Molecular evidence for hybridization between invasive <Emphasis Type="Italic">Solidago canadensis</Emphasis> and native <Emphasis Type="Italic">S</Emphasis>. <Emphasis Type="Italic">virgaurea</Emphasis>

Hybridization between alien and native species is biologically very important and could lead to genetic erosion of native taxa. Solidago × niederederi was discovered over a century ago in Austria and described by Khek as a natural hybrid between the alien (nowadays regarded also as invasive) S. canadensis and native S. virgaurea. Although interspecific hybridization in the genus Solidago is considered to be relatively common, hybrid nature of S. × niederederi has not been independently proven using molecular tools, to date. Because proper identification of the parentage for the hybrid Solidago individuals solely based on morphological features can be misleading, in this paper we report an additive polymorphism pattern expressed in the ITS sequences obtained from individuals representing S. × niederederi, and confirm the previous hypothesis that the parental species of this hybrid are S. canadensis and S. virgaurea. Additionally, based on variability at the cpDNA rpl32-trnL locus, we showed that in natural populations hybridization occurs in both directions.     

Molecular identification of <Emphasis Type="Italic">Schoenoplectiella</Emphasis> species (Cyperaceae) by use of microsatellite markers

Over the past few decades, use of molecular markers for species delimitation has drastically increased. Schoenoplectiella Lye has been recognized as a taxonomically difficult genus because of its morphological simplicity and frequent interspecific hybridization. The main objective of this study was to clarify the taxonomic identities of eight Schoenoplectiella species by use of molecular markers. We also evaluated the genetic relationships among S. × trapezoidea, known as a natural hybrid, and its close relatives. We used six microsatellite markers for 44 individuals from 31 natural populations of eight Schoenoplectiella species in South Korea. Six microsatellite marker combinations generated 59 amplification-detectable bands, of which 33 were specifically detected in one or more individuals of each species. Cluster analysis revealed that the grouping was consistent with the taxonomically recognized species. Our results do not support the hybrid origin of S. × trapezoidea. Rather, they suggest that this species is more closely related to S. hotarui. The informative microsatellite markers enabled us to clarify the distinctions among Schoenoplectiella species from South Korea and to identify the genetic relationships among these species. The molecular signatures found suitable for accurate identification of Schoenoplectiella species can be reliably used for studies of the phylogeny and evolution of this genus.     

Hares in Corsica: high prevalence of <Emphasis Type="Italic">Lepus corsicanus</Emphasis> and hybridization with introduced <Emphasis Type="Italic">L. europaeus</Emphasis> and <Emphasis Type="Italic">L. granatensis</Emphasis>

The Italian hare, Lepus corsicanus, was first described in Corsica more than 100 years ago, but the knowledge on the status of the species in this island remains scarce. Moreover, frequent introductions of thousands of individuals from other hare species, namely Lepus europaeus and Lepus granatensis, into Corsica are known to have occurred and an updated assessment of the prevalence of L. corsicanus in Corsica is therefore of utmost importance. Here, to estimate the relative prevalence of the hare species present in Corsica, we conducted a molecular analysis on 67 samples collected by hunters between 2002 and 2007 in 36 Corsican communes. Sequencing of portions of the nuclear gene transferrin and of the control region of the mitochondrial DNA allowed classifying most of the collected samples as belonging to L. corsicanus (70.1%). Of the sampled Corsican communes, 86.1% contained this species, while only in 11.1%, L. europaeus was present. Three of the analyzed specimens showed an inconsistent molecular assignment between markers suggesting a hybrid origin: L. corsicanus × L. europaeus, L. corsicanus × L. granatensis, and L. europaeus × L. granatensis. The first two cases of hybridization had never been described in nature, even in studies focusing on hares from Italy where L. corsicanus and L. europaeus are often sympatric. These results stress the real risk of corrosion of the native gene pool of L. corsicanus via hybridization with introduced species. We highlight the need of urgently rethinking the management plan of hare populations in Corsica.     

Genetic population structure of <Emphasis Type="Italic">Osmunda japonica</Emphasis>, rheophilous <Emphasis Type="Italic">Osmunda lancea</Emphasis> and their hybrids

Rheophilous Osmunda lancea often hybridizes with a dryland ally, Osmunda japonica, to produce O. × intermedia, forming zonation in riverbanks and the adjacent dryland along flooding frequency clines. This study examined the genetic structure of populations consisting of O. × intermedia and the two parental species by analyzing ten nuclear DNA markers [six cleaved amplified polymorphic sequence (CAPS) markers and three simple sequence repeat (SSR) markers developed from an expressed sequence tag (EST) library, and the sequence of the glyceraldehyde-3-phosphate dehydrogenase gene GapCp] and chloroplast DNA sequences. The results suggest that the nuclear genes of O. japonica and O. lancea are genetically differentiated despite shared polymorphism in their chloroplast DNA sequences. This discrepancy may be attributable to natural selection and recent introgression, although it is not evident if introgression occurs between O. japonica and O. lancea in the examined populations. Our findings of putative F2 hybrids in O. × intermedia support its partial reproducibility, and also suggest that formation of later-generation hybrids generates morphological variation in O. × intermedia. O. lancea plants collected from geographically distant localities were genetically very similar, and it is suggested that O. lancea originated monotopically.     

The ectopic expression of <Emphasis Type="Italic">PttKN1</Emphasis> gene causes pleiotropic alternation of morphology in transgenic carnation (<Emphasis Type="Italic">Dianthus caryophyllus</Emphasis> L<Emphasis Type="Italic">.</Emphasis>)

Carnation (Dianthus caryophyllus L.) is one of the most important ornamental plants in the world. Though morphological modification of carnation is very important to its commercial value, there have been no relevant reports until now. PttKN1 (Populus tremula × Tremuoides knotted1), isolated from the vascular cambial region of hybrid aspen, is a novel member of KNOX gene family. In this paper, we transformed 35S:PttKN1 to carnation via Agrobacterium tumefaciens. All primary transformants subsequently obtained were placed into phenotypic categories and self-pollinated. A total of 32 T0 progeny with aberrant phenotypes were obtained. PCR assay proved the validity of these transgenic plants. Phenotypes of 32 35S:PttKN1 plants were distinct from those of wild-type plants, including: (1) modification of phyllotaxis (15/32): wild-type carnation was with typical opposite phyllotaxis, while transgenic plants displayed tricussate whorled and multiple-cussate whorled phyllotaxis. Irregular modification of phyllotaxis was also observed; (2) modification of stem (9/32): wild-type stems were round; however, some transgenic plants exhibited much thicker and flatter stem; (3) the whole transgenic plants of carnation (8/32) became dwarf. These morphological modifications of carnation indicate that we have successfully attained some novel lines of carnation. These lines can have potential practical applications. In conclusion, the selection of stably genetic lines is discussed.     

Using genetic and morphological analysis to distinguish endangered taxa from their hybrids with the cultivated exotic pest plant <Emphasis Type="Italic">Lantana strigocamara</Emphasis> (syn: <Emphasis Type="Italic">Lantana camara</Emphasis>)

Because highly invasive species can rapidly assimilate rare taxa, we questioned whether two Florida endangered Lantana depressa varieties existed 21 years after Sanders documented their widespread hybridization with exotic Lantana strigocamara, and whether morphological traits could accurately discriminate genetic individuals. Stepwise discriminant analysis of morphological characters discriminated the three taxa, correctly classifying 98, 91, 89% of L. strigocamara, L. depressa var. depressa, and var. floridana. Hybrids blurred taxonomic distinctions of varieties and reduced classification accuracy by 7–17%. Species-specific Random Fragment Length Polymorphism (RFLP-PCR) confirmed hybridization has occurred. Intersimple Sequence Repeat (ISSR) fingerprints analyzed with STRUCTURE identified three groups indicating introgression. Morphological traits significantly, but weakly correlated with q ratios (P = 0.0001; r ² = 0.45). Although L. strigocamara introgression is widespread and ongoing, wild populations contain individuals that are predominantly L. depressa genome, supporting actions to remove adventive L. strigocamara, prevent its sale, and promote sales of genetically confirmed natives.     

Genetic Variability of Thermal <Emphasis Type="Italic">Nymphaea</Emphasis> (Nymphaeaceae) Populations Based on ISSR Markers: Implications on Relationships,Hybridization, and Conservation

The globally widespread genus Nymphaea exhibits a wide range of morphological and taxonomical diversity. The intrusion of a cultivated variety by progressive propagation and its affect on aquatic habitat is demonstrated in this case study. We have studied the genetic diversity, population, and stand structure of the neophyte Nymphaea × ‘Panama Pacific’ as well as other species found in Lake Hévíz and dikes nearby using inter-simple sequence repeat (ISSR) markers. The ISSR assay revealed a low genetic variability for the small populations of Nymphaea caerulea, Nymphaea lotus var. thermalis, and a medium level for Nymphaea alba, Nymphaea rubra var. longiflora, and Nymphaea × ‘Panama Pacific’. The evolutionary genetic status of individuals found in the overlapping cultivation area of Nymphaea × ‘Panama Pacific’ and N. caerulea was affirmed to be of hybrid origin by reticulate network analysis and with morphological parameters. The Bayesian analysis of hybrid classes and the segregation of the ISSR markers also confirmed the hybrid origin of the individuals in question and revealed that they are falling into F2 or latter genotype frequency classes, indicating the viability and fertility of the hybrids. The set of analyzed species by phylogenetic network analysis of ISSR data has been divided into three major groups according to their evolutionary patterns (subg. Barachyceras, Lotos, and Nymphaea). Our results are in accordance with these three major subgenera within Nymphaea.     

Fecundity and feeding of <Emphasis Type="Italic">Galerucella calmariensis</Emphasis> and <Emphasis Type="Italic">G. pusilla</Emphasis> on <Emphasis Type="Italic">Lythrum salicaria</Emphasis>

Fecundity and feeding of two introduced sibling biological control species, Galerucella calmariensis and G. pusilla (Coleoptera: Chrysomelidae) on purple loosestrife, Lythrum salicaria L. (Lythraceae) were compared at constant temperatures of 12.5, 15, 20, 25, and 27.5 °C. Larval feeding was also carried out at 30 °C, but at this temperature, larvae developed only to the L2 stage and none pupated. Thus, data for this temperature were not used in the analysis. There were significant species × temperature interactions in fecundity. Of the two species, Galerucella pusilla laid more eggs. Although egg production of both species was lowest at 12.5 °C and increased to 20 °C, at higher temperatures, the two species reacted differently. From 25 to 27.5 °C, egg production decreased for G. pusilla, but G. calmariensis fecundity peaked at 27.5 °C. Significant temperature × species × life-stage interactions were also observed in feeding. For each species, the amount of feeding varied with temperature and stage of development. Galerucella pusilla adults consumed more foliage at 15, 20, and 27.5 °C. However, at 12.5 °C G. calmariensis adults fed more than G. pusilla. G. pusilla larvae consumed an average of 25% less foliage than G. calmariensis. The lower larval consumption of G. pusilla suggests that when food is limited, G. pusilla larvae may have a higher survival rate because of its ability to complete larval development with less food and produce more progeny due to its greater fecundity. When food is not limited neither species would have a competitive advantage and both species could coexist temporally and spatially. However, since G. calmariensis larvae consumed more leaf material, the larval stage of this species would have a greater impact on purple loosestrife than G. pusilla.     

An efficient method for <Emphasis Type="Italic">in vitro</Emphasis> regeneration of leafy spurge (<Emphasis Type="Italic">Euphorbia esula</Emphasis> L.)

Leafy spurge (Euphorbia esula L.) is a perennial, invasive weed used as a model to study invasive plant behavior, because molecular tools (such as a deep expressed sequence tag database and deoxyribonucleic acid microarrays) have been developed for this species. However, the lack of effective in vitro regeneration and genetic transformation systems has hampered molecular approaches to study leafy spurge. In this study, we describe an efficient in vitro regeneration system. Three highly regenerative lines were selected by screening the in vitro regeneration capabilities of stem explants of 162 seedlings. The effects of various culture conditions on in vitro regeneration were then evaluated based on explant competence to form calluses and shoots. High rates of shoot regeneration can be obtained using a growth medium containing 1× woody plant basal medium and 1× Murashige and Skoog (MS) basal salts, 1× MS vitamins, 1.11 μM 6-benzylaminopurine, 1.97 μM indole-3-butyric acid, and 3% sucrose, pH 5.6–5.8. After 30 d culture, multiple shoots formed either directly from the stem or indirectly from the callus. This method is a requisite for the development of genetic transformation systems for leafy spurge and may be used to develop in vitro regeneration techniques for other species in the Euphorbiaceae.     

Reclassification of Two <Emphasis Type="Italic">Peronospora</Emphasis> Species Parasitic on <Emphasis Type="Italic">Draba</Emphasis> in <Emphasis Type="Italic">Hyaloperonospora</Emphasis> Based on Morphological and Molecular Phylogenetic Data

On the family Brassicaceae, the causal agent responsible for downy mildew disease was originally regarded as a single species, Peronospora parasitica (now under Hyaloperonospora), but it was recently reconsidered to consist of many distinct species. In this study, 11 specimens of Peronospora drabae and P. norvegica parasitic on the genus Draba were investigated morphologically and molecularly. Pronounced differences in conidial sizes (P. drabae: 14–20 × 12.5–15.5 μm; P. norvegica: 20–29 × 15.5–22 μm) and 7.8% sequence distance between their ITS1-5.8S-ITS2 rDNA sequences confirmed their status as distinct species. Based on ITS phylogeny and morphology (monopodially branching conidiophores, flexuous to sigmoid ultimate branchlets, hyaline conidia and lobate haustoria), the two species unequivocally belong to the genus Hyaloperonospora and not to Peronospora to which they were previously assigned. Therefore, two new combinations, Hyaloperonospora drabae and H. norvegica, are proposed. The two taxa are illustrated and compared using the type specimen for H. norvegica and authentic specimens for H. drabae, which is lectotypified.     

Gene expression analysis of the ovary of hybrid females of <Emphasis Type="Italic">Xenopus laevis</Emphasis> and <Emphasis Type="Italic">X. muelleri</Emphasis>

Background  

Interspecific hybrids of frogs of the genus Xenopus result in sterile hybrid males and fertile hybrid females. Previous work has demonstrated a dramatic asymmetrical pattern of misexpression in hybrid males compared to the two parental species with relatively few genes misexpressed in comparisons of hybrids and the maternal species (X. laevis) and dramatically more genes misexpressed in hybrids compared to the paternal species (X. muelleri). In this work, we examine the gene expression pattern in hybrid females of X. laevis × X. muelleri to determine if this asymmetrical pattern of expression also occurs in hybrid females.     

Genetic Relationship of <Emphasis Type="Italic">Curcuma</Emphasis> Species from Northeast India Using PCR-Based Markers

Molecular genetic fingerprints of nine Curcuma species from Northeast India were developed using PCR-based markers. The aim involves elucidating there intra- and inter-specific genetic diversity important for utilization, management, and conservation. Twelve random amplified polymorphic DNA (RAPD), 19 Inter simple sequence repeats (ISSRs), and four amplified fragment length polymorphism (AFLP) primers produced 266 polymorphic fragments. ISSR confirmed maximum polymorphism of 98.55% whereas RAPD and AFLP showed 93.22 and 97.27%, respectively. Marker index and polymorphic information content varied in the range of 8.64–48.1, 19.75–48.14, and 25–28 and 0.17–0.48, 0.19–0.48, and 0.25–0.29 for RAPD, ISSR, and AFLP markers, respectively. The average value of number of observed alleles, number of effective alleles, mean Nei’s gene diversity, and Shannon’s information index were 1.93–1.98, 1.37–1.62, 0.23–0.36, and 0.38–0.50, respectively, for three DNA markers used. Dendrograms based on three molecular data using unweighted pair group method with arithmetic mean (UPGMA) was congruent and classified the Curcuma species into two major clusters. Cophenetic correlation coefficient between dendrogram and original similarity matrix were significant for RAPD (r = 0.96), ISSR (r = 0.94), and AFLP (r = 0.97). Clustering was further supported by principle coordinate analysis. High genetic polymorphism documented is significant for conservation and further improvement of Curcuma species.     

Population genetic structure of the endangered and endemic medicinal plant <Emphasis Type="Italic">Commiphora</Emphasis><Emphasis Type="Italic">wightii</Emphasis>

Commiphora wightii is a medicinally important endangered species endemic to the Thar Desert of Rajasthan, India and adjoining areas of Pakistan. The populations of this species are declining sharply because of its extensive use as a natural herb. Random amplified polymorphic DNA analysis was conducted to find the genetic variation among 7 populations of C. wightii. Of the 100 random primers screened, 44 primers yielded 220 loci. Statistical analysis indicated low genetic diversity (H _pop = 0.0958; I = 0.1498; mean polymorphic loci = 14.28%), and high genetic differentiation among the populations (G _ST = 0.3990; AMOVA Φ _ST of 0.3390; Bayesian θ ^(II) = 0.3002). The low genetic diversity may be due to geographic isolation and restricted gene flow (N _m = 0.7533) between the fragmented populations. Unsustainable utilization of the plant has fragmented the population continuum which served the purpose of genetic exchange between populations. Mantel’s test was performed which revealed a highly significant positive correlation between genetic and geographic distance (r ² = 0.614, P = 0.023) among the populations studied. Low variation can also be attributed to poor seed setting and the slow growth pattern of the species, which is also an apomict. In UPGMA dendrogram the Commiphora wightii samples were divided into two major and one minor cluster. These findings can serve as a guide to preserving the genetic resources of this medicinal plant species.     

Genetic relationships of <Emphasis Type="Italic">Osmanthus</Emphasis> based on ISSR-PCR

Osmanthus is an important gardening plant, but determination of germplasm resources, number of species and bipship among species of Osmanthus are still not clear. In this article, genetic diversity of Osmanthus was first studied by ISSR molecular marker. Based on 20 primers screened from 74 primers, 361 bands were obtained from 17 species, and 95.29 % bands were polymorphic. Furthermore, it was found that the genetic distance between O. attenuatus and O. heterophyllus was the highest (0.5559), and was the lowest (0.2351) between O. ×fortunei and O. heterophyllus. As shown in the dendrogram obtained by UPGMA method, 17 species of Osmanthus could be divided into group I and group II, the group I was composed of O. matsumuraanus, O. americanus, O. yunnanensis and O. attenuatus, and the other species belonged to the group II that was further divided into two subgroups apparently, which were not consistent with the classic classification system, but mostly similar with traditional viewpoint about bipship among species. Therefore, it was feasible and effective to study genetic diversity among species of Osmanthus by ISSR molecular marker.     

RAPD and microsatellite transferability studies in selected species of <Emphasis Type="Italic">Prosopis</Emphasis> (section Algarobia) with emphasis on <Emphasis Type="Italic">Prosopis juliflora</Emphasis> and <Emphasis Type="Italic">P. pallida</Emphasis>

The genus Prosopis (Leguminosae, Mimosoideae), comprises 44 species widely distributed in arid and semi-arid zones. Prosopis pallida (Humb. & Bonpl. ex Willd.) Kunth and P. juliflora (Sw.) DC. are the two species that are truly tropical apart from P. africana, which is native to tropical Africa (Pasiecznik et al. 2004), and they have been introduced widely beyond their native ranges. However, taxonomic confusion within the genus has hampered exploitation and better management of the species. The present study focusses primarily on evaluating the genetic relationship between Prosopis species from the section Algarobia, containing most species of economic importance, though P. tamarugo from section Strombocarpa is also included for comparison. In total, 12 Prosopis species and a putative P. pallida × P. chilensis hybrid were assessed for their genetic relationships based on RAPD markers and microsatellite transferability. The results show that P. pallida and P. juliflora are not closely related despite some morphological similarity. Evidence also agrees with previous studies which suggest that the grouping of series in section Algarobia is artificial.     

Early development and larval behaviour of two clingfishes, <Emphasis Type="Italic">Lepadogaster purpurea</Emphasis> and <Emphasis Type="Italic">Lepadogaster lepadogaster</Emphasis> (Pisces: Gobiesocidae)

The recent revision on the taxonomic status of Lepadogaster lepadogaster resulted in the division of this species into L. lepadogaster and L. purpurea, the clarification of each species’ distribution ranges and the elimination of L. zebrina (now in synonymy with L. lepadogaster). This new taxonomic status led to the need of clarifying the early development of the two species. Embryonic development lasted 21 days in L. purpurea at a mean temperature of 14.2°C, and 16 days in L. lepadogaster at a mean temperature of 16.5°C. Newly hatched larvae of both species measured 5.2 mm, had the mouth and anus opened, pigmented eyes and almost no yolk. At hatching and throughout development the two species can be distinguished by the ventral pigmentation which is absent in L. purpurea. The change to a benthic mode of life was gradual in both species, with larvae increasingly spending more time close to the bottom until definitely settling. Larval development lasted 33 days in L. purpurea at a mean temperature of 14.6°C and 18 days in L. lepadogaster at a mean temperature of 16.5°C. Locomotion and foraging behaviours are described for both species. L. lepadogaster showed a higher frequency of swimming and foraging behaviour when compared with L. purpurea.     

An intersubspecific genetic map of<Emphasis Type="Italic"> Lens</Emphasis>

A Lens map was developed based on the segregational analysis of five kinds of molecular and morphological genetic markers in 113 F₂ plants obtained from a single hybrid of Lens culinaris ssp. culinaris × L. c. ssp. orientalis. A total of 200 markers were used on the F₂ population, including 71 RAPDs, 39 ISSRs, 83 AFLPs, two SSRs and five morphological loci. The AFLP technique generated more polymorphic markers than any of the others, although AFLP markers also showed the highest proportion (29.1%) of distorted segregation. At a LOD score of 3.0, 161 markers were grouped into ten linkage groups covering 2,172.4 cM, with an average distance between markers of 15.87 cM. There were six large groups with 12 or more markers each, and four small groups with two or three markers each. Thirty-nine markers were unlinked. A tendency for markers to cluster in the central regions of large linkage groups was observed. Likewise, clusters of AFLP, ISSR or RAPD markers were also observed in some linkage groups, although RAPD markers were more evenly spaced along the linkage groups. In addition, two SSR, three RAPD and one ISSR markers segregated as codominant. ISSR markers are valuable tools for Lens genetic mapping and they have a high potential in the generation of saturated Lens maps.Communicated by H.C. Becker     

<Emphasis Type="Italic">Agrobacterium</Emphasis><Emphasis Type="Italic">tumefaciens</Emphasis>-mediated transformation of callus cells of <Emphasis Type="Italic">Crataegus</Emphasis><Emphasis Type="Italic">aronia</Emphasis>

A genetic transformation system has been developed for callus cells of Crataegus aronia using Agrobacterium tumefaciens. Callus culture was established from internodal stem segments incubated on Murashige and Skoog (MS) medium supplemented with 5 mg l⁻¹ Indole-3-butyric acid (IBA) and 0.5 mg l⁻¹ 6-benzyladenine (BA). In order to optimize the callus culture system with respect to callus growth and coloration, different types and concentrations of plant growth regulators were tested. Results indicated that the best average fresh weight of red colored callus was obtained on MS medium supplemented with 2 mg l⁻¹ 2,4-dichlorophenoxyacetic acid (2,4-D) and 1.5 mg l⁻¹ kinetin (Kin) (callus maintenance medium). Callus cells were co-cultivated with Agrobacterium harboring the binary plasmid pCAMBIA1302 carrying the mgfp5 and hygromycin phosphotransferase (hptII) genes conferring green fluorescent protein (GFP) activity and hygromycin resistance, respectively. Putative transgenic calli were obtained 4 weeks after incubation of the co-cultivated explants onto maintenance medium supplemented with 50 mg l⁻¹ hygromycin. Molecular analysis confirmed the integration of the transgenes in transformed callus. To our knowledge, this is the first time to report an Agrobacterium-mediated transformation system in Crataegus aronia.     

Post-pollination hybridization barriers in <Emphasis Type="Italic">Nicotiana</Emphasis> section <Emphasis Type="Italic">Alatae</Emphasis>

Nicotiana section Alatae contains eight species with variable flower sizes and morphologies. Section members readily hybridize in the glasshouse, but no hybrids have been observed in natural sympatric and parapatric populations. To investigate interspecific crossing relationships with respect to mechanisms preventing hybridization, all members of section Alatae were intercrossed in a complete diallel. We found positive correlation between the pistil length of the pollen donor and interspecific seed set relative to the conspecific cross. Pollen tube growth rate and pollen donor pistil length were positively correlated as well. Furthermore, pollen from short-pistil members of section Alatae could only grow a maximum distance proportional to, but greater than, their own pistil lengths. Our results show that pollen tube growth capacity (i.e., rate and distance), provides a hybridization barrier in long-pistil species × short-pistil species crosses. We also found another hybridization barrier not specifically related to pollen tube growth capacity in short-pistil species × long-pistil species. Taken together, these barriers can generally be described by a ‘pistil-length mismatch’ rule; in section Alatae, pollen has the most success fertilizing ovules from species with pistil lengths similar to their own. This rule could contribute to hybridization barriers in Section Alatae because the species display dramatically different pistil lengths. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.