Breeding systems in Cyrtanthus (Amaryllidaceae): variation in self‐sterility and potential for ovule discounting

• Breeding systems of plants determine their reliance on pollinators and ability to produce seeds following self‐pollination. Self‐sterility, where ovules that are penetrated by self‐pollen tubes that do not develop into seeds, is usually considered to represent either a system of late‐acting self‐incompatibility or strong early inbreeding depression. Importantly, it can lead to impaired female function through ovule or seed discounting when stigmas receive mixtures of self and cross pollen, unless cross pollen is able to reach the ovary ahead of self pollen (‘prepotency’). Self‐sterility associated with ovule penetration by self‐pollen tubes appears to be widespread among the Amaryllidaceae.
• We tested for self‐sterility in three Cyrtanthus species – C. contractus, C. ventricosus and C. mackenii – by means of controlled hand‐pollination experiments. To determine the growth rates and frequency of ovule penetration by self‐ versus cross‐pollen tubes, we used fluorescence microscopy to examine flowers of C. contractus harvested 24, 48 and 72 h after pollination, in conjunction with a novel method of processing these images digitally. To test the potential for ovule discounting (loss of cross‐fertilisation opportunities when ovules are disabled by self‐pollination), we pollinated flowers of C. contractus and C. mackenii with mixtures of self‐ and cross pollen.
• We recorded full self‐sterility for C. contractus and C. ventricosus, and partial self‐sterility for C. mackenii. In C. contractus, we found no differences in the growth rates of self‐ and cross‐pollen tubes, nor in the proportions of ovules penetrated by self‐ and cross‐pollen tubes. In this species, seed set was depressed (relative to cross‐pollinated controls) when flowers received a mixture of self and cross pollen, but this was not the case for C. mackenii.
• These results reveal variation in breeding systems among Cyrtanthus species and highlight the potential for gender conflict in self‐sterile species in which ovules are penetrated and disabled by pollen tubes from self pollen.

     

Post‐pollination process in a partially self‐compatible distylous plant,Primula sieboldii (Primulaceae)

Pollen germination and pollen‐tube growth under natural conditions were observed in a population of a distylous species, Primula sieboldii, in which partial self‐compatibility has been demonstrated in some long‐styled genets. We observed post‐pollination processes microscopically in styles collected after self‐morph and inter‐morph hand pollination (with standardized pollen load on the stigmas) in four genets each from the following three ‘genet types’: self‐incompatible long‐styled (SI), partially self‐compatible long‐styled (SC) and self‐incompatible short‐styled morph genets. Irrespective of the genet type, pollen germination began within 24 h after pollination and tubes of pollen reached to the style base with 48–96 h after inter‐morph pollination. Although pollen tubes germinated after self‐pollination in the SC genets, the number of germinated pollen tubes was significantly lower than in the case of inter‐morph pollination. Few pollen tubes germinated after self‐pollination of the SI or short‐styled genets. In SC genets, the rate of pollen‐tube growth did not differ between self‐morph and inter‐morph pollination (～1.9 mm/day). Therefore, differences in self‐compatibility between SC and SI genets in P. sieboldii are likely to be attributable to differential pollen germination rates rather than to differential pollen‐tube growth rates.     

The Skp1‐like protein SSK1 is required for cross‐pollen compatibility in S‐RNase‐based self‐incompatibility

The self‐incompatibility (SI) response occurs widely in flowering plants as a means of preventing self‐fertilization. In these self/non‐self discrimination systems, plant pistils reject self or genetically related pollen. In the Solanaceae, Plantaginaceae and Rosaceae, pistil‐secreted S‐RNases enter the pollen tube and function as cytotoxins to specifically arrest self‐pollen tube growth. Recent studies have revealed that the S‐locus F‐box (SLF) protein controls the pollen expression of SI in these families. However, the precise role of SLF remains largely unknown. Here we report that PhSSK1 (Petunia hybrida SLF‐interacting Skp1‐like1), an equivalent of AhSSK1 of Antirrhinum hispanicum, is expressed specifically in pollen and acts as an adaptor in an SCF(Skp1‐Cullin1‐F‐box)^SLF complex, indicating that this pollen‐specific SSK1‐SLF interaction occurs in both Petunia and Antirrhinum, two species from the Solanaceae and Plantaginaceae, respectively. Substantial reduction of PhSSK1 in pollen reduced cross‐pollen compatibility (CPC) in the S‐RNase‐based SI response, suggesting that the pollen S determinant contributes to inhibiting rather than protecting the S‐RNase activity, at least in solanaceous plants. Furthermore, our results provide an example that a specific Skp1‐like protein other than the known conserved ones can be recruited into a canonical SCF complex as an adaptor.     

Pollen S–locus F–box proteins of Petunia involved in S–RNase‐based self‐incompatibility are themselves subject to ubiquitin‐mediated degradation

Many flowering plants show self‐incompatibility, an intra‐specific reproductive barrier by which pistils reject self‐pollen to prevent inbreeding and accept non‐self pollen to promote out‐crossing. In Petunia, the polymorphic S–locus determines self/non‐self recognition. The locus contains a gene encoding an S–RNase, which controls pistil specificity, and multiple S‐locus F‐box (SLF) genes that collectively control pollen specificity. Each SLF is a component of an SCF (Skp1/Cullin/F‐box) complex that is responsible for mediating degradation of non‐self S‐RNase(s), with which the SLF interacts, via the ubiquitin–26S proteasome pathway. A complete set of SLFs is required to detoxify all non‐self S‐RNases to allow cross‐compatible pollination. Here, we show that SLF1 of Petunia inflata is itself subject to degradation via the ubiquitin–26S proteasome pathway, and identify an 18 amino acid sequence in the C‐terminal region of S₂‐SLF1 (SLF1 of S₂ haplotype) that contains a degradation motif. Seven of the 18 amino acids are conserved among all 17 SLF proteins of S₂ haplotype and S₃ haplotype involved in pollen specificity, suggesting that all SLF proteins are probably subject to similar degradation. Deleting the 18 amino acid sequence from S₂‐SLF1 stabilized the protein but abolished its function in self‐incompatibility, suggesting that dynamic cycling of SLF proteins is an integral part of their function in self‐incompatibility.     

Cryptic self‐incompatibility and distyly in Hedyotis acutangula Champ. (Rubiaceae)

Distyly, floral polymorphism frequently associated with reciprocal herkogamy, self‐ and intramorph incompatibility and secondary dimorphism, constitutes an important sexual system in the Rubiaceae. Here we report an unusual kind of distyly associated with self‐ and/or intramorph compatibility in a perennial herb, Hedyotis acutangula. Floral morphology, ancillary dimorphisms and compatibility of the two morphs were studied. H. acutangula did not exhibit precise reciprocal herkogamy, but this did not affect the equality of floral morphs in the population, as usually found in distylous plants. Both pin and thrum pollen retained relatively high viability for 8 h. The pollen to ovule ratio was 72.5 in pin flowers and 54.4 in thrum flowers. Pistils of pin flowers remained receptive for longer than those of thrum flowers. No apparent difference in the germination rate of pin and thrum pollen grains was observed when cultured in vitro, although growth of thrum pollen tubes was much faster than that of pin pollen tubes. Artificial pollination revealed that pollen tube growth in legitimate intermorph crosses was faster than in either intramorph crosses or self‐pollination, suggesting the occurrence of cryptic self‐incompatibility in this species. Cryptic self‐incompatibility functioned differently in the two morphs, with pollen tube growth rates after legitimate and illegitimate pollination much more highly differentiated in pin flowers than in thrum flowers. No fruit was produced in emasculated netted flowers, suggesting the absence of apomixis. Our results indicate that H. acutangula is distylous, with a cryptic self‐incompatibility breeding system.     

Ubiquitin–proteasome‐mediated degradation of S‐RNase in a solanaceous cross‐compatibility reaction

Many plants have a self‐incompatibility (SI) system in which the rejection of self‐pollen is determined by multiple haplotypes at a single locus, termed S. In the Solanaceae, each haplotype encodes a single ribonuclease (S‐RNase) and multiple S‐locus F‐box proteins (SLFs), which function as the pistil and pollen SI determinants, respectively. S‐RNase is cytotoxic to self‐pollen, whereas SLFs are thought to collaboratively recognize non‐self S‐RNases in cross‐pollen and detoxify them via the ubiquitination pathway. However, the actual mechanism of detoxification remains unknown. Here we isolate the components of a SCF^SLF (SCF = SKP1‐CUL1‐F‐box‐RBX1) from Petunia pollen. The SCF^SLF polyubiquitinates a subset of non‐self S‐RNases in vitro. The polyubiquitinated S‐RNases are degraded in the pollen extract, which is attenuated by a proteasome inhibitor. Our findings suggest that multiple SCF^SLF complexes in cross‐pollen polyubiquitinate non‐self S‐RNases, resulting in their degradation by the proteasome.     

Electrostatic potentials of the S‐locus F‐box proteins contribute to the pollen S specificity in self‐incompatibility in Petunia hybrida

Self‐incompatibility (SI) is a self/non‐self discrimination system found widely in angiosperms and, in many species, is controlled by a single polymorphic S‐locus. In the Solanaceae, Rosaceae and Plantaginaceae, the S‐locus encodes a single S‐RNase and a cluster of S‐locus F‐box (SLF) proteins to control the pistil and pollen expression of SI, respectively. Previous studies have shown that their cytosolic interactions determine their recognition specificity, but the physical force between their interactions remains unclear. In this study, we show that the electrostatic potentials of SLF contribute to the pollen S specificity through a physical mechanism of ‘like charges repel and unlike charges attract’ between SLFs and S‐RNases in Petunia hybrida. Strikingly, the alteration of a single C‐terminal amino acid of SLF reversed its surface electrostatic potentials and subsequently the pollen S specificity. Collectively, our results reveal that the electrostatic potentials act as a major physical force between cytosolic SLFs and S‐RNases, providing a mechanistic insight into the self/non‐self discrimination between cytosolic proteins in angiosperms.     

Apple MdABCF assists in the transportation of S‐RNase into pollen tubes

Self‐incompatibility (SI) is a reproductive isolation mechanism in flowering plants. Plants in the Solanaceae, Rosaceae and Plantaginaceae belong to the gametophytic self‐incompatibility type. S‐RNase, which is encoded by a female‐specific gene located at the S locus, degrades RNA in the pollen tube and causes SI. Recent studies have provided evidence that S‐RNase is transported non‐selectively into the pollen tube, but have not specified how this transportation is accomplished. We show here that the apple (Malus domestica) MdABCF protein, which belongs to group F of the ABC transporter family, assists in transportation of S‐RNase into the pollen tube. MdABCF is located in the pollen tube membrane and interacts with S‐RNase. S‐RNase was unable to enter the pollen tube when MdABCF was silenced by antisense oligonucleotide transfection. Our results show that MdABCF assists in transportation of either self or non‐self S‐RNase into the pollen tube. Moreover, MdABCF coordinates with the cytoskeleton to transport S‐RNase. Blockage of S‐RNase transport disrupts self‐incompatibility in this system.     

Characterization of self-compatibility in sweet cherry varieties by crossing experiments and molecular genetic analysis

Self-compatibility is a major breeding objective in sweet cherry. The identification and characterization of new sources of self-compatibility will be useful for breeding and research purposes. In this work, self-compatibility of four local Spanish sweet cherry varieties was investigated by crossing experiments and molecular genetic analysis of two self-incompatibility loci. Crossing experiments included self- and cross-pollinations in the laboratory followed by microscopic observation of pollen tube growth and fruit set assay in the field. After crossing experiments, two accessions, ‘Son Miró’ and ‘Talegal Ahín’, were self-compatible while the other two were self-incompatible. Inheritance of S-locus and microsatellite EMPaS02 (linked to self-compatibility, Sc) were investigated in self-pollination progeny of both self-compatible genotypes. Results indicate that self-compatibility in ‘Talegal Ahín’ is similar to self-compatibility described in sweet cherry ‘Cristobalina’ and may be caused by the same mutation. That is a pollen part mutation not linked to the S-locus but linked to microsatellite EMPaS02 in cherry LG3. In ‘Son Miró’ self-compatibility seems more complex, affecting pollen and style function, and probably involving more than one mutation not described previously in sweet cherry. Together with ‘Cristobalina’, the newly described self-compatible varieties ‘Son Miró’ and ‘Talegal Ahín’ confirm the existence of unique self-compatible plant material in local germplasm from Spain that should be conserved and characterized for its use in breeding and research.     

Evolution towards self‐compatibility when mates are limited

Theory of plant mating system evolution predicts the spread of self‐compatibility (SC) in a predominantly self‐incompatible population when inbreeding depression (ID; the decline in fitness because of selfing) is small and when compatible mates are limited. I tested these two predictions by measuring the occurrence of SC in 13 natural populations of Ranunculus reptans L. that varied in ID and frequency of cross‐incompatible mates. Enforced selfing experiments were conducted in 2 years. In the first year, self‐pollination was applied at two flower ages to investigate the occurrence of delayed SC. I found that SC was not uncommon across all populations, but self‐compatible plants usually produced few seeds. There was no evidence for delayed SC. The occurrence of SC was not associated with population‐level ID, but populations with more limited availability of compatible mates had a significantly higher frequency of plants that were at least partially self‐compatible. The results indicate that, in R. reptans, a shortage of available mates in small populations may cause the evolution of partial SC and mixed mating.     

Genetic and molecular analysis in Cristobalina sweet cherry,a spontaneous self-compatible mutant

Self-compatibility in a naturally self-incompatible species like sweet cherry is a highly interesting trait for breeding purposes and a powerful tool with which to investigate the basis of the self-incompatible reaction in gametophytic systems. However, natural self-compatibility in sweet cherry is a very rare phenomenon. Cristobalina is a local Spanish sweet cherry cultivar that has proven to be spontaneously self-compatible. In this work, the nature of the self-compatibility in Cristobalina has been studied using genetic and molecular approaches. Pollination studies and microscopic observations of pollen tube growth were carried out to confirm the self-compatible character and the results obtained indicate that self-compatibility is caused by a failure of the pollen and not the style factor. Polymerase chain reaction (PCR) analysis of progenies derived from Cristobalina revealed that self-compatibility in this genotype is not related uniquely to one of the two pollen S alleles, but that pollen grains carrying either of the two haplotypes can overcome the incompatibility barrier. Moreover, PCR analysis and microscopic observation of pollen tube growth in progeny derived from Cristobalina also confirmed that the self-compatible descendants can carry either of the two S haplotypes of their progenitor. Isolation and sequencing of the style S-RNases and pollen SFBs revealed that the DNA sequences of these factors are the same as those described in other self-incompatible sweet cherry cultivars with the same S alleles. Possible mechanisms to explain self-compatibility in Cristobalina are discussed.     

Homomorphic Self‐Incompatibility,with Stylar Rejection Site,in Luehea grandiflora Mart. and Zucc. (Tiliaceae‐Malvaceae s.l.)

Abstract: The breeding system of Luehea grandiflora (Tiliaceae‐Malvaceae s.l.) was investigated using hand pollinations and fluorescence microscopy studies of pollen tube growth. Although selfed flowers persisted for some 10 days, our study indicates that L. grandiflora is self‐incompatible, with self pollen tube inhibition in the upper style, as occurs in many taxa with homomorphic, gametophytic self‐incompatibility (GSI). L. grandiflora is only the second species reported within the Malvales with homomorphic stylar inhibition. This result is discussed within the context of a report for self‐compatibility in this species, and we also consider the phylogenetic implications for the occurrence of GSI in the family Malvaceae s.l.     

Paternal‐specific S‐allele transmission in sweet cherry (Prunus avium L.): the potential for sexual selection

Homomorphic self‐incompatibility is a well‐studied example of a physiological process that is thought to increase population diversity and reduce the expression of inbreeding depression. Whereas theoretical models predict the presence of a large number of S‐haplotypes with equal frequencies at equilibrium, unequal allele frequencies have been repeatedly reported and attributed to sampling effects, population structure, demographic perturbation, sheltered deleterious mutations or selection pressure on linked genes. However, it is unclear to what extent unequal segregations are the results of gametophytic or sexual selection. Although these two forces are difficult to disentangle, testing S‐alleles in the offspring of controlled crosses provides an opportunity to separate these two phenomena. In this work, segregation and transmission of S‐alleles have been characterized in progenies of mixed donors and fully compatible pollinations under field conditions in Prunus avium. Seed set patterns and pollen performance have also been characterized. The results reveal paternal‐specific distorted transmission of S‐alleles in most of the crosses. Interestingly, S‐allele segregation within any given paternal or maternal S‐locus was random. Observations on pollen germination, pollen tube growth rate, pollen tube cohort size, seed set dynamics and transmission patterns strongly suggest post‐pollination, prezygotic sexual selection, with male–male competition as the most likely mechanism. According to these results, post‐pollination sexual selection takes precedence over frequency‐dependent selection in explaining unequal S‐haplotype frequencies.     

Effects of pollen availability and the mutation bias on the fixation of mutations disabling the male specificity of self‐incompatibility

The evolution of self‐compatibility (SC) by the loss of self‐incompatibility (SI) is regarded as one of the most frequent transitions in flowering plants. SI systems are generally characterized by specific interactions between the male and female specificity genes encoded at the S‐locus. Recent empirical studies have revealed that the evolution of SC is often driven by male SC‐conferring mutations at the S‐locus rather than by female mutations. In this study, using a forward simulation model, we compared the fixation probabilities of male vs. female SC‐conferring mutations at the S‐locus. We explicitly considered the effects of pollen availability in the population and bias in the occurrence of SC‐conferring mutations on the male and female specificity genes. We found that male SC‐conferring mutations were indeed more likely to be fixed than were female SC‐conferring mutations in a wide range of parameters. This pattern was particularly strong when pollen availability was relatively high. Under such a condition, even if the occurrence of mutations was biased strongly towards the female specificity gene, male SC‐conferring mutations were much more often fixed. Our study demonstrates that fixation probabilities of those two types of mutation vary strongly depending on ecological and genetic conditions, although both types result in the same evolutionary consequence—the loss of SI.     

Reproductive system and fitness of Vriesea friburgensis,a self‐sterile bromeliad species

Reproductive biology and plant fertility are directly related to many aspects of plant evolution and conservation biology. Vriesea friburgensis is an epiphytic and terrestrial bromeliad endemic to the Brazilian Atlantic rainforest. Hand‐pollination experiments were used to examine the reproductive system in a wild population of V. friburgensis. Plant fertility was assigned considering flower production, fruit and seed set, seed germination, and pollen viability. Self‐sterility observed from spontaneous selfing and manual self‐pollination treatments may be the consequence of late‐acting self‐incompatibility. Hand‐pollination results indicated no pollen limitation in the population studied. Floral biology features such as a few daily open flowers, nectar production, and sugar concentration corroborate hummingbirds as effective pollinators, although bees were also documented as pollinators. Components of fitness such as high flower, fruit, and seed production together with high seed and pollen viability indicate that this wild population is viable. From a conservation point of view, we highlight that this self‐sterile species depends on pollinator services to maintain its population fitness and viability through cross‐pollination. Currently, pollinators are not limited in this population of V. friburgensis. Conversely, the maintenance and continuous conservation of this community is essential for preserving this plant–pollinator mutualism.     

Self‐pollination and parthenocarpic ability in developing ovaries of self‐incompatible Clementine mandarins (Citrus clementina)

This study aimed to determine if self‐pollination is needed to trigger facultative parthenocarpy in self‐incompatible Clementine mandarins (Citrus clementina Hort. ex Tan.). ‘Marisol’ and ‘Clemenules’ mandarins were selected, and self‐pollinated and un‐pollinated flowers from both cultivars were used for comparison. These mandarins are always seedless after self‐pollination and show high and low ability to develop substantial parthenocarpic fruits, respectively. The time‐course for pollen grain germination, tube growth and ovule abortion was analyzed as well as that for carbohydrates, active gibberellins (GA₁ and GA₄), auxin (IAA) and abscisic acid (ABA) content in the ovary. ‘Clemenules’ showed higher pollen grain germination, but pollen tube development was arrested in the upper style 9 days after pollination in both cultivars. Self‐pollination did not stimulate parthenocarpy, whereas both un‐pollinated and self‐pollinated ovaries set fruit regardless of the cultivar. On the other hand, ‘Marisol’ un‐pollinated flowers showed greater parthenocarpic ovary growth than ‘Clemenules’ un‐pollinated flowers, i.e. higher ovule abortion rate (+21%), higher fruit set (+44%) and higher fruit weight (+50%). Further, the greater parthenocarpic ability of ‘Marisol’ paralleled higher levels of GA₁ in the ovary (+34% at anthesis). ‘Marisol’ ovary also showed higher hexoses and starch mobilization, but lower ABA levels (?64% at anthesis). Self‐pollination did not modify carbohydrates or GA content in the ovary compared to un‐pollination. Results indicate that parthenocarpy in the Clementine mandarin is pollination‐independent with its ability to set depending on the ovary hormone levels. These findings suggest that parthenocarpy in fertile self‐incompatible mandarins is constitutively regulated.     

Autonomous self‐pollination in the nectarless orchid Pogonia minor

The pollination biology of the nectarless orchid Pogonia minor was investigated in central Japan. The investigation revealed that the solitary flowers failed to attract pollinators, while high rates of fruit set were observed in the natural population. Comparable levels of fruit set were obtained in bagged, artificial self‐pollinated and artificial cross‐pollinated plants, indicating that the species is not pollinator‐limited for fruit set under natural conditions. Autonomous self‐pollination in P. minor resulted from a reduced rostellum, which allowed contact between the pollinia and the stigma. Self‐pollination is thought to be an adaptive response that provides reproductive assurance under conditions of pollinator limitation. Since pollen limitation is generally known to be frequent among deceptive orchids, strong pollen limitation is probably a driving force in the autonomous self‐pollination mechanism in the nectarless orchid P. minor.     

Breeding systems in Clivia (Amaryllidaceae): late‐acting self‐incompatibility and its functional consequences

Late‐acting (ovarian) self‐incompatibility, characterized by minimal or zero seed production following self‐pollen tube growth to the ovules, is expected to show phylogenetic clustering, but can otherwise be difficult to distinguish from early‐acting inbreeding depression. In Amaryllidaceae, late‐acting self‐incompatibility has been proposed for Narcissus (Narcisseae) and Cyrtanthus (Cyrtantheae). Here, we investigate whether it occurs in the horticulturally important genus Clivia (Haemantheae) and test whether species in this genus experience ovule discounting in wild populations. Seed‐set results following controlled hand pollinations revealed that Clivia miniata and C. gardenii are largely self‐sterile. Self‐ and cross‐pollinated flowers of both species had similar proportions of pollen tubes entering the ovary, and those of C. gardenii also did not differ in the proportions of pollen tubes that penetrated ovules, thus ruling out classical gametophytic self‐incompatibility acting in the style, but not early inbreeding depression. Flowers that received equal mixtures of self‐ and cross‐pollen set fewer seeds than those that received cross‐pollen only, but it was unclear whether this effect was a result of ovule discounting or interactions on the stigma. The prevention of self‐pollination by the emasculation of either single flowers or whole inflorescences in wild populations did not affect seed set, suggesting that ovule discounting is not a major natural limitation on seed production. Flowers typically produce one to three large fleshy seeds from approximately 16 available ovules, even when supplementally hand pollinated, suggesting that fecundity is mostly resource limited. The results of this study suggest that Clivia spp. are largely self‐sterile as a result of either a late‐acting self‐incompatibility system or severe early inbreeding depression, but ovule discounting caused by self‐pollination is not a major constraint on fecundity. © 2014 The Linnean Society of London, Botanical Journal of the Linnean Society, 2014, 175 , 155–168.     

The differential contributions of herkogamy and dichogamy as mechanisms of avoiding self‐interference in four self‐incompatible Epimedium species

Self‐interference is one of the most important selective forces in shaping floral evolution. Herkogamy and dichogamy both can achieve reductions in the extent of self‐interference, but they may have different roles in minimizing self‐interference in a single species. We used four self‐incompatible Epimedium species to explore the roles of herkogamy and dichogamy in avoiding self‐interference and to test the hypothesis that herkogamy and dichogamy may be separated and become selected preferentially in the taxa. Two species (E. franchetii and E. mikinorii) expressed strong herkogamy and weak protogyny (adichogamy), whereas another two species (E. sutchuenense and E. leptorrhizum) expressed slight herkogamy and partial protandry. Field investigations indicated that there was no physical self‐interference between male function and female function regarding pollen removal and pollen deposition in all species. Self‐pollination (autonomous or facilitated) was greater in species with slight herkogamy than in those with strong herkogamy. Artificial pollination treatments revealed that self‐pollination could reduce outcrossed female fertility in all species, and we found evidence that self‐interference reduced seed set in E. sutchuenense and E. leptorrhizum in the field, but not in E. franchetii and E. mikinorii. These results indicate that well‐developed herkogamy is more effective compared with dichogamy in avoiding self‐interference in the four species. In genus Epimedium, herkogamy instead of dichogamy should be selected preferentially and evolved as an effective mechanism for avoiding self‐interference and might not need to evolve linked with dichogamy.