Pollen as the link between floral phenotype and fitness

Pollen plays a key role in plant reproductive biology. Despite the long history of research on pollen and pollination, recent advances in pollen-tracking methods and statistical approaches to linking plant phenotype, pollination performance, and reproductive fitness yield a steady flow of exciting new insights. In this introduction to the Special Issue “Pollen as the Link Between Phenotype and Fitness,” we start by describing a general conceptual model linking functional classes of floral phenotypic traits to pollination-related performance metrics and reproductive fitness. We use this model as a framework for synthesizing the relevant literature, highlighting the studies included in the Special Issue, and identifying gaps in our understanding and opportunities for further development of the field. The papers that follow in this Special Issue provide new insights into the relationships between pollen production, presentation, flower morphology, and pollination performance (e.g., pollen deposition onto stigmas), the role of pollinators in pollen transfer, and the consequences of heterospecific pollen deposition. Several of the studies demonstrate exciting experimental and analytical approaches that should pave the way for continued work addressing the intriguing role of pollen in linking plant phenotypes to reproductive fitness.     

FLORAL PAEDOMORPHY LEADS TO SECONDARY SPECIALIZATION IN POLLINATION OF MADAGASCAR DALECHAMPIA (EUPHORBIACEAE)

The traditional evolutionary interpretation of Von Baer's “laws” of embryology is that retention of early developmental forms into adulthood (paedomorphosis) leads to the evolution of simpler or more generalized morphology and ecology. Here we show that paedomorphosis can also be involved in an increase in ecological specialization, in this case of plant–pollinator relationships. A paedomorphic transition from generalized pollination (by several functional types of pollinators) to specialized pollination (by one or a few species in one functional type) occurred in a clade of endemic Madagascar vines (Dalechampia spp., Euphorbiaceae). This evolutionary transition involved staminate flowers that fail to develop “normally,” instead holding mature pollen inside virtually unopened, bud‐like flowers. This paedomorphic morphology restricts reward access to “buzz‐pollinating” bees, including Xylocopa species (carpenter bees), which can remove pollen by sonication. This is one of very few reports of paedomorphic specialization, and, as far as we are aware, the first documented case of a rapid reversal to specialized pollination in a lineage of plants that had previously switched from specialized to generalized pollination in conjunction with dispersing to a new region.     

POLLINATION DROP IN RELATION TO CONE MORPHOLOGY IN PODOCARPACEAE: A NOVEL REPRODUCTIVE MECHANISM

Observation of ovulate cones at the time of pollination in the southern coniferous family Podocarpaceae demonstrates a distinctive method of pollen capture, involving an extended pollination drop. Ovules in all genera of the family are orthotropous and single within the axil of each fertile bract. In Microstrobus and Phyllocladus ovules are erect (i.e., the micropyle directed away from the cone axis) and are not associated with an ovule-supporting structure (epimatium). Pollen in these two genera must land directly on the pollination drop in the way usual for gymnosperms, as observed in Phyllocladus. In all other genera, the ovule is inverted (i.e., the micropyle is directed toward the cone axis) and supported by a specialized ovule-supporting structure (epimatium). In Saxegothaea there is no pollination drop and gametes are delivered to the ovule by pollen tube growth. Pollination drops were observed in seven of the remaining genera. In these genera the drop extends over the adjacent bract surface or cone axis and can retain pollen that has arrived prior to drop secretion (“pollen scavenging”). The pollen floats upward into the micropylar cavity. The configuration of the cone in other genera in which a pollination drop has not yet been observed directly suggests that pollen scavenging is general within the family and may increase pollination efficiency by extending pollination in space and time. Increased pollination efficiency may relate to the reduction of ovule number in each cone, often to one in many genera, a derived condition. A biological perspective suggests that animal dispersal of large seeds may be the ultimate adaptive driving force that has generated the need for greater pollination efficiency.     

REPRODUCTIVE BIOLOGY OF APIACEAE. II. CRYPTIC SPECIALIZATION AND FLORAL EVOLUTION IN THASPIUM AND ZIZIA

The classification of specialized floral syndromes has imposed a bias in the interpretation of pollination systems which may be either more generalized, or more specialized, than we have universally acknowledged. An analysis of floral biology in two umbellifer genera, Thaspium and Zizia, was undertaken in order to determine the extent to which cryptic floral or inflorescence variations determine pollination specialization despite a broad visitor spectrum and open reward system. Separate analyses were made of the primary attractants, nectar and pollen, and the secondary attractants, floral color and floral and inflorescence structure in conjunction with analyses of pollinator movements, stigmatic pollen loads and fruit set. All data support the conclusion that cryptic variation in floral and inflorescence characters enhance specialization for pollination by solitary bees or syrphid flies. In addition, evidence is presented for the importance of the oligolectic relationship between taxa of Thaspium and Zizia and the solitary bee, Andrena ziziae. The need for more experimental work both to further define the oligolectic relationship and to understand how floral and inflorescence color and structure affect insect movements is especially revealed by this study. The degree of pollination specialization in Thaspium and Zizia is not uncommon in Apiaceae and has important implications for floral evolution in this family and other plant groups with pollination systems categorized as “promiscuous.”     

Intersexual mimicry and imperfect deceit of a threatened aquatic herb Ottelia acuminata

Mimicry of non-rewarding flowers to rewarding flowers has been accepted as a strategy to improve pollination success in angiosperms. It has been proposed that this mechanism depends on whether potential pollinators can discriminate between the flowers. In this study, the intersexual mimicry and deceit pollination were studied in a threatened dioecious aquatic herb, Ottelia acuminata. Its female flowers resemble male flowers in morphology and odor compounds, to avoid discrimination by pollinators and outcompete male flowers in attracting the pollinators using stronger scents and bigger flowers. However, an obvious visit bias of its pollinator (Apis cerana) to male flowers was detected, suggesting that bees can distinguish the rewarding males from non-rewarding females. Although the deceit was not successful, pollination was not seriously undermined because pollen limitation was found to be low in the sampled natural population. We speculate that, due to “accidental” visits on female flowers and “mistake” pollinations, pollen limitation could be mitigated by a high frequency of pollen donors, and is correlated with the size and sex ratio of a population. Ottelia acuminata is a threatened dioecious aquatic herb. We suggest that developing multi-stakeholder coalitions should be encouraged to save the threatened edible and ornamental plant species in China. We hope this study could provide new insights into understanding of the role of intersexual mimicry in other flowering plants.     

Does the morphological fit between flowers and pollinators affect pollen deposition? An experimental test in a buzz‐pollinated species with anther dimorphism

Some pollination systems, such as buzz‐pollination, are associated with floral morphologies that require a close physical interaction between floral sexual organs and insect visitors. In these systems, a pollinator's size relative to the flower may be an important feature determining whether the visitor touches both male and female sexual organs and thus transfers pollen between plants efficiently. To date, few studies have addressed whether in fact the “fit” between flower and pollinator influences pollen transfer, particularly among buzz‐pollinated species. Here we use Solanum rostratum, a buzz‐pollinated plant with dimorphic anthers and mirror‐image flowers, to investigate whether the morphological fit between the pollinator's body and floral morphology influences pollen deposition. We hypothesized that when the size of the pollinator matches the separation between the sexual organs in a flower, more pollen should be transferred to the stigma than when the visitor is either too small or too big relative to the flower. To test this hypothesis, we exposed flowers of S. rostratum with varying levels of separation between sexual organs, to bumblebees (Bombus terrestris) of different sizes. We recorded the number of visits received, pollen deposition, and fruit and seed production. We found higher pollen deposition when bees were the same size or bigger than the separation between anther and stigma within a flower. We found a similar, but not statistically significant pattern for fruit set. In contrast, seed set was more likely to occur when the size of the flower exceeded the size of the bee, suggesting that other postpollination processes may be important in translating pollen receipt to seed set. Our results suggest that the fit between flower and pollinator significantly influences pollen deposition in this buzz‐pollinated species. We speculate that in buzz‐pollinated species where floral morphology and pollinators interact closely, variation in the visitor's size may determine whether it acts mainly as a pollinator or as a pollen thief (i.e., removing pollen rewards but contributing little to pollen deposition and fertilization).     

Pollen:ovule ratio and its relationship with other floral traits in Papilionoideae (Leguminosae): an evaluation with Argentine species

The pollen:ovule ratio (P/O) has traditionally been used as a rough estimator of plant breeding systems. It has been shown that plant breeding systems are associated with particular floral traits. In this study, we determined the P/O in 21 Leguminosae species from Argentina and explored relationships between P/O and taxonomic position, flower size, floral rewards, pollen presentation and pollination mechanisms. According to the results, 15 out of the 21 species classified were obligate xenogamous, although some of them have been recorded as facultative xenogamous in previous studies. There was a significant effect of taxonomic position (genus), reward type and pollination mechanism on P/O. Species offering only nectar as a floral reward (which were species with a brush mechanism) had a significantly lower P/O than species offering pollen or pollen and nectar. Species with the brush pollination mechanism had the lowest P/O, while species with valvular and pump mechanism had the highest P/O. However, pollen presentation (primary and secondary) and flower size did not have a significant effect on P/O. Our results demonstrate that P/O variability is determined by taxonomic position and pollination mechanism in this plant group.     

AIRFLOW PATTERNS AROUND SOME EARLY SEED PLANT OVULES AND CUPULES: IMPLICATIONS CONCERNING EFFICIENCY IN WIND POLLINATION

Aerodynamic analyses showing characteristic airflow patterns and the potential for wind-mediated pollination are presented for models of Paleozoic (Carboniferous) ovules and ovulate cupules (i.e., Genomosperma kidstoni, G. latens, Salpingostoma dasu, Physostoma elegans, Eurystoma angulare, and Stamnostoma huttonense). Lobes on ovules and cupules are shown to produce localized regions of turbulent flow with a concomitant reduction in airflow velocity. Data based upon models that mimic the characteristics of windborne pollen (= pseudopollen) show that these regions of turbulent flow correspond to those in which suspended pseudopollen impact with ovule and/or cupule surfaces. These data have bearing on a sequence of ovule morphologies purported to show the evolution of the integument by the progressive reduction in length of “preintegumentary” lobes and their acropetal fusion. As the preintegumentary lobes of the models studied consolidate around the megasporangium, regions of turbulent flow and high pseudopollen impact become localized around the pollen chamber or salpinx. The general morphologic trend envisioned for the evolution of the ovule is seen to be associated with an aerodynamic streamlining and an increased potential for wind-mediated pollination. Data for hair-bearing ovules and for ovulate cupules are discussed within the context of possible selective pressures favouring streamlining.     

MORPHOLOGICAL STUDIES OF THE NYMPHAEACEAE. XI. THE FLORAL BIOLOGY OF NELUMBO PENTAPETALA

The floral biology of Nelumbo pentapetala (Walter) Fernald, the American lotus, native to Texas, was investigated. Anthesis occurs over three consecutive days with flowers opening each morning and closing around noon. First-day flowers are protogynous with the perianth parts partially expanded so that pollen-covered insects which are attracted by floral color and the intense “fruity” odor (diffused with the aid of increased floral temperature) are directed on to the flattened receptacle (= carpellary receptacle) from which the receptive stigmas protrude, thus accomplishing pollination. During the second morning anther dehiscence begins and insects which visit and forage within the flower become covered with pollen and typically crawl over the still receptive stigmas achieving “facilitated” self-pollination (indirect autogamy). By mid-morning of the second day the stigmas dry and become non-receptive to pollen. During the third day of anthesis perianth and staminal parts quickly abscise and over the period of a few weeks the receptacle and enclosed fruits mature. In most populations studied, Hymenoptera (e.g., Lusioglossum spp., and Apis mellifera) were the most abundant and effective pollinators. In some populations, however, Coleoptera (e.g., Chauliognathus) were also numerous and effective pollinators. It is suggested that the overall floral structure (e.g., large numbers of stamens, masses of pollen, staminal appendages) are adaptations which facilitate the pollination of Nelumbo by beetles.     

THE MOTION OF WINDBORNE POLLEN GRAINS AROUND CONIFER OVULATE CONES: IMPLICATIONS ON WIND POLLINATION

Comparisons are presented between the three-dimensional airflow patterns created around and by a scale model of a conifer ovulate cone and the trajectories of windborne pollen grains around Picea, Larix, and Pinus ovulate cones. Three general components of the airflow pattern around an ovulate cone model are 1) doldrum-like eddies, rotating over the adaxial surfaces of cone scales and directed toward attached ovules, 2) airflow spiralling around the cone axis along cone scale orthostichies and parastichies, and 3) a complex pattern of vortices (“umbilicus”) directed toward the leeward surface of the ovulate cone. The observed trajectories of pollen grains around cones of Picea, Larix, and Pinus conform to two of these three airflow components: 1) pollen grains are seen to roll along cone scales toward the distal scale margin and to become reentrained in airflow directed backward toward attached ovules, and 2) pollen grains passing around the cone are deflected into the “umbilicus” airflow pattern, where they either settle on or impact with cone scales (approach trajectories), or where they approach the leeward cone surface but are deflected away by airflow passing under the cone (Z-shaped trajectories). Vectoral analyses of pollen grain motion reveal a complex pattern of trajectories influenced by boundary layer conditions defined by ovulate cone geometry and ambient airflow speed. Wind tunnel studies of ovulate cones subtended by leaves and stem indicate that leaves circumscribing the cone act as a snowfence, deflecting windborne pollen toward the cone. Vectoral analyses of airflow patterns and pollen grain trajectories close to ovulate cones indicate that wind pollination in conifers is a non-stochastic aerodynamic process influenced by cone-leaf morphology and the behavior of pollen grains as windborne particles.     

STARCH IN ANGIOSPERM POLLEN GRAINS AND ITS EVOLUTIONARY SIGNIFICANCE

Pollen grains of 990 species were examined and measured to test four predictions: (1) “Primitive” angiosperms will have starch-containing pollen; more advanced families will have starchless pollen. (2) Where Hymenoptera and Diptera use pollen nutritionally, there will be selection of starchless (oilcontaining) pollen, particularly where pollen is the only reward for visitors. (3) Conversely, in autogamous species, anemophilous species, and those pollinated by Lepidoptera or birds, the energetically more economical accumulation of starch will be seen. (4) Small pollen grains will be more likely to be starchless (oil-rich); starch-containing grains will tend to be larger. Prediction 1 is only partially supported by the results; contemporary representatives of primitive families often show starchless grains, and so do most “advanced” families. Strong intra-familial resemblances are found (and some between groups of families), but other families show mixtures of “starchy” and “starchless” species. The latter are more likely to show a fit between food reserve and pollination system. Prediction 2 is well supported. Prediction 3 is supported in that autogamous and anemophilous species show relatively high proportions of starchy grains, but the picture is complicated for them by apparent selection for starchlessness (oil-richness) in small pollen grains (prediction 4), and for the lepidopteran- and bird-pollinated species by selection for large starchless (or very large starchy grains) in species whose pollen tubes must traverse long styles.     

Floral reward presentation favored the expression of male function in the pollen‐only flower Melastoma malabathricum

Abstract Animal‐pollinated plant species modulate the presentation of pollinator rewards to maximize reproductive success. In plants providing pollen as the only reward for pollinators, it is usually difficult to unravel the dual roles of reward presentation and the realization of male and female functions (pollen removal and deposition). Exploiting the two types of anther in the androecia of Melastoma malabathricum L., we examined whether the removal of pollen for reward is regulated primarily to favor male function or female function. Pollen removal by carpenter bees from the feeding and pollination anthers, as well as pollen deposition on the stigmas, were quantified during anthesis of M. malabathricum. There was no significant difference in pollen removal rates from the feeding and pollination anthers of M. malabathricum between the onset of anthesis and flower wilting. The stigmatic pollen loads exceeded the ovule number after three sonication bouts, and female function was satisfied earlier than male function. The results support the hypothesis that the presentation of pollination reward in this species is regulated primarily to favor the expression of male function, rather than female function, in agreement with the pollen‐donation hypothesis. A cooperative relationship between the feeding and pollination anthers was demonstrated in heterantherous flowers, which optimizes the balance in investments between pollinator rewards and “functional pollen” for gene transfer.     

Pollen transfer dynamics and the evolution of gametophytic self-incompatibility

Recent studies of mating system evolution have attempted to include aspects of pollination biology in analysis of both theoretical models and experimental systems. In light of this growing trend, we propose a simple population genetic model for the evolution of gametophytic self-incompatibility, incorporating parameters for pollen discounting and pollen export/capture. In this model, we consider several cases that span the spectrum for dominance of the mutant self-incompatibility allele and for the degree of incompatibility conferred by the allele. We confirm earlier results that inbreeding depression is required for successful invasion of the self-incompatibility allele and we demonstrate that, unless pollen discounting is very low, the level of inbreeding depression must be very high for an allele conferring self-incompatibility to become established. Finally, we show that the dominance of the mutant allele has a greater impact on the fate of a newly arisen self-incompatibility allele than the strength of the incompatibility conferred by the allele. In particular, the more recessive the self-incompatibility expression in heterozygote stigmas and the weaker the response induced, the easier it is for a self-incompatibility allele to invade.     

Using geometric morphometrics to determine the “fittest” floral shape: A case study in large-flowered,buzz-pollinated Meriania hernandoi

Premise

Floral shape (relative arrangement and position of floral organs) is critical in mediating fit with pollinators and maximizing conspecific pollen transfer particularly in functionally specialized systems. To date, however, few studies have attempted to quantify flowers as the inherently three-dimensional (3D) structures they are and determine the effect of intraspecific shape variation on pollen transfer. We here addressed this research gap using a functionally specialized system, buzz pollination, in which bees extract pollen through vibrations, as a model. Our study species, Meriania hernandoi (Melastomataceae), undergoes a floral shape change from pseudocampanulate corollas with more actinomorphically arranged stamens (first day) to open corollas with a more zygomorphic androecium (second day) over anthesis, providing a natural experiment to test how variation in floral shape affects pollination performance.

Methods

In one population of M. hernandoi, we bagged 51 pre-anthetic flowers and exposed half of them to bee pollinators when they were in either stage of their shape transition. We then collected flowers, obtained 3D flower models through x-ray computed tomography for 3D geometric morphometric analyses, and counted the pollen grains remaining per stamen (male pollination performance) and stigmatic pollen loads (female pollination performance).

Results

Male pollination performance was significantly higher in open flowers with zygomorphic androecia than in pseudo-campanulate flowers. Female pollination performance did not differ among floral shapes.

Conclusions

These results suggest that there is an “optimal” shape for male pollination performance, while the movement of bees around the flower when buzzing the spread-out stamens results in sufficient pollen deposition regardless of floral shape.

     

VISCIN THREADS,POLLINATION EFFICIENCY AND LOW POLLEN-OVULE RATIOS

Some Onagraceae are characterized by quite low pollen-ovule ratios. We suggest they are an evolutionary response to highly efficient pollination, mediated through viscin threads, which hold large numbers of pollen grains together, and by stigmas, which are large relative to the area of the pollinator over which the pollen is spread. The pollen grains of some Leguminosae are joined by “exinal connections” which are chemically and morphologically similar to viscin threads. In the legumes the area of the stigma is small relative to the area of pollen on the pollinator and pollination is facilitated by short connections which hold small numbers of pollen grains together.     

Flower morphology and pollen germination in the genus Capsella (Brassicaceae)

Evolutive changes in mating systems are often accompanied by changes in flower morphology, such as the reduction in size or even loss of petals, changes in production of volatiles, pollen/ovule ratio, the position between anthers and stigma and the germination time of pollen after pollination. These changes have been merged under the term “selfing syndrome” and often result in new taxonomic species. The evolutionary shift frequently happens parallel within many families and genera, for example within the Brassicaceae family. Within the genus Capsella, which is closely related to the molecular model species pair Arabidopsis lyrata (SI)/A. thaliana (SC), we studied self-incompatible and self-compatible species. SC species C. rubella and C. bursa-pastoris produce in comparison with the SI species C. grandiflora (i) smaller petals as the result of decreased cell division and only less of decreasing cell volume, (ii) less production of pollen in one flower, (iii) show a lesser incision between the two valves of the fruits, in combination with a shorter style, and (iv) have a much quicker fertilization of SC pollen after pollination. Crossing success between the diploid species, between different provenances of the tetraploid C. bursa-pastoris, and between the two diploid species and particular individuals of the self-incompatible C. grandiflora has been proven.     

Cover Caption

The Western honey bee, Apis mellifera (L.), is perhaps the most beneficial insect we know, mainly because of the pollination services it provides to fruits and vegetables. Honey bee workers show changes in behaviors as they age. Young bees typically are “nurses” and perform in‐hive tasks such as feeding larvae and take care of the queen, old bees become foragers and bring in food sources (nectar, pollen, and water) or propolis. Methoprene has been known to accelerate worker development so bees become foragers earlier, but its mechanisms were not known. In this study Huang et al. show that most likely methoprene works directly on hormone receptors to mimic juvenile hormone, in causing bees to forage early (pages 235–240). Photo by Zachary Y. Huang. [Correction added on 17 April 2018, after first online publication: Cover caption has been revised.]     

INCOMPATIBILITY AND THE POLLEN ECONOMY OF JEPSONIA PARRYI

Jepsonia parryi (Saxifragaceae) has heterostylous flowers and is strongly self-incompatible. Pin flowers have long styles, large stigmas, short stamens, and numerous, small pollen grains with finely sculptured walls. Thrum flowers have short styles, small stigmas, long stamens, and fewer, larger pollen grains with coarsely sculptured walls. Pin plants and thrum plants occur in a 1:1 ratio in field populations. Although the insect pollinators of J. parryi transfer ample compatible pollen to pin and thrum stigmas to account for full seed production, much of the pollen deposited on stigmas is incompatible. Analysis of the pollen deposits on stigmas collected from field populations indicates that compatible “legitimate” pollination of pin and thrum flowers is essentially random and is not obviously aided by floral dimorphism. It is suggested that although heterostyly had a positive adaptive value in the past evolutionary history of Jepsonia it is no longer adaptive under the present pollination regime, although it is maintained because of its strong genetic fixity.     

The role of electrostatic forces in pollination

This paper reviews research on the role of electrostatic forces in pollination, both in natural and in agricultural systems. Researchers from various fields of biological studies have reported phenomena which they related to electrostatic forces. The theory of electrostatically mediated pollen transfer between insect pollinators and the flowers they visit is described, including recent studies which confirmed that the accumulated charges on airborne honey bees are sufficient for non-contact pollen detachment by electrostatic forces (i.e., electrostatic pollination). The most important morphological features in flower adaptiveness to electrostatic pollination were determined by means of two theoretical models of a flower exposed to an approaching charged cloud of pollen; they are style length and flower opening. Supplementary pollination by using electrostatic techniques is reported, and its possible importance in modern agriculture is discussed.     

A biophysical model for buzz pollination in angiosperms

The stamens of most of the world's flowering plants are longitudinally dehiscent, releasing their pollen passively, whereupon floral visitors may collect it. In nearly 400 genera in 65 plant families, the anthers dehisce by means of short apical slits or true pores. In these forms, the small light pollen can only be efficiently released by native bees capable of vibrating these stamens. This intrafloral behavior propels pollen out of the pores striking the bees on their venters. It is then collected for use in larval cell provisions. Aspects of the historical development of this novel pollination syndrome, known as “buzz” or vibratile (equals vibrational) pollination, are presented including a discussion and figures of a poricidal anther, a buzzing bee and the model system.A biophysical model for the pollen/locule wall interactions resulting in pollen expulsion upon bee or artificial vibration is developed. The model was created with the morphology of anthers of Solanum (Solanaceae) in mind, but the results obtained are generally applicable to any apically dehiscent flower which is vibrated by bees to release pollen.The anthers were modeled as a tall rectangular box with an apical pore and containing numerous small particles. As the box vibrates, particles striking the walls rebound elastically. If a pollen grain strikes a receding wall, it loses energy. If a grain strikes an advancing wall, it gains energy in the collision. In each oscillation, there is a net gain in the energy of the particles. As the anther (box) is shaken, vibrational energy is transmitted from the pterothorax of the bees to the flower, the pollen grains gaining significant energy. As the energy increases and the particles begin to move about more and more vigorously, they will begin to escape through the hole in the box (or stamina] pore). The rate at which particles leave the box and time required to empty the box are calculated as functions of the geometry of the model system and the frequency of vibration.In order to test the influence of air currents, Bernolli effects and viscous drag, the flowers were mecahnically vibrated in vacuum. The pollen cloud thus produced was virtually unchanged ans so it seems unlikely that air plays any significant role in the phenomenon of vibrational pollen release.Finally, variables such as: inelastic interactions, electrostatic forces, slightly sticky pollen due to presence of “pollenkitt”, duration and types of bee buzzes are discussed in relation to the mathematical model presented.