Variation In Flower Morphology Research Articles

First buckthorn (Rhamnaceae) fossil flowers from India

AbstractFossil leaves, fruits, and woods assigned to Rhamnaceae have been recorded from India, however, there are no previous reports of rhamnaceous fossil flowers from India. Here, we report the first fossil flowers in appreciable numbers from the early Eocene (Palana Formation) sedimentary sequences of Gurha Lignite Mine, Rajasthan, western India, and show that they are attributable to Rhamnaceae. We examined variation in flower morphology among extant and extinct rhamnaceous species as a basis for interpreting our fossil flowers. The specimens are small star‐like, pentamerous, actinomorphic, gamosepalous flowers with triangular, keeled sepals with acute apices, spathulate, short‐clawed petals alternating with sepals, shallow, five‐lobed nectary disc with prominent pentagonal outer margin, and a centrally placed globose ovary with stigma scar. The flowers co‐occur with ziziphoid leaves and are recognized as a new species Eopaliura indica Patel, Rana & Khan sp. nov. Because of aforesaid characteristics, fossil flowers are easily identifiable as Rhamnaceae and should provide a reliable means of documenting the evolutionary history of this family during the Cenozoic.

Flower morphology evolution in Jaborosa (Solanaceae): shape and size variation associated with contrasting pollination modes.

Flower morphology is considered an important factor in species diversification because it may influence the efficiency of pollination in different ways (e.g. attraction and mechanical fit with different groups of pollinators). In the present study, we quantified the variation in flower morphology (i.e. shape and size) of the diverse South American genus Jaborosa Juss. (Solanaceae) in relation to contrasting pollination modes: rewarding pollination either by moths or by generalist small insects versus brood-site deceptive pollination by saprophilous flies. We examined variations of flower morphology in frontal (pollinator attraction) and sagittal (functional fit with pollinators) views in 12 Jaborosa species using geometric morphometric methods and comparative approaches to infer whether flower shape evolution, not attributable to flower allometry or phylogenetic relationship, is associated with shifts in pollination modes. We found remarkable variation in flower morphology among both Jaborosa species and pollination modes, largely in sagittal view. Evolutionary trends in shape of fly-pollinated flowers were mainly attributable to changes in developmental trajectories. Variation in flower architecture facilitated differential pollen placement - on the proboscis of moths, and either on the back or ventral region of saprophilous flies - promoting diversification of the genus. Diversification of shape, independent of size, in most of the studied Jaborosa species would indicate adaptation to contrasting pollination modes.

Flower morphology variation in five species of Penstemon (Plantaginaceae) displaying Hymenoptera pollination syndrome

Background: Geographic distance promotes phenotypic variation by facilitating environmental distance, limiting gene flow, and exposing plants to different pollen vectors. Therefore, understanding how plant morphology changes across a geographic range improves our understanding of the drivers of morphological diversification both on a macro- and micro-evolutionary scale.
 Questions: 1) How do geographic location and abiotic factors affect flower morphology between populations? 2) Is there a geographic pattern of flower morphology variation? and 3) How does yearly variation in temperature and precipitation affect flower morphology within populations?
 Studied species: Penstemon albidus, P. fruticosus, P. glandulosus, P. speciosus, and P. whippleanus
 Study site and dates: The continental USA, summers of 2017 and 2018
 Methods: Fifty-seven populations and 496 individuals were selected at random to measure ten floral traits. Bioclimatic variables were extracted from the WorldClim database and NOAA. Linear models, partial least squares regression, Mantel tests and canonical correlation analysis were used to analyze the data.
 Results: Geographic variables alone explained a significant portion of the variation in flower morphology in two species, while in others, flower morphology did not vary despite large geographic distances. Penstemon albidus and P. whippleanus flowers increase in size from south-north, while P. glandulosus and P. speciosus exhibited an east-west increasing trend. Additionally, mean annual precipitation was the most important variable influencing P. glandulosusflower morphology.
 Conclusions: Geographic distance facilitates isolation-by-distance and isolation-by-phenology as well as abiotic differences between populations; however, other factors such as pollinators might be keeping populations morphologically homogeneous despite large geographic distance.

Delayed differentiation of epidermal cells walls can underlie pedomorphosis in plants: the case of pedomorphic petals in the hummingbird-pollinated Caiophora hibiscifolia (Loasaceae, subfam. Loasoideae) species

BackgroundUnderstanding the relationship between macroevolutionary diversity and variation in organism development is an important goal of evolutionary biology. Variation in the morphology of several plant and animal lineages is attributed to pedomorphosis, a case of heterochrony, where an ancestral juvenile shape is retained in an adult descendant. Pedomorphosis facilitated morphological adaptation in different plant lineages, but its cellular and molecular basis needs further exploration. Plant development differs from animal development in that cells are enclosed by cell walls and do not migrate. Moreover, in many plant lineages, the differentiated epidermis of leaves, and leaf-derived structures, such as petals, limits organ growth. We, therefore, proposed that pedomorphosis in leaves, and in leaf-derived structures, results from delayed differentiation of epidermal cells with respect to reproductive maturity. This idea was explored for petal evolution, given the importance of corolla morphology for angiosperm reproductive success.ResultsBy comparing cell morphology and transcriptional profiles between 5 mm flower buds and mature flowers of an entomophile and an ornitophile Loasoideae species (a lineage that experienced transitions from bee- to hummingbird-pollination), we show that evolution of pedomorphic petals of the ornithophile species likely involved delayed differentiation of epidermal cells with respect to flower maturity. We also found that developmental mechanisms other than pedomorphosis might have contributed to evolution of corolla morphology.ConclusionsOur results highlight a need for considering alternatives to the flower-centric perspective when studying the origin of variation in flower morphology, as this can be generated by developmental processes that are also shared with leaves.Graphical

Isolation and characterization of microsatellites markers in Centaurium grandiflorum ssp. boissieri.

Estimating outcrossing/selfing rates and characterizing genetic diversity with microsatellite markers are crucial to understanding the evolution of plant mating systems. We developed, optimized and characterized eight new primer pairs for Centaurium grandiflorum ssp. boissieri and transferred them to three subspecies of Centaurium quadrifolium. Two SSR loci were transferred from Sabatia campestris to the four Centaurium taxa. Polymorphisms, He, Ho and H-W deviations were estimated in two populations of C. grandiflorum ssp. boissieri and in seven individuals each of C. quadrifolium ssp. barrelieri, C. quadrifolium ssp. parviflorum and C. quadrifolium ssp. quadrifolium. A total of 80 individuals was used in these experiments. The number of polymorphic loci varied among species from one to ten. A total of 127 alleles was scored. The average number of alleles per locus was 12.7. He was higher than Ho in all sampled populations. Hardy-Weinberg equilibrium was found for some loci in different species. This is the first report of microsatellites successfully amplified in the whole Centaurium genus. They will be valuable for estimating mating system parameters and genetic diversity and exploring their relationships with the wide variation in flower morphology in the genus, especially anther-stigma separation.

Morphospace exploration reveals divergent fitness optima between plants and pollinators.

The obligate mutualism and exquisite specificity of many plant-pollinator interactions lead to the expectation that flower phenotypes (e.g., corolla tube length) and corresponding pollinator traits (e.g., hawkmoth proboscis length) are congruent as a result of coevolution by natural selection. However, the effect of variation in flower morphology on the fitness of plants and their pollinators has not been quantified systematically. In this study, we employed the theoretical morphospace paradigm using a combination of 3D printing, electronic sensing, and machine vision technologies to determine the influence of two flower morphological features (corolla curvature and nectary diameter) on the fitness of both parties: the artificial flower and its hawkmoth pollinator. Contrary to the expectation that the same flower morphology maximizes the fitness of both plant and pollinator, we found that the two parties have divergent optima for corolla curvature, with non-overlapping fitness peaks in flower morphospace. The divergent fitness optima between plants and pollinators could lead to evolutionary diversification in both groups.

Pollinator availability, mating system and variation in flower morphology in a tropical savanna tree

ABSTRACT Widely distributed organisms face different ecological scenarios throughout their range, which can potentially lead to micro-evolutionary differentiation at specific localities. Mating systems of animal pollinated plants are supposed to evolve in response to the availability of local pollinators, with consequent changes in flower morphology. We tested the relationship among pollination , mating system, and flower morphology over a large spatial scale in Brazilian savannas using the tree Curatella americana (Dilleniaceae). We compared fruit set with and without pollinators in the field, and analyzed pollen tube growth from self- and cross-pollinated flowers in different populations. Populations with higher natural fruit set also had lower fruit set in bagged flowers, suggesting stronger barriers to self-fertilization. Furthermore, higher levels of autogamy in field experiments were associated with more pollen tubes reaching ovules in self-pollinated flowers. Morphometric studies of floral and leaf traits indicate closer-set reproductive organs, larger stigmas and smaller anthers in populations with more autogamy. We show that the spatial variation in mating system, flower morphology and pollination previously described for herbs also applies to long-lived, perennial tropical trees, thus reemphasizing that mating systems are a population-based attribute that vary according to the ecological scenario where the plants occur.

Evolution of the selfing syndrome: Anther orientation and herkogamy together determine reproductive assurance in a self-compatible plant.

Capacity for autonomous self-fertilization provides reproductive assurance, has evolved repeatedly in the plant kingdom, and typically involves several changes in flower morphology and development (the selfing syndrome). Yet, the relative importance of different traits and trait combinations for efficient selfing and reproductive success in pollinator-poor environments is poorly known. In a series of experiments, we tested the importance of anther-stigma distance and the less studied trait anther orientation for efficiency of selfing in the perennial herb Arabis alpina. Variation in flower morphology among eight self-compatible European populations was correlated with efficiency of self-pollination and with pollen limitation in a common-garden experiment. To examine whether anther-stigma distance and anther orientation are subject to directional and/or correlational selection, and whether this is because these traits affect pollination success, we planted a segregating F2 population at two native field sites. Selection strongly favored a combination of introrse anthers and reduced anther-stigma distance at a site where pollinator activity was low, and supplemental hand-pollination demonstrated that this was largely because of their effect on securing self-pollination. The results suggest that concurrent shifts in more than one trait can be crucial for the evolution of efficient self-pollination and reproductive assurance in pollinator-poor habitats.

Morphological and chemical variation of Stemona tuberosa from southern China - Evidence for heterogeneity of this medicinal plant species.

The occurrence of bioactive alkaloids and tocopherols was studied in 15 different provenances of Stemona tuberosa Lour. collected in southern China, to examine chemical variation of individuals that show notable differences in flower characteristics. Morphological variations stimulated examination of chemical characteristics of these individuals. Methanolic root extracts of 15 individuals of S.tuberosa were comparatively assessed with HPLC-UV-DAD/ELSD. Five of seven compounds were co-chromatographically identified. Two compounds were isolated and their structure elucidated using NMR and MS. Amounts of alkaloids and tocopherols were determined using HPLC-UV-DAD/ELSD with the external standard method. Five alkaloids, tuberostemonine (1), tuberostemonine A (2), neotuberostemonine (3), tuberostemonine N (4), stemoninine (5) and two 3,4-dehydrotocopherol derivatives were identified. Within S.tuberosa alkaloid accumulation tends either towards tuberostemonine (1) or stemoninine (5). All individuals show a notable co-occurrence of compounds 1 or 5 and 3,4-dehydro-δ-tocopherol (6). These results coincide with differences in flower morphology of S.tuberosa. Stemona tuberosa, as defined in the Flora of China, shows a remarkable variation in flower morphology and additionally in the accumulation of alkaloids. The obtained data show the need for future species delimitation to either species or subspecies level.

Characterizing Floral Symmetry in the Core Goodeniaceae with Geometric Morphometrics.

Core Goodeniaceae is a clade of ~330 species primarily distributed in Australia. Considerable variation in flower morphology exists within this group and we aim to use geometric morphometrics to characterize this variation across the two major subclades: Scaevola sensu lato (s.l.) and Goodenia s.l., the latter of which was hypothesized to exhibit greater variability in floral symmetry form. We test the hypothesis that floral morphological variation can be adequately characterized by our morphometric approach, and that discrete groups of floral symmetry morphologies exist, which broadly correlate with subjectively determined groups. From 335 images of 44 species in the Core Goodeniaceae, two principal components were computed that describe >98% of variation in all datasets. Increasing values of PC1 ventralize the dorsal petals (increasing the angle between them), whereas increasing values of PC2 primarily ventralize the lateral petals (decreasing the angle between them). Manipulation of these two morphological “axes” alone was sufficient to recreate any of the general floral symmetry patterns in the Core Goodeniaceae. Goodenia s.l. exhibits greater variance than Scaevola s.l. in PC1 and PC2, and has a significantly lower mean value for PC1. Clustering clearly separates fan-flowers (with dorsal petals at least 120° separated) from the others, whereas the distinction between pseudo-radial and bilabiate clusters is less clear and may form a continuum rather than two distinct groups. Transitioning from the average fan-flower to the average non-fan-flower is described almost exclusively by PC1, whereas PC2 partially describes the transition between bilabiate and pseudo-radial morphologies. Our geometric morphometric method accurately models Core Goodeniaceae floral symmetry diversity.

Disruption of the distylous syndrome in Primula veris.

Distyly is a floral polymorphism characterized by the presence of two discrete morphs with reciprocal positioning of anthers and stigmas in flowers on different plants within the same population. Although reciprocal herkogamy and associated floral traits are generally thought to be discrete and strict polymorphisms, little is known about variation in floral traits related to the distylous syndrome within and among populations of a single species. In this study, variation in floral morphology and reciprocal positioning of the sexual organs in the distylous Primula veris (cowslip) is quantified. Data were collected in ten populations occurring in two contrasting habitat types (grasslands and forests), and for each population the average level of reciprocity was assessed, the strength of the self-incompatibility system was determined, and seed production under natural conditions was quantified. In grassland populations, flowers showed clear distyly with low and symmetric reciprocity indices at both the lower and upper level. In forests, P. veris produced larger flowers that showed strong deviations in stigma-anther separation, especially in the L-morph. This deviation was mainly driven by variation in stigma height, resulting in high and asymmetric reciprocity indices and the occurrence of several short-styled homostylous plants. Self-incompatibility was, however, strict in both habitats, and morph ratios did not differ significantly from isoplethy. The observed shift in reciprocity in forest populations was associated with a significant reduction in seed production in the L-morph. The results indicate that populations of P. veris show habitat-specific variation in flower morphology. Deviations from perfect reciprocal positioning of stigmas and anthers also translate into reduced seed production, suggesting that small changes in sexual organ reciprocity can have far-reaching ecological and evolutionary implications.

Covariation of flower traits and bird pollinator assemblages among populations of Kniphofia linearifolia (Asphodelaceae)

Globally, bird-pollinated plants can be separated into two groups, one consisting of species pollinated by specialist nectarivores, and the other of plants pollinated by occasional nectarivores. There are marked differences in nectar properties among the two groups, implying that there has been pollinator-mediated selection on these traits. This raises the possibility that variation in bird assemblages among populations of a plant species could lead to the evolution of intraspecific variation in floral traits. We examined this hypothesis in Kniphofia linearifolia, a common and widespread plant in southern Africa. Although bees are common visitors to flowers of this species, exclusion of birds from inflorescences led to significant reductions in seed set, indicating that the species is primarily bird-pollinated. We showed that bird pollinator assemblages differ markedly between five different populations of K. linearifolia, and that variation in flower morphology and nectar properties between these populations are associated with the dominant guild of bird visitors at each population. We identified two distinct morphotypes, based on corolla length, nectar volume and nectar concentration, which reflect the bird assemblages found in each type. Further work is needed to establish if a natural geographic mosaic of bird assemblages are the ultimate cause of differentiation in floral traits in this species.

Genetic differentiation and reproductive isolation of a naturally occurring floral homeotic mutant within a wild‐type population of Capsella bursa‐pastoris (Brassicaceae)

Apart from the common floral architecture in Brassicaceae, variation in flower morphology occurs in several genera within the family and is considered to affect speciation processes. We analysed genetic differentiation and flowering time variation of two floral variants of Capsella bursa-pastoris, the Spe variant and the wild-type, which occur sympatrically in a vineyard in southwest Germany. The Spe variant is characterized by an additional whorl of stamens instead of petals and was formerly classified as an independent taxon 'Capsella apetala' Opiz. Amplified fragment length polymorphism and allozyme analysis revealed a substantial genetic differentiation of the two floral variants and a higher genetic variation within the wild-type subpopulation compared with the Spe subpopulation. The low genetic variation in the mutant provided evidence of a recent local origin or recent introduction. Flowering time analysis indicated that, within the analysed population, the Spe variant flowers significantly later than the wild-type (P < 0.001). We conclude that the evolution and persistence of Spe within a wild-type population is facilitated by high selfing rates and been enhanced by a shift in flowering phenology. Hence, our data provide substantial evidence that the Spe phenotype has established itself as an isolated entity within a wild-type population and may thus serve as a model for the analysis of the evolutionary significance of homeotic mutants in wild populations.

Flavonoids chemistry of the adonis amurensis complex in eastern Asia

Flavonoids were isolated and identified from flowers and/or leaves of the Adonis amurensis complex. This complex has been a persistent source of taxonomic confusion due to the exclusive use of continuous variation in flower morphology for species definition and recognition. Nineteen flavonoids, including C‐glycosylflavones, O‐glycosylflavones, and flavonol O‐glycosides were characterized. The flavonoids isolated from flowers and leaves of the A. amurensis complex were not the same. The phenogram indicated three major clusters; A. multiflora, A. ramosa‐A. pseudoamurensis, and A. amurensis. A. amurensis was characterized by a higher flavonoid diversity composed of a variety of C‐glycosylflavones, while A. multiflora was generally characterized by a reduced flavonoid profiles. It is probable that evolutionary advancement is reflected in loss of some of C‐glycosylflavone and all of O‐glycosylflavone synthesis, resulting in A. multiflora within this series. The present study has demonstrated that the flavonoid chemistry may indeed be as variable as the morphological features, and no significant flavonoid differences at the species level were observed. It was suggested that parallel and/or radiate modes of evolution could have taken place within this complex.

Cold tolerance during early reproductive growth of chickpea ( Cicer arietinum L.): genetic variation in gamete development and function

Chilling temperatures at flowering cause floral abortion in most chickpea cultivars. Recent evaluations of germplasm showed distinct genotypic differences in pod and seed set at low temperature but the morpho-physiological basis for such variation is unclear. Observations in the field during cold spells of December and January, and at 15/5°C and 15/0°C (day/night) regimes in growth rooms showed distinct genetic variation in flower morphology, gamete development (viability and size of pollen and ovules) and function (pollen germination and tube growth, ovule viability and fertilization, etc.). The greater pod-setting ability of tolerant lines (ICCVs 88502 and 88503) than the sensitive cultivars (Annigeri, Pant G 114, etc.) was associated with a higher pollen vigour (germination and tube growth) and ovule viability at low temperature in the former. The number of ovules was not affected by cold stress in all cultivars/lines but pollen size and viability were reduced in Annigeri. The reduced ovule fertilization, associated with decline in pollen tube growth and ovule viability, was the major cause for poor seed set at low temperatures. The magnitude of effects on gamete function varied with cultivar/line and severity of stress. Function of both pollen and ovules was adversely affected in Annigeri in 15/5° and 15/0°C regimes. In contrast, the reduction in pollen vigour was more than in ovule viability in ICCV 88510 and Pant G 114 at 15/5°C. At 15/0°C, however, both pollen vigour and ovule viability were reduced. A small increase in pod set of sensitive genotypes from manual pollinations with pollen from plants in the warm regime further suggested that pollen function was more adversely affected than pistil function at moderately low temperatures.

Flower morphometry of Rhizophora mangle (Rhizophoraceae): geographical variation in Mexican populations

We explored the patterns of intra- and interpopulation variation in flower morphology of the red mangrove, Rhizophora mangle. Twelve populations in Mexico were studied: five from the Gulf of Mexico and the Caribbean Sea, and seven from the Pacific Coast. Six metric floral attributes were measured from a sample of 1370 flowers. Significant differences among populations were found for all six attributes. Because floral attributes were all correlated, scores derived from principal factor analysis were used to describe the variation in flower morphology. Two factors explained essentially all of the variance in flower morphology. Corolla and calyx size had a strong effect on factor 1, while gynoecium size had the higher effect on factor 2. Nested analyses of variance on the scores from both factors revealed significant differences among coasts, among populations within coasts, and among plants within populations. Nonetheless, this variation cannot be explained as a result of clinal variation, as indicated by a series of regression analyses. Cluster analysis (UPGMA) showed that a population from the Pacific coast was clustered together with those of the Atlantic, and the arrangement of populations within each coast showed no evident geographical pattern. We propose that frequent events of extinction and recolonization by a few individuals, followed by selfing, may produce differentiation among populations of red mangrove.

A New Name in Mexican Asclepias (Asclepiadaceae)

Asclepias lynchiana Fishbein is proposed as a new name for the species commonly known as A. grandiflora E. Fournier, a later homonym of A. grandiflora L. f. The other available name for this species, A. insignis (Brandegee) Woodson, is a later homonym ofA. insignis (Schlechter) Schlechter. Asclepias lynchiana has among the largest flowers of any North American species of Asclepias and is widely distributed from central to southern Mexico. In the course of phylogenetic study of Asclepias (Fishbein, 1996), it became apparent that the species commonly known as A. grandflora E. Fournier (see Woodson, 1954) lacks a legitimate name. Therefore, the following new name is proposed. Asclepias lynchiana Fishbein, nom. nov. Replaced name: Asclepias grandiflora E. Fournier, Ann. Sci. Nat. Bot., ser. 6, 14: 379. 1882; not Asclepias grandiflora L. f., Suppl. Plant. 170. 1781. TYPE: Mexico, Hahn s.n. (holotype, P not seen). Asclepiodora insignis Brandegee, Zoe 5: 253. 1908. Asclepias insignis (Brandegee) Woodson, Ann. Missouri Bot. Gard. 28: 207. 1941; not Asclepias insignis (Schlechter) Schlechter, J. Bot. 34: 455. 1896. TYPE: Mexico, Purpus 2624 (holotype, UC not seen; isotype, MO not seen). Asclepias lynchiana grows on rocky slopes, occasionally on limestone, in tropical deciduous forest, oak woodland, and pastures. It has been found in the Mexican states of Jalisco and Puebla south to Michoacin, Guerrero, and Oaxaca. Although widely distributed, populations of this species are apparently highly dispersed, judging by the relatively few specimens in herbaria. Some variation in flower morphology has been observed, especially in flower size (S. Lynch, pers. comm.), but no infraspecific taxa have been recognized. The species is closely related to A. glaucescens HBK and A. crocea Woodson, as noted by Woodson (1954), as well as to A. elata Bentham and A. mirifica Woodson (Fishbein, 1996). This species has been known as A. grandiflora E. Fournier; however, use of the specific epithet was predated by A. grandiflora L. f. [= Pachycarpus grandiflorus (L. f.) E. Meyer]. In recognition of the illegitimate status ofA. grandiflora E. Fournier, specimens at some herbaria have been annotated using the only other available name, A. insignis (Brandegee) Woodson. Unfortunately, this combination was published previously as A. insignis (Schlechter) Schlechter [= Pachycarpus transversalis (Schlechter) N. E. Brown; see Smith, 1988]. No other validly published names for this species have been found, necessitating the publication of the new name. It is appropriate to rename this magnificent milkweed in honor of Steven P. Lynch, Louisiana State University at Shreveport. Lynch is the foremost student of Mexican Asclepias, and his assistance and encouragement have been invaluable to my research on the genus. Acknowledgments. I thank John R. Reeder for his patient advice regarding taxonomic issues, and the staff at the University of Arizona Herbarium (ARIZ) for assistance in many aspects of my work. I am also grateful to the curators at MEXU, MO, and other herbaria for loans of specimens. This work was supported in part by funding from the Research Training Group in the Analysis of Biological Diversification, University of Arizona.

Induced changes in growth and flowering of chrysanthemum after irradiation and in vitro culture of pedicels and petal epidermis

To increase the genetic variation for yield in Chrysanthemum morifolium cv. White Spider, pedicel segments and petal epidermis were induced in vitro to regenerate adventitious buds either directly from the original tissue or indirectly via callus. Besides, pedicels were irradiated before in vitro culture with an X-ray dose of 8 Gy. All treatments yielded variation in flower morphology. Changes in flower colour were largely restricted to treatments involving irradiation. Treatments yielding many morphological variants also yielded most production variants. Plants regenerated from petals differed from those originating from pedicels. Clones were found that had retained the morphology of ‘White Spider’, but outperformed the controls in flowering time and flower number.

Corresponding patterns of geographic variation among populations ofSilene latifolia (=S. alba =S. pratensis) (Caryophyllaceae)

Morphological and biochemical data were analysed from 30 greenhouse-grown populations of EuropeanSilene latifolia. Six separate character sets (flavones, seed, pollen, capsules, male and female flower morphology) were used in the analyses. There was broad-scale congruence between trends of geographic variation in most character sets, with the populations being assigned to western (or southern and western) and eastern clusters. The eastern and western clusters abut along a transition zone that runs roughly from Belgium to the northern Balkans; this zone represents a region of relatively rapid change and contains populations intermediate between the eastern and western clusters. Variation in flower morphology was weak and discordant with variation in the other character sets. The origin and maintenance of the variation pattern is discussed in terms of migrational history and “hybrid zones”.