Evol Biol (2010) 37:19–28 DOI 10.1007/s11692-010-9082-7
RESEARCH ARTICLE
Phylogenetic Community Context Influences Pollen Delivery to Allium cernuum Elissa M. Schuett • Jana C. Vamosi
Received: 28 December 2009 / Accepted: 26 February 2010 / Published online: 13 March 2010 Ó Springer Science+Business Media, LLC 2010
Abstract Few studies have examined how the number and identity of species in the neighbouring community influences the reproductive success of particular focal species. Pollen delivery, an important component of fitness of sexual plants, is a function of not just the floral traits of any particular individual, but of features of the population and community as it depends on pollinator abundance and preferences. Many pollinators in flowering communities will prefer patches with high floral abundance or diversity yet may exhibit lower floral constancy when more flowering species are present. Thus, pollination efficiency could increase or decrease with increased species richness and this will alter the selection pressures placed upon the floral traits (such as floral colour or reward) of any member of a particular community. Moreover, recent studies have indicated that plant-pollinator networks are phylogenetically structured (pollinators visit related plant species more than expected by chance) and this may be an important factor structuring flowering plant communities. Thus, the sheer number of species within a patch may be less important than the number of closely-related species. We investigate whether species richness or phylogenetic proximity of coflowering species influences the amount and proportion of conspecific pollen delivered to nodding onion, Allium cernuum, in fragment patches of Garry Oak meadows in South Western British Columbia, Canada. We find that pollen delivery depended upon the presence of close relatives far more than on species richness or population density, indicating a central role of the community structure on pollination in flowering plant communities. E. M. Schuett J. C. Vamosi (&) Department of Biological Sciences, University of Calgary, 2500 University Dr. NW, Calgary, AB, Canada e-mail:
[email protected]
Insofar as pollen delivery relates to the relative number of seeds produced by members of the community, pollination may in turn determine the community structure of the next generation. Keywords Allium cernuum Conspecific pollen delivery Species richness Phylogenetic diversity
Introduction There is emerging interest in how interacting neighbours may influence the fitness of a particular species (Thompson 2005). When the community composition changes, this alters the dynamics within hosting communities with regard to species diversity, phylogenetic community structure, and the selection pressures on neighbouring competing and/ or mutualistic species (Webb 2000; Harmon et al. 2009). Interestingly, such a change in selection pressures can then produce increasingly heterogeneous landscapes that encourage further diversification (Hansen et al. 2000; Vamosi et al. 2006). From a more applied perspective, when the community is altered by the inclusion of invading species the selection pressures on the residing species in the native community may change in much the same way (Strauss et al. 2006; Vellend et al. 2007; Morales and Traveset 2009). Further study of these processes requires knowledge of how members of the community may influence the reproductive success of a focal species (Sargent and Ackerly 2008). Within flowering plant communities, competition between coflowering species has been identified as a factor influencing population dynamics (Rathcke 1988; Waites and Agren 2004) and the evolution of pollination systems
123
20
Fig. 1 Diagram portraying how pollen delivery may be affected when patches differ in terms of species richness. Species (or phylogenetic) diversity may increase conspecific pollen delivery due to increasing the sheer number of pollinators recruited to the neighborhood. However, if population density of a focal species is the predominant factor influencing conspecific pollen delivery, then diversity may be associated with lower degrees of pollen delivery. Furthermore, if pollinator sharing increases with species richness, the net effect of increased species or phylogenetic delivery may be negative because a reduced proportion of the visits are concentrated on any given species
(Sargent and Otto 2006). Pollinator-mediated competition may influence the prevalence and structure of flowering species within a community. The majority of plant species are not specialized to a specific plant-pollinator interaction and can be pollinated by several different pollinator species (Sargent and Otto 2006), which in turn pollinate many different plant species. While generalized plant-pollinator interactions may prove to be beneficial for pollinator facilitation in small communities or for rare plants (Ghazoul 2006), competition for shared pollinators may negatively influence reproductive success of a focal species through pollen limitation (Fig. 1). The degree to which competition versus facilitation occurs depends on how many coflowering species are present and their respective identities. Facilitation of pollination can occur for rare species because the pollinator visitation rate to a local patch increases or pollinators disproportionately visit varied patches for optimal foraging of diverse resources (e.g., pollen and nectar). Facilitating effects have been observed to operate with the addition of one other species into the neighbourhood but the addition of a third species did not appear to increase the degree of facilitation (Ghazoul 2006) and no studies have examined the effect of the presence of realistic numbers of coflowering species under natural conditions (C4 species). Our knowledge of how community composition (the identities rather than the number of species present) affects the outcome of pollination is even more limited but there is some indication that the presence of closely related species may result in competition (Caruso 2000; Brown et al. 2002) while more distantly related species may have a facilitating effect. Community composition may influence pollen receipt, and thus reproductive success, due to a number of factors. The sharing of pollinators may bring about a decline in pollen delivery success through (1) decreased visitation and (2) increased pollen transfer between individuals of different species, referred to as interspecific pollen transfer (Waser 1978). A decrease in visitation may occur if
123
Evol Biol (2010) 37:19–28
pollinators prefer to visit other reward-producing species in the community other than the focal species, resulting in a decline in the number of conspecific pollen grains delivered between individuals of a focal species (Thomson 1986; Wilson and Stine 1996). Furthermore, generalist pollinators may visit a mixture of flowering species within a single foraging bout (Gegear 2005; Morales and Traveset 2008) and this may occur more often if more closelyrelated species are present within the community. This lack of constancy within a foraging bout can lead to increased interspecific pollen transfer when the community is composed of more similar species. Interspecific pollen transfer influences reproductive success through heterospecific (different species) pollen deposition, which reduces reproductive success by chemical prevention of tube formation, ovule abortion and/or stigma clogging of the recipient (Kohn and Waser 1985; Flanagan et al. 2009) and is often referred to as the ‘‘quality’’ component of pollen delivery (Herrera 1987). Taken together, the diversity of the community can therefore present benefits (by increasing pollinator abundance) or losses (through increased competition), and a further understanding of the net effect in terms of pollen delivery (Fig. 1) will help us rank the destruction of native habitats due to habitat fragmentation and exotic plant invasions (Wilcock and Neiland 2002). Much of the work on how the community context alters pollen delivery has focused the population density of certain focal species (Bosch and Waser 1999; Waites and Agren 2004; Zorn-Arnold and Howe 2007), with conspecific pollen delivery often increasing with population size (but see Bosch and Waser 2001). With increasing species richness, there is a general trend for a decrease in population size of each species in the community (Jakobsson et al. 2008). If species richness is negatively related to the population size/density (Fig. 1), then we may expect that conspecific pollen delivery will be negatively related with the species richness of a patch. While a complete analysis of all of the potential pathways would be enlightening, the most pertinent piece of missing information is the general net effect of diversity on pollen delivery. If the number and identity of co-flowering species within the community has a positive net effect, then populations of a focal species within diversity-rich habitats may evolve to reduce their floral rewards (as their pollination is facilitated by the increased diversity). Should the opposite be true, population of a focal species should evolve increased floral rewards in diversity-rich environments. An understanding of the effect of species and phylogenetic diversity on the pollination of a constituent species is critical to our knowledge of community dynamics and has implications on how an altering geographical landscape (such as the introduction of invasives) could place divergent selection pressures upon a species (Thompson 2005).
Evol Biol (2010) 37:19–28
In this study, we analyzed pollen delivery within the areas representative of the Garry Oak ecosystem on Vancouver Island and the Gulf Islands, British Columbia. The impact of habitat fragmentation and introduction of invasives makes the Garry Oak ecosystem one of the most at-risk habitats in Canada. Many species of endangered status are found solely within this habitat. Invasion of exotic species, such as the European Scotch broom (Cytisus scoparius), have spread to approximately 82% of the herbaceous cover of this ecosystem while less than 5% of the original habitat remains, with many native species facing extinction (Fuchs 2001). With a select few native flowering species inhabiting each outcrop of this ecosystem on the Gulf Islands, this habitat provides various study sites of differing diversity of species to measure heterospecific pollen delivery to a focal species native to the Garry Oak ecosystem. Within these outcrops, we investigated the heterospecific pollen delivery to Allium cernuum (nodding onion) from sites of different species richness. Allium cernuum was chosen as the study species due to its widespread distribution in the Garry Oak ecosystem, the reliance of this species on generalist pollinators for sexual reproduction, and the relative ease with which its pollen could be distinguished from other members of the community. Within these sites, we investigated (1) whether there exists a negative correlation between the species richness of a site and the quality of delivery to the species stigma and (2) whether phylogeny plays a role in conspecific and heterospecific pollen delivery within a community.
Methods Data Collection We gathered information from the UBC, UVic, Royal BC Museum, and Malaspina herbariums on locations of Allium cernuum on Vancouver Island and the Gulf Islands. We also identified possible sites based on the presence of the Garry Oak habitat, featuring Garry Oak, rock outcrops and grassland habitat. At each site, A. cernuum was identified, and the population size and density were estimated by counting number of individuals per square metre. At each site (11 in total), a single stigma was removed from each of five random A. cernuum individuals. Because pollinators are thought to search and choose among the locally-available floral resources (Olesen et al. 2008), we examined the coflowering species within a 3 m radius of the A. cernuum central individual (*28 m2 in area). Fragments of Garry Oak meadow were small and visual searches within a larger radius (*78 m2) revealed few, or no, additional species. Because we were interested in the dynamics within a tractable neighbourhood of plant species and the size of the
21
larger patches was observed to influence the density of individuals within our 28 m2 neighborhoods (in general, larger Garry Oak fragments were ‘‘good quality’’ patches supporting higher densities of A. cernuum), we examined the population density of A. cernuum for a given area (the residuals of the regression of total area on density); hereafter ‘‘corrected population density’’). This procedure effectively removes the effect of total area in the analysis. All coflowering species were identified and pollen samples were taken to create a pollen reference library based on microscopic, family specific, exine patterns. Within the 11 sites species richness ranged from 1 to 11 coflowering species. Generalist bee pollinators (largely Bombus sp. and Andrena sp.) were observed moving between flowers of A. cernuum on the same plant, on different plants, and between A. cernuum and surrounding species. However, as the degree of visitation and pollinator sharing often exhibits a mismatch between the pollinators observed (Alarcon 2010) and the identity of pollen actually delivered, we identified the pollen (conspecific and heterospecific) delivered to A. cernuum more directly. Pollen from coflowering species was dusted onto microscope slides, the exine distinguished with the use of basic fuchsin stain, and a digital image produced for future identification. The stigma pollen load of Allium cernuum flowers was observed by softening stigmas with a 10% ethanol solution and then, once dried, applied with the basic fucsion stain onto the microscope slide. The size and exine characteristics of pollen allowed us to distinguish and quantify heterospecific and conspecific pollen grains under 10009 magnification. In the case of monocot pollen, we identified to species level (whenever possible), while in the case of other heterospecific pollen we identified to species level when possible but often found pollen not matching any species in our reference collection. When pollen did not match any of the species in our pollen reference collection, we identified the heterospecific pollen identity to family level using online sources (e.g., www.paldat.org). Data Analysis The mean number of, and ratio between, conspecific and heterospecific pollen grains were calculated for each population and the identity of heterospecific pollen grains noted, where possible. Data that are counts (such as the number of pollen grains) typically violate the assumptions of constant variance and normal errors central to standard least squares regression methods (Crawley 2007). We therefore used the ‘‘glm’’ procedure in R (version 2.8.1; www.R-project.org) software to apply generalized linear models with Poisson errors to investigate associations between the number of pollen grains delivered and
123
22
species richness, population density, and/or phylogenetic distance. Exploratory analyses found evidence for overdispersion in the initial model (i.e., residual deviance was greater than residual degrees of freedom). Therefore, we applied quasipoisson errors, which account for statistical overdispersion (Crawley 2007). Because species richness may in turn exhibit a relationship with population density (e.g., Bock et al. 2007) and both may affect pollen delivery to A. cernuum, the most complex model included interactions among species richness and population density. Because our hypothesis (Fig. 1) predicts changes in the quality (proportion of pollen that is conspecific) as well as the quantity (total conspecific pollen delivery), we also examined the effects of population density, species richness, on the proportion of pollen delivered that was conspecific. These models were examined with the ‘‘glm’’ procedure in R using binomial errors, fitting the main effects and their interactions. Phylogeny and Pollen Delivery A phylogeny of the species found coflowering in the Garry Oak ecosystem during July, 2008 was constructed using Phylomatic (Webb and Donoghue 2005) to examine the effect of phylogenetic distance on pollen delivery. In order to address whether overall phylogenetic similarity of other members of the community to a focal species are important in determining pollen delivery, we used Phylocom (Webb et al. 2008) to calculate the total branch length between A. cernuum and the coflowering species within each patch and then generated a mean phylogenetic distance (MPD) between A. cernuum and all of its coflowering neighbors at each site. We again used the ‘‘glm’’ procedure in R (version 2.8.1; www.R-project.org) software to apply generalized linear models with Poisson errors to examine relationships between pollen delivery and the relatedness of the neighbouring communities to A. cernuum. Because species richness values are nested within the phylogenetic distance metric (most species in the area are not closely related to A. cernuum so mean phylogenetic distance increases as species richness increases), the phylogenetic distance and species richness analyses were run separately. Because the foraging choices of pollinators may vary with relative density of closely-related species there may be nonlinear relationships between pollen delivered, population density and phylogenetic distance of neighbours. We therefore fit the complex (main effects of phylogenetic distance and population density and their interactions). Exploratory analyses again found evidence for overdispersion in the initial model (i.e., residual deviance was greater than residual degrees of freedom) and we therefore reverted to quasipoisson errors to account for statistical overdispersion (Crawley 2007).
123
Evol Biol (2010) 37:19–28
Results Geographic Mosaic of Allium cernuum We located flowering Allium cernuum in Garry Oak communities at several locations on Gabriola Island, Quadra Island, and Vancouver Island of British Columbia (Table 1) from late-June-July 2008, starting our search for field sites in the South and moving northward such that the A. cernuum populations were found in roughly the same phenological stage. Population sizes ranged from 9 individuals per site in Elk Falls, Vancouver Island, to 16900 individuals per site on small islands in the Gowland Harbour (Quadra Island). Population density of Allium cernuum ranged between 9 plants/m2 and 130 plants/m2, however was most frequently between 9 and 40 plants/m2 (Table 1). The size of Garry Oak forest fragments ranged in area, making our total population size estimates of A. cernuum range from 9 to 16,900 individuals. Populations were separated by at least 500 m, minimizing the role of spatial autocorrelation in pollinator visitation (i.e., pollinators were assumed to fly between patches very infrequently). The number of species coflowering with Allium cernnum ranged from 1 to 11 and included: Medicago lupulina, Prunella vulgaris, Hypochaeris radicata, Rubus leucodermis, Achillea millefolium, Mimulus guttatus, Saxifraga ferruginea, Montia parvifolia, Brodiaea hyacinthina, Grindelia integrifolia, Rosa nutkana, Plantago lanceolata, Cerastium arvense, Rumex acetosella, Allium acuminatum, Sedum oreganum, Lactuca muralis, Taraxacum officinale, Epilobium cilatum, Physocarpus capitata, Leucanthemum vulgare, Lonicera hispidula, Trifolium repens. Species richness in A. cernuum neighbourhoods exhibited a negative relationship with the corrected population density of A. cernuum (t1,10 = -0.33; P = 0.0073). In examining the phylogenetic distribution of neighbours to A. cernuum, we found that the MPD was divided into two discrete groupings based on whether another monocot (Allium acuminatum or Brodiaea hyacinthina) was present (in 5 sites) or absent (6 sites) in the patch, all other co-flowering species belonging to the distantly related eudicots (Fig. 2). Total Stigmatic Pollen Load to Allium cernuum One stigma was damaged in transit and had to be excluded from the analysis, leaving a total of 54 stigmas analyzed for pollen load. The elliptic, boat-shaped pollen of Allium cernuum was relatively easy to distinguish from pollen of other species. Stigmatic pollen load varied from 0 to 1215 conspecific pollen grains per stigma and 0 to 46 heterospecific pollen grains per stigma. Despite this variance, our collected stigmas per population gave us a reasonable
Evol Biol (2010) 37:19–28
23
Table 1 Locations, coflowering species richness (SR), population density, total population size, and identities of heterospecific pollen found on the stigmata of Allium cernuum in all 11 study sites SR
Pop density (per m2)
Pop size
Heterospecific pollen
Location
Coordinates
Nanaimo
49°05.680 N, 123°55.884 W
7
12
360
Hypochaeris, Achillea, Fabaceae, Rosaceae, Lilium, unknown family
Neck Point Park
49°14.136 N, 123°57.815 W
5
30
180
Brodiaea, Rosaceae, Fabaceae
Pipers Lagoon Park
49°13.688 N, 123°57.147 W
4
50
900
Lilium, Fabaceae, Rosaceae, Asteraceae, Rumex, Plantago
Quadra, Gowland Harbour S
50°04.748 N, 125°13.176 W
2
28
28
Quadra, Gowland Harbour N
50°04.808 N, 125°13.272 W
3
130
16900
Quadra, Gowland Harbour W Elk Falls
50°04.765 N, 125°13.748 W 50°02.502 N, 125°19.861 W
2 8
130 3
6240 9
Rosaceae, Fabaceae, Liliaceae, Polygonaceae, unknown family Fabaceae, Rosaceae, Rumex, unknown Sedum, Rumex, Asteraceae, Rosaceae, Fabaceae Physocarpus, Fabaceae, Polygonaceae, Plantago
Gabriola Island N
49°11.988 N, 123°49.726 W
1
18
18
Rosaceae, Fabaceae, Mimulus
Gabriola Island SE
49°11.984 N, 123°49.752 W
2
15
30
Rosaceae, Fabaceae, Geranium
Gabriola Island SW
49°10.402 N, 123°51.030 W
11
30
90
Cirsium, Hypochaeris, Ericaceae, Fabaceae, Rosaceae
Gabriola Island NW
49°10.992 N, 123°51.798 W
3
40
120
Hypochaeris, Sedum, Rosaceae, Fabaceae, Polygonaceae
Fig. 2 Reconstructed phylogeny of all species found coflowering with A. cernuum in the 11 Garry Oak sites using Phylomatic. Axis indicates dimensionless proportion of total branch length between members of the community
estimate of the pollen delivery within a patch (mean conspecific pollen delivery per site ranged from 45.6 ± 22.4 to 372.4 ± 215.9). The pollen transport network (Table 1) indicates that Allium cernuum shares pollinator with 3–6 other members. Polygonaceae and Plataginaceae were
noted despite reports that they are wind-pollinated because their pollen may still usurp conspecific pollen placement. The average stigmatic pollen load per population ranged from 45 to 372 conspecific pollen grains and 1–16 heterospecific pollen grains per stigma. While the overall
123
24
Evol Biol (2010) 37:19–28
average conspecific pollen load (134.1 pollen grains per stigma) was well above the number of ovules available (6 per flower), previous reports indicate that Allium requires [ 60 pollen grains to achieve full seed-set (Molano-Flores et al. 1999) such that 4 out of the 11 populations (36%) may be expected to produce suboptimal seed set (assuming some failure rate of pollen tube formation). While work is currently underway to more formally assess pollen limitation in this species, we expect that Allium cernuum would experience increased fitness with increased stigmatic pollen load far beyond the number of ovules through increased pollen competition, as has been shown in other species (Niesenbaum 1999; Flanagan et al. 2009). Finally, we should note that 2008 was cooler and had more precipitation than the average expected for the area (Average June/July precipitation 32 mm/month and average temperature 22°C compared to 2008 with 41 mm/ month and 15.5°C; http://www.climate.weatheroffice.ec. gc.ca/climate_normals) and these factors may have influenced pollinator foraging behaviour. Heterospecific pollen grains composed only a small proportion of the total stigmatic pollen load, an overall average of 3.4% of all pollen grains. Using the pollen image reference library amassed from the area from late May to July, we found that many heterospecific pollen grains (*70%) were belonging to species not present in the immediate neighbourhood of A. cernuum (see Table 1), such that there was no relationship between the species richness of the neighbourhood patches and the number of different heterospecific species found on the stigma (F1,9 = 1.34; P = 0.276). There was a significant positive
relationship between the number of conspecific and heterospecific pollen grains delivered (Table 2), indicating additional visits by pollinators resulted in the addition of both con- and heterospecific pollen. Mean total stigmatic pollen load (conspecific ? heterospecific) of the population did not decrease as species richness increased or as population density decreased indicating that average visitation rate to A. cernuum per individual was not a simple reflection of its relative abundance in a given neighbourhood. Total stigmatic pollen load decreased with decreasing MPD (i.e., when close relatives were present) indicating that close relatives could have outcompeted A. cernuum for visits (Mitchell et al. 2009).
Table 2 Coefficients obtained from generalized linear models of the effects of population density, species richness (SR), phylogenetic distance (MPD), and the presence of coflowering monocots on total
pollen receipt (conspecific ? heterospecific), conspecific pollen receipt, and proportion of pollen received that is conspecific
Model
Community Species Richness and Composition Effects on Conspecific Pollen Delivery We expected species richness to alter the quantity and proportion of conspecific pollen grains per stigma (see Fig. 1) but observed little support for this hypothesis (Table 2; Fig. 3). There was a marginal effect of corrected population density as well as a marginally significant interaction between population density and species richness (Table 2). The interaction was in the direction to imply that patches that had a high number of species for a given population density experienced lower conspecific pollen delivery. Conversely, we observed that increased MPD increased the quantity and quality of pollen delivered to A. cernuum (Table 2). We did not find any effect between population density and MPD (t1,9 = -0.01, P = 0.86). The results were robust to reducing MPD to a
Total pollen receipt
Conspecific pollen receipt
Proportion pollen that is conspecific
-0.3304
-0.3406
-0.3922
Model 1 Log (SR) Population density (corrected)
1.6099
1.6618
1.1007*
Population ensity * Log (SR)
-0.9068
-0.9367
-0.6574*
1.9954*
2.0687*
Model 2 MPD Population density (corrected) Population density * MPD
1.4374**
1.6152 -0.7522
1.6412 -0.7635
0.6226 -0.2677
Model 3 Monocot
-1.0043*
-1.0367*
-0.82637**
Population density (corrected)
0.1129
0.1163
0.09401
Population density * Monocot
0.2647
0.2705
0.04451
Degrees of freedom = 7 for all models. Quasipoisson error structure was applied to the response variables of total pollen receipt and conspecific pollen receipt, while the binomial error structure was applied to the proportion of pollen that was conspecific. Significance codes: *** \0.001, ** \0.01, * \0.05, \0.1
123
Evol Biol (2010) 37:19–28
25
Fig. 3 Mean conspecific pollen delivery in response to a coflowering species richness, b population density (corrected), and c phylogenetic distance (MPD) and d the presence -1- or absence -0- of close relatives (from monocot lineage). Pollen grains averaged from 5 stigmas extracted from A. cernuum per site at 11 sites. Results were qualitatively the same (yet stronger) when the outlier was removed from the analysis (all results reported are with all 11 sites included)
binary character of whether another monocot species was present are not (Fig. 3; Table 2).
Discussion Community Dynamics and Pollen Delivery The influence of community composition on pollination is often ignored and can provide insight as to how community structure influences the selection pressures on a focal species (Armbruster 1985; Rathcke 1988; Caruso 2000; Bell et al. 2005). Analysis of conspecific and heterospecific pollen delivery to Allium cernuum in communities of varying species richness revealed that species diversity was not a significant influence on the quantity or quality of pollen delivered. Our results also indicated that pollinators did not visit A. cernuum in direct proportion to its population density, nor did increased plant species richness provide a general facilitating effect on conspecific pollen delivery (Ghazoul 2006; Lazaro et al. 2009). We did however find a strong role of phylogenetic similarity of the coflowering community members in that A. cernuum received a decrease in quantity and proportion of conspecific pollen when it cohabitated with congenerics
A. acuminatum or confamilial Brodiaea hyacintha. These results would suggest that phylogenetic community structure does indeed influence the fitness of resident plants (Webb 2000; Lazaro et al. 2009). Few studies thus far have considered phylogeny as an aspect of community structure when measuring pollen delivery to a focal species. Ghazoul (2006) investigated the effects of species richness on increasing pollinator visitation rates, and inferred that increasing species richness recruited more pollinators because of the increased diversity in food sources available (e.g., nectar and pollen). However, because related species are inferred to share traits through common descent (Felsenstein 1985) related species would thus have similar floral characteristics (Wright and Calderon 1995) which may increase competition for pollinators (Campbell and Motten 1985; Bell et al. 2005). Nevertheless, when additional pollinators can be recruited to the general area, facilitative effects have been observed (Moeller 2004; Moeller and Geber 2005) indicating that overall outcome is likely to depend on the degree of floral overlap and whether pollinators are limiting. Thus, we may have observed the effects of competition between relatives in A. cernuum, because all of the related coflowering monocots in the area have a similar floral morphology.
123
26
There was some support for opposing influences of species richness and population density on pollen delivery in the way we originally envisioned (Fig. 1), in that we found the expected negative relationship between species richness and population density. Conspecific pollen delivery increased with population density for a given species richness. Density has been observed to positively influence reproductive success both due to the increased chances of foraging constancy (Feinsinger et al. 1991; Kunin 1997; Waites and Agren 2004) and becoming the optimal foraging choice by pollinators (Geber and Moeller 2006). We did not however find any evidence for a general facilitating effect of high species richness. Our results may indicate that the increased recruitment of pollinators to a patch through increased species diversity may be balanced by decreased visitation to any particular species through decreased population density, such that generally no net effect is observed between conspecific pollen delivery and either variable. Of course, sample size may have reduced our ability to find strong effects, as only 11 sites of A. cernuum could be found. A more comprehensive study with a more frequent species, complete with pollinator visitation data, is required to confirm our findings regarding the opposing effects of species richness and population density. Interestingly, we also did not find a positive relationship between population density and MPD, as might be expected if close relatives competitively exclude one another, indicating that other factors other than competition (e.g., edaphic factors) may be influencing the community composition in Garry Oak fragments. While we did receive some support for the notion that quality of pollen received declines with the phylogenetic distance between coflowerers (i.e., when close relatives are present), this was the result of decreased quantity of conspecific pollen grains, while heterospecific pollen delivery was not influenced by phylogeny. The heterospecific pollen received by A. cernuum did not often belong to close relatives. A. cernuum received a low transfer of heterospecific pollen grains (3.4% of total pollen load) in comparison to similar studies, which accounted for heterospecific pollen delivery up to 80% of pollen load (Waites and Agren 2004). This resulting low deposition of heterospecific pollen to A. cernuum may be an indication of high pollinator constancy (i.e., pollinators either perform an entire (or nearly so) foraging bout on either A. cernuum or a relative, but not both). Alternatively, sufficient interspecific variation in morphology and placement of sexual organs on the flowers visited by the generalist pollinator may have affected placement of pollen on the pollinator’s body (Armbruster 1985; Geber and Moeller 2006). Regardless of the reasons why heterospecific pollen load was low, it is unlikely that heterospecific pollen delivery has a large effect on fitness in this species. We expect that fitness is
123
Evol Biol (2010) 37:19–28
determined more by A. cernuum successfully competing for the number of visits it requires for full seed set. The degree to which relatedness influences the degree of competition will also depend on the extent to which past competition has shaped the floral traits of species (Fenster et al. 2004). If the continued presence of coflowering relatives has resulted in character displacement between competing species (Caruso 2000), the influence of relatedness will decline. In this light, we may expect different results in studies that examine competition for pollen delivery in natural communities, where character displacement may have differentiated close relatives, and more applied studies examining the effects of fitness in terms of native seed set upon the arrival of invasive species (Brown and Mitchell 2001; Brown et al. 2002). While attractive invasive species may increase visitation to the community by attracting irregular pollinators, there appears to be more evidence that invasive species decrease the number of pollinator visits to native species by drawing generalist pollinators away from natives and disrupting mutalistic interactions built within the community (Traveset and Richardson 2006). Other studies have shown that the relationship between conspecific pollen delivery and introduction of invasives varies with native species examined and imply that a consideration of the relatedness between potential competing species may have a significant impact on pollinator preference and the competitive edge of invasive species (Strauss et al. 2006; Gilbert and Webb 2007; Morales and Traveset 2008; Morales and Traveset 2009). Because no previous coexistence between invasive and native congeners has taken place, they are especially likely to share important floral traits in common. As well as exhibiting the utility of the evolutionary perspective for examining modern day environmental problems, these applied studies will become increasingly relevant to a fundamental understanding of how rapidly, and under what circumstances, species evolve in response to their community (Thompson 2005). Acknowledgements We thank C. Hepp for field assistance, S. Vamosi for statistical consultation and abundant help on R programming, and two anonymous reviewers for their helpful suggestions. This study was funded by an NSERC Discovery Grant to JCV and a PURE award to EMS.
References Alarcon, R. (2010). Congruence between visitation and pollentransport networks in a California plant-pollinator community. Oikos, 119, 35–44. Armbruster, W. S. (1985). Patterns of character divergence and the evolution of reproductive ecotypes of Dalechampia scandens (Euphorbiaceae). Evolution, 39, 733–752.
Evol Biol (2010) 37:19–28 Bell, J. M., Karron, J. D., & Mitchell, R. J. (2005). Interspecific competition for pollination lowers seed production and outcrossing in Mimulus ringens. Ecology, 86, 762–771. Bock, C. E., Jones, Z. F., & Bock, J. H. (2007). Relationships between species richness, evenness, and abundance in a Southwestern savanna. Ecology, 88, 1322–1327. Bosch, M., & Waser, N. M. (1999). Effects of local density of pollination and reproduction in Delphinium nuttallianum and Aconitum columbianum (Ranunculaceae). American Journal of Botany, 86, 871–879. Bosch, M., & Waser, N. M. (2001). Experimental manipulation of plant density and its effect on pollination and reproduction of two confamilial montane herbs. Oecologia, 126, 76–83. Brown, B. J., & Mitchell, R. J. (2001). Competition for pollination: Effects of pollen of an invasive plant on seed set of a native congener. Oecologia, 129, 43–49. Brown, B. J., Mitchell, R. J., & Graham, S. A. (2002). Competition for pollination between an invasive species (purple loosestrife) and a native congener. Ecology, 83, 2328–2336. Campbell, D. R., & Motten, A. F. (1985). The mechanism of competition for pollination between two forest herbs. Ecology, 66, 554–563. Caruso, C. M. (2000). Competition for pollination influences selection on floral traits of Ipomopsis aggregata. Evolution, 54, 1546–1557. Crawley, M. J. (2007). The R Book. New York: John Wiley. Feinsinger, P., Tiebout, H. M., & Young, B. E. (1991). Do tropical bird-pollinated plants exhibit density-dependent interactions? Field experiments. Ecology, 72, 1953–1963. Felsenstein, J. (1985). Phylogenies and the comparative method. The American Naturalist, 125, 1–15. Fenster, C. B., Armbruster, W. S., Wilson, P., Thomson, J. D., & Dudash, M. R. (2004). Pollination syndromes and floral specialization. Annual Review of Ecology and Systematics, 35, 375–403. Flanagan, R. J., Mitchell, R. J., Knutowski, D., & Karron, J. D. (2009). Interspecific pollinator movements reduce pollen deposition and seed production in MImulus ringens (Phrymaceae). American Journal of Botany, 96, 809–815. Fuchs, M. A. (2001). Towards a recovery strategy for Garry Oak and associated ecosystems in Canada: Ecological assessment and literature review. Environment Canada, Canadian Wildlife Service, Pacific and Yukon Region. Geber, M. A., & Moeller, D. (2006). Pollinator responses to plant communities and implications for reproductive character evolution. In L. Harder & S. C. H. Barrett (Eds.), Ecology and evolution of flowers. Oxford: Oxford University Press. Gegear, R. J. (2005). Multicomponent floral signals elicit selective foraging in bumblebees. Naturwissenschaften, 92, 269–271. Ghazoul, J. (2006). Floral diversity and the facilitation of pollination. Journal of Ecology, 94, 295–304. Gilbert, G. S., & Webb, C. O. (2007). Phylogenetic signal in plant pathogen-host range. Proceedings of the National Academy of Sciences of the United States of America, 104, 4979–4983. Hansen, D. M., Armbruster, W. S., & Antonsen, L. (2000). Comparative analysis of character displacement and spatial adaptations as illustrated by the evolution of Dalechampia blossoms. The American Naturalist, 156, S17–S34. Harmon, L. J., Matthews, B., DesRoches, S., Chase, J., Shurin, J., & Schluter, D. (2009). Evolutionary diversification in stickleback affects ecosystem functioning. Nature, 458, 1167–1170. Herrera, C. M. (1987). Components of pollinator ‘quality’: Comparative analysis of a diverse insect assemblage. Oikos, 50, 79–90. Jakobsson, A., Padron, B., & Traveset, A. (2008). Pollen transfer from invasive Carpobrotus spp. to natives—A study of pollinator behaviour and reproductive success. Biological Conservation, 141, 136–145.
27 Kohn, J. R., & Waser, K. M. (1985). The effect of Delphinium nelsonii pollen on seed set in Ipomopsis aggregate, a competitor for hummingbird pollination. American Journal of Botany, 72, 1144–1148. Kunin, W. (1997). Population size and density effects in pollination: Pollinator foraging and plant reproductive success in experimental arrays of Brassica kaber. Journal of Ecology, 85, 225– 234. Lazaro, A., Lundren, R., & Totland, O. (2009). Co-flowering neighbors influence the diversity and identity of pollinator groups visiting plant species. Oikos, 118, 691–702. Mitchell, R. J., Irwin, R. J., Brown, B. J., Waser, N. M., & Karron, J. D. (2009). New frontiers in competition for pollination. Annals of Botany, 103, 1403–1413. Moeller, D. (2004). Faciliative interactions among plants via shared pollinators. Ecology, 85, 3289–3301. Moeller, D., & Geber, M. A. (2005). Ecological context of the evolution of self-pollination in Clarkia xantiana: Populations size, plant communities, and reproductive assurance. Evolution, 59, 786–799. Molano-Flores, B., Hendrix, S. D., & Heard, S. B. (1999). The effect of population size on stigma pollen load, fruit set, and seed set in Allium stellatum Ker. (Liliaceae). International Journal of Plant Sciences, 160, 753–757. Morales, C. L., & Traveset, A. (2008). Interspecific pollen transfer: Magnitude, prevalence, and consequences for plant fitness. Critical Reviews in Plant Science, 27, 221–238. Morales, C. L., & Traveset, A. (2009). A meta-analysis of impacts of alien vs. native plants on pollinator visitation and reproductive success of co-flowering native plants. Ecological Letters, 12, 716–728. Niesenbaum, R. A. (1999). The effects of pollen load size and donor diversity on pollen performance, selective abortion, and progeny vigor in Mirabilis jalapa (Nyctaginaceae). American Journal of Botany, 86, 261–268. Olesen, J. M., Bascompte, J., Elberling, H., & Jordano, P. (2008). Temporal dynamics of a pollination network. Ecology, 89, 1573– 1582. Rathcke, B. (1988). Interactions for pollination among co-flowering shrubs. Ecology, 69, 446–457. Sargent, R. D., & Ackerly, D. D. (2008). Plant-pollinator interactions and the assembly of plant communities. Trends in Ecology & Evolution, 23, 123–130. Sargent, R. D., & Otto, S. P. (2006). The role of local species abundance in the evolution of pollinator attraction in flowering plants. The American Naturalist, 167, 67–80. Strauss, S. Y., Webb, C. O., & Salamin, N. (2006). Exotic taxa less related to native species are more invasive. Proceedings of the National Academy of Sciences of the United States of America, 103, 5841–5845. Thompson, J. N. (2005). The geographic mosaic of coevolution. Chicago: Chicago University Press. Thomson, J. D. (1986). Pollen transport and deposition by bumble bees in Erythronium: Influences of floral nectar and bee grooming. Journal of Ecology, 74, 329–341. Traveset, A., & Richardson, D. M. (2006). Biological invasions as disruptors of plant reproductive mutualisms. Trends in Ecology & Evolution, 21, 208–216. Vamosi, J., Knight, T., Steets, J., Mazer, S., Burd, M., & Ashman, T.-L. (2006). Pollination decays in biodiversity hotspots. Proceedings of the National Academy of Sciences of the United States of America, 103, 956–961. Vellend, M., Harmon, L. J., Lockwood, J. L., Mayfield, M. M., Hughes, A. R., Wares, J. P., et al. (2007). Effects of exotic species on evolutionary diversification. Trends in Ecology & Evolution, 22, 481–488.
123
28 Waites, R., & Agren, J. (2004). Pollinator visitation, stigmatic pollen loads, and among-population variatin in seed set in Lythrum salicaria. Journal of Ecology, 92, 512–526. Waser, N. M. (1978). Interspecific pollen transfer and competition between co-occurring plant species. Oecologia, 36, 223–236. Webb, C. O. (2000). Exploring the phylogenetic structure of ecological communities: An example for rain forest trees. The American Naturalist, 156, 145–155. Webb, C. O., Ackerly, D., & Kembel, S. (2008) Software for the analysis of community phylogenetic structure and character evolution. Version 4.0.1. http://www.phylodiversity. net/phylocom. Webb, C. O., & Donoghue, M. J. (2005). Phylomatic: Tree assembly for applied phylogenetics. Molecular Ecology Notes, 5, 181 http://www.phylodiversity.net/phylomatic/.
123
Evol Biol (2010) 37:19–28 Wilcock, C. C., & Neiland, R. (2002). Pollination failure in plants: Why it happens and when it matters. Trends in Plant Science, 7, 1360–1385. Wilson, P., & Stine, M. (1996). Floral constancy in bumble bees: Handling efficiency or perceptual conditioning? Oecologia, 106, 493–499. Wright, S. J., & Calderon, O. (1995). Phylogenetic patterns among tropical flowering phenologies. Journal of Ecology, 83, 937–948. Zorn-Arnold, B., & Howe, H. F. (2007). Density and seed set in a self-compatible forb, Penstemon digitalis (Plantaginaceae), with multiple pollinators. American Journal of Botany, 94, 1594– 1602.
