Skip to main content
NCBI home page
Annals of Botany logoLink to Annals of Botany
. 2018 Jun 7;123(2):373–380. doi: 10.1093/aob/mcy102

Nectar supplementation changes pollinator behaviour and pollination mode in Pedicularis dichotoma: implications for evolutionary transitions

Ze-Yu Tong 1, Xiang-Ping Wang 2, Ling-Yun Wu 1, Shuang-Quan Huang 1,✉
PMCID: PMC6344217  PMID: 29878060

Abstract

Backgrounds and Aims

Gain or loss of floral nectar, an innovation in floral traits, has occurred in diverse lineages of flowering plants, but the causes of reverse transitions (gain of nectar) remain unclear. Phylogenetic studies show multiple gains and losses of floral nectar in the species-rich genus Pedicularis. Here we explore how experimental addition of nectar to a supposedly nectarless species, P. dichotoma, influences pollinator foraging behaviour.

Methods

The liquid (nectar) at the base of the corolla tube in P. dichotoma was investigated during anthesis. Sugar components were measured by high-performance liquid chromatography. To understand evolutionary transitions of nectar, artificial nectar was added to corolla tubes and the reactions of bumble-bee pollinators to extra nectar were examined.

Key Results

A quarter of unmanipulated P. dichotoma plants contained measurable nectar, with 0.01–0.49 μL per flower and sugar concentrations ranging from 4 to 39 %. The liquid surrounding the ovaries in the corolla tubes was sucrose-dominant nectar, as in two sympatric nectariferous Pedicularis species. Bumble-bees collected only pollen from control (unmanipulated) flowers of P. dichotoma, adopting a sternotribic pollination mode, but switched to foraging for nectar in manipulated (nectar-supplemented) flowers, adopting a nototribic pollination mode as in nectariferous species. This altered foraging behaviour did not place pollen on the ventral side of the bees, and sternotribic pollination also decreased.

Conclusion

Our study is the first to quantify variation in nectar production in a supposedly ‘nectarless’ Pedicularis species. Flower manipulations by adding nectar suggested that gain (or loss) of nectar would quickly result in an adaptive behavioural shift in the pollinator, producing a new location for pollen deposition and stigma contact without a shift to other pollinators. Frequent gains of nectar in Pedicularis species would be beneficial by enhancing pollinator attraction in unpredictable pollination environments.

Keywords: Bumble-bee, buzz pollination, foraging behaviours, loss of nectar, nectar supplementation, Pedicularis, pollination niche

INTRODUCTION

Nectar as a common floral reward is widespread in various plant groups, including a few gymnosperms and most angiosperms (Koptur et al., 1982; Pacini et al., 2003; De la Barrera and Nobel, 2004; Willmer, 2011). Nectar production may be a labile evolutionary trait, which has been repeatedly gained and lost in angiosperms (Nicolson et al., 2007). For instance, a survey of nectar in 111 orchid species on a well-resolved phylogeny showed that absence of nectar was the ancestral condition, and nectar production evolved at least nine times but nectar was lost at least once (Johnson et al., 2013).

Nectar production could turn a visitor’s attention away from the pollen as food, and offer an ideal fuel in the form of sugars, and toxic chemical components within nectar could filter unwanted or inefficient pollinators (Adler, 2000; Irwin et al., 2004). Given the energetic costs of this type of floral secretion for some pollinators (Galen and Geib, 2007; Rojas-Nossa et al., 2016), it is not surprising that many plants have lost nectar production, being pollen-only providers, such as plants with a buzz-pollination system (i.e. Solanum flowers) in diverse lineages (Vallejo-Marín et al., 2009, 2010; De Luca and Vallejo-Marín, 2013; Cardinal et al., 2018). It has been shown that loss of nectar in orchids could lead to pollinators visiting fewer flowers on the same plant, reducing intra-plant selfing and promoting pollen export (Johnson and Nilsson, 1999; Johnson et al., 2004; Jersáková and Johnson, 2006); and could allow a plant to allocate more resources for other vegetative or reproductive functions (Southwick, 1984; Bell, 1986; Pyke, 1991; Mattila and Kuitunen, 2000). Compared with nectarless flowers, nectariferous flowers with additional reward would be preferred by pollinators and receive high rates of pollination (Neiland and Wilcock, 1998; Smithson and Gigord, 2001; Aizen et al., 2011; Natalis and Wesselingh, 2012).

The presence or absence of nectar production could be linked to pollinator behaviour, probably in plant species with bilaterally symmetrical flowers. For example, pollen and nectar rewards are located on the two opposing sides of the horizontal axis of labiate flowers. If a pollinator harvests both types of rewards during foraging, it has to adopt two postures: (1) while it is upright sucking nectar, pollen will be passively deposited on the dorsal side of its body (i.e. nototribic pollination mode); and (2) when it is upside down to collect pollen, pollen will be placed on the ventral side (sternotribic pollination mode), as observed in Rhinanthoideae (Kwak, 1979; Natalis and Wesseligh, 2012, 2013) or in Pedicularis flowers (Macior, 1982; Robart, 2005; Huang and Shi, 2013; Armbruster et al., 2014). The two different foraging postures would potentially affect plant reproductive success, given that pollen collected for food provision is lost to plant sexual reproduction (Kwak, 1979; Robart, 2005; Tong and Huang, 2018). Furthermore, the use of different sites of pollen placement on the pollinator body can partition pollen from different species with shared pollinators into ‘ecological niches’, reducing interspecific pollen competition. Numerous clades in the genus Pedicularis have evolved gain or loss of nectar, and the two pollination modes are often exploited in co-flowering-related species (Eaton et al., 2012). The type of floral isolation mediated by pollinator behaviour is likely to limit interspecific pollen flow and contributes to pre-pollination reproductive isolation, described as the ‘Pedicularis type’ by Grant (1994).

Although nectar production can be an innovative floral trait (Crepet and Niklas, 2008), evolutionary analyses usually use some correlated traits reflecting the status of nectar indirectly. For example, morphological measurement of nectaries is often used instead of direct measurement of nectar production (Vallejo-Marín et al., 2010; Liu et al., 2015). There are >600 species of Pedicularis in the North Temperate Zone, with most species in the Himalayas. Pedicularis flowers are characterized by diverse corolla morphologies; especially variable are the upper lips and the lengths of corolla tubes (Li, 1951; Macior, 1982; Yang et al., 1998). Some species have typically labiate corollas (beakless shorted-tubed corollas; Li 1951), but the upper lips in many species are folded into a galea enclosing the four anthers, with a beak- or elephant trunk-shaped extension. The beak has been regarded as an adaptation for buzz pollination, functionally mimicking poricidal anthers, though the anthers of most Pedicularis species are laterally dehisced (Harder, 1990; Huang and Shi, 2013). Unlike short-tubed beakless species, which are usually pollinated by bumble-bees collecting nectar, species with beaked corollas with either short or long tubes are usually observed to be pollinated by pollen-collecting bumble-bees (Macior, 1982). Phylogenetic analyses in Pedicularis suggest that loss was more frequent than gain of the beak (Ree, 2005b; Yu et al., 2015). An association of a beaked corolla with lack of nectar has long been noted in this superdiverse genus, implying an interesting reverse transition: gain, rather than loss, of nectar.

Although nectary remnants of different sizes have been observed in various beaked Pedicularis species (Huang and Fenster, 2007; Liu et al., 2015), there have been no systematic direct measurements of nectar properties or examination of the function of nectar in pollination. Some Pedicularis species which were formerly regarded as nectarless do secrete dilute but measurable nectar (Liu et al., 2016; Tong and Huang, 2016). To test the effects of nectar production on pollinator foraging behaviour and pollination mode, here we used a typically beaked, buzz-pollinated, ‘nectarless’ species, Pedicularis dichotoma, as an experimental model flower. In a recent phylogenetic analysis, P. dichotoma and most of its relatives (seven out of nine) in the same clade were nectarless species with beaks and short tubes, and the closest related monophyletic sister species P. batangensis is a nectarless species with a long tube and a beak, suggesting that P. dichotoma was probably derived from a nectarless species (Yu et al., 2015). We first measured its nectar properties and floral and individual variation in nectar volume and sugar concentration. Then we examined pollinator responses to flowers with added artificial nectar (Johnson and Nilsson, 1999; Johnson et al., 2004) and the effect on pollination modes.

MATERIALS AND METHODS

Study species and sites

Pedicularis dichotoma Bonati is a perennial alpine herb endemic to south-west China, usually flowering from late July to early September. The hermaphroditic flower is adichogamous, with female and male functioning simultaneously. Flowering individuals are up to 30 cm high, and the erect stem is somewhat woody (Yang et al., 1998). It is a typically ‘beaked’ species with a short corolla tube (1.25 ± 0.12 cm, n = 45, mean ± s.e.), and the upper two corolla lobes are fused forming a hoodlike galea that encloses the four anthers. The beak is pink to purple, while the lower lip changes from white to purple over the 3 d of anthesis. At the base, the corolla tubes are enclosed by pinkish brown bracts (Fig. 1). Wild plants usually grow in rocky or dry slopes, but can be seen in relatively wet meadows. The study site was in a field station, Shangri-La Alpine Botanic Garden (SABG, 27°54’5’’N, 99°38’17’’E, 3300–3350 m above sea level), from early August to early September in 2017 with about 1000 flowering individuals of P. dichotoma.

Fig. 1.

Fig. 1.

Open in a new tab

(A) Lateral view of a Pedicularis dichotoma flower, with the white arrow indicating the level of natural nectar around the ovary at the base of the floral tube, and the pink arrow indicating the level of added artificial nectar. (B) A bumble-bee is collecting P. dichotoma pollen in a sternotribic position, and the black arrow indicates the direction of probing used by a bumble-bee sucking nectar.

Measurements of nectar properties

To examine variation in nectar production and the nectar secretion schedule during the day or in the dark, we tagged and numbered 60 P. dichotoma individuals at two sites and bagged the whole inflorescence on each plant to exclude pollinator visits. Thirty plants were from one rocky slope which was relatively drier than the meadow from which the other 30 plants were obtained. We sampled nectar at 07.00 and 19.00 h on 13, 15 and 20 August, to reveal the day/night schedule. On each nectar sampling occasion, one flower from each individual was removed from the pedicel and was carefully taken back to the laboratory (within 600 m). Microcapillary tubes of 0.1 mm diameter were used to extract nectar from the base of the floral tube. To avoid sample contamination, particular care was taken to avoid tissue damage (which could cause other tissue fluids to leak into the nectar) (Herrera et al., 2006). The length of the liquid in the microcapillary tube (L) was measured, and the volume of the liquid was thus calculated as L × π × (0.1/2)2. Then the liquids were blown out by means of a rubber pipette bulb onto the measuring platform of a handheld refractometer (Eclipse 45–81, measuring range 0–50 %; Corbet, 2003), to measure sugar concentration. A total of 360 flowers (30 plants × 2 sites × 2 sampling flowers per day × 3 d) were sampled.

To measure sugar components of P. dichotoma, we withdrew nectar from a total of approx. 200 bagged flowers using microcapillary tubes, and spotted nectar onto filter paper strips, with air-drying at room temperature (Galetto and Bernardello, 2005). Each paper strip was then stored in a clean centrifuge tube. In the laboratory, the sugars were eluted from the dried filter paper with 50 μL of distilled water. Sugars were identified and their relative masses were quantified by high-performance liquid chromatography (HPLC) using a liquid chromatograph (Waters Corporation, Milford, MA, USA) with a refractive index detector and an Agilent Zorbax (Agilent Technologies, Santa Clara, CA, USA) carbohydrate analysis column, 843300–908. The solvent was an acetonitrile:water system (70:30 v/v) at a flow rate of 1 mL min–1 and a temperature of 35°C. Injected volumes were 20 μL for calibration and 20 μL for the nectar samples. For calibration, regression analyses were used based on the response peak to standard sugar solutions. The coefficient of determination of the regression was ≥0.9995. Quantities of each sugar (glucose, fructose and sucrose) were determined by comparison with the standards using the regression equations and were then expressed as relative percentage by mass.

Pollinator response to additional nectar

To investigate whether the presence or absence of nectar affects bumble-bees’ visitation behaviour, we compared pollinator foraging behaviour on flowers with artificial nectar or control (unmanipulated) flowers at two sites with abundant P. dichotoma individuals in August 2012. In each site, within a 5 × 5 m patch, using a microlitre syringe, we added artificial nectar (30 % sucrose solution) to newly open flowers on 30 plants which were randomly selected, while the other (approx. 100) plants mixed in the patch did not receive added nectar as controls. The numbers of unmanipulated and manipulated flowers at the two sites were 403 and 106, and 310 and 106. About 1–2 μL of artificial nectar was injected into the upper part of the corolla tube, carefully to avoid damaging the corolla (Fig. 1A; Thairu and Brunet, 2015). As there is a curve in the corolla tube, nectar solution remained attached to the inside of the tube wall above the curve and could not be syringed into the middle part of the corolla tube. Sugar concentration in the artificial nectar was similar to that in two nearby Pedicularis species, 28.5 ± 0.7% (n = 175 flowers) in P. densispica Franchet ex Maximowicz and 28.6 ± 0.7 (n = 190) in P. rex C. B. Clarke, generally with standing crops of about 0.3–1.3 μL. These two nectariferous species were usually nototribically pollinated by nectar-collecting workers of Bombus friseanus Skorikov, sharing the same pollinator with P. dichotoma (Huang and Shi, 2013; Corbet and Huang, 2014). We recorded pollinator visitation frequency and their foraging behaviour in these two sites, noting in particular whether the bumble-bee was collecting pollen or nectar and whether bee bodies were in contact with pollen or stigmas dorsally or ventrally, i.e. nototribic or sternotribic pollination mode, respectively. Flower manipulations and pollinator observations were conducted on three consecutive sunny days, with a total of 59 censuses, each lasting 15 min.

Data analysis

To detect a difference in nectar volume or concentration among individuals (as six flowers per plant were sampled), we used a generalized linear model (GLM) with normal distribution and an identity link function, fixing plant as a random factor. The effect of habitat on nectar traits of individuals (averaged by flowers with measurable nectar) was also tested by using a GLM with normal distribution and an identity link function, fixing site (dry or wet habitat) as the main factor.

To detect a possible relationship between the mean nectar concentration/volume and the number of flowers with measurable nectar, we used Spearman’s rank correlation.

The effect of adding nectar on flower visitation frequency was examined using a GLM with normal distribution and a logit-link function, fixing treatment (control or nectar added), reward type (for pollen or for nectar) and their interactions as factors. Pairwise comparisons for treatment and reward type foraged were also conducted within the model.

All the analyses were conducted in SPSS 20.0 (IBM Inc., New York).

RESULTS

Nectar production in unmanipulated flowers

Variation among flowers

Among the total of 360 flowers sampled, 131 flowers (36.4 %) presented liquid ranging from 0.01 to 0.49 μL (mean 0.07 μL). The liquid from 95 flowers (26.4 %) was enough for measurement by the refractometer; sugar concentration ranged from 4.0 to 39.0 % (mean 14.6 %). Among these 95 flowers, 54.7 % (52/95) secreted nectar with sugar concentration from 8 to 15 % (Fig. 2A). For those flowers secreting measurable nectar, nectar volume was negatively correlated with sugar concentration (n = 95, r = –0.296, P = 0.004).

Fig. 2.

Fig. 2.

Open in a new tab

Frequency distribution of (A) sugar concentration in sampled unmanipulated flowers (n = 360) and (B) numbers of flowers that secreted nectar in each individual (unmanipulated) plant (n = 60) in Pedicularis dichotoma. Note that 229 flowers which do not secrete liquids and 36 flowers secreting liquids but at unmeasurable levels are treated as 0 in (A).

Variation among individuals

Among the 60 tagged P. dichotoma individuals, 83.3 % (50/60) secreted liquid and 75.0 % (45/60) secreted measurable nectar in the corolla tube in at least one flower. Only 6.7 % of individuals (4/60) had more than three flowers with measurable nectar (Fig. 2B). Most (54/60, 90 %) individuals secreted 0–0.05 μL (mean volume from six flowers) of nectar (Fig. 3A), and the sugar concentration of 70 % (42/60) of individuals was from 0 to 5 % (Fig. 3B).

Fig. 3.

Fig. 3.

Open in a new tab

Frequency distribution of nectar volume (A) and sugar concentration (B) of 60 sampled individuals (mean values of six flowers) in Pedicularis dichotoma.

Both nectar volume (Wald χ2 = 107.28, d.f. = 59, P < 0.001) and sugar concentration (Wald χ2 = 100.74, d.f. = 59, P <0.001) differed significantly among individuals. Twenty-five individuals (41.7 %) had flowers with quite concentrated nectar (higher than the mean concentration of 14.6 %). For these individuals, on average 66.1 ± 5.6 % of sampled flowers on each individual secreted high concentration nectar (>14.6 %), indicating that flowers which produced concentrated nectar were more likely to appear on the same plant (Fig. 2).

Mean nectar volume (averaged for six flowers per plant) and mean sugar concentration were positively correlated (n = 60, r = 0.776, P < 0.001) but not if zero values of either volume or concentration were excluded (n = 60, r = 0.248, P = 0.056). The number of flowers secreting measurable nectar per plant was positively correlated with mean nectar volume (n = 60, r = 0.660, P <0.001) but not correlated with mean sugar concentration (averaged from the flowers with measurable nectar) (n = 60, r = 0.100, P = 0.446).

Nectar volume of individuals (averaged from the flowers with measurable nectar) was smaller (Wald χ2 = 64.16, P = 0.030) in the dry habitat (0.043 ± 0.010 μL) than in the wet habitat (0.075 ± 0.010 μL), but the sugar concentration did not differ significantly (9.21 ± 1.33 % vs. 12.72 ± 1.35 %, Wald χ2 = 133.38, P = 0.065).

The HPLC results showed that the liquid sampled in the corolla tube consisted of fructose (30.30 %, mass percentage), glucose (4.92 %) and sucrose (64.89 %). The sucrose to hexose ratio was 1.84. It can be classified as sucrose-dominant nectar following the scale of Baker and Baker (1983).

Bumble-bee responses to flowers with nectar added

Field pollinator observations comprised 59 censuses (total 885 min). A total of 69 Bombus friseanus individuals made 308 visits to flowers to which nectar had been added and 44 bees made 245 visits to unmanipulated flowers. When visiting unmanipulated (control) flowers, a bumble-bee grasped the beak and performed buzzes; pollen grains were seen to be released from the tip of the beak, and deposited on the ventral surface of the bee (sternotribic pollination mode). On flowers with nectar added, bumble-bees usually sucked nectar, first landing on the lower lip, and probing the corolla opening with the tongue from the left side of the flower (Fig. 1B), i.e. adopting a nototribic pollination mode (but not causing pollen transfer). The bee quickly learned where the nectar was located and foraged continuously for artificial nectar in the patch during most of the foraging bouts. We did not note shifts of foraging behaviour between nectared and natural plants within a foraging bout. While the bee sucked nectar, the tip of the beak was above the dorsal side of the bumble-bee, and the stigma would sometimes contact the bee laterally or ventrally, where no pollen had been placed. This change in foraging behaviour prevented the stigma from contacting the ventral side of the bee. Bumble-bees were occasionally observed to collect pollen from to which nectar had been added, performing sternotribic pollination as in unmanipulated flowers.

On manipulated flowers with additional nectar, the visitation frequency of B. friseanus (0.2 ± 0.02 visits per flower h–1) was nearly five times (P < 0.001) that on unmanipulated flowers (0.04 ± 0.01 visits per flower h–1) (Fig. 4). On unmanipulated flowers with little, dilute nectar, bumble-bees were only observed collecting pollen (0.043 ± 0.010 visits per flower h–1); however, on flowers with artificial nectar, most foraging bumble-bees switched to sucking added nectar, and the frequency of pollen collection decreased significantly (P = 0.027). The frequency of nectar collection (0.188 ± 0.019 visits per flower h–1) was significantly higher (P < 0.001) than that of pollen collection (0.009 ± 0.005 visits per flower h–1), showing the bees’ preference for nectar over pollen rewards (Fig. 4).

Fig. 4.

Fig. 4.

Open in a new tab

Frequency of bumble-bee visits for pollen and for nectar in natural flowers (control) and flowers to which nectar had been added in Pedicularis dichotoma. P-values <0.05 for comparison between groups are marked with asterisks.

DISCUSSION

Our investigation of nectar production in Pedicularis dichotoma showed that both nectar volume and sugar concentration varied greatly within and between plants. This species has been considered as a nectarless, short-tubed, beaked species, given that it was observed to be effectively pollinated by pollen-collecting bumble-bees (Huang and Shi, 2013; Corbet and Huang, 2014). Our flower manipulation showed that bumble-bee pollinator foraging behaviour changed with supplemental nectar in which the volume and sugar concentration were similar to nearby nectariferous Pedicularis species. The shifts from pollen to nectar preference and from sternotribic to nototribic pollination mode suggest the importance of nectar production in the evolution of diverse corolla morphologies in Pedicularis species, highlighting flower divergence mediated by the same pollinator species through a diverse use of pollen placement sites.

In a recent survey of nectary anatomy and nectar production in flowers of Pedicularis section Cyathophora, Liu et al. (2016) observed that two nectariferous, beakless species usually contained about 1 μL of standing crop of nectar with a sugar concentration of around 30 %, and two out of three beaked species produced dilute nectar with a much lower sugar concentration: 0.16 and 1.37 μL of nectar with 6 and 7 % sugar concentration in P. superba and P. cyathophylla (n = 20 flowers), respectively (Liu et al., 2016). These authors found substantial nectary structure and accumulated starch grains in the nectary tissues in nectariferous species, but variation in nectar production data among flowers or individuals was not present (see Herrera, 2006), particularly in beaked species. We observed that the liquid detected at the base of corolla tubes in P. dichotoma was indeed nectar, but measurable nectar was detected in 75 % of individuals and in only 26.4 % of flowers, unlike the two sympatric nectariferous species that we sampled, P. densispica and P. rex, in which it was present in 100 % of 32 individuals (32 × 6 = 192 flowers) in both species and in 95.2 % (175/192) and 99.0 % (190/192) of flowers, respectively.

Plants of P. dichotoma whose flowers secreted measurable nectar per plant tended to produce a large volume of nectar, and some plants tended to produce relatively concentrated nectar on the same individual. These observations suggest that differentiation in nectar production could be involved in this species, although the low volume of nectar was temporarily non-functional in that the nectar was not observed to be used by any floral visitors. We observed that nectar volume of individuals in the dry habitat was smaller than that in the wet habitat, but the sugar concentration was similar. Flowers in wetter conditions were observed to contain more dilute nectar in larger volumes (Wyatt et al., 1992).

Nectar properties including sucrose:hexose ratios and sugar concentration were used to predict pollinator functional groups (Baker and Baker, 1983, 1990). Our HPLC results showed that nectar in P. dichotoma was sucrose dominant, and the sucrose:hexose ratio was 1.84, predicting pollination by long-tongued bees or hawkmoths (see Baker and Baker, 1983). Interestingly, the other two nectariferous species produced similar sucrose-dominant nectar, in which fructose, glucose and sucrose were 11.45, 8.58 and 79.97 % in P. densispica, and 17.27, 3.44 and 79.30 % in P. rex. To track the evolutionary history of nectar properties in Pedicularis species, a study involving more species in various clades is needed given that sugar components in the genus have not been analysed by HPLC previously.

Due to the curved shape of corolla tubes in P. dichotoma, additional nectar was not injected at the level of the nectary, the bottom of corolla tubes where ovaries normally secrete a little nectar. This high nectar position may in some way exaggerate the possibility of a bee’s behaviour in exploiting nectar instead of collecting pollen, but the bees were not naïve during the flowering period (late August) and were skillful at collecting pollen. The bumble-bees seemed able to detect invisible nectar in P. rex in which the basal yellow corolla tubes were submerged by rainwater (Sun et al., 2016). The presence of nectar in a nectarless Pedicularis species induced a shift of pollinator behaviours from pollen to nectar collecting, but the underlying mechanisms of nectar detection in the study system requires further studies.

Gain or loss of nectar?

The presence/absence of nectar has been regarded as one of the most important innovations in Pedicularis flowers (Macior, 1982; Ree, 2005a, b; Huang and Fenster, 2007; Liu et al., 2015, 2016), but experimental manipulation of nectar status, such as nectar supplementation or depletion, has not previously been attempted. Early investigations suggested that beaked (nectarless) species were likely to be derived from nectariferous species given that the basal group in this genus comprised nectared, beakless species (Li, 1949, 1951; Macior, 1982). However, recent molecular phylogenies suggested that gain of beakless (nectared) species frequently occurred rather than loss of nectar in this genus (Ree, 2005b; Yu et al., 2015). Ree (2005b) asked under what ecological conditions beakless floral forms might regain nectar production. In an investigation of nectar production in three beaked species of Pedicularis section Cyathophora, two species produced dilute nectar and one produced no measurable nectar. The transition of nectar production in this section was considered as a process of losing nectar in the evolution from nectariferous to nectarless species (Liu et al., 2016). Alternatively, it could be an intermediate stage of a reverse transition from nectarless to nectariferous species.

Plants offer nectar as a reward for pollinators which deliver pollen during nectar collection (Brandenburg et al., 2009). Their interactions largely depend on the reward quantity and quality offered by plants and energetic costs of floral visitors (Bell, 1986; Corbet et al., 1995). Studies have shown that pollinators alter their foraging patterns to switch to more rewarding flowers if they encounter empty or less rewarding flowers (Brandenburg et al., 2009), so that nectar-rich species or individual plants (i.e. orchids) benefit from higher pollination rates than rewardless forms (Smithson and Gigord, 2001; Tremblay et al., 2005). Nectarless flowers are unlikely to be visited continuously by nectar-collecting pollinators. Consistent with previous studies of nectar supplementation (e.g. Johnson et al., 2004; Jersáková and Johnson, 2006), bumble-bees visited P. dichotoma flowers to which nectar had been added significantly more frequently than unmanipulated flowers. The bees seemed able to detect added nectar without contacting the flower as they continuously probed flowers with artificial nectar in the experimental populations. Furthermore, we found that a change of bumble-bee foraging preference resulted in a shift between two pollination modes which are extensively exploited by two types of plant species: nectared or nectarless species (Macior, 1982). Both pollination modes were occasionally observed in some nectariferous but not beaked Pedicularis species (Wang and Li, 1998; Robart, 2005; Tang et al., 2007; Huang and Shi, 2013).

Changes of pollinator preference to nectared flowers have been shown in previous manipulations of nectar (e.g. Johnson et al., 2004; Jersáková and Johnson, 2006), but we have shown here a change of pollination mode as well. The finding may contribute to our understanding of floral divergence mediated by selection of the same pollinator group. Studies in diverse lineages have shown that transitions in floral traits are usually associated with shifts of pollinator functional groups (Fenster et al., 2004; Whittall and Hodges, 2007; Thomson and Wilson, 2008; van der Niet and Johnson, 2012), but it remains a challenge to understand flower divergence without a shift of pollinator groups (Wilson and Thomson, 1996).

It has been proposed that in Pedicularis flowers, changes in the galea from a simple domed form to forms in which the tip of the galea is extended into a pointed beak or long rostrum of Pedicularis flowers are accompanied by shifts of bumble-bees from nototribic (dorsal stigma contact) to sternotribic (ventral stigma contact) pollination (Macior, 1982; Robart, 2005). The beak structure has been considered as a key factor that initiated and mediated shifts of bumble-bee foraging behaviour in Pedicularis (Macior, 1982; Macior and Sood, 1991; Ree, 2005b; Robart, 2005; Yu et al., 2013) given that the beak is closely involved in buzz pollination mediated by large pollen-collecting bees (Buchman, 1983; De Luca and Vallejo-Marín, 2013; Corbet and Huang, 2014). However, our flower manipulations with supplemental nectar showed that nectar rather than the beak seemed to be the key factor driving a shift of bumble-bee foraging behaviour. When nectar was placed into beaked flowers in P. dichotoma, the bumble-bees collected both nectar and pollen, adopting a nototribic pollination mode. The effects of bumble-bee behaviours on pollination were not examined in the nectar supplementation experiment, although this foraging behaviour was unlikely to deliver pollen effectively given that pollen grains were rarely placed on the ventral side of the bee. On the other hand, bumble-bee visits for pollen, in which flowers could be sternotribically pollinated, greatly decreased. The changes of bumble-bee foraging preference and behaviour would consequently reduce pollen transfer efficiency in plants with nectar supplementation, suggesting a potentially detrimental effect of nectar production in beaked species.

Over two-thirds of Pedicularis species are nectarless, providing only a pollen reward for bumble-bees, and are sternotribically pollinated. Community surveys of co-flowering Pedicularis species have often found the two pollination modes exploited by different plant species, with the use of diverse pollen placement niches on pollinator bodies facilitating the coexistence of plant species with shared pollinators (Macior and Sood, 1991; Macior and Tang, 1997; Eaton et al., 2012; Huang and Shi, 2013). Theoretical models suggest that nectar presentation can be an important resource partitioned among flower visitors in multispecies communities; populations can be easily invaded by mutants exhibiting variable degrees of nectar concealment or exposure (Rodríguez-Gironés and Santamaría, 2005). The species diversity centre of the genus is located in the East Himalayas, which harbour >300 Pedicularis species. The reproductive success of flowering species in mountainous areas is generally subject to pollinator limitation (Phillipp et al., 1996; Totland, 1997), as is evident in some Pedicularis species that exhibit autogamy (Sun et al., 2005). The fact that evolutionary transitions involving gains of nectar occurred more frequently than losses could be that nectariferous species benefit in terms of pollinator attraction, gaining higher visitation rates than nectarless species, as has been observed in several co-flowering species (Harder, 1990; Tong and Huang, 2018) as well as in the nectar supplementation experiments shown here. We did not see bumble-bees buzzing and releasing pollen from the tip of the beak when they were sucking artificial nectar on the lower lips of P. dichotoma. This is surprising given that bumble-bees were buzzing when they sucked nectar from two nearby nectariferous Pedicularis species (Corbet and Huang, 2014). If no pollen is released from the beak without buzzing, then a shift back to nectar foraging could not be realized without an accompanying (or even preceding) shift in pollen release mechanisms that would remove the need for buzzing. This could explain why nectarless species are often associated with the presence of corollas with beaks in Pedicularis species. Species involving loss of nectar are likely to experience a low number of pollinator visits. Under such a circumstance, selection for presenting large pollen doses per visit via actively collecting pollen by bees (sternotribic pollination) would be favoured (Thomson et al., 2000). Further measurements of buzz frequency in nectar-supplemented flowers and differences in pollen transfer efficiency resulting from a shift of pollination mode may provide insights into flower divergence mediated by shared pollinators.

ACKNOWLEDGEMENTS

We thank Zhi-Xi Tian for field assistance, Zhen-Dong Fang and the staff of Shangri-La Garden for logistical support, participants at a symposium ‘Evolution of floral traits’ who kindly provided comments on this work presented by the first author during the International Botanical Congress in Shenzhen, China, and Sarah Corbet, Jeffrey Karron, James Thomson, Renate A. Wesselingh and an anonymous reviewer for valuable suggestions on improving this manuscript. This work was supported by the National Science Foundation of China (grant nos. 31730012, U1402267) to S.-Q.H., National Postdoctoral Program for Innovative Talents (grant no. BX201600059) and China Postdoctoral Science Foundation (grant no. 2017M610482) to Z.-Y.T.

LITERATURE CITED

  1. Adler LS. 2000. The ecological significance of toxic nectar. Oikos 91: 409–420. [Google Scholar]
  2. Aizen MA, Lozada M, Morales CL. 2011. Comparative nectar-foraging behaviors and efficiencies of an alien and a native bumble bee. Biological Invasions 13: 2901–2909. [Google Scholar]
  3. Armbruster WS, Shi XQ, Huang SQ. 2014. Do specialized flowers promote reproductive isolation? Realized pollination accuracy of three sympatric Pedicularis species. Annals of Botany 113: 331–340. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Baker HG, Baker I. 1983. Floral nectar sugar constituents in relation to pollinator type. In: Jones CE, Little RJ, eds. Handbook of experimental pollination ecology. New York: Van Nostrand Reinhold, 117–141. [Google Scholar]
  5. Baker HG, Baker I. 1990. The predictive value of nectar chemistry to the recognition of pollinator types. Israel Journal of Botany 39: 159–166. [Google Scholar]
  6. Bell G. 1986. The evolution of empty flowers. Journal of Theoretical Biology 118: 253–258. [Google Scholar]
  7. Brandenburg A, Dell’Olivo A, Bshary R, Kuhlemeier C. 2009. The sweetest thing: advances in nectar research. Current Opinion in Plant Biology 12: 486–490. [DOI] [PubMed] [Google Scholar]
  8. Buchmann SL. 1983. Buzz pollination in angiosperms. In: Jones CE, Little RJ, eds. Handbook of experimental pollination biology. New York: Van Nostrand Reinhold, 73–113. [Google Scholar]
  9. Cardinal S, Buchmann SL, Russell AL. 2018. The evolution of floral sonication, a pollen foraging behavior used by bees (Anthophila). Evolution 72: 590–600. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Corbet SA. 2003. Nectar sugar content: estimating standing crop and secretion rate in the field. Apidologie 34: 1–10. [Google Scholar]
  11. Corbet SA, Huang S-Q. 2014. Buzz pollination in eight bumblebee-pollinated Pedicularis species: does it involve vibration-induced triboelectric charging of pollen grains?Annals of Botany 114: 1665–1674. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Corbet SA, Saville NM, Fussell M, Prys-Jones OE, Unwin DM. 1995. Bees, nectar and the competition box. Journal of Applied Ecology 32: 707–719. [Google Scholar]
  13. Crepet WL, Niklas KJ. 2008. Darwin’s second ‘abominable mystery’: why are there so many angiosperm species?American Journal of Botany 96: 366–381. [DOI] [PubMed] [Google Scholar]
  14. De la Barrera E, Nobel PS. 2004. Nectar: properties, floral aspects, and speculations on origin. Trends in Plant Sciences 9: 65–69. [DOI] [PubMed] [Google Scholar]
  15. De Luca PA, Vallejo-Marín M. 2013. What’s the ‘buzz’ about? The ecology and evolutionary significance of buzz-pollination. Current Opinion in Plant Biology 16: 429–435. [DOI] [PubMed] [Google Scholar]
  16. Eaton DA, Fenster CB, Hereford J, Huang SQ, Ree RH. 2012. Floral diversity and community structure in Pedicularis (Orobanchaceae). Ecology 93: S182–S194. [Google Scholar]
  17. Fenster CB, Armbruster WS, Wilson P, Dudash MR, Thomson JD. 2004. Pollination syndromes and floral specialization. Annual Review of Ecology, Evolution, and Systematics 35: 375–403. [Google Scholar]
  18. Galen C, Geib JC. 2007. Density-dependent effects of ants on selection for bumblebee pollination in Polemonium viscosum. Ecology 88: 1202–1209. [DOI] [PubMed] [Google Scholar]
  19. Galetto L, Bernardello G. 2005. Rewards in flowers: nectar. In: Dafni A, Kevan PG, Husband BC, eds. Practical pollination biology. Ontario, Canada: Enviroquest Ltd, 261–313. [Google Scholar]
  20. Grant V. 1994. Modes and origins of mechanical and ethological isolation in angiosperms. Proceedings of the National Academy of Sciences, USA 91: 3–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Harder LD. 1990. Pollen removal by bumble bees and its implications for pollen dispersal. Ecology 71: 1110–1125. [Google Scholar]
  22. Herrera CM, Pérez R, Alonso C. 2006. Extreme intraplant variation in nectar sugar composition in an insect-pollinated perennial herb. American Journal of Botany 93: 575–581. [DOI] [PubMed] [Google Scholar]
  23. Huang SQ, Fenster CB. 2007. Absence of long-proboscid pollinators for long-corolla-tubed Himalayan Pedicularis species: implications for the evolution of corolla length. International Journal of Plant Sciences 168: 325–331. [Google Scholar]
  24. Huang SQ, Shi XQ. 2013. Floral isolation in Pedicularis: how do congeners with shared pollinators minimize reproductive interference?New Phytologist 199: 858–865. [DOI] [PubMed] [Google Scholar]
  25. Irwin RE, Adler LS, Brody AK. 2004. The dual role of floral traits: pollinator attraction and plant defense. Ecology 85: 1503–1511. [Google Scholar]
  26. Jersáková J, Johnson SD. 2006. Lack of floral nectar reduces self-pollination in a fly-pollinated orchid. Oecologia 147: 60–68. [DOI] [PubMed] [Google Scholar]
  27. Johnson SD, Nilsson LA. 1999. Pollen carryover, geitonogamy, and the evolution of deceptive pollination system in orchids. Ecology 80: 2607–2619. [Google Scholar]
  28. Johnson SD, Peter CI, Agren J. 2004. The effects of nectar addition on pollen removal and geitonogamy in the non-rewarding orchid Anacamptis morio. Proceedings of the Royal Society B: Biological Sciences 271: 803–809. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Johnson SD, Hobbhahn N, Bytebier B. 2013. Ancestral deceit and labile evolution of nectar production in the African orchid genus Disa. Biology Letters 9: 20130500. doi: 10.1098/rsbl.2013.0500. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Koptur S, Smith AR, Baker I. 1982. Nectaries in some neotropical species of Polypodium (Polypodiaceae): preliminary observations and analyses. Biotropica 14: 108–113. [Google Scholar]
  31. Kwak MM. 1979. Effects of bumblebee visits on the seed set of Pedicularis, Rhinanthus and Melampyrum (Scrophulariaceae) in the Netherlands. Plant Biology 28: 177–195. [Google Scholar]
  32. Li HL. 1949. A revision of the genus Pedicularis in China. Part II. Proceedings of the Academy of Natural Sciences of Philadelphia 101: 1–214. [Google Scholar]
  33. Li HL. 1951. Evolution in the flowers of Pedicularis. Evolution 5: 158–164. [Google Scholar]
  34. Liu YN, Li Y, Yang FS, Wang XQ. 2016. Floral nectary, nectar production dynamics, and floral reproductive isolation among closely related species of Pedicularis. Journal of Integrative Plant Biology 58: 178–187. [DOI] [PubMed] [Google Scholar]
  35. Liu ML, Yu WB, Kuss P, Li DZ, Wang H. 2015. Floral nectary morphology and evolution in Pedicularis (Orobanchaceae). Botanical Journal of the Linnean Society 178: 592–607. [Google Scholar]
  36. Macior LW. 1982. Plant community and pollinator dynamics in the evolution of pollination mechanisms in Pedicularis (Scrophulariaceae). In: Armstrong JA, Powell JM, Richards AJ, eds. Pollination and evolution. Sydney: Royal Botanic Gardens, 29–45. [Google Scholar]
  37. Macior LW, Sood SK. 1991. Pollination ecology of Pedicularis megalantha D. Don (Scrophulariaceae) in the Himachal Himalaya. Plant Species Biology 6: 75–81. [Google Scholar]
  38. Macior LW, Tang Y. 1997. A preliminary study of the pollination ecology of Pedicularis in the Chinese Himalaya. Plant Species Biology 12: 1–7. [Google Scholar]
  39. Mattila E, Kuitunen M. 2000. Nutrient vs. pollination limitation in Platanthera bifolia and Dactylorhiza incarnata (Orchidaceae). Oikos 89: 360–366. [Google Scholar]
  40. Natalis LC, Wesselingh RA. 2012. Shared pollinators and pollen transfer dynamics in two hybridizing species, Rhinanthus minor and R. angustifolius. Oecologia 170: 709–721. [DOI] [PubMed] [Google Scholar]
  41. Natalis LC, Wesselingh RA. 2013. Parental frequencies and spatial configuration shape bumblebee behavior and floral isolation in hybridizing Rhinanthus.Evolution 67: 1692–1705. [DOI] [PubMed] [Google Scholar]
  42. Neiland MRM, Wilcock CC. 1998. Fruit set, nectar reward, and rarity in the Orchidaceae. American Journal of Botany 85: 1657–1671. [PubMed] [Google Scholar]
  43. Nicolson SW, Nepi M, Pacini E 2007. Nectaries and nectar. Dordrecht: Springer Netherlands. [Google Scholar]
  44. van der Niet T, Johnson SD. 2012. Phylogenetic evidence for pollinator-driven diversification of angiosperms. Trends in Ecology and Evolution 27: 353–361. [DOI] [PubMed] [Google Scholar]
  45. Pacini E, Nepi M, Vesprini JL. 2003. Nectar biodiversity: a short review. Plant Systematics and Evolution 238: 7–21. [Google Scholar]
  46. Phillipp M, Woddell SRJ, Bocher J, Mattson O. 1996. Reproductive biology of four species of Pedicularis (Scrophulariaceae) in west Greenland. Arctic Alpine Research 28: 403–413. [Google Scholar]
  47. Pyke GH. 1991. What does it cost a plant to produce floral nectar?Nature 350: 58–59. [Google Scholar]
  48. Ree RH. 2005a Detecting the historical signature of key innovations using stochastic models of character evolution and cladogenesis. Evolution 59: 257–265. [PubMed] [Google Scholar]
  49. Ree RH. 2005b Phylogeny and the evolution of floral diversity in Pedicularis (Orobanchaceae). International Journal of Plant Sciences 166: 595–613. [Google Scholar]
  50. Robart BW. 2005. Morphological diversification and taxonomy among the varieties of Pedicularis bracteosa Benth. (Orobanchaceae). Systematic Botany 30: 644–656. [Google Scholar]
  51. Rodríguezgironés MA, Santamaría L. 2005. Resource partitioning among flower visitors and evolution of nectar concealment in multi-species communities. Proceedings of the Royal Society B: Biological Sciences 272: 187–192. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Rojas-Nossa SV, Sánchez JM, Navarro L. 2016. Nectar robbing: a common phenomenon mainly determined by accessibility constraints, nectar volume and density of energy rewards. Oikos 125: 1044–1055. [Google Scholar]
  53. Smithson A, Gigord LDB. 2001. Are there fitness advantages in being a rewardless orchid? Reward supplementation experiments with Barlia robertiana. Proceedings of the Royal Society B: Biological Sciences 268: 1435–1441. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Southwick EE. 1984. Photosynthate allocation to floral nectar: a neglected energy investment. Ecology 65: 1775–1775. [Google Scholar]
  55. Sun SG, Guo Y-H, Gituru RW, Huang SQ. 2005. Corolla wilting facilitates delayed autonomous self-pollination in Pedicularis dunniana (Orobanchaceae). Plant Systematics and Evolution 251: 229–237. [Google Scholar]
  56. Sun SG, Armbruster WS, Huang S-Q. 2016. Geographic consistency and variation in conflicting selection generated by pollinators and seed predators. Annals of Botany 118: 227–237. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Tang Y, Xie JS, Sun H. 2007. The pollination ecology of Pedicularis rex subsp. lipkyana and P. rex subsp. rex (Orobanchaceae) from Sichuan, southwestern China. Flora 202: 209–217. [Google Scholar]
  58. Thairu MW, Brunet J. 2015. The role of pollinators in maintaining variation in flower colour in the Rocky Mountain columbine, Aquilegia coerulea. Annals of Botany 115: 971–979. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Thomson JD, Wilson P. 2008. Explaining evolutionary shifts between bee and hummingbird pollination: convergence, divergence, and directionality. International Journal of Plant Sciences 169: 23–38. [Google Scholar]
  60. Thomson JD, Wilson P, Valenzuela M, Malzone M. 2000. Pollen presentation and pollination syndromes, with special reference to Penstemon. Plant Species Biology 15: 11–29. [Google Scholar]
  61. Tong ZY, Huang SQ. 2016. Pre- and post-pollination interaction between six co-flowering Pedicularis species via heterospecific pollen transfer. New Phytologist 211: 1452–1461. [DOI] [PubMed] [Google Scholar]
  62. Tong ZY, Huang SQ. 2018. Safe sites of pollen placement: a conflict of interest between plants and bees?Oecologia 186: 163–171. [DOI] [PubMed] [Google Scholar]
  63. Totland Ø. 1997. Limitations on reproduction in alpine Ranunculus acris. Canadian Journal of Botany 75: 137–144. [Google Scholar]
  64. Tremblay R, Ackerman J, Zimmerman J, Calvo R. 2005. Variation in sexual reproduction in orchids and its evolutionary consequences: a spasmodic journey to diversification. Biological Journal of the Linnean Society 84: 1–54. [Google Scholar]
  65. Vallejo-Marín M, Manson JS, Thomson JD, Barrett SC. 2009. Division of labour within flowers: heteranthery, a floral strategy to reconcile contrasting pollen fates. Journal of Evolutionary Biology 22: 828–839. [DOI] [PubMed] [Google Scholar]
  66. Vallejo-Marín M, Da Silva EM, Sargent RD, Barrett SCH. 2010. Trait correlates and functional significance of heteranthery in flowering plants. New Phytologist 188: 418–425. [DOI] [PubMed] [Google Scholar]
  67. Wang H, Li DZ. 1998. A preliminary study of pollination biology of Pedicularis (Scrophulariaceae) in Northwest Yunnan, China. Acta Botanica Sinica 40: 204–210. [Google Scholar]
  68. Whittall JB, Hodges SA. 2007. Pollinator shifts drive increasingly long nectar spurs in columbine flowers. Nature 447: 706–709. [DOI] [PubMed] [Google Scholar]
  69. Willmer P. 2011. Pollination and floral ecology. Princeton, NJ: Princeton University Press. [Google Scholar]
  70. Wilson P, Thomson JD. 1996. How do flowers diverge? In: Lloyd DG, Barrett SCH, eds. Floral biology: studies on floral evolution in animal-pollinated plants. Boston, MA: Springer, 88–111. [Google Scholar]
  71. Wyatt R, Broyles SB, Derda GS. 1992. Environmental influences on nectar production in milkweeds (Asclepias syriaca and A. exaltata). American Journal of Botany 6: 636–642. [Google Scholar]
  72. Yang HB, Holmgren NH, Mill RR. 1998. Pedicularis L. In: Wu Z-Y, Raven PH, Hong DY, eds. Flora of China, Vol. 18 Beijing/St. Louis: Science Press/Missouri Botanical Garden Press, 97–209. [Google Scholar]
  73. Yu WB, Cai J, Li DZ, Mill RR, Wang H. 2013. Floral ontogeny of Pedicularis (Orobanchaceae), with an emphasis on the corolla upper lip. Journal of Systematics and Evolution 51: 435–450. [Google Scholar]
  74. Yu WB, Liu ML, Wang H, Mill RR, Ree RH, Yang JB, Li DZ. 2015. Towards a comprehensive phylogeny of the large temperate genus Pedicularis (Orobanchaceae), with an emphasis on species from the Himalaya–Hengduan Mountains. BMC Plant Biology 15: 176. doi: 10.1186/s12870-015-0547-9. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Annals of Botany are provided here courtesy of Oxford University Press

RESOURCES