Kidwell, 2002 ). Mechanisms resulting in a decrease in genome size include unequal crossing over, illegitimate recombination, a higher overall rate of deletions than insertions, and selection against transposable elements (Morgan, 2001 ; Devos et al., 2002 ; Petrov, 2002 ; Wendel et al., 2002 ; Ma et al., 2004 ; Bennetzen et al., 2005 ).

The evolutionary significance of genome size changes is not well understood. Bennett (1971 , 1972 ) coined the term "nucleotype" to describe those properties of the nuclear DNA (mass and volume) that affect the phenotype independently of its encoded informational content. Such nucleotypic effects include an increase in nucleus and cell size, prolonged duration of both mitotic cycles and meiotic divisions, and consequences of these changes (Bennett, 1972 , 1977 ; Price et al., 1973 ; Kenton et al., 1986 ; Wyman et al., 1997 ). The drawbacks of the nucleotype model are that a direct influence of DNA content has not yet been conclusively established and that a satisfactory mechanistic model explaining this effect does not exist (Gregory, 2001 ; Petrov, 2001 ; for a criticism of the nucleotype model, see Cavalier-Smith, 2005 ). Correlations between genome size and specific life traits, most importantly life history and breeding system (e.g., Rees and Hazarika, 1969 ; Bennett, 1972 ; Price, 1976 ; Govindaraju and Cullis, 1991 ; Torrell and Vallès, 2001 ; Matzk et al., 2003 ; Vinogradov, 2003 ; Albach and Greilhuber, 2004 ; Bure et al., 2004 ; Grotkopp et al., 2004 ; Jakob et al., 2004 ; Chase et al., 2005 ; Price et al., 2005 ; but see Ohri and Pistrick, 2001 ), suggest that genome size does in fact have important evolutionary effects. Relationships between genome size and ecological factors, however, are less clear and might be confounded by other non-environmental factors such as life form (Jakob et al., 2004 ), or they might be stronger for species with higher C values (Knight and Ackerly, 2002 ). Additional complications may arise from phylogenetic dependence, e.g., closely related species have similar C values due to shared ancestry rather than shared ecology. Several approaches are used to address this problem, including generalized least squares methods (Pagel, 1997 , 1999 ) and independent contrasts (Felsenstein, 1985 ), the latter being a special case of the former (Pagel, 2004 ). Nevertheless, rigorous analyses in plants using a phylogenetic approach are still scarce (Bharathan, 1996 ; Cox et al., 1998 ; Kellogg, 1998 ; Albach and Greilhuber, 2004 ; Grotkopp et al., 2004 ; Jakob et al., 2004 ).

C-values have been estimated for c. 4100 angiosperm species (Bennett and Leitch, 2004 , 2005 ), representing more than 1% of flowering plant species and roughly 50% of all angiosperm families (sensu APG, 1998 ; APG II, 2003 ). This wide taxonomic coverage together with the availability of sound hypotheses on angiosperm phylogeny (e.g., Qiu et al., 1999 ; Soltis et al., 2000 , 2003a ) has only recently allowed rigorous testing of genome size evolution in angiosperms (Soltis et al., 2003b ; Leitch et al., 2005 ). Nevertheless, for some biologically interesting groups, data on genome size are still scarce. Among these poorly studied groups are parasitic plants (Bennett and Leitch, 2005 ). Of the three major families which contain parasitic species, only Loranthaceae (including Viscaceae) has a large data set in the Plant C-value Database, while Santalaceae and Orobanchaceae (sensu Young et al., 1999 ) are represented by only a few entries each (Bennett and Leitch, 2004 ).

Orobanche is the largest genus among the holoparasitic members of Orobanchaceae and comprises approximately 170 species distributed predominantly in the northern hemisphere. Of these, a few species, e.g., O. ramosa and O. crenata, are economically important weeds, causing severe yield losses mostly in Mediterranean areas (Parker and Riches, 1993 ; Wegmann, 1998 ; Musselman et al., 2001 ). Following Beck-Mannagetta (1930) , most authors divide Orobanche into four sections: Orobanche and Trionychon, both in the Old World, and Gymnocaulis and Myzorrhiza, confined to the New World. Some authors treat these sections as separate genera (Soják, 1972 ; Holub, 1977 , 1990 ; Teryokhin et al., 1993 ). Teryokhin et al. (1993) even consider Orobanche sect. Orobanche (as separate genus Orobanche) to be more closely related to Phelypaea and Cistanche than to O. sect. Trionychon (as separate genus Phelipanche). Recent molecular phylogenetic studies suggest two major lineages, informally called the Orobanche-group and the Phelipanche-group (Schneeweiss et al., 2004a ). The first includes Orobanche sect. Orobanche and the small SW Asian genus Phelypaea (listed as Diphelypaea in Schneeweiss et al., 2004a ), while the second includes the remaining sections of Orobanche (sects. Gymnocaulis, Myzorrhiza, and Trionychon). The circumscription of these groups is well supported by the distribution of chromosome base numbers, these being x = 19 in the Orobanche-group and x = 12 in the Phelipanche-group (Schneeweiss et al., 2004b ).

In this study, we provide estimates of C-values for 76 accessions representing 40 taxa of Orobanche and two taxa each of the related genera Phelypaea and Cistanche, providing the first data set for any group of nonphotosynthetic angiosperms. For all accessions investigated, chromosome numbers and thus ploidy level have been determined (Schneeweiss et al., 2004b ; this study), allowing us to calculate Cx-values for all accessions. The distribution of C-values in Orobanche and related genera is used to evaluate existing phylogenetic–taxonomic concepts. Due to the presence of several polyploid lineages and groups (Schneeweiss et al., 2004b ), emphasis is put on the evolution of genome size in polyploid groups to test whether the trend of genome size reduction in polyploids, as generally observed in angiosperms (Leitch and Bennett, 2004 ), is also found in Orobanche and related genera. In the core group of the Orobanche group, the evolution of C-values in diploids is analyzed in more detail using a generalized least squares method, which takes phylogenetic dependence into account, in order to (1) infer the ancestral genome size and (2) investigate whether there is evidence for adaptive radiation or accelerated evolution of genome size.

MATERIALS AND METHODS

Plant material
We investigated material from 68 accessions of 40 Orobanche taxa, including at least one representative of each of the major lineages within the Orobanche and the Phelipanche groups. We also investigated three accessions of two species of Phelypaea, a member of the Orobanche group, and five accessions of two subspecies of Cistanche phelypaea, resulting in a total of 76 accessions (Appendix).

Feulgen densitometry
Young flower buds collected in the field (by H. W.-S. or G. M. S., unless otherwise noted in the Appendix) were fixed in ethanol–acetic acid (3 : 1) for at least 24 h and then stored at –20°C until further use. Some of the samples were stored for an extended time (up to 2 years). If stored properly, this storage does not negatively affect the estimation of C-values with Feulgen densitometry (Greilhuber and Temsch, 2001 ). So we were able to include estimates obtained from these samples. Nuclear C-values were determined using Feulgen densitometry using video-based image analysis with CIRES (Cell Image Retrieval and Evaluation System, version 3.1, Kontron, Munich, Germany), as described in Greilhuber and Ebert (1994) , Dimitrova et al. (1999) , and Vilhar et al. (2001) , and following recommendations for best laboratory practice as outlined in Greilhuber and Temsch (2001) . As an internal standard, Pisum sativum cv. Kleine Rheinländerin (2C = 8.84 pg; Greilhuber and Ebert, 1994 ) was used, using one standard for 3– 7 accessions of Orobanche. Fixed material of the accessions and the standard were hydrolyzed synchronously, but in separate test tubes, for 60 min in 5 M HCl at 20°C in a water bath and then rinsed thoroughly. Hydrolyzed material was stained in Schiff's reagent for 1.5 h in the dark and afterwards washed in SO₂ water (0.5 g potassium bisulphite in 100 mL 0.05 N HCl) at room temperature for 45 min, changing the water four times. The material was then squashed in 45% acetic acid as a softening medium. After cover slip removal over a cold plate, the slides were air dried and measured on the following days. For each accession, 15 anaphases or early telophases on each of five slides were measured. Measurements of the integrated optical density (IOD) per nucleus were performed using a monochromatic green filter and a 63x oil immersion objective without application of a cover slip. The green image channel was selected, and the shading correction during image acquisition was activated. The correct light intensity (grey value 200 of an empty background image) was controlled before measuring each slide. Selected nuclei were segmented from the background and measured with the "interactive online nucleus segmentation" mode, with "local" and "object specific" background determination, i.e., around each single nucleus to be measured. Only C-values of accessions with a standard deviation of less than 5% between different slides are reported. New chromosome counts were done on the same slides as C-value estimation.

C-value statistics
Nuclear DNA-values (2C) for each accession are calculated as mean values from the 5 x 15 measurements. Diploid and tetraploid accessions were analyzed separately to avoid confounding the effects of real intraspecific variability with those of genome evolution after polyploidization (see also Results, C- and Cx-values). All statistical tests were conducted with SPSS 11.0.0 (SPSS Inc., Chicago, Illinois, USA). Normal distribution of data was tested for with the Shapiro–Wilkes statistics. Visual inspection of genome size distribution in diploids suggested that most of the phylogenetically distinct groups (see Schneeweiss et al., 2004a ) also had distinct genome size ranges. This was tested using Welch's analysis of variance (Welch's ANOVA due to heterogeneity of variances [Levene's statistic: P = 0.004]), with Dunnett's C post hoc test to test for significant differences among Phelypaea, Orobanche anatolica, O. sect. Orobanche pro parte, and O. sect. Trionychon. Cistanche was not included in this comparison, because its genome size range has no overlap with that of either Phelypaea or Orobanche. ANOVA or Welch's ANOVA (with Scheffé- and Dunnett's C post hoc tests, respectively, where appropriate) were conducted to test for significant differences within Cistanche, Phelypaea, Orobanche sect. Trionychon, O. anatolica, and O. sect. Orobanche (excluding O. anatolica). Accessions with non-normally distributed data (hirtiflora acc. 1, ramosa acc. 1) were excluded. Sequential Mann– Whitney tests for these accessions were not performed, because the Bonferroni-corrected significance level of = 0.00238 is below the lowest possible value of 0.008 for comparisons of two groups with N = 5. Although no significant differences were found among the polyploids in the phylogenetically distinct groups O. latisquama, O. sect. Orobanche pro parte, and O. sect. Myzorrhiza (Kruskal–Wallis test, P = 0.278), these three groups were analyzed separately as described earlier. Accessions with non-normally distributed data (gracilis acc. 4; transcaucasica acc. 2) were excluded. Again, sequential Mann–Whitney tests were not performed due to the low Bonferroni corrected significance level of = 0.005.

Sequence alignment and phylogenetic analysis
An updated phylogenetic hypothesis for the Orobanche-group based on nuclear ITS-sequences was obtained using the data set from Schneeweiss et al. (2004a) plus eight sequences from Carlón et al. (2005) . Alignment was done manually in the program BioEdit 5.0.9 (Hall, 1999 ). Insertions present in less than 20% of sequences were removed (in total 23 bp). Phylogenetic relationships were inferred using maximum likelihood as implemented in PAUP* 4.0.b10 (Swofford, 2001 ). The best-fit substitution model was determined using the Akaike information criterion (Akaike, 1974 ), as implemented in the program Modeltest 3.06 (Posada and Crandall, 1998 ), to be a general time reversible model with substitution rate heterogeneity accounted for by a gamma distribution (GTR + ). Using the model parameters suggested by Modeltest, a heuristic search was conducted with starting tree obtained via neighbor joining and subsequent tree-bisection-reconnection (TBR) branch swapping, keeping no more than 10 trees. The values of the model parameters were newly estimated on the most likely trees obtained, and the heuristic search was repeated using the new parameter values and the most likely trees as starting trees. This procedure was repeated until no change in likelihood scores in two consecutive runs was found. Nodal support was estimated via nonparametric bootstrap employing 100 replicates with the same settings as described.

Evolution of genome size
Although the relatively small number of accessions with available data for both genome size and ITS sequences warrants caution in interpreting the results, we believe that hypotheses can still be formulated that are worthy of further investigation. Because we had only one estimate for the genome size for a species in the basal lineages within the Phelipanche group (O. purpurea) and because the clade containing the majority of species (e.g., O. rosmarina and O. ramosa; Schneeweiss et al., 2004a ) was poorly resolved, we refrained from analyzing this group in a manner similar to that for the Orobanche group.

Evolution of genome size was investigated in the Orobanche group on one randomly chosen most likely tree using the generalized least squares method as implemented in the program Continuous (Pagel, 2004 ). Prior to analysis, we excluded Phelypaea and O. latisquama, the latter a polyploid with strongly reduced genome size (see Results, C- and Cx-values). Taxa for which no 2C-values are available, as well as polyploid accessions, were pruned from the tree. The following analyses were conducted on two different data sets (Fig. 3B): (1) one containing only accessions with both 2C-values and ITS sequences available and (2) one containing taxa with both 2C-values and ITS-sequences available. In case of more than one accession being measured for the 2C-value, mean 2C-values were used in the second data set, with the exception of O. anatolica, for which the respective 2C-value estimates were used for both accessions with sequence data. A polytomy involving Orobanche crenata, O. minor, O. owerinii, O. picridis, and O. transcaucasica was randomly resolved by introducing branches of length 10^–6. To test whether there is a dominant direction of change for a trait, two models of trait evolution are compared using the likelihood ratio statistics (Huelsenbeck and Rannala, 1997 ). Model A (constant-variance random walk model [= drift model] of evolution) has a single parameter, the instantaneous variance of evolution, while Model B (directional random walk model [= directional model] of evolution) additionally has a directional change parameter. The directional change parameter effectively measures the regression of trait values across species against total path length from the root of the tree to the tips and can thus detect any general trends towards a dominant direction of evolutionary change (e.g., species developed a larger genome). The program Continuous allows estimation of three scaling parameters for a given data set and phylogeny, denoted by lambda (), kappa (), and delta (), which describe phylogenetic associations, mode, and tempo of trait evolution, respectively. The parameter reveals whether the phylogeny correctly predicts the patterns of covariance among species on a given trait: if it is significantly different from zero, then some sort of phylogenetic correction is necessary. The parameter scales branch lengths and can be used to test punctual vs. gradual modes of trait evolution: if = 0, then trait evolution is independent of the length of the branch (punctual mode). Otherwise, > 0, some form of gradualism is suggested. The parameter scales overall path lengths in the phylogeny and can be used to test for adaptive radiation: if < 1, then shorter paths (earlier evolution in the phylogeny) contribute disproportionately to trait evolution (i.e., a rapid early evolution is followed by slower rates of change among closely related species), thus indicating adaptive radiation. If > 1, then longer paths contribute more to trait evolution, suggesting accelerated evolution. Again, these parameters are evaluated using likelihood ratio test statistics. Ancestral genome size was estimated using Model A (as suggested by the likelihood ratio test) and the maximum likelihood estimates of the transformation parameters , , and .

View larger version (47K): [in this window] [in a new window]   Fig. 3. Phylogenetic relationships of Orobanche sect. Orobanche inferred from maximum likelihood using Phelypaea as outgroup. (A) One of five equally most likely trees (score –ln = 3268.66219). Numbers at branches are bootstrap values >50. Names in bold indicate polyploid accessions (ploidy level of O. crinita is not yet known). Asterisks indicate accessions for which genome size data are available. (B) One randomly chosen most likely tree (outgroup taxa, exclusively polyploid taxa, and taxa without genome size data pruned from the tree) with polytomies randomly resolved and equal branch lengths. Numbers are 2C-values used in the analyses of the larger/smaller data sets (see Materials and Methods, Evolution of genome size, for details)

C- and Cx-values
Among group variation
The estimated 2C-values range from 2.9 pg in Orobanche cernua var. cumana to 19.9 pg in Cistanche phelypaea subsp. lutea. This approximately seven-fold difference is due to the large 2C-values of the genus Cistanche (16.8–19.9 pg), and reduces to approximately four-fold (3.8-fold) when only Orobanche species are considered (tetraploid O. transcaucasica with 2C-value of 11.6 pg and diploid O. anatolica with 2C-value of 10.9 pg). The 2C-values in Cistanche lie outside those of all other investigated taxa. Among the remaining lineages, pronounced gaps exist between O. anatolica and O. sect. Orobanche and between O. sect. Orobanche and sect. Trionychon (Welch's ANOVA, P = 0.001, Dunnett's C post hoc test). The 2C-values for Phelypaea are highly disparate, and no significant differences from any of the other groups are inferred (Fig. 1).

View larger version (30K): [in this window] [in a new window]   Fig. 1. Distribution of genome size (2Cx-values) in Orobanche and related genera. Note the gap in 2Cx-values between Cistanche and the remaining taxa

Phelypaea is a small SW Asian genus with only 2–3 species. It is closely related to Orobanche sect. Orobanche (Teryokhin et al., 1993 ; Schneeweiss et al., 2004a ), with which it shares the chromosome base number x = 19 (Schneeweiss et al., 2004b ), but from which it differs morphologically by its large, single, long pedicelled flowers with a shiny red corolla. The 2C-values of the two investigated species, P. coccinea and P. tournefortii, differ significantly: that of P. tournefortii is about twice as large as that of P. coccinea (10.52 vs. 4.85 and 5.85 pg [Appendix], the latter two being significantly different [Appendix S2, see online Supplemental Data]). This result is particularly puzzling, because both species are diploid (Schneeweiss et al., 2004b ; this study [Fig. 2]). We would expect the two species to have substantially different chromosome sizes. This cannot be proven with our data, because only meiotic plates are available for P. tournefortii, whereas in P. coccinea, mitotic plates could also be investigated. Further studies on P. tournefortii are necessary to firmly establish its relatively high 2C-value and to discover what causes it to increase which, without knowledge of chromosome number, would be highly suggestive of polyploidy.

Of the two American sections of Orobanche, material was only available from the larger sect. Myzorrhiza. All three taxa investigated in this survey are tetraploid (2n = 4x = 48; Schneeweiss et al., 2004b ). Their 2C-values range from 4.31– 5.06 pg, the 1Cx-values accordingly from 1.08–1.26 pg (Appendix). Although the 2C-values are statistically significantly different (Appendix S3, see Supplemental Data with online version of this article), the number of investigated taxa and accessions is too low to draw conclusions on interspecific differences.

The 2C-values and 2Cx-values in the exclusively diploid species (2n = 2x = 24; Schneeweiss et al., 2004b ) of sect. Trionychon range from 6.75–10.83 pg (Appendix), i.e., a 1.6-fold difference. 2C-values for accessions of one species are often not significantly different from 2C-values for accessions of other species (Appendix S4, see online Supplemental). Underrepresented in our survey are accessions of the basal lineages within this section, e.g., O. arenaria and O. nowackiana, and only one estimate for O. purpurea subsp. purpurea is available (8.84 pg), which lies within the range of 2C-values found in O. ramosa and its relatives.

The 2C-values in diploid members (2n = 2x = 38; Schneeweiss et al., 2004b ) of the largest group, Orobanche sect. Orobanche, range from 2.90 pg in O. cernua var. cumana to 7.13 pg in O. raddeana, i.e., a 2.5-fold difference (Appendix). Significant differences between accessions can be found both between species and within species (Appendix S5, see Supplemental Data with online version of this article). For most species, however, only one accession could be measured, warranting caution in the interpretation of these differences (see also Discussion). If polyploid species are included (2n = 4x = 76, 2n = 6x = 114; Schneeweiss et al., 2004b ), the 2C-value differences increase to four-fold due to the 2C-value of 11.61 pg in tetraploid O. transcaucasica. The majority of polyploid species found in the clade, including O. gracilis and relatives (Schneeweiss et al., 2004a ; this study), are among those taxa with the lowest 2C-values (3.31–4.89 pg) and, therefore, the lowest Cx-values within sect. Orobanche (1Cx = 0.83–1.22 pg in 4x, 0.62 pg in 6x). This is in contrast to the tetraploid accessions of O. transcaucasica, which have the largest 2C-values in this group (10.84–11.61 pg) but Cx-values similar to those of the diploid accessions of the same species (1Cx = 2.71–2.90 in 4x vs. 3.19 pg in 2x). Within each of these two groups (O. transcaucasica and O. gracilis and relatives), 2C-values are very similar and differences, if present, are not confined to between-species comparisons (Appendix S6, see online Supplemental Data).

A phylogenetically distinct lineage includes the Near Asian species O. anatolica and O. colorata (Schneeweiss et al., 2004a ), which are traditionally considered members of O. sect. Orobanche. Their distinctness is supported by having larger chromosomes (Schneeweiss et al., 2004b ) and correspondingly larger 2C-values, and—because both are diploid (2n = 2x = 38)—also have larger 2Cx-values of 9.37–10.88 pg (Appendix S7, see online Supplemental Data). A second phylogenetically distinct lineage traditionally placed in O. sect. Orobanche includes the SW-Mediterranean O. latisquama (= O. macrolepis; Schneeweiss et al., 2004a ). This species is exclusively tetraploid (Schneeweiss et al., 2004b ) and has a 2C-value of 6.39–6.76 pg (1Cx = 1.60–1.69 pg; Appendix S8, see online Supplemental Data), thus lying within the range of species in sect. Orobanche.

Phylogenetic relationships
An updated hypothesis on the phylogenetic relationships within the Orobanche-group is provided in Fig. 3A. In contrast to Schneeweiss et al. (2004a) , we analyzed this group separately from other lineages such as Cistanche or the Phelipanche group. Adjustment of the alignment by removal of indels not relevant for the respective group resulted in an aligned matrix of 602 bp, of which 255 are variable and 219 are parsimony-informative.

Phylogenetic relationships inferred from maximum likelihood are nearly identical to those obtained by Bayesian analysis in the previous study by Schneeweiss et al. (2004a) . One exception is the insufficiently supported position of Orobanche latisquama (= O. macrolepis) as sister to O. anatolica/colorata (bootstrap value 62). This position is due to the choice of Phelypaea as outgroup: if more distant outgroup taxa, such as Cistanche or Epifagus, are included, O. latisquama is suggested as sister to the remaining species of O. sect. Orobanche (data not shown). Most of the accessions not included in Schneeweiss et al. (2004a) belong to already represented species and fall into their respective groups, e.g., O. cernua subsp. desertorum with other accessions of O. cernua s. l. Sequences from one plant with shiny, dark-red corollas that was originally determined as O. crinita var. occidentalis (see Foley, 2001 ) are indistinguishable from the yellow-flowering O. densiflora. This, together with the co-occurrence of yellow- and red-flowered plants in the same population, even on the same host individual, supports the view of Carlón et al. (2005) that these represent color morphs of the same species rather than different species. Orobanche ballotae has only recently been described from southern Spain (Pujadas Salvà, 1997 ) and is restricted to Ballota hirsuta (Lamiaceae) as host. Interestingly, the southern Iberian O. ballotae appears to be phylogenetically distinct, if only weakly supported (bootstrap value 69), from O. minor and its relatives, while a morphologically and ecologically very similar race from Greece (designated as O. minor s. l. in Fig. 3A) growing on Ballota acetabulosa is not.

Evolution of genome size in the Orobanche group
Analyses of both data sets result in very similar parameter values. A directional model of evolution (model B) does not result in significantly higher likelihood scores than the drift model of evolution (model A; data not shown) indicating that there is no general trend toward genome size increase or decrease. Thus, the parameter values were estimated using model A. For both data sets, the maximum likelihood value of lambda () is 1, indicating that the phylogeny correctly predicts the patterns of covariance among species on a given trait. For both data sets, the maximum likelihood value of kappa () is 0, suggesting that trait evolution is independent of branch length (punctual mode of evolution). Finally, the maximum likelihood values for delta () are 0.235 for the larger and 0.280 for the smaller data set and in both cases they are significantly different from 1 (likelihood ratio test, df = 1, P < 0.05), suggestive of adaptive radiation. Ancestral 2C-values (i.e., the parameter alpha []) are estimated as 7.397 pg (variance of 4.079) and 7.409 pg (variance of 3.904) for the large and small data sets, respectively.

DISCUSSION

Reliability of measurements
Some of the material was stored over an extended period (up to 2 years) before being processed. Although this extended storage might have had negative effects on the accuracy of the obtained values, we are confident that this is not the case for our data for two reasons: (1) If properly stored, DNA leaking in fixed material is minimal (Greilhuber, 2005 ), and (2) in some cases, measurements of older accessions are higher than those from younger ones (e.g., O. tunetana). Nevertheless, we refrain from interpreting differences found between species and/or accessions within species, even if intraspecific genome size variation might be indicative of incipient speciation (Murray, 2005 ). In any case, measurements of more taxa and more accessions are clearly desirable.

Genome size and phylogeny
Genome size distribution is highly congruent with phylogenetically defined lineages (Fig. 1). In particular, this is the case for Orobanche sect. Orobanche and O. sect. Trionychon, which have very little overlap in their genome size ranges. The only exception is O. anatolica, which has considerably higher Cx-values than the rest of sect. Orobanche (Appendix), supporting its phylogenetic distinctness, as previously indicated by ITS sequence data (Schneeweiss et al., 2004a ). On the other hand, the Cx-value of the second phylogenetically distinct lineage within sect. Orobanche, O. latisquama, lies well within the range of the section (Appendix). However, the tendency of polyploid genomes to considerably reduce their genome size (see next paragraph) makes it difficult to evaluate its phylogenetic value for O. latisquama.

Evolution of genome size in polyploids
In both the Phelipanche and Orobanche groups, polyploids have not only smaller 1Cx-values than their diploid relatives, but even smaller 2C-values. This could be due to either an increase of genome size in diploids or a decrease of genome size in polyploids. The large ancestral genome size (2C) of 7.4 pg inferred for the Orobanche group clearly favors the latter hypothesis. Downsizing of the genome after polyploidization appears to be a general trend in angiosperms (Kellogg and Bennetzen, 2004 ; Leitch and Bennett, 2004 ). Mechanisms leading to a decrease in genome size include unequal crossing over (Wendel et al., 2002 ), illegitimate crossing over (Devos et al., 2002 ), a higher overall rate of deletions than insertions (Comeron, 2001 ; Petrov, 2001 , 2002 ), and selection against transposable elements (Wright and Schoen, 1999 ; Morgan, 2001 ).

An exception to this downsizing pattern is the tetraploid cytotype of O. transcaucasica, having 2C-values nearly twice those of the diploid accessions. This suggests that polyploidization in this species has occurred relatively recently and that there has not yet been time for substantial reductions in nuclear DNA-content. This hypothesis is supported by the lack of sequence divergence between diploid and tetraploid accessions (Fig. 3A). Alternatively, mechanisms causing major reductions in other polyploids might not be active in this species or only to a minor extent. Given the predominance of genome size reduction in polyploid lineages of Orobanche in general, this hypothesis appears less likely. Another alternative might be that the larger genome size of the tetraploid plants has selective advantage. At this point, we should mention that diploid O. transcaucasica parasitizes Rhus coriacea (Anacardiaceae) in dry shrublands, while tetraploid O. transcaucasica grows on Carpinus betulus (Betulaceae) in mesic broadleaved forests, suggesting substantial ecological differentiation.

As discussed in Schneeweiss et al. (2004b) , the basic chromosome number of Cistanche, x = 20, is likely to be of paleopolyploid origin. Deducing the putative genome size (2Cx) of an ancestor with x = 10 from the values found in current species, the value of roughly 8–10 pg would be in the range of Orobanche sect. Trionychon and not much higher than the ancestral genome size inferred for O. sect. Orobanche. It is therefore not possible to decide whether polyploidization in Cistanche has affected its genome size substantially and if so, in which direction.

Evolution of genome size in Orobanche sect. Orobanche
Although the small number of taxa included in this analysis evidently calls for caution in the interpretation of its results, it appears that, despite the lack of a general trend toward genome size increase or decrease, genome size evolved more rapidly in earlier phases of evolution of sect. Orobanche than in later ones. This is what would be intuitively expected given that the basal lineages within sect. Orobanche have the highest (O. anatolica) and the lowest (O. cernua s. l.) genome size values of the whole section. A rapid change of a trait in early evolutionary phases and a decreased rate of change in later phases might be indicative of adaptive radiation (Pagel, 2004 ). This might also be the case in O. sect. Orobanche, the most species-rich group of Orobanche, although a hypothesis on the evolutionary significance of nuclear DNA content in the early evolutionary phases of O. sect. Orobanche is still missing. Furthermore, genome size evolved in a punctual rather than in a gradual mode, thus fitting a scenario of early adaptive radiation. It must be kept in mind, however, that a similar distribution of trait evolution might have causes other than adaptive radiation, such as phylogenetically grouped differential susceptibility of species toward genome size changes via, e.g., retrotransposon activities or indel formation (Gregory, 2003 ). The inclusion of further species is necessary to test if this trend can be corroborated.

Adaptive role of genome size?
Most species of O. sect. Orobanche grow on perennial hosts; therefore, they are at least potentially perennial themselves. A few species, however, grow on annual hosts and must (at least facultatively) be annual themselves. When comparing appropriate sister-taxa in sect. Orobanche (O. cernua var. cernua vs. O. cernua var. cumana and O. owerinii vs. O. crenata), facultatively annual species have lower C-values, suggesting that a shift to an annual life history might be accompanied by a reduction in genome size (e.g., Gregory, 2002 ; Albach and Greilhuber, 2004 ). However, in sect. Trionychon (O. pulchella vs. O. hirtiflora, O. lavandulacea vs. O. mutelii) this trend is not observed. A denser sampling that is focused on these species pairs is necessary to evaluate how genome size changes in response to changing life history.

Conclusion
We present the first data set on genome size of one of the most important groups of holoparasitic plants, Orobanche, and related genera. Distribution of genome size agrees with phylogenetic lineages identified by analyses of sequence data, in particular the distinctness of O. anatolica from the rest of O. sect. Orobanche. Polyploid taxa are among those with the smallest 2C-values, suggesting substantial genome downsizing after polyploidization. The only exception is tetraploid O. transcaucasica with a DNA content nearly twice as high as that of the diploid race, which might be due to a recent origin of this polyploid. In O. sect. Orobanche, genome size evolved more rapidly in earlier stages of evolution than in later stages, which might indicate adaptive radiation. It will be rewarding to extend genome size measurements to include more taxa and accessions in order to test this hypothesis.

Taxon: Chromosome no.; 2C ± SD; 1Cx; Locality (herbarium: voucher number); Collector (unless one of the authors).

Cistanche phelypaea (L.) Coutinho subsp. lutea (Desj.) F. Casas & M. Lainz (acc. 1): 40; 18.108 ± 0.544; 9.054; Spain (WU: S&T 7656). C. phelypaea (L.) Coutinho subsp. lutea (Desj.) F. Casas & M. Lainz (acc. 2): 40; 18.372 ± 0.210; 9.186; Spain (WU: S&T 8744). C. phelypaea (L.) Coutinho subsp. lutea (Desj.) F. Casas & M. Lainz (acc. 3): 40*; 19.867 ± 0.413; 9.935; Spain (WU: S&T 8805). C. phelypaea (L.) Coutinho subsp. phelypaea (acc. 1): 40*; 16.841 ± 0.158, 8.420; Morocco (WU: S. Talavera et al. 19/03 M), K. Tremetsberger. C. phelypaea (L.) Coutinho subsp. phelypaea (acc. 2): 40; 17.851 ± 0.173; 8.926; Spain (WU: S&T 8734).

Phelypaea^a coccinea Poir. (acc. 1): 38; 4.851 ± 0.035; 2.425; Georgia (WU: S&T 7817). P.^a coccinea Poir. (acc. 2): 38; 5.849 ± 0.128; 2.925; Georgia (WU: S&T 8196). P.^a tournefortii Desf.: 38; 10.519 ± 0.132; 5.260; Turkey (WU: S&T 7764).

Orobanche californica Cham. & Schlecht. subsp. californica: 48; 4.313 ± 0.055; 1.078; USA (WTU: AC 02-37). O. californica Cham. & Schlecht. subsp. grandis Heckard: 48; 5.057 ± 0.121; 1.264; USA (WTU: AC 00-44). O. pinorum Geyer ex Hook.: 48; 4.640 ± 0.009; 1.160; USA (WU: S&T 8709). O. bungeana G. Beck: 24; 10.826 ± 0.063; 5.413; Turkey (WU: S&T 7768). O. cf. coelestis (Reut.) G. Beck: 24; 9.236 ± 0.079; 4.618; Turkey (WU: S&T 7767). O. coelestis (Reut.) G. Beck (acc. 1): 24; 8.750 ± 0.098; 4.375; Turkey (WU: S&T 7766). O. coelestis (Reut.) G. Beck (acc. 2): 24; 8.946 ± 0.108; 4.473; Turkey (WU: S&T 7769). O. cf. graciosa G. Beck: 24; 8.410 ± 0.043; 4.205; Canary Islands (WU: S&T 8757). O. hirtiflora (Reut.) Tzvel.^b (acc. 1): 24; 9.124 [9.072] ± 0.105; 4.562 [4.536]; Georgia (WU: S&T 6629). O. hirtiflora (Reut.) Tzvel.^b (acc. 2): 24; 9.599 ± 0.074; 4.800; Georgia (WU: S&T 6611). O. lavandulacea Rchb. (acc. 1): 24; 8.764 ± 0.086; 4.382; Italy (WU: S&T 7504). O. lavandulacea Rchb. (acc. 2): 24; 8.930 ± 0.033; 4.465; Italy (WU: S&T 7510). O. mutelii F. Schultz (acc. 1): 24; 8.386 ± 0.051; 4.193; Italy (WU: S&T 7505). O. mutelii F. Schultz (acc. 2): 24; 9.123 ± 0.167; 4.562; Spain (WU: S&T 8742). O. mutelii F. Schultz (acc. 3): 24; 9.774 ± 0.202; 4.887; Spain (WU: S&T 8728). O. nana (Noe ex Reut.) Nyman (acc. 1): 24; 6.750 ± 0.030; 3.375; Italy (WU: S&T 7484). O. nana (Noe ex Reut.) Nyman (acc. 2): 24; 6.862 0.055; 3.431; Italy (WU: S&T 7488). O. portoilicitana A. Pujadas & M. B. Crespo^c: 24; 8.881 ± 0.049; 4.440; Spain (WU: S&T 8782). O. pulchella (C. A. Mey.) Novopokr.: 24; 8.537 ± 0.076; 4.268; Georgia (WU: S&T 8581). O. purpurea Jacq.: 24; 8.840 ± 0.086; 4.420; Georgia (WU: S&T 8107). O. ramosa L. (acc. 1): 24; 8.685 [8.671] ± 0.063; 4.343 [4.336]; Georgia (WU: S&T 6880). O. ramosa L. (acc. 2): 24*; 8.756 ± 0.090; 4.378; Tunesia (WU: S&T 10694), L. Schratt-Ehrendorfer. O. ramosa L. (acc. 3): 24*; 9.421 ± 0.182; 4.710; Georgia (WU: S&T 6636). O. rosmarina G. Beck: 24; 7.092 ± 0.085; 3.546; Spain (WU: S&T 7615). O. tunetana G. Beck (acc. 1): 24; 7.238 ± 0.052; 3.619; Spain (WU: S&T 8745). O. tunetana G. Beck (acc. 2): 24; 7.997 ± 0.096; 3.998; Spain (WU: S&T 7558). O. alba Steph. ex Willd. (acc. 1): 38; 5.575 ± 0.172; 2.787; Croatia (Herb. Gutermann: 36563). O. alba Steph. ex Willd. (acc. 2): 38; 6.342 ± 0.126; 3.171; Austria (WU: S&T 6563). O. alsatica Kirschl. subsp. libanotidis (Rupr.) Pusch: 38; 5.215 ± 0.072; 2.607; Croatia (WU: S&T 6250). O. amethystea Thuill. (acc. 1): 38; 3.591 ± 0.074; 1.796; Spain (WU: S&T 7565). O. amethystea Thuill. (acc. 2): 38; 4.425 ± 0.110; 2.213; Spain (WU: S&T 7550). O. anatolica Reut. (acc. 1): 38*; 9.374 ± 0.128; 4.687; Turkey (WU: S&T 7761). O. anatolica Reut. (acc. 2): 38; 10.095 ± 0.289; 5.047; Turkey (WU: S&T 7762). O. anatolica Reut. (acc. 3): 38; 10.876 ± 0.094; 5.438; Turkey (WU: S&T 7765). O. austrohispanica M. J. Y. Foley: 76; 3.780 ± 0.053; 0.945; Spain (WU: S&T 7659). O. caryophyllacea Sm. (acc. 1): 38; 7.097 ± 0.166; 3.549; Austria (WU: S&T 6565). O. caryophyllacea Sm. (acc. 2): 38; 7.126 ± 0.088; 3.563; Georgia (WU: S&T 6653). O. cernua Loefl. var. cernua: 38; 3.424 ± 0.043; 1.712; Spain (WU: S&T 7548). O. cernua Loefl. var. cumana (Wallr.) G. Beck: 38; 2.900 ± 0.050; 1.450; Georgia (WU: S&T 6609). O. crenata Forsk.: 38*; 5.687 ± 0.065; 2.843; Spain (Herb. Gutermann: 37342). O. densiflora Salzm. ex Reut. (acc. 1): 76; 3.784 ± 0.130; 0.946; Spain (WU: S&T 8733). O. densiflora Salzm. ex Reut. (red-flowered form) (acc. 2): 76*; 3.991 ± 0.077; 0.998; Spain (Herb. Gutermann: 37439). O. densiflora Salzm. ex Reut. (acc. 3): 76*; 4.245 ± 0.089; 1.061; Morocco (WU: S. Talavera et al. 18bis/03 M), K. Tremetsberger. O. densiflora Salzm. ex Reut. (acc. 4): 76*; 4.400 ± 0.042; 1.100; Spain (Herb. Gutermann: 37440). O. foetida Poir.: 76*; 3.583 ± 0.055; 0.896; Morocco (WU: S. Talavera et al. 82/03 M), K. Tremetsberger. O. gracilis Sm. (tetraploid acc. 1): 76; 3.314 ± 0.041; 0.829; Croatia (WU: S&T 6244). O. gracilis Sm. (tetraploid acc. 2): 76; 3.605 ± 0.086; 0.901; Italy (WU: S&T 6368). O. gracilis Sm. (tetraploid acc. 3): 76*; 3.994 ± 0.054; 0.998; Spain (WU: S&T 8803). O. gracilis Sm. (tetraploid acc. 4): 76; 4.892 [4.902] ± 0.015; 1.223 [1.226]; Georgia (WU: S&T 7856). O. gracilis Sm. (hexaploid): 114; 3.733 ± 0.041; 0.622; Spain (WU: S&T 8746). O. grossheimii Novopokr.: 38; 7.087 ± 0.068; 3.544; Georgia (WU: S&T 8273). O. hederae Duby: 38; 4.652 ± 0.091; 2.326; Georgia (WU: S&T 7889). O. latisquama (F. Schultz) Batt.^d (acc. 1): 76*; 6.394 ± 0.077; 1.598; Spain (WU: S&T 8804). O. latisquama (F. Schultz) Batt.^d (acc. 2): 76 ; 6.759 ± 0.122; 1.690; Spain (WU: S&T 8743). O. lutea Baumg.: 38; 4.948 ± 0.052; 2.474; Georgia (WU: S&T 6966). O. minor Sutt. (acc. 1): 38; 3.658 ± 0.054; 1.829; Italy (WU: S&T 7495). O. minor Sutt. (acc. 2): 38; 4.731 ± 0.131; 2.366; Italy (WU: S&T 7497). O. minor Sutt. (acc. 3): 38; 5.446 ± 0.108; 2.723; Italy (WU: S&T 7496). O. owerinii (G. Beck) G. Beck (acc. 1): 38; 6.006 ± 0.144; 3.003; Georgia (WU: S&T 8262). O. owerinii (G. Beck) G. Beck (acc. 2): 38; 6.364 ± 0.071; 3.182; Georgia (WU: S&T 8475). O. picridis F. Schultz: 38; 5.570 ± 0.207; 2.785; Turkey (WU: S&T 7750). O. raddeana G. Beck: 38; 7.128 ± 0.040; 3.564; Georgia (WU: S&T 8041). O. rapum-genistae Thuill.: 38; 5.132 ± 0.022; 2.566; France (WU: S&T 6404). O. teucrii Holandre: 38; 6.946 ± 0.043; 3.473; Austria (WU: S&T 6564). O. transcaucasica Tzvel. (diploid): 38; 6.371 ± 0.052; 3.185; Turkey (WU: S&T 7753). O. transcaucasica Tzvel. (tetraploid acc. 1): 76; 10.843 ± 0.025; 2.711; Georgia (WU: S&T 7890). O. transcaucasica Tzvel. (tetraploid acc. 2): 76; 11.592 [11.580] ± 0.031; 2.898 [2.895]; Georgia (WU: S&T 7877). O. transcaucasica Tzvel. (tetraploid acc. 3): 76; 11.605 ± 0.041; 2.901; Georgia (WU: S&T 7860).

^a Published under its synonym Diphelypaea in Schneeweiss et al. (2004a , b ).

^b Published as O. cf. aegyptiaca in Schneeweiss et al. (2004a , b ).

^c Published as O. olbiensis in Schneeweiss et al. (2004b) ; see Pujadas-Salvà and Crespo (2004) .

^d Published under its synonym O. macrolepis in Schneeweiss et al. (2004a , b ).

View larger version (124K): [in this window] [in a new window]   Fig. 2. Light micrograph of meiotic chromosomes (diakinesis) of Phelypaea tournefortii. 2n = 38. Scale bar = 10 µm

¹

The authors thank everyone who helped collect material in the field, T. F. Stuessy for continuous support, A. Luck for improving the English, and two anonymous reviewers for helpful comments. This study was financially supported by the Austrian Science Fund (FWF P14352-BIO).

⁴ E-mail: gerald.schneeweiss{at}univie.ac.at

LITERATURE CITED

Akaike H. 1974 A new look at statistical model identification. IEEE Transactions on Automatic Control AC-19 716-723

Albach D. A. J. Greilhuber 2004 Genome size variation and evolution in Veronica. Annals of Botany 94: 897-911[Abstract/Free Full Text]

APG [Angiosperm Phylogeny Group]. 1998 An ordinal classification for the families of flowering plants. Annals of the Missouri Botanical Garden 85: 531-553[CrossRef][ISI]

APG II [Angiosperm Phylogeny Group II]. 2003 An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG II. Botanical Journal of the Linnean Society 141: 399-436[CrossRef][ISI]

Arabidopsis Genome Initiative. 2000 Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408: 796-815[CrossRef][Medline]

Beck-Mannagetta G. 1930 IV. 261. Orobanchaceae. In A. Engler [ed.], Das Pflanzenreich. Regni Vegetabilis Conspectus. Wilhelm Engelmann, Leipzig, Germany

Bennett M. D. 1971 The duration of meiosis. Proceedings of the Royal Society of London, B, Biological Sciences 178: 277-299

Bennett M. D. 1972 Nuclear DNA content and minimum generation time in herbaceous plants. Proceedings of the Royal Society of London, B, Biological Sciences 181: 109-135

Bennett M. D. 1977 The time and duration of meiosis. Philosophical Transactions of the Royal Society of London, B, Biological Sciences 277: 201-277[CrossRef]

Bennett M. D. I. J. Leitch 2004 Plant DNA C-values database, release 3.0. [online, December 2004], website: http://www.rbgkew.org.uk/cval/homepage.html

Bennett M. D. I. J. Leitch 2005 Nuclear DNA amounts in angiosperms: progress, problems and prospects. Annals of Botany 95: 45-90[Abstract/Free Full Text]

Bennett M. D. J. B. Smith 1976 Nuclear DNA amounts in angiosperms. Philosophical Transactions of the Royal Society of London, B, Biological Sciences 274: 227-274[CrossRef]

Bennett M. D. J. B. Smith 1991 Nuclear DNA amounts in angiosperms. Philosophical Transactions of the Royal Society of London, B, Biological Sciences 334: 309-345[CrossRef]

Bennett M. D. I. J. Leitch L. Hanson 1998 DNA amounts in two samples of angiosperm weeds. Annals of Botany 82: (Supplement A) 121-134[Abstract/Free Full Text]

Bennetzen J. L. 2000 Transposable element contributions to plant gene and genome evolution. Plant Molecular Biology 42: 251-269[CrossRef][ISI][Medline]

Bennetzen J. L. J. X. Ma K. M. Devos 2005 Mechanisms of recent genome size variation in flowering plants. Annals of Botany 95: 127-132[Abstract/Free Full Text]

Bharathan G. 1996 Reproductive development and nuclear DNA content in angiosperms. American Journal of Botany 83: 440-451[CrossRef][ISI]

Bowers J. E. B. A. Chapman J. Rong A. H. Paterson 2003 Unravelling angiosperm genome evolution by phylogenetic analysis of chromosomal duplication events. Nature 422: 433-438[CrossRef][Medline]

Bure P. Y.-F. Wang L. Horová J. Suda 2004 Genome size variation in Central European species of Cirsium (Compositae) and their natural hybrids. Annals of Botany 94: 353-363[Abstract/Free Full Text]

Carlón L. G. Gómez Casares M. Laínz G. Moreno Moral Ó. Sánchez Pedraja G. M. Schneeweiss 2005 Más, a propósito de algunas Orobanche L. y Phelipanche Pomel (Orobanchaceae) del oeste del Paleártico. Documentos Jardín Botánico Atlántico Gijón 3: 1-71

Cavalier-Smith T. 2005 Economy, speed and size matter: evolutionary forces driving nuclear genome miniaturization and expansion. Annals of Botany 95: 147-175[Abstract/Free Full Text]

Chase M. W. L. Hanson V. A. Albert W. M. Whitten N. H. Williams 2005 Life history evolution and genome size in subtribe Oncidiinae (Orchidaceae). Annals of Botany 95: 191-199[Abstract/Free Full Text]

Comeron J. M. 2001 What controls the length of noncoding DNA?. Current Opinion in Genetics and Development 11: 652-659[CrossRef][ISI][Medline]

Cox A. V. G. J. Abdelnour M. D. Bennett I. J. Leitch 1998 Genome size and karyotype evolution in the slipper orchids (Cypripedioideae: Orchidaceae). American Journal of Botany 85: 681-687[Abstract]

Devos K. M. J. K. M. Brown J. L. Bennetzen 2002 Genome size reduction through illegitimate recombination counteracts genome expansion in Arabidopsis. Genome Research 12: 1075-1079[Abstract/Free Full Text]

Dimitrova D. I. Ebert J. Greilhuber S. Kozhuharov 1999 Karyotype constancy and genome size variation in Bulgarian Crepis foetida s. l. (Asteraceae). Plant Systematics and Evolution 217: 245-257[CrossRef][ISI]

Felsenstein J. 1985 Phylogenies and the comparative method. American Naturalist 125: 1-15[CrossRef][ISI]

Foley M. J. Y. 2001 Orobanche L. In S. Castroviejo [ed.], Flora Iberica XIV, 32–72. Real Jardín Botánico, CSIC, Madrid, Spain

Govindaraju D. R. C. A. Cullis 1991 Modulation of genome size in plants: the influence of breeding systems and neighbourhood size. Evolutionary Trends in Plants 5: 43-51[ISI]

Gregory T. R. 2001 Coincidence, coevolution or causation? DNA content, cell size, and the C-value enigma. Biological Review 76: 65-101

Gregory T. R. 2002 Genome size and developmental complexity. Genetica 115: 131-146[CrossRef][ISI][Medline]

Gregory T. R. 2003 Insertion–deletion biases and the evolution of genome size. Gene 324: 15-34[CrossRef][ISI]

Gregory T. R. 2005 The C-value enigma in plants and animals: a review of parallels and an appeal for partnership. Annals of Botany 95: 133-146[Abstract/Free Full Text]

Greilhuber J. 2005 Intraspecific variation in genome size in angiosperms: identifying its existence. Annals of Botany 95: 91-98[Abstract/Free Full Text]

Greilhuber J. I. Ebert 1994 Genome size variation in Pisum sativum. Genome 37: 646-655[Medline]

Greilhuber J. E. M. Temsch 2001 Feulgen densitometry: some observations relevant to best practice in quantitative nuclear DNA content determination. Acta Botanica Croatica 60: 285-298

Greilhuber J. J. Doleel M. A. Lysák M. D. Bennett 2005 The origin, evolution and proposed stabilization of the terms ‘genome size’ and ‘C-value’ to describe nuclear DNA contents. Annals of Botany 95: 255-260[Abstract/Free Full Text]

Grotkopp E. M. Rejmánek M. J. Sanderson T. L. Rost 2004 Evolution of genome size in pines (Pinus) and its life-history correlates: supertree analyses. Evolution 58: 1705-1729[CrossRef][ISI][Medline]

Hall T. A. 1999 BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series 41: 95-98

Holub J. 1977 New names in Phanerogamae 6. Folia Geobotanica et Phytotaxonomica 12: 417-432

Holub J. 1990 Some taxonomic and nomenclatural changes within Orobanche s. l. (Orobanchaceae). Preslia 62: 193-198

Huelsenbeck J. P. B. Rannala 1997 Phylogenetic methods come of age: testing hypotheses in an evolutionary context. Science 276: 227-232[CrossRef][ISI][Medline]

Jakob S. S. A. Meister F. R. Blattner 2004 The considerable genome size variation in Hordeum species (Poaceae) is linked to phylogeny, life form, ecology, and speciation rates. Molecular Biology and Evolution 21: 860-869[Abstract/Free Full Text]

Kellogg E. A. 1998 Relationships of cereal crops and other grasses. Proceedings of the National Academy of Sciences, USA 95: 2005-2010[Abstract/Free Full Text]

Kellogg E. A. J. L. Bennetzen 2004 The evolution of nuclear genome structure in seed plants. American Journal of Botany 91: 1709-1725[Abstract/Free Full Text]

Kenton A. Y. P. J. Rudall A. R. Johnson 1986 Genome size variation in Sisyrinchium L. (Iridaceae) and the relationship to phenotype and habitat. Botanical Gazette 147: 342-354

Kidwell M. G. 2002 Transposable elements and the evolution of genome size in eukaryotes. Genetica 115: 49-63[CrossRef][ISI][Medline]

Knight C. A. D. D. Ackerly 2002 Variation in nuclear DNA content across environmental gradients: a quantile regression analysis. Ecology Letters 5: 66-76[CrossRef][ISI]

Leitch I. J. M. D. Bennett 1997 Polyploidy in angiosperms. Trends in Plant Science 2: 470-476[CrossRef][ISI]

Leitch I. J. M. D. Bennett 2004 Genome downsizing in polyploid plants. Biological Journal of the Linnean Society 82: 651-663[CrossRef][ISI]

Leitch I. J. D. E. Soltis P. S. Soltis M. D. Bennett 2005 Evolution of DNA amounts across land plants (Embryophyta). Annals of Botany 95: 207-217[Abstract/Free Full Text]

Ma J. K. M. Devos J. L. Bennetzen 2004 Analyses of LTR-retrotransposon structures reveal recent and rapid genomic DNA loss in rice. Genome Research 14: 860-869[Abstract/Free Full Text]

Matzk F. K. Hammer I. Schubert 2003 Coevolution of apomixis and genome size within the genus Hypericum. Sexual Plant Reproduction 16: 51-58[CrossRef][ISI]

Morgan M. T. 2001 Transposable element number in mixed mating populations. Genetical Research 77: 261-275[CrossRef][ISI][Medline]

Murray B. G. 2005 When does intraspecific C-value variation become taxonomically significant?. Annals of Botany 95: 119-125[Abstract/Free Full Text]

Musselman L. J. J. I. Yoder J. H. Westwood 2001 Parasitic plants: major problem to food crops. Science 293: 1434[Medline]

Ohri D. K. Pistrick 2001 Phenology and genome size in Allium L.: a tight correlation?. Plant Biology 3: 654-660[CrossRef]

Otto S. P. J. Whitton 2000 Polyploidy incidence and evolution. Annual Review of Genetics 34: 401-437[CrossRef][ISI][Medline]

Pagel M. 1997 Inferring evolutionary processes from phylogenies. Zoologica Scripta 26: 331-348[CrossRef][ISI]

Pagel M. 1999 Inferring the historical patterns of biological evolution. Nature 401: 877-884[CrossRef]

Pagel M. 2004 Continuous. Computer program and documentation available at website, http://www.rubic.rdg.ac.uk/meade/Mark/

Parker C. C. R. Riches 1993 Parasitic weeds of the world: biology and control. CAB International, Wallingford, UK

Petrov D. A. 2001 Evolution of genome size: new approaches to an old problem. Trends in Genetics 17: 23-28[CrossRef][ISI][Medline]

Petrov D. A. 2002 DNA loss and evolution of genome size in Drosophila. Genetica 115: 81-91[CrossRef][ISI][Medline]

Posada D. K. A. Crandall 1998 MODELTEST: testing the model of DNA substitution. Bioinformatics 14: 817-818[Abstract/Free Full Text]

Price H. J. 1976 Evolution of DNA content in higher plants. Botanical Review 42: 27-82

Price H. J. A. G. Sparrow A. F. Nauman 1973 Correlations between nuclear volume, cell volume, and DNA content in meristematic cells of herbaceous angiosperms. Experientia 29: 1028-1029[CrossRef][ISI]

Price H. J. S. L. Dillon G. Hodnett W. L. Rooney L. Ross J. S. Johnston 2005 Genome evolution in the genus Sorghum (Poaceae). Annals of Botany 95: 219-227[Abstract/Free Full Text]

Pujadas Salvà A. 1997 Orobanche ballotae A. Pujadas (Orobanchaceae), especie nueva. Acta Botanica Malacitana 22: 29-34

Pujadas Salvà A. M. B. Crespo 2004 A new species of Orobanche (Orobanchaceae) from south-eastern Spain. Botanical Journal of the Linnean Society 146: 97-102[CrossRef][ISI]

Qiu Y.-L. J. Lee F. Bernasconi-Quadroni D. E. Soltis P. S. Soltis M. Zanis E. A. Zimmer Z. Chen V. Savolainen M. W. Chase 1999 The earliest angiosperms: evidence from mitochondrial, plastid and nuclear genomes. Nature 402: 404-407[CrossRef]

Rees H. M. H. Hazarika 1969 Chromosome evolution in Lathyrus. In C. D. Darlington and K. R. Lewis [eds.], Chromosomes today, 158– 165. Oliver and Boyd, Edinburgh, UK

Schneeweiss G. M. A. Colwell J.-M. Park C.-G. Jang T. F. Stuessy 2004a Phylogeny of holoparasitic Orobanche (Orobanchaceae) inferred from nuclear ITS-sequences. Molecular Phylogenetics and Evolution 30: 465-478[CrossRef][ISI][Medline]

Schneeweiss G. M. T. Palomeque A. E. Colwell H. Weiss-Schneeweiss 2004b Chromosome numbers and karyotype evolution in holoparasitic Orobanche (Orobanchaceae) and related genera. American Journal of Botany 91: 439-448[Abstract/Free Full Text]

Soják J. 1972 Nomenklatorické poznámky (Phanerogamae). asopis Ná rodniho Muzea, odd