Reproductive versatility in legumes: the case of amphicarpy in Trifolium polymorphum

Amphicarpy is a fascinating reproductive strategy, defined as fruit produced both below the soil surface and as aerial fruit on the same plant. Trifolium polymorphum is a grassland species subject to herbivory that combines amphicarpy with vegetative reproduction through stolons. Underground flowers have been described as obligate autogamous and aerial ones as self‐compatible allogamous, with aerial floral traits favouring cross‐pollination. In the present work we performed different pollination treatments on aerial flowers to analyse rates of pollen tube development and offspring fitness, measured as fruit set, seed production and germination percentage. This last variable was compared to that of seeds produced underground. No significant differences were found between fruit set in self‐ and cross‐pollinations. Seed production was higher in self‐pollinations, which is consistent with the higher rate of pollen tube development observed in self‐crosses. Spontaneous self‐pollination is limited in aerial flowers; thus pollen transfer by means of a vector is required even within the same flower. Germination tests showed that aerial seeds produced after self‐ and cross‐pollination did not differ in fitness, but underground seeds had higher germination percentage than aerial ones. Thus, we conclude that T. polymorphum has a mixed mating system. In grasslands with heavy grazing pressure, clonal propagation and underground seed production ensure persistence in the field. An intermediate level of selfing in aerial flowers ensures offspring, but morphological (herkogamy) and functional (dicogamy) floral traits maintain a window to incorporate genetic variability, allowing the species to tolerate temporal and spatial pressures.     

The reproductive assurance benefit of selfing: importance of flower size and population size

Autonomous selfing can provide reproductive assurance (RA) for flowering plants that are unattractive to pollinators or in environments that are pollen limited. Pollen limitation may result from the breakdown of once-continuous habitat into smaller, more isolated patches (habitat fragmentation) if fragmentation negatively impacts pollinator populations. Here we quantify the levels of pollen limitation and RA among large and small populations of Collinsia parviflora, a wildflower with inter-population variation in flower size. We found that none of the populations were pollen limited, as pollen-supplemented and intact flowers did not differ in seed production. There was a significant effect of flower size on RA; intact flowers (can self) produced significantly more seeds than emasculated flowers (require pollen delivery) in small-flowered plants but not large-flowered plants. Population size nested within flower size did not significantly affect RA, but there was a large difference between our two replicate populations for large-flowered, small populations and small-flowered, large populations that appears related to a more variable pollination environment under these conditions. In fact, levels of RA were strongly negatively correlated with rates of pollinator visitation, whereby infrequent visitation by pollinators yielded high levels of RA via autonomous selfing, but there was no benefit of autonomous selfing when visitation rates were high. These results suggest that autonomous selfing may be adaptive in fragmented habitats or other ecological circumstances that affect pollinator visitation rates.     

Delayed autonomous self-pollination in Hibiscus laevis (Malvaceae)

Delayed autonomous self-pollination allows outcrossing to occur while also ensuring that seeds are produced in the absence of pollen vectors. We investigated variation in the efficacy of this pollination mechanism in populations of Hibiscus laevis. Recurvature of stylar branches occurred after 1 d of anthesis, and in plants from Ohio, Illinois, Arkansas, and Oklahoma this behavior resulted in autonomous selfing (surprisingly, stylar movement was facultative in that it did not take place when the stigmas were already pollinated). In contrast to these more northern populations, the distance between anthers and stigmas was too great to allow autonomous selfing in plants from Texas, Mississippi, and Louisiana. Greenhouse studies of plants from Ohio demonstrated that autonomous selfing resulted in an average of 29.5 seeds per flower, as compared to 59.9 seeds per flower from hand-pollination of stigmas with self pollen. In an assessment of the possible significance of this selfing mode, emasculated flowers did not set significantly fewer seeds in a natural stand in Ohio, suggesting that few seeds resulted from autonomous selfing at that site. Modest inbreeding depression was detected at this population. Our results suggest that delayed autonomous selfing is more common in northern populations, where it may facilitate population establishment and persistence at times when pollinators are scarce.     

Ovule pre-emption and pollen limitation in a self-fertile perennial herb (Blandfordia grandiflora,Liliaceae)

The extent, frequency and causes of pollenlimited seed production were examined in partially selffertile populations of Blandfordia grandiflora for 2 years. Percentage seed set of open-pollinated plants (50–57%) did not differ within or between years, and was about 19% less than experimentally cross-pollinated plants (70–75%). Floral visits by honeybees did not differ through the flowering season and the number of pollen grains deposited on stigmas within 12 h of flowers opening exceeded the number of ovules per flower, indicating that the quality rather than the quantity of pollination limited seed set. Pollen limitation was caused by concurrent self- and cross-pollination and the subsequent abortion of some selfed ovules due to inbreeding depression. Natural seed set (55%) was intermediate between selfed (43%) and crossed (75%) flowers and was not increased when flowers that had been available to pollinators for 24 h were hand cross-pollinated, suggesting that ovules were already fertilized. Similarly, experimental pollination with both cross and self pollen within 24 h of flowers opening did not increase seed set relative to natural seed set, indicating that both cross- and self-fertilizations had occurred. In contrast, when selfing followed crossing by 48 h, or vice versa, seed set did not differ from crossed-only or selfed-only flowers, respectively, indicating that ovules were pre-empted by the first pollination. Collectively, these results indicate that under natural conditions self pollen pre-empts ovules, rendering them unavailable for cross-fertilization. This selfing reduces fecundity by 50%, as estimated from the natural production of cross seeds when selfing was prevented. Consequently, selection should favour floral traits, such as increased stigma-anther separation or protandry, that reduce interference between male and female functions that leads to selfing.     

Does stigma curvature promote delayed selfing? An experimental investigation in Triodanis perfoliata (Campanulaceae)

• Self‐fertilisation that is delayed until after opportunities for outcrossing have ceased has been argued to provide both the reproductive assurance benefits of selfing and the genetic advantages of outcrossing. In the Campanulaceae, presentation of pollen on stylar hairs and progressive stigma curvature have been hypothesised to facilitate delayed selfing, but experimental tests are lacking. Stigma curvature is common in Campanula, a genus largely characterised by self‐incompatibility, and therefore is unlikely to have initially evolved to promote self‐fertilisation. In derived self‐compatible species, however, stigma curvature might serve the secondary function of delayed selfing.
• We investigated delayed selfing in Triodanis perfoliata, a self‐compatible relative of Campanula. Using floral manipulation experiments and pollen tube observations, we quantified the extent and timing of self‐pollination. Further, we hypothesised that, if stigma curvature provides the benefit of delayed selfing in Triodanis, selection should have favoured retention of self‐pollen through the loss of a stylar hair retraction mechanism.
• Results of a stigma removal experiment indicated that autonomous selfing produces partial seed set, but only some selfing was delayed. Pollen tube observations and a flower senescence assay also supported the finding of partial delayed selfing. Scanning electron microscopy revealed that pollen‐collecting hairs retract during anthesis, which may limit the extent of delayed selfing.
• Delayed selfing appeared to be only partially effective in T. perfoliata. The stylar hair retraction in this species would seem to contradict selection for selfing. We suggest that caution and rigour are needed in interpreting floral traits as adaptive mechanisms for delayed selfing.

     

Asymmetric reproductive interference: The consequences of cross‐pollination on reproductive success in sexual–apomictic populations of Potentilla puberula (Rosaceae)

Apomixis evolves from a sexual background and usually is linked to polyploidization. Pseudogamous gametophytic apomicts, which require a fertilization to initiate seed development, of various ploidy levels frequently co‐occur with their lower‐ploid sexual ancestors, but the stability of such mixed populations is affected by reproductive interferences mediated by cross‐pollination. Thereby, reproductive success of crosses depends on the difference in ploidy levels of mating partners, that is, on tolerance of deviation from the balanced ratio of maternal versus paternal genomes. Quality of pollen can further affect reproductive success in intercytotype pollinations. Cross‐fertilization, however, can be avoided by selfing which may be induced upon pollination with mixtures of self‐ and cross‐pollen (i.e., mentor effects). We tested for reproductive compatibility of naturally co‐occurring tetraploid sexuals and penta‐ to octoploid apomicts in the rosaceous species Potentilla puberula by means of controlled crosses. We estimated the role of selfing as a crossing barrier and effects of self‐ and cross‐pollen quality as well as maternal: paternal genomic ratios in the endosperm on reproductive success. Cross‐fertilization of sexuals by apomicts was not blocked by selfing, and seed set was reduced in hetero‐ compared to homoploid crosses. Thereby, seed set was negatively related to deviations from balanced parental genomic ratios in the endosperm. In contrast, seed set in the apomictic cytotypes was not reduced in hetero‐ compared to homoploid crosses. Thus, apomictic cytotypes either avoided intercytotype cross‐fertilization through selfing, tolerated intercytotype cross‐fertilizations without negative effects on reproductive success, or even benefitted from higher pollen quality in intercytotype pollinations. Our experiment provides evidence for asymmetric reproductive interference, in favor of the apomicts, with significantly reduced seed set of sexuals in cytologically mixed populations, whereas seed set in apomicts was not affected. Incompleteness of crossing barriers further indicated at least partial losses of a parental genomic endosperm balance requirement.     

Nonrandom mating in Clarkia gracilis (Onagraceae): a case of cryptic self-incompatibility

In plants capable of both self-fertilization and outcrossing, the selfing rate depends on the proportion of self pollen in pollen loads and on the relative postpollination success of self pollen in siring offspring. While the composition of pollen loads is subject to unpredictable variation, paternity success of self vs. outcross pollen following pollen deposition may be controlled by maternal plants. This study examined postpollination paternity success in Clarkia gracilis ssp. sonomensis, in which deposition of self pollen is common. Pure loads of self and outcross pollen produced similar numbers of mature seeds, but equal mixtures of self and outcross pollen yielded more than three times as many outcrossed offspring as selfed offspring. The finding that the paternity success of self pollen depends on whether it is in competition with outcross pollen helps to explain an earlier finding that the selfing rate in experimental populations was highest when pollinator activity was lowest. Cryptic self-incompatibility allows paternity by self pollen when outcross pollen is unavailable.     

POLLEN INTERFERENCE AND CRYPTIC SELF-FERTILITY IN CAMPSIS RADICANS

The effect of self pollen on the success of cross pollinations was studied in Campsis radicans (L.) Seem., a species previously considered self-sterile. The application of self pollen to stigmas in combination with or preceding the application of cross pollen markedly reduced the likelihood of fruit production. This effect was not due to the dilution of cross pollen nor to physical blocking of the stigmatic surface, since mixtures of talc and cross pollen were as effective as pure cross pollen in causing fruit production. Pollen tubes produced by self pollen grew at rates similar to those from cross pollen, and penetrated the ovary. The interference caused by self pollen appears to take place in the ovary, although it cannot be stated to what degree it is prezygotic or postzygotic. The marked protandry occurring in C. radicans is presumed to be due at least partly to selection for avoidance of these negative effects on female reproductive success. Up to 33% of viable seeds from mixed self + cross pollinations were sired by self pollen. The term “cryptic self-fertility” is coined to describe this phenomenon where pollination with loads of pure self pollen rarely or never yields fruit, but pollination using mixtures of self and cross pollen yields fruit containing considerable numbers of selfed seed. Estimates of selfing frequency that are based on pollination using loads of purely self pollen will be in error for species possessing cryptic self-fertility.     

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.     

Shifts to earlier selfing in sympatry may reduce costs of pollinator sharing

Coexisting plant congeners often experience strong competition for resources. Competition for pollinators can result in direct fitness costs via reduced seed set or indirect costs via heterospecific pollen transfer (HPT), causing subsequent gamete loss and unfit hybrid offspring production. Autonomous selfing may alleviate these costs, but to preempt HPT, selfing should occur early, before opportunities for HPT occur (i.e., “preemptive selfing hypothesis”). We evaluated conditions for this hypothesis in Collinsia sister species, C. linearis and C. rattanii. In field studies, we found virtually identical flowering times and pollinator sharing between congeners in sympatric populations. Compared to allopatric populations, sympatric C. linearis populations enjoyed higher pollinator visitation rates, whereas visitation to C. rattanii did not differ in sympatry. Importantly, the risk of HPT to each species in sympatry was strongly asymmetrical; interspecies visits comprised 40% of all flower‐to‐flower visits involving C. rattanii compared to just 4% involving C. linearis. Additionally, our greenhouse experiment demonstrated a strong cost of hybridization when C. rattanii was the pollen donor. Together, these results suggest that C. rattanii pays the greatest cost of pollinator sharing. Matching predictions of the preemptive selfing hypothesis, C. rattanii exhibit significantly earlier selfing in sympatric relative to allopatric populations.     

Characterization of six microsatellite loci in Echium vulgare (Boraginaceae)

Echium vulgare is a tetraploid plant with a very low selfing rate in the field. We suspect that cryptic self incompatibility plays a role in this species. In order to show lower success of self pollen/selfed embryos, after pollination with a mixture of self and outcross pollen, a paternity analysis has to be done. For the purpose of such analysis we developed microsatellites in E. vulgare. In this study, we report on six microsatellite loci which are easy to score, polymorphic, with a number of alleles per locus ranging from two to eight and, therefore, suitable for paternity analysis.     

Direct and indirect estimates of the selfing rate in small and large individuals of the bumblebee pollinated Cynoglossum officinale L (Boraginaceae)

We measured the relationship between selfing rates and flower number in an experimental population of bumblebee pollinated Cynoglossum officinale , with plants differing in flower number. Results were compared with the prediction of a model based on pollen dynamics and pollinator behaviour. The selfing rate, as measured by multilocus oligonucleotide DNA fingerprinting, increased with flower number and ranged from 0% to 70%. Flowers on large plants received an equal number of visits from bumblebees as flowers on small plants. On large plants more flowers in a row were visited, inducing geitonogamy. The overall relationship between selfing rate and number of flowers can be explained by pollen dynamics and pollinator behaviour without invoking postpollination processes such as differential pollen tube growth and abortion.     

Analysis of the disassortative mating pattern in a heterodichogamous plant, <Emphasis Type="Italic">Acer mono</Emphasis> Maxim. using microsatellite markers

Heterodichogamy is a form of sex expression in which protandrous and protogynous individuals coexist, and is considered to be a mechanism that avoids selfing and promotes disassortative mating. We examined mating patterns in a heterodichogamous maple, Acer mono, using microsatellite markers. Parentage analysis revealed a selfing rate of only 9.8%. Disassortative mating between flowering types significantly exceeded within-type mating, but the mating patterns were better explained by flowering phenology (i.e., the temporal overlap between the female and male stages). Heterodichogamy in A. mono thus appears to promote outcrossing without requiring obligate self- or cross-incompatibility systems, although it did not guarantee disassortative mating. Multiple-regression analysis suggested that successful reproduction of pollen parents significantly increased with increased flower production and reciprocal flowering synchrony, but decreased only marginally with mating distance, although the distribution of mating distances suggested leptokurtic dispersal of pollen.     

Effect of pollen and resource limitation on reproduction of Zygophyllum xanthoxylum in fragmented habitats

Limitations on pollen and resources may significantly affect plant reproduction in fragmented habitats. In this study, phenology and pollinator frequency and activity were investigated to estimate the role of pollinators in Zygophyllum xanthoxylum reproduction, and this species is ecologically important in northwest China. In addition, the relative impact of restrictive amounts of pollen and resources on the seed set per flower was evaluated. It was found that adding pollen boosted the size of the seed set per flower, but had no significant effect on the number of flowers. By contrast, the addition of resources increased flower numbers as well as had a slight impact on the seed set per flower. These results indicate the amount of available pollen is a limiting factor for reproductive success. Moreover, Apis mellifera was identified as the most effective pollinator of Z. xanthoxylum, and there were more overall pollinators and visitations in the control than in the fragmented habitats. Furthermore, the limitations in pollen were more restrictive in the fragmented area than in the control. This was due to increased pollinator visitations in the control that could ameliorate the effects of lower pollen levels. When there is a limited availability of suitable pollinators, self‐pollination is critical in fragmented habitats. Z. xanthoxylum has reproductive strategies that aid in adapting to harsh environments, including protogyny and delayed selfing.     

Pollinator visitation patterns strongly influence among-flower variation in selfing rate

Background and Aims

Adjacent flowers on Mimulus ringens floral displays often vary markedly in selfing rate. We hypothesized that this fine-scale variation in mating system reflects the tendency of bumble-bee pollinators to probe several flowers consecutively on multiflower displays. When a pollinator approaches a display, the first flower probed is likely to receive substantial outcross pollen. However, since pollen carryover in this species is limited, receipt of self pollen should increase rapidly for later flowers. Here the first direct experimental test of this hypothesis is described.

Methods

In order to link floral visitation sequences with selfing rates of individual flowers, replicate linear arrays were established, each composed of plants with unique genetic markers. This facilitated unambiguous assignment of paternity to all sampled progeny. A single wild bumble-bee was permitted to forage on each linear array, recording the order of floral visits on each display. Once fruits had matured, 120 fruits were harvested (four flowers from each of five floral displays in each of six arrays). Twenty-five seedlings from each fruit were genotyped and paternity was unambiguously assigned to all 3000 genotyped progeny.

Key Results

The order of pollinator probes on Mimulus floral displays strongly and significantly influenced selfing rates of individual fruits. Mean selfing rates increased from 21 % for initial probes to 78 % for the fourth flower probed on each display.

Conclusions

Striking among-flower differences in selfing rate result from increased deposition of geitonogamous (among-flower, within-display) self pollen as bumble-bees probe consecutive flowers on each floral display. The resulting heterogeneity in the genetic composition of sibships may influence seedling competition and the expression of inbreeding depression.Key words: Autogamy, bee, Bombus fervidus, floral display, geitonogamy, mating system, monkeyflower, Mimulus ringens, paternity analysis, pollen carryover, pollinator visitation sequence, self-fertilization     

Selfing rates in natural populations of Echium vulgare: a combined empirical and model approach

1. We quantified geitonogamous selfing in Echium vulgare , a self-compatible, bumble-bee pollinated plant. A maximum estimate of selfing was determined using a paternity analysis with RAPDs. In the first experiment, bumble-bees visited a sequence of virgin flowers. The percentage selfing increased rapidly from 12% in the first flower visited, up to 50% in the 15th flower visited in the sequence. In the second experiment, when bees visited plants in a natural population, the average selfing of plants increased with the number of open flowers from 0% to maximally 33%.
2. The results obtained in both experiments are consistently lower than predicted by our model on pollen dynamics ( Rademaker, de Jong & Klinkhamer 1997 ). We modified the model on pollen dynamics to link it more to the field situation with observations on flower stage, flower opening and bumble-bee preference, so that the bumble-bees encounter a variable number of pollen grains per flower. We also adjusted the parameters. If less pollen adheres to the bee (25% instead of 50%) after removal from the anthers, or if bees arrive at a plant with more pollen grains (6000 instead of 4448), the predictions of the model in regard to selfing could be improved but were still high compared with the observed selfing rates measured with RAPDs.
3. We suggest that the model is consistent with pollen dynamics in the field. However, post-pollination processes like selective abortion could play a role in E. vulgare .     

The effect of crossing distances and population subdivision on floral morphology inScleranthus annuus (Caryophyllaceae), a selfing annual

I used a discontinuous population ofScleranthus annuus (Caryophyllaceae) to study the effect of crossing distances on flower morphology of the progeny. Four types of progeny were produced by artificial selfing, crossing with pollen-donors from the same patch in the population, crossing with pollen from donors from other patches in the population and inter-population crosses. The size of gynoecium parts and 12 sepal characters in this petal-lacking species were significantly influenced by the type of cross and the patch in the population from where the seed-parents originated. All comparisons of progeny types except selfed vs progeny produced by within-patch crosses were significantly separated from each other, while all four seed-parent patches used were significantly separated in a multidimensional space.     

The effect of pollination mode on seed performance of Gentiana lutea: a laboratory evaluation of seed germinability

Gentiana lutea L. (yellow gentian, Gentianaceae) is a protected orophyte of central and southern Europe, mainly threatened by the uncontrolled collection of its rhizome, used in traditional medicine and for liquor production. The species is self‐compatible, but outcrossing mediated by pollinators is needed to obtain a viable progeny. In this study, we considered five natural populations belonging to the four subspecies of G. lutea. We performed controlled pollinations in the field (self‐ versus cross‐pollination) followed by seed germination tests in laboratory conditions, adding a solution of gibberellic acid, in order to evaluate the seed performance. A cumulative index of inbreeding depression was calculated considering maternal reproductive output as well as seed performance traits. Seed weight and seed germination performance was similar between seeds resulting from naturally pollinated and pollen‐augmented flowers and higher compared to selfed flowers, highlighting a disadvantage of selfing and the importance of cross‐pollen transfer in natural conditions. However, in the small and isolated population of G. lutea subsp. symphyandra on Mt Grande we found a general reduction in seed germination rate, likely due to increased selfing or mating among close relatives as a consequence of a severe bottleneck. We discuss our results with regards to implications for conservation practices.     

Cryptic self-incompatibility in tristylous Decodon verticillatus (Lythraceae)

There is growing evidence that many self-compatible plants control the level of self-fertilization with postpollination processes that give a siring advantage to cross pollen over self pollen through "cryptic self-incompatibility" (CSI). Previous marker-gene experiments with self-compatible, tristylous Decodon verticillatus (Lythraceae) have demonstrated a siring advantage to cross pollen, though the extent to which this advantage results from prezygotic discrimination vs. early acting inbreeding depression is not clear. Here, we provide evidence that prezygotic mechanisms are involved in this siring advantage by comparing pollen tube numbers at various times following cross- and self-pollination conducted in a natural population. In the 24 h following pollination, cross pollen yielded almost twice as many pollen tubes at various positions in the style compared to self pollen. After 36 and 48 h, the difference between pollen types had disappeared, suggesting that the advantage to cross pollen results from differences in the rate of pollen germination and;clor tube growth rather than pollen tube attrition. Comparison of tube numbers after legitimate vs. illegitimate cross-pollination did not reveal any difference, suggesting that D. verticillatus possesses CSI unrelated to heteromorphic self- and intramorph-incompatibility found in other heterostylous members of the Lythraceae. CSI resulting from differential pollen tube growth may minimize geitonogamous selfing when cross pollen is abundant, while maximizing fecundity when cross pollen is scarce due to local clonal spread.     

Evolutionary dynamics of mating system shifts in Arabidopsis lyrata

Outcrossing is the prevalent mode of reproduction in plants and animals despite its substantial costs, while selfing and mixed mating occur at much lower frequency. Comparative research on plants has demonstrated the lability of self‐incompatibility, but there is little information about the transition on a within‐species level from self‐incompatibility to predominant selfing. We studied variation in mating system among 18 populations of Arabidopsis lyrata within a phylogenetic context to shed light on the evolution of selfing. Realized and potential mating systems were assessed by genetic analysis with microsatellite markers and hand‐self‐pollinations on 30 plants from each population. The fraction of self‐incompatible plants in a population was highly correlated with the outcrossing rate, showing that the spread of self‐compatibility is accompanied by or soon followed by an increase in the rate of selfing. The four predominantly selfing populations (outcrossing rates < 0.25) fell into more than one phylogenetic cluster, suggesting that the transition to selfing occurred more than once independently. Hence, A. lyrata offers an opportunity for the comparative analysis of outcrossing as a predominant mode of reproduction in plants and of the causes of the shift to selfing.