Floral roles in hummingbirds-mediated indirect plant interactions in Tropical Andean communities.

Floral roles in hummingbirds-mediated indirect plant interactions in Tropical Andean communities. | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL Ecology and Evolution This is a preprint and has not been peer reviewed. Data may be preliminary. 3 September 2025 V1 Latest version Share on Floral roles in hummingbirds-mediated indirect plant interactions in Tropical Andean communities. Authors : Ann Frías-Romero 0009-0003-5456-408X [email protected] , Boris Tinoco 0000-0003-1054-8945 , Bryan Rojas , Ariana Vélez , Samara Zeas-Bermeo , and Catherine Graham 0000-0001-9267-7948 Authors Info & Affiliations https://doi.org/10.22541/au.175692134.44757608/v1 Published Ecology and Evolution Version of record Peer review timeline 230 views 201 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract In pollination networks, indirect plant-plant interactions mediated by their shared pollinators, can shape community dynamics and species fitness; yet, the influence of floral traits on species roles remains unclear, particularly in diverse ecosystems like the tropical Andes. We studied hummingbird-mediated interactions among 31 flowering plants in three high-elevation shrubby habitats located in southern Ecuador. During August and November 2022, and January 2023, we collected stigma samples and constructed weighted interaction networks linked by heterospecific pollen grains. Species roles were determined by defining if they were overall donors or receptors of pollen. We also explored the association between the abundance of flowers and different floral traits and species roles. Finally, we assessed the potential influence of floral roles on species fitness by calculating the ratio of conspecific-to-heterospecific pollen grains observed in each species. We found that the identity of donor and receptor species was highly dynamic across sites and time. Donor species were characterized by being highly abundant, and having long and wide corollas, while receptor species had low abundance, and short and narrow corollas. Donor species received less heterospecific pollen than conspecific pollen, indicating that the pollen sharing roles of species could have fitness consequences. Our findings highlight the importance of floral traits and abundance in shaping floral roles and their potential fitness consequences in hummingbird-mediated indirect interactions. Floral roles in hummingbirds-mediated indirect plant interactions in Tropical Andean communities. Authors: Ann Frías-Romero 1 , Boris A. Tinoco 1,2 , Bryan G. Rojas 1 , Ariana Vélez 2 , Samara Zeas-Bermeo 1 and Catherine H. Graham 3 . Corresponding author: Ann Frías Romero Contact information: [email protected] Author addresses: 1. Programa de Recursos Naturales Renovables, Universidad del Azuay, Ecuador. 2. Escuela de Biología, Universidad del Azuay, Ecuador. 3. Swiss Federal Institute for Forest, Snow and Landscape Research WSL, Zürcherstrasse 111, Birmensdorf, Switzerland Author contributions: Ann Frías-Romero and Boris A. Tinoco conceived the idea and designed the project. Ann Frías-Romero, Bryan G. Rojas, Ariana Vélez and Samara Zeas-Bermeo collected the data. Ann Frías-Romero conducted the analysis and generated figures with help from Boris A. Tinoco, Catherine H. Graham and Bryan G. Rojas. Ann Frías-Romero wrote the manuscript with help from Catherine H. Graham and Boris A. Tinoco. All authors read and commented on the manuscript. Acknowledgements: Swiss National fund project Experimental Network Ecology and Restoration (Exper-net), grant number IZSTZ0_199379 granted to Graham. To Pedro J. Bergamo, Angie Bermeo, Franchesco Bresciani, Diego Gómez, Daniel González, Ever Guamán for their invaluable contributions and support. Data statement: Once the article is approved, I will upload my databases and statistical programming code to make them available to everyone. For now, I uploaded my databases as supportive information, which will later be improved for the publication of the article. Abstract. In pollination networks, indirect plant-plant interactions mediated by their shared pollinators, can shape community dynamics and species fitness; yet the influence of floral traits on species roles remains unclear, particularly in diverse ecosystems like the tropical Andes. We studied hummingbird-mediated interactions among 31 flowering plants in three high-elevation shrubby habitats located in southern Ecuador. During August and November 2022, and January 2023, we collected pollen from stigma samples and constructed weighted interaction networks linked by heterospecific pollen grains. Species roles were determined by defining if they were donors or receivers of pollen, measured by node degree out and node degree in respectively. We also explored the association between the abundance of flowers and different floral traits and species roles. Finally, we assessed the potential influence of floral roles on species fitness by calculating the ratio of conspecific-to-heterospecific pollen grains observed in each species. We found that the identity of donor and receiver species was highly dynamic across habitats and time. Receiver species were characterized by being highly abundant, while donor species presented high levels of nectar production. Receiver species received more heterospecific pollen than conspecific pollen, indicating that the pollen sharing roles of species could have fitness consequences. Our findings highlight the importance of some floral traits and abundance in shaping floral roles and their potential fitness consequences in hummingbird-mediated indirect interactions. Key words: Pollination networks, plant-plant interactions, hummingbird-mediated networks, floral traits, stigmas samples, pollen grains, tropical Andes. math_shortcuts Introduction. Indirect interactions among species are known to have important fitness consequences (Strauss, 1991; Werner & Peacor, 2003), yet they remain less explored in comparison to direct interactions. In pollination, indirect interactions occur between co-occurring plants that share the same pollinators (Carvalheiro et al., 2014). These interactions are mediated by plants competing to attract pollinators and the consequences of heterospecific pollen deposited by shared pollinators (Arceo-Gómez, 2021) . While recently there have been major advances for understanding the structure of indirect interactions using network analysis tools (Bergamo et al., 2021; Carvalheiro et al., 2014; Fang & Huang, 2013), the influence of flower traits on plant species indirect interactions is still unclear, especially in species-rich systems such as the tropical Andes. Flowering plants can have different floral roles in indirect interaction networks and function as overall receivers or donors of pollen within communities. Receiver species obtain high amounts of heterospecific pollen from other species in the community, while donor species deliver their pollen to other species of the community (Lanuza et al., 2021). Indirect interaction networks often exhibit highly asymmetric patterns (Vázquez et al., 2007), where only a few species act as an overall receiver or donor of pollen (Fang & Huang, 2013; Peuker et al., 2020). Such asymmetry fosters cohesion and resilience within the network, promoting the persistence of rare species, and minimizing competition among species (Bascompte et al., 2003; Bastolla et al., 2009) . However, these asymmetric patterns also suggest that pollination interactions may be structured by a select group of plants that can have an important role for plant community dynamics and stability (Suárez-Mariño et al., 2019). The pollen sharing roles of species can be influenced by the abundance of flowers and the functional traits of the interacting species. For instance, the abundance of flowers can determine the amount of pollen potentially available in the community (Wei et al., 2021) and thus, may influence the role of the species in the community. While highly abundant flowers have high frequency of interactions (Vázquez et al., 2007), and are thus likely to donate pollen to the community (Arceo-Gómez, 2021; Fang & Huang, 2013), rare species benefit from being in dense conspecific patches and receiving pollen through interspecific facilitation with abundant species (Bergamo et al., 2020). Moreover, floral traits can filter the pollinator visiting a plant species and determine the interactions among co-flowering plants (Bergamo et al., 2019; Lunau et al., 2011). For instance, floral tube length and corolla width can influence the type of pollinators visiting a flower (E‐Vojtkó et al., 2020) and the amount of heterospecific pollen shared among species (Muchhala & Thomson, 2012). Flowers with short floral tubes and wide corolla flowers often have a receiver role because they are visited by multiple pollinators (Muchhala, 2007; Stang et al., 2006). Conversely, long-tubed flowers and narrow corollas attract mostly specialized pollinators (Aigner, 2004; Dalsgaard et al., 2021; Sonne et al., 2020), reducing the diversity of pollinator visits (Benadi & Pauw, 2018); as a result, they may receive less pollen and are likely to donate more pollen to the community. Another potentially important floral trait is the exsertion of stamens and stigmas. Stamens contain pollen grains to be shared with pollinators, while stigmas represent the receptive tip of flowers where pollen arrives. Increased exposure of stamens may promote pollen donation (Li et al., 2022), whereas a exposed stigma can lead to high amounts of pollen reception. Finally, the sugar concentration and the volume of nectar in a flower can determine pollinator visitation rates and influence pollen reception or donation (Heiling et al., 2021; Kim et al., 2011). Species with highly rewarding nectar for pollinators may have high visitation rates and have donor roles in the community, while those with less attractive nectar may act as receivers due to low visitation rates. The roles of species within networks can have fitness consequences, particularly in plant–plant interactions, where the ultimate outcome depends on the number of conspecific pollen deposited on the stigmas that are able to fertilize ovules (Arceo-Gómez, 2021; Lopes et al., 2022). As a result of obtaining pollen grains, receiver species may accumulate heterospecific pollen, which can have negative fitness costs (Streher et al., 2020). Heterospecific pollen can prevent the full development of conspecific pollen and reduce reproductive success (Morales & Traveset, 2008; Moreira-Hernández & Muchhala, 2019). For donor species, widely distributing pollen increases the chances of their pollen reaching another individual of the same species (Tur et al., 2016). Nonetheless, being a donor species can also entail fitness costs related to pollen waste (Muchhala et al., 2010; Muchhala & Thomson, 2012). In the tropical Andes, hummingbirds are highly diverse and specialized nectar-feeding birds, playing a crucial role in pollinating a significant number of plant species (Barreto et al., 2024; Stiles, 1981). Thus, the plant-hummingbird system has been used as a model to investigate the evolutionary, historical and ecological factors that shape the assembly of species within networks on interacting species (Sonne et al., 2020; Tinoco et al., 2017; Vizentin-Bugoni et al., 2014). We studied the indirect interactions among flowering plants that share hummingbird pollinators in high elevation shrubby habitats located in the southern Ecuadorian Andes. We built indirect interaction networks in order to 1) identify the floral roles (receivers or donors) of the species, 2) evaluate the importance of floral abundance and floral traits in these roles, and 3) explore the potential fitness consequences for plant species with different roles. We collected 1454 samples of gynoecium from among 31 flower species, from which we expected to find (1) some species playing a defined floral role, but not the majority, (2) low-abundance flowers functioning as a receivers, and high-abundance flowers as a donors, (3) donor species with floral traits related specifically to hummingbird pollination (i.e long corollas and narrow corollas), and receiver species with floral traits that allow pollen deposition from a variety of groups of pollinators, (4) negative fitness consequences for receiver species due to the accumulation of heterospecific pollen grains. Methodology. Study area. Fieldwork was conducted in three sites located between 2900 and 3100 m.a.s.l. in the Southern Andes of Ecuador: El Gullán (3000 m asl), Aguarongo (3160 m asl) and La Tranca (3060 m asl) (Fig. 1), located at least 10 km apart. Previous studies in the tropical Andes detect significant floristic turnover over similar spatial scales, driven by elevation, edaphic variation and landscape heterogeneity (Cuesta et al., 2017), supporting the expectation that distinct floral communities occur at each site, justifying their treatment as ecologically independent units. The average temperature in the region ranges from 14° to 20° C (IGM, 2013), while average rainfall varies across sites: El Gullán from 400 to 600 mm per year, Aguarongo from 700 to 800 mm per year, and La Tranca 1200 to 1500 mm per year (IGM, 2012). The region experiences a bimodal rainfall pattern, characterized by two periods of substantial precipitation, from March to April and October to December, and one season with comparatively lower rainfall from June to August (Celleri et al., 2007). Field work was carried out in three sampling periods in each site: August, November 2022 and January 2023. All sites are dominated by native shrubs, including species that are known to be important nectar resources for hummingbirds such as Oreocallis grandiflora, Barnadesia arborea, Fuchsia sp., Salvia sp., Macleania rupestris, Brachyotum confertum, Tillandsia sp. or Viola arguta (Crespo et al., 2022). The sites chosen were mostly undisturbed by human activities; however, in Aguarongo and La Tranca there are a few cattle trails that crossed our sampling transects. Figure 1. Location of the study area and sampling sites in southern Ecuador. Panel A shows the location of Ecuador within South America, panel B highlights the Azuay province within Ecuador, panel C shows a digital elevation model of Azuay province showing the altitudinal gradient and the location of the three study sites: El Gullán, Aguarongo and La Tranca. Elevation ranges are represented in different shades of green, from lowlands (light green) to highlands (dark green). A scale bar and north arrow are included for spatial reference. Study design and sampling. This study was based on interactions among flowering plants measured by the presence and abundance of heterospecific pollen found on flower stigmas of co-flowering plant species visited by hummingbirds. We focused on species known to interact with hummingbirds in the montane forests of Azuay, drawing on local studies in pollination ecology (Crespo et al., 2022; Knowlton et al., 2022; Tinoco et al., 2017; Vélez et al., 2025). Although these species may also be visited by other pollinators, most exhibit traits associated with the ornithophilous pollination syndrome, suggesting that hummingbirds are their primary pollinators. Consequently, we assume that indirect interactions among these plants are mainly mediated by hummingbirds. This assumption is supported by previous findings indicating that species with ornithophilous traits predominantly interact with hummingbirds, and indirect interactions among them are likely mediated by these interactions (Bergamo et al., 2017). The identification of species was helped by the use of a local field guide for plants (Arias et al., 2022). At each study site we placed three 100 m x 10 m transects, located at least 100 m apart, where we counted the flowers of each individual of all flowering plant species found in the transect. This information was used later to determine the number of stigmas to be sampled per flowering species by assigning each species to an abundance quartile based on its overall abundance at each sampling period and site. The least abundant species occupied quartile one and the most abundant occupied quartile three. Our samples were the stigma of the flowers, and we took three samples per individual according to abundance quartiles. For quartile one, species with low abundance, we took samples from 2 individuals (6 samples); for quartile two, species with medium abundance, we took samples from 3 individuals (9 samples); and for quartile three, we took samples from 5 individuals (15 samples). Stigmas were collected directly in the field and placed in slides coated with fuchsin dye, which were later examined under a microscope for pollen grain identification. Each slide was observed under 40 x magnification, with the cover slip (measuring 25 x 25 mm) serving as the area to be examined for pollen. The field of view of the microscope was swept across the entire cover slip from left to right, covering each millimeter of the slip. To facilitate pollen grain identification, we compiled a library consisting of plant species present at our study sites. This library incorporated information from previous work in the study area (Jaramillo Espinosa, 2019; Nieto Orellana & Silva Alemán, 2012), complemented by our work. Network construction and species’ roles. We constructed nine indirect interaction networks, three networks per site and per each sampling period. The network nodes represent species, genus or families, depending on the level of identification we were able to reach, and the arrows connecting nodes (plant taxa) demonstrate the direction and prevalence of interspecific pollen transfer as determined from heterospecific pollen grains encountered on the stigmas of each plant species (Function graph_from_adjacency_matrix in R (Csárdi et al., 2024)). The networks included up to 38 nodes: 27 identified at the species level, 8 at the genus level and 3 at the family level. To weight networks, we used the median value for each distinct heterospecific pollen type encountered in the stigmas of each plant species. The median value was used because it allowed us to capture a central tendency in the asymmetric distribution of pollen grain counts. Using the constructed networks, we created chord diagrams (Function Chord diagram in R (Gu et al., 2014)) to enhance the visualization of indirect interactions. To determine the pollen sharing roles of the different species we calculated the weighted node degree in and weighted node degree out, considering the number of links entering and leaving a node, respectively (Foster et al., 2010; Leicht & Newman, 2008). These metrics were based exclusively on heterospecific pollen grains found in flower’s stigma. Species with higher values of weighted node degree-in than others in the community can be identified as receiver species, while species with higher weighted node degree-out compared to others can be considered donor species (Fang & Huang, 2013). Floral traits. For the sampled species ( Appendix table 2 ) we gathered information of floral traits that are known to influence patterns of pollen sharing. This data was taken from a local database with at least 10 individual measurements for each species (Fernández & Frías Romero, 2021). Corolla opening can function as a morphological barrier to pollinators (Muchhala, 2007), and was measured at the widest part of the corollas, where the flower restricts the access to pollinators. Floral tube length is a morphological barrier to pollinators (Sonne et al., 2020), and was measured as the distance from the base of the flower to the tip of the corolla. Exsertion of stamen is the difference between stamen length and floral tube length. Exsertion of stigma is the difference between stigma length and floral tube length. In these last traits, positive values represent stamen or stigmas extending beyond the floral tube, whereas negative values indicate that the stamen or stigma is situated inside the floral tube. Finally, Nectar production and nectar concentration represent the reward offered to pollinators (Hainsworth & Wolf, 1972; Kingsolver & Daniel, 1983). Nectar production was measured using microcapillary tubes by recording the length of nectar drawn into the tube. Nectar concentration was determined with a handheld refractometer (%w/w). To obtain standardized measurements, flowers were first depleted from nectar and then enclosed with fine mesh (tulle) to exclude pollinators. After 24 hours, nectar volume and concentration were measured. Another important floral trait that could influence polling sharing is the amount of pollen produced by each species and how the pollen is placed on the pollinator’s body. However, we were not able to obtain data on these traits due to the amount of work required to count pollen grains and assess placement patterns. As indicated in the section Network construction and species’ roles, some species were not identified at the species level using pollen samples. For those species we used mean trait data of the genus. That was the case for eight species ( Fuchsia spp., Rubus spp., Tillandsia spp., Centropogon spp., Castilleja spp., Brugmansia spp., Berberis spp., Nasa spp.). We are aware that there could be large variation in floral traits within species of the same genus, but we minimized that variation by using information of species within the genus of interest that are present in each study area, which was never more than three species. For another five species, including Brugmansia spp. , Castilleja spp. , Chuquiraga jussieui , Vaccinium floribundum and Berberis spp. , nectar data were not available, although the rest of the floral traits were included in the analysis. Moreover, we lacked floral trait data for Stenomesson auriantiacum . Data analysis. We first performed a community ordination analysis to evaluate differences in community composition across sites and sampling periods. This analysis was performed using a non-metric multidimensional scaling based on Bray-Curtis dissimilarities. We used the number of flowers of each species found in each site and each sampling period. In addition, we conducted a PERMANOVA to statistically test for differences in community composition among sites and sampling periods. We assessed the associations between node degree - in and out - of species and their abundances and floral traits using linear mixed models. The node degree of the different species was used as the response variable, while the floral traits (stigma and stamen exsertion, corolla opening, floral tube length, nectar concentration and nectar production) and floral abundance were used as the fixed factors; we used network identity as a random factor to consider the potential correlation of species that are part of the same network. We conducted univariate variables models using the lmer function from lme4 package in R (Bates et al., 2015). In these models, we excluded Passiflora cumbalensis; its flower up to 20 cm in length, which is 2 times larger than the second largest species; therefore, including this species would have influenced the normality of the distribution of the fixed factors in the models. We also examined pairwise correlations among floral traits. These correlations revealed functional trade-offs and associations, such as a positive relationship between tube length and nectar production, and a decoupling between stigma and stamen exertion ( Appendix figure 1 ). Finally, in order to assess the potential implications of floral roles on fitness of the different species, we calculated conspecific-to-heterospecific pollen grains ratio (CP/HP ratio). This was done by computing the mean values of conspecific and heterospecific pollen grains for each sample of each species. An increase in the CP/HP ratio suggests a higher proportion of conspecific versus heterospecific pollen grains, potentially indicating positive fitness outcomes. Conversely, a decrease in the ratio implies a higher proportion of heterospecific versus conspecific pollen grains, indicating negative consequences for fitness of the pollen sharing role of the species. We did linear mixed models using node degree in as the response variable and CP/HP ratio as a fixed factor; locality was a random factor grouping species that are part of the same network. The use of CP/HP ratio should only be considered a broad indicator of the fitness consequences of sharing pollen, because there are other measures that are direct indicators of fitness effects (e.g. pollen tube growth, seed set). However, this measure can be a useful indicator across species with the premise that it will be always beneficial for a species to receive more conspecific pollen than heterospecific pollen (Lanuza et al., 2021), considering that even small amounts of heterospecific pollen can reduce reproductive success (Thomson et al., 1982). We did not model node degree out with CP/HP ratio because the fitness consequences of donor roles should be mostly related to male fitness (pollen waste), a fitness measure out of the scope of our analysis. Results Overall, we collected 1454 stigmas of 31 flower species. On these stigmas we identified 5341 hetero-specific pollen grains from 14 families and 38 species; 75.98% of the hetero-specific pollen grains were identified to the species level, 15% to the genus level and 7.62% to family level; 1.6% of the heterospecific pollen were not identified, and thus not used in later analysis. Of all the pollen grains observed, 97.2% were con-specific (CP) and 2.8% were hetero-specific (HP). Additionally, on average each species received 191 conspecific pollen grains and only 3.02 heterospecific pollen grains (Appendix table 1). La Tranca network was built with 25 nodes connected by 116 arrows, with 24.14% of the arrows being reciprocal. The Aguarongo network had 25 nodes connected to 114 arrows, showing 28% of reciprocal arrows. The El Gullán network had 20 nodes connected to 90 arrows (Fig. 2), of which 20% were reciprocal. To further explore differences in community structure and their potential effects on floral roles, we performed a NMDS analysis which revealed clear separation among communities across sites and sampling periods. This visual pattern was confirmed by PERMANOVA results, which indicated significant variation in species composition (R2 = 0.432, F= 2.29, p=0.003). These results suggest that variation in species presence and abundance across space and time likely influenced the floral roles in different networks (Appendix table 3, Appendix figure 2). Figure 2. Chord diagrams of indirect interaction networks mediated by hummingbirds in three localities of the southern Andes of Ecuador: La Tranca (A, B & C), Aguarongo (D, E & F), and El Gullán (G, H & I). Each locality was sampled during three periods: August 2022 (A, D & G), November 2022, (B, E & H), and January 2023 (C, F & I). The size of the nodes and arrows denote the prevalence of interspecific pollen transfer as determined from the median number of heterospecific pollen grains encountered on the stigmas of each plant species. The direction of the tips of the arrows indicate the direction of the interaction. Species names and codes are also presented in Appendix table 1. Figure 3. Frequency of species of flowering plants that share pollen grouped by node degree In (A, C, E) and node degree Out (B, D, F) in localities of La Tranca (A, B), El Gullán (C, D) and Aguarongo (E, F). The data comes from indirect interaction networks mediated by hummingbirds in shrubby habitats of southern Ecuador: Figure 4. Values of node degree in and node degree out of plant species in three localities La Tranca (A, B), El Gullán (C, D) and Aguarongo (E, F) studied in three different sampling periods (different colors of the bars). The data comes from indirect interaction networks mediated by hummingbirds in three shrubby habitats of southern Ecuador. For each plot, the six species with the highest node degree in and out values were selected and displayed. Floral roles. We found that only a few species exhibited multiple connections to many other species, functioning either as receivers or donors. Meanwhile, the majority of species displayed low node degree in and out values, indicating limited connections to few other species (Fig. 3). There was large variation in the roles of species among sites, with most species being either an important donor or receiver only in one site, while other species functioned exclusively as either donors and receivers, depending on the location and sampling period (Fig 4). For example, Macleania rupestris and Gaultheria reticulata acted exclusively as receivers in both La Tranca and El Gullan (Fig. 4A and 4C) and Bomarea uncifolia acted as a donor in La Tranca and Aguarongo (4B and 4F). In contrast, a few species acted simultaneously as both donors and receivers within the same locality, such as Viola arguta in La Tranca, Oreocallis grandiflora in El Gullan, Brachyotum confertum and Stenomesson aurantiacum in Aguarongo (Fig. 4). The dynamic of species roles was supported by the weak correlation found between node degree in and out ( S = 103, p-value = 0.021, rho = - 0.173 ) suggesting that acting as a donor did not necessarily imply a reciprocal role as a receiver (Appendix table 5). Floral traits and floral abundance. Despite including all floral traits, only abundance and nectar production showed significant associations with floral roles. Flower abundance was an important predictor of node degree in, indicating that the most abundant species were receptors (Fig 5). Conversely, species producing high amounts of nectar tended to have higher node degree out, reflecting a stronger role as donor species (Fig 6). Figure 5. Associations between different floral traits and node degree in of flower species that are part of indirect interactions in shrubby habitats of the southern Andes of Ecuador. (A) for floral abundance, (B) for floral tube length, (C) for corolla opening, (D) nectar concentration, (E) for stigma exsertion, and (F) for stamen exsertion. All predictor variables are in log scale, except for stigma and stamen exsertion. Fitted lines and confidence intervals (shaded area) represent fitted values from linear mixed-effects regression model. Figure 6. Associations between different floral traits and node degree out values of flower species that are part of indirect interactions in shrubby habitats of the southern Andes of Ecuador. (A) for floral abundance, (B) for floral tube length, (C) for corolla opening, (D) nectar concentration, (E) for stigma exsertion, and (F) for stamen exsertion. All predictor variables are in log scale, except for stigma and stamen exsertion. Fitted lines and confidence intervals (shaded area) represent fitted values from linear mixed-effects regression model. math_shortcuts Potential fitness consequences. We found a significant association between the node degree in of the species and CP/HP ratio (df = 114, t-value =- 3.31, P = 0.001) (Fig. 7). This indicates that being a receiver species could lead to negative fitness consequences. Figure 7. Relationships between amount of con-specific (CP) per heterospecific (HP) pollen grains (CP/HP ratio) found in stigmas and node degree in of plant species that are part of indirect interaction networks in the southern Andes of Ecuador. Fitted line and confidence intervals (shaded area) represent fitted values from linear mixed-effects regression model. Discussions Identifying the roles of plant species that share pollen, and how these roles are influenced by floral traits in indirect plant–plant networks can provide insight into interaction dynamics that potentially impact plant reproductive success and community structure. We found that the role of species in indirect plant-plant networks is highly dynamic in space and time in the tropical Andes. Moreover, the pollen sharing roles were explained by the abundance of flowers and nectar production. Additionally, we found that species that act as pollen receivers could potentially have negative fitness consequences. In our study, we found an average of 3.13 heterospecific pollen grains per flower, a relative low number compared to the global patterns of heterospecific pollen on stigmas reported by Arceo-Gómez et al. (2019). However, the spatial coverage of that global study was low, particularly in tropical regions, where more data are needed for a better understanding of patterns of heterospecific pollen loads. Nonetheless, the relatively low number of heterospecific pollen arriving at the stigmas in our study site suggests high levels of pollinator niche partitioning among plant species. This pattern could be related to high specialization of tropical Andean hummingbirds in their floral visits (Dalsgaard et al., 2021; Sonne et al., 2016; Weinstein & Graham, 2017), which likely reduces pollen mixing (Lopes et al., 2022). Additionally, floral morphological traits, such as the position of anthers, may facilitate pollen placement on distinct body parts of pollinators, further limiting heterospecific pollen transfer (Arteaga-Chávez et al., 2025; Stewart & Dudash, 2016). More studies are needed to identify the mechanisms that influence heterospecific pollen transfer (Arceo-Gómez et al., 2019; Lopes et al., 2022), but our findings suggest reduced pollen mixing among co-flowering plants, which may minimize competition for pollinators and facilitate the coexistence in the diverse plant communities of the tropical Andes. Few species acted as receivers or donors of pollen, and these species were connected to multiple species that had weak roles. This pattern fits our predictions and has been observed in other plant-plant interaction networks (Fang & Huang, 2013; Tur et al., 2016). An asymmetric assembly of species in networks, where a few species are connected to many species and most species are connected to a few species (Bascompte, 2007), is a common pattern in plant-animal mutualistic networks (Jordano et al., 2009), and can be scaled to indirect interaction networks. This asymmetric assembly increases redundancy in ecological networks (Bascompte & Scheffer, 2023). If one donor or receiver were to disappear, there are other species to fulfill similar functions, enhancing the stability of the network to perturbations (Thébault & Fontaine, 2010). The asymmetric structure in species interactions indicates a hierarchical structure and cohesive organization of networks around a central role of a few species, a pattern that promotes network stability (Bascompte & Jordano, 2006). Flowers exhibited dynamic roles as donors or receivers across sampling locations and sampling period, indicating that floral roles are context dependent, influenced by variation in the presence and abundance of other species in space and time. This result was also supported by our finding of a weak correlation between the node degree in and out. Spatial variation in floral roles, observed even among proximate areas, can be attributed to a high beta diversity of plants in the tropical Andes (Egawa et al., 2020; Ogishima et al., 2022; Van Der Niet et al., 2014), as it was also the case in our study sites, where the composition of flowers varied among sites (Appendix figure 2, Appendix table 3). Moreover, floral phenology, predominantly dictated by rainfall-related events (Günter et al., 2008), shapes the temporal abundance of flowers available for pollinators, consequently affecting the patterns of pollen sharing among plants (Gallagher & Campbell, 2020). It is important to mention that this study focuses on plant species known to interact mainly with hummingbirds, and the local roles found may change if a broader range of flowering plant species were included. Nonetheless, despite this limitation, the observed fluctuations in floral roles highlight the dynamic nature of ecological interactions in the tropical Andes. Different floral traits influenced the donor and receiver roles of species in these high-Andean plant communities. In the case of receiver species, this role was associated with floral abundance. This result was contrary to our expectation that abundant species would primarily act as donors due to their numerical dominance. Nonetheless, abundant flowers can attract a wide diversity of pollinators (Crespo et al., 2022; Fründ et al., 2010), which may deposit a high diversity of pollen in the stigmas of these species (Peuker et al., 2020). Moreover, donor species were associated with high nectar production. Hummingbirds have high energetic demands compared to other pollinators; thus, they prefer to visit flowers offering high nectar volumes (Stiles & Freeman, 1993). Accordingly, these plant species with high nectar production, particularly attractive to hummingbirds, are exporting their pollen to other hummingbird visited plants in the community. This finding aligns with our expectation that donor species would possess traits specialized for hummingbird pollination. In general, our results indicate that while floral abundance may promote generalized pollinator visitation and increase pollen reception, nectar production appears to enhance targeted pollen export through hummingbird visitation. These trait-based differences help explain asymmetries in pollen flow and highlight the importance of considering both the quantity and quality of floral rewards in determining floral roles. There are other floral traits that have been found to be important predictors of floral roles in other studies. For instance, flowers with exserted stigmas tend to accumulate a high richness of heterospecific pollen (Ashman & Wei, 2024; Fang & Huang, 2013; Lanuza et al., 2021). Moreover, floral traits such as corolla opening and floral tube length can filter the species of pollinators that visit a flower, and influence the pollen sharing role of species (Fang & Huang, 2013; Zhao et al., 2025). However, our study focused on a specific subset of flowers mainly visited by hummingbirds, which may explain the lack of a clear relationship between some floral traits with floral roles. Nonetheless, the traits explaining reception and donor roles can be highly variable within communities (Arceo-Gómez et al., 2019), with studies even finding weak predictive power of floral traits determining floral roles (Peuker et al., 2020). We found a significant negative relationship between node degree in and the CP/HP ratio, indicating that receiver species tend to accumulate heterospecific pollen loads. Heterospecific pollen can interfere with pollen tube growth, clog stigmatic surfaces, or outcompete conspecific grains, ultimately reducing fertilization success and fitness (Morales & Traveset, 2008; Streher et al., 2020). Receiving heterospecific pollen represents a “service fee” for sharing pollinators with other species (Tur et al., 2016). However, this cost may be offset if pollinator sharing increases the likelihood of receiving enough conspecific pollen to achieve ovule fertilization (Lopes et al., 2022; Tur et al., 2016). Thus, the balance between heterospecific interference and conspecific pollen delivery should determine the fitness consequences of the reception role. Moreover, our measure of fitness costs does not capture all aspects of reproductive success-such as pollen germination, pollen tube growth, or seed set- which are essential for evaluating fitness outcomes of pollen sharing (Németh & Smith‐Huerta, 2003). Future studies incorporating these direct fitness measures could help to better understand the implications of floral roles in pollen-sharing networks. Conclusions. Understanding the role of pollinators mediating indirect interactions among co-flowering plants is crucial for gaining insights into community structure and the mechanisms influencing coexistence. Our research from the tropical Andes uncovered an asymmetric assembly pattern within hummingbird-mediated indirect interactions, revealing a cohesive organization around a small set of interactions. Moreover, floral roles in plant-plant interactions seem to be highly dynamic in space and time, where the spatial and temporal abundance of the flowers can play an important role in the patterns of pollen sharing among flowering plants. Importantly, floral roles could impact plant reproductive success, particularly for receiver species, which can likely influence the reproductive strategy of plants. Identifying floral roles enables the identification of species and structures of networks that can contribute to the maintenance and diversity of flowering plants in species rich systems. Bibliography Aigner, P. A. (2004). Floral specialization without trade-offs: Optimal corolla flare in contrasting pollination environments. Ecology , 85 (9), 2560-2569. https://doi.org/10.1890/03-0815Arceo-Gómez, G. (2021). Spatial variation in the intensity of interactions via heterospecific pollen transfer may contribute to local and global patterns of plant diversity. Annals of Botany , 128 (4), 383-394. https://doi.org/10.1093/aob/mcab082Arceo-Gómez, G., Schroeder, A., Albor, C., Ashman, T.-L., Knight, T. M., Bennett, J. M., Suarez, B., & Parra-Tabla, V. (2019). Global geographic patterns of heterospecific pollen receipt help uncover potential ecological and evolutionary impacts across plant communities worldwide. Scientific Reports , 9 (1), 8086. https://doi.org/10.1038/s41598-019-44626-0Arias, R., Alarcon, I., Graham, C., Espinosa, C., & Tinoco, B. A. (2022). Plantas visitadas por colibríes y pinchaflores en Bosques Montanos de las microcuencas: Tomebamba, Tarqui y Machangara. Azuay-Ecuador . Escuela de Biología, Univerisdad del Azuay.Arteaga-Chávez, W. A., Graham, C. H., Guevara, E. A., & Tinoco, B. A. (2025). Plants partition the pollinator niche by depositing pollen on different parts of the pollinator body. PLOS One , 20 (5), e0323577. https://doi.org/10.1371/journal.pone.0323577Ashman, T., & Wei, N. (2024). Evaluating the influences of floral traits and pollinator generalism on α and β diversity of heterospecific pollen on stigmas. Functional Ecology , 38 (2), 465-476. https://doi.org/10.1111/1365-2435.14465Barreto, E., Boehm, M. M. A., Ogutcen, E., Abrahamczyk, S., Kessler, M., Bascompte, J., Dellinger, A. S., Bello, C., Dehling, D. M., Duchenne, F., Kaehler, M., Lagomarsino, L. P., Lohmann, L. G., Maglianesi, M. A., Morlon, H., Muchhala, N., Ornelas, J. F., Perret, M., Salinas, N. R., … Graham, C. H. (2024). Macroevolution of the plant–hummingbird pollination system. Biological Reviews , brv.13094. https://doi.org/10.1111/brv.13094Bascompte, J. (2007). Networks in ecology. Basic and Applied Ecology , 8 (6), 485-490. https://doi.org/10.1016/j.baae.2007.06.003Bascompte, J., & Jordano, P. (2006). The structure of plant-animal mutualistic networks. Ecological networks: linking structure to dynamics in food webs , 143-159.Bascompte, J., Jordano, P., Melián, C. J., & Olesen, J. M. (2003). The nested assembly of plant–animal mutualistic networks. Proceedings of the National Academy of Sciences , 100 (16), 9383-9387. https://doi.org/10.1073/pnas.1633576100Bascompte, J., & Scheffer, M. (2023). The resilience of plant–pollinator networks. Annual Review of Entomology , 68 (1), 363-380. https://doi.org/10.1146/annurev-ento-120120-102424Bastolla, U., Fortuna, M. A., Pascual-García, A., Ferrera, A., Luque, B., & Bascompte, J. (2009). The architecture of mutualistic networks minimizes competition and increases biodiversity. Nature , 458 (7241), 1018-1020. https://doi.org/10.1038/nature07950Benadi, G., & Pauw, A. (2018). Frequency dependence of pollinator visitation rates suggests that pollination niches can allow plant species coexistence. Journal of Ecology , 106 (5), 1892-1901. https://doi.org/10.1111/1365-2745.13025Bergamo, P. J., Susin Streher, N., Traveset, A., Wolowski, M., & Sazima, M. (2020). Pollination outcomes reveal negative density‐dependence coupled with interspecific facilitation among plants. Ecology Letters , 23 (1), 129-139. https://doi.org/10.1111/ele.13415Bergamo, P. J., Traveset, A., & Lázaro, A. (2021). Pollinator-mediated indirect effects on plant fecundity revealed by network indices. The American Naturalist , 198 (6), 734-749. https://doi.org/10.1086/716896Bergamo, P. J., Wolowski, M., Maruyama, P. K., Vizentin‐Bugoni, J., Carvalheiro, L. G., & Sazima, M. (2017). The potential indirect effects among plants via shared hummingbird pollinators are structured by phenotypic similarity. Ecology , 98 (7), 1849-1858. https://doi.org/10.1002/ecy.1859Bergamo, P. J., Wolowski, M., Telles, F. J., De Brito, V. L. G., Varassin, I. G., & Sazima, M. (2019). Bracts and long-tube flowers of hummingbird-pollinated plants are conspicuous to hummingbirds but not to bees. Biological Journal of the Linnean Society , 126 (3), 533-544. https://doi.org/10.1093/biolinnean/bly217Carvalheiro, L. G., Biesmeijer, J. C., Benadi, G., Fründ, J., Stang, M., Bartomeus, I., Kaiser-Bunbury, C. N., Baude, M., Gomes, S. I. F., Merckx, V., Baldock, K. C. R., Bennett, A. T. D., Boada, R., Bommarco, R., Cartar, R., Chacoff, N., Dänhardt, J., Dicks, L. V., Dormann, C. F., … Kunin, W. E. (2014). The potential for indirect effects between co-flowering plants via shared pollinators depends on resource abundance, accessibility and relatedness. Ecology Letters , 17 (11), 1389-1399. https://doi.org/10.1111/ele.12342Celleri, R., Willems, P., Buytaert, W., & Feyen, J. (2007). Space–time rainfall variability in the Paute basin, Ecuadorian Andes. Hydrological Processes , 21 (24), 3316-3327. https://doi.org/10.1002/hyp.6575Crespo, A., Aguilar, J. M., Pintado, K., & Tinoco, B. A. (2022). Key plant species to restore plant–hummingbird pollinator communities in the southern Andes of Ecuador. Restoration Ecology , 30 (4). https://doi.org/10.1111/rec.13557Csárdi, G., Nepusz, T., Traag, V., Horvát, S., Zanini, F., Noom, D., & Müller, K. (2024). igraph: Network analysis and visualization in R. Zenodo , 2.0.2 . https://doi.org/doi:10.5281/zenodo.7682609Cuesta, F., Muriel, P., Llambí, L. D., Halloy, S., Aguirre, N., Beck, S., Carilla, J., Meneses, R. I., Cuello, S., Grau, A., Gámez, L. E., Irazábal, J., Jácome, J., Jaramillo, R., Ramírez, L., Samaniego, N., Suárez‐Duque, D., Thompson, N., Tupayachi, A., … Gosling, W. D. (2017). Latitudinal and altitudinal patterns of plant community diversity on mountain summits across the tropical Andes. Ecography , 40 (12), 1381-1394. https://doi.org/10.1111/ecog.02567Dalsgaard, B., Maruyama, P. K., Sonne, J., Hansen, K., Zanata, T. B., Abrahamczyk, S., Alarcón, R., Araujo, A. C., Araújo, F. P., Buzato, S., Chávez‐González, E., Coelho, A. G., Cotton, P. A., Díaz‐Valenzuela, R., Dufke, M. F., Enríquez, P. L., Martins Dias Filho, M., Fischer, E., Kohler, G., … Martín González, A. M. (2021). The influence of biogeographical and evolutionary histories on morphological trait‐matching and resource specialization in mutualistic hummingbird–plant networks. Functional Ecology , 35 (5), 1120-1133. https://doi.org/10.1111/1365-2435.13784Egawa, S., Hirose, K., & Itino, T. (2020). Geographic changes in pollinator species composition affect the corolla tube length of self‐heal ( Prunela vulgaris L.): Evidence from three elevational gradients. Ecological Research , 35 (5), 819-825. https://doi.org/10.1111/1440-1703.12146E‐Vojtkó, A., De Bello, F., Durka, W., Kühn, I., & Götzenberger, L. (2020). The neglected importance of floral traits in trait‐based plant community assembly. Journal of Vegetation Science , 31 (4), 529-539. https://doi.org/10.1111/jvs.12877Fang, Q., & Huang, S.-Q. (2013). A directed network analysis of heterospecific pollen transfer in a biodiverse community. Ecology , 94 (5), 1176-1185. https://doi.org/10.1890/12-1634.1Fernández, N., & Frías Romero, A. (2021). Caracterización de recursos florales aprovechados por colibríes en una gradiente de disturbio en bosques montanos del cantón de Cuenca [Universidad del Azuay]. https://dspace.uazuay.edu.ec/handle/datos/11226Foster, J. G., Foster, D. V., Grassberger, P., & Paczuski, M. (2010). Edge direction and the structure of networks. Proceedings of the National Academy of Sciences , 107 (24), 10815-10820. https://doi.org/10.1073/pnas.0912671107Fründ, J., Linsenmair, K. E., & Blüthgen, N. (2010). Pollinator diversity and specialization in relation to flower diversity. Oikos , 119 (10), 1581-1590. https://doi.org/10.1111/j.1600-0706.2010.18450.xGallagher, M. K., & Campbell, D. R. (2020). Pollinator visitation rate and effectiveness vary with flowering phenology. American Journal of Botany , 107 (3), 445-455. https://doi.org/10.1002/ajb2.1439Gu, Z., Gu, L., Eils, R., Schlesner, M., & Brors, B. (2014). Circlize implements and enhances circular visualization in R. Bioinformatics , 30 , 2811-2812.Günter, S., Stimm, B., Cabrera, M., Diaz, M. L., Lojan, M., Ordoñez, E., Richter, M., & Weber, M. (2008). Tree phenology in montane forests of southern Ecuador can be explained by precipitation, radiation and photoperiodic control. Journal of Tropical Ecology , 24 (3), 247-258. https://doi.org/10.1017/S0266467408005063Hainsworth, F. R., & Wolf, L. L. (1972). Crop volume, nectar concentration and hummingbird energetics. Comparative Biochemistry and Physiology Part A: Physiology , 42 (2), 359-366. https://doi.org/10.1016/0300-9629(72)90117-XHeiling, J. M., Bronstein, J. L., & Irwin, R. E. (2021). Nectar addition changes pollinator behavior but not plant reproduction in pollen‐rewarding Lupinus argenteus . American Journal of Botany , 108 (3), 402-410. https://doi.org/10.1002/ajb2.1613Jordano, P., Vázquez, D., & Bascompte, J. (2009). Redes complejas de interacciones mutualistas planta-animal. In Ecología y Evolución de Interacciones Planta-Animal (Editorial Universitaria, pp. 17–41).Kim, W., Gilet, T., & Bush, J. W. M. (2011). Optimal concentrations in nectar feeding. Proceedings of the National Academy of Sciences , 108 (40), 16618-16621. https://doi.org/10.1073/pnas.1108642108Kingsolver, J. G., & Daniel, T. L. (1983). Mechanical determinants of nectar feeding strategy in hummingbirds: Energetics, tongue morphology, and licking behavior. Oecologia , 60 (2), 214-226. https://doi.org/10.1007/BF00379523Knowlton, J. L., Crafford, R. E., Tinoco, B. A., Padrón, P. S., & Wilson Rankin, E. E. (2022). High foraging fidelity and plant-pollinator network dominance of non-native honeybees ( Apis mellifera) in the ecuadorian Andes. Neotropical Entomology , 51 (5), 795-800. https://doi.org/10.1007/s13744-022-00967-6Lanuza, J. B., Bartomeus, I., Ashman, T., Bible, G., & Rader, R. (2021). Recipient and donor characteristics govern the hierarchical structure of heterospecific pollen competition networks. Journal of Ecology , 109 (6), 2329-2341. https://doi.org/10.1111/1365-2745.13640Leicht, E. A., & Newman, M. E. J. (2008). Community structure in directed networks. Physical Review Letters , 100 (11), 118703. https://doi.org/10.1103/PhysRevLett.100.118703Li, D.-F., Han, W.-L., Renner, S. S., & Huang, S.-Q. (2022). Touch-sensitive stamens enhance pollen dispersal by scaring away visitors. eLife , 11 , e81449. https://doi.org/10.7554/eLife.81449Lopes, S. A., Bergamo, P. J., Najara Pinho Queiroz, S., Ollerton, J., Santos, T., & Rech, A. R. (2022). Heterospecific pollen deposition is positively associated with reproductive success in a diverse hummingbird-pollinated plant community. Oikos . https://doi.org/10.1111/oik.08714Lunau, K., Papiorek, S., Eltz, T., & Sazima, M. (2011). Avoidance of achromatic colours by bees provides a private niche for hummingbirds. Journal of Experimental Biology , 214 (9), 1607-1612. https://doi.org/10.1242/jeb.052688Morales, C. L., & Traveset, A. (2008). Interspecific pollen transfer: Magnitude, prevalence and consequences for plant fitness. Critical Reviews in Plant Sciences , 27 (4), 221-238. https://doi.org/10.1080/07352680802205631Moreira-Hernández, J. I., & Muchhala, N. (2019). Importance of pollinator-mediated interspecific pollen transfer for Angiosperm evolution. Annual Review of Ecology, Evolution, and Systematics , 50 (1), 191-217. https://doi.org/10.1146/annurev-ecolsys-110218-024804Muchhala, N. (2007). Adaptive trade‐off in floral morphology mediates specialization for flowers pollinated by bats and hummingbirds. The American Naturalist , 169 (4), 494-504. https://doi.org/10.1086/512047Muchhala, N., Brown, Z., Armbruster, W. S., & Potts, M. D. (2010). Competition drives specialization in pollination systems through costs to male fitness. The American Naturalist , 176 (6), 732-743. https://doi.org/10.1086/657049Muchhala, N., & Thomson, J. D. (2012). Interspecific competition in pollination systems: Costs to male fitness via pollen misplacement. Functional Ecology , 26 (2), 476-482. https://doi.org/10.1111/j.1365-2435.2011.01950.xNémeth, M. B., & Smith‐Huerta, N. L. (2003). Pollen deposition, pollen tube growth, seed production, and seedling performance in natural populations of Clarkia unguiculata (Onagraceae). International Journal of Plant Sciences , 164 (1), 153-164. https://doi.org/10.1086/344549Ogishima, M., Hoshino, Y., Horie, S., Yamashiro, T., Maki, M., Suzuki, K., & Dohzono, I. (2022). Secondary contact and adaptation to local pollinator assemblages mediate geographical variation in corolla length in Isodon shikokianus . Plant Species Biology , 37 (3), 222-230. https://doi.org/10.1111/1442-1984.12370Peuker, M. A., Burger, H., Krausch, S., Neumüller, U., Ayasse, M., & Kuppler, J. (2020). Floral traits are associated with the quality but not quantity of heterospecific stigmatic pollen loads. BMC Ecology , 20 (1), 54. https://doi.org/10.1186/s12898-020-00323-5Sonne, J., Martín González, A. M., Maruyama, P. K., Sandel, B., Vizentin-Bugoni, J., Schleuning, M., Abrahamczyk, S., Alarcón, R., Araujo, A. C., Araújo, F. P., Mendes De Azevedo, S., Baquero, A. C., Cotton, P. A., Ingversen, T. T., Kohler, G., Lara, C., Guedes Las-Casas, F. M., Machado, A. O., Machado, C. G., … Dalsgaard, B. (2016). High proportion of smaller ranged hummingbird species coincides with ecological specialization across the Americas. Proceedings of the Royal Society B: Biological Sciences , 283 (1824), 20152512. https://doi.org/10.1098/rspb.2015.2512Sonne, J., Vizentin-Bugoni, J., Maruyama, P. K., Araujo, A. C., Chávez-González, E., Coelho, A. G., Cotton, P. A., Marín-Gómez, O. H., Lara, C., Lasprilla, L. R., Machado, C. G., Maglianesi, M. A., Malucelli, T. S., González, A. M. M., Oliveira, G. M., Oliveira, P. E., Ortiz-Pulido, R., Rocca, M. A., Rodrigues, L. C., … Dalsgaard, B. (2020). Ecological mechanisms explaining interactions within plant–hummingbird networks: Morphological matching increases towards lower latitudes. Proceedings of the Royal Society B: Biological Sciences , 287 (1922), 20192873. https://doi.org/10.1098/rspb.2019.2873Stang, M., Klinkhamer, P. G. L., & Van Der Meijden, E. (2006). Size constraints and flower abundance determine the number of interactions in a plant-flower visitor web. Oikos , 112 (1), 111-121. https://doi.org/10.1111/j.0030-1299.2006.14199.xStewart, A. B., & Dudash, M. R. (2016). Differential pollen placement on an Old World nectar bat increases pollination efficiency. Annals of Botany, 117(1), 145–152. https://doi.org/10.1093/aob/mcv163Stiles, F. G. (1981). Geographical aspects of bird-flower coevolution, with particular reference to Central America. Annals of the Missouri Botanical Garden , 68 (2), 323. https://doi.org/10.2307/2398801Stiles, F. G., & Freeman, C. E. (1993). Patterns in floral nectar characteristics of some bird-visited plant species from Costa Rica. Biotropica , 25 (2), 191. https://doi.org/10.2307/2389183Strauss, S. Y. (1991). Indirect effects in community ecology: Their definition, study and importance. Trends in Ecology & Evolution, 6(7), 206–210. https://doi.org/doi.org/10.1016/0169-5347(91)90023-Q.Streher, N. S., Bergamo, P. J., Ashman, T.-L., Wolowski, M., & Sazima, M. (2020). Effect of heterospecific pollen deposition on pollen tube growth depends on the phylogenetic relatedness between donor and recipient. AoB PLANTS , 12 (4), plaa016. https://doi.org/10.1093/aobpla/plaa016Suárez-Mariño, A., Arceo-Gómez, G., Sosenski, P., & Parra‐Tabla, V. (2019). Patterns and effects of heterospecific pollen transfer between an invasive and two native plant species: The importance of pollen arrival time to the stigma. American Journal of Botany , 106 (10), 1308-1315.Thébault, E., & Fontaine, C. (2010). Stability of ecological communities and the architecture of mutualistic and trophic networks. Science , 329 (5993), 853-856. https://doi.org/10.1126/science.1188321Thomson, J. D., Andrews, B. J., & Plowright, R. C. (1982). The Effect of a Foreign Pollen on Ovule Development in Diervilla lonicera (Caprifoliaceae). New Phytologist , 90 (4), 777-783.Tinoco, B. A., Graham, C. H., Aguilar, J. M., & Schleuning, M. (2017). Effects of hummingbird morphology on specialization in pollination networks vary with resource availability. Oikos , 126 (1), 52-60. https://doi.org/10.1111/oik.02998Tur, C., Sáez, A., Traveset, A., & Aizen, M. A. (2016). Evaluating the effects of pollinator-mediated interactions using pollen transfer networks: Evidence of widespread facilitation in south Andean plant communities. Ecology Letters , 19 (5), 576-586. https://doi.org/10.1111/ele.12594Van Der Niet, T., Peakall, R., & Johnson, S. D. (2014). Pollinator-driven ecological speciation in plants: New evidence and future perspectives. Annals of Botany , 113 (2), 199-212. https://doi.org/10.1093/aob/mct290Vázquez, D. P., Melián, C. J., Williams, N. M., Blüthgen, N., Krasnov, B. R., & Poulin, R. (2007). Species abundance and asymmetric interaction strength in ecological networks. Oikos , 116 (7), 1120-1127. https://doi.org/10.1111/j.0030-1299.2007.15828.xVélez, A., Graham, C. H., Alarcón, I. P., Arias, R., Cárdenas-Calle, S., Landázuri, O., Espinosa, C. I., & Tinoco, B. A. (2025). Changes in functional traits and resources reduce the specialization of hummingbirds in fragmented landscapes. Global Ecology and Conservation , 58 , e03469. https://doi.org/10.1016/j.gecco.2025.e03469Vizentin-Bugoni, J., Maruyama, P. K., & Sazima, M. (2014). Processes entangling interactions in communities: forbidden links are more important than abundance in a hummingbird–plant network. Proceedings of the Royal Society B: Biological Sciences, 281(1780), 20132397.Wei, N., Kaczorowski, R. L., Arceo-Gómez, G., O’Neill, E. M., Hayes, R. A., & Ashman, T.-L. (2021). Pollinators contribute to the maintenance of flowering plant diversity. Nature , 597 (7878), 688-692. https://doi.org/10.1038/s41586-021-03890-9Weinstein, B. G., & Graham, C. H. (2017). Persistent bill and corolla matching despite shifting temporal resources in tropical hummingbird‐plant interactions. Ecology Letters , 20 (3), 326-335. https://doi.org/10.1111/ele.12730Werner, E. E., & Peacor, S. D. (2003). A review of trait-mediated indirect interactions in ecological communities. Ecology , 84 (5), 1083-1100. https://doi.org/10.1890/0012-9658(2003)084[1083:AROTII]2.0.CO;2Zhao, C., Ba, S., Zhang, T., Li, Y., & Fang, Q. (2025). Floral trait dissimilarity indirectly decreases heterospecific pollen transfer through pollinator sharing in a subalpine meadow. Oecologia , 207 (4), 55. https://doi.org/10.1007/s00442-025-05694-4 Appendix. Appendix table 1. List of node names with their corresponding plotcode for identification of the species present in the networks of figure 2. Species in bold are the species that lack data. math_shortcuts Plotcode Node names Plotcode Node names sp01 Bomarea uncifolia sp20 Salvia corrugata sp02 Brachyotum confertum sp21 Disterigma empetrifolium sp03 Ericaceae sp22 Barnadesia arborea sp04 Viola arguta sp23 Lamiaceae sp05 Castilleja sp. sp24 Axinaeae merianiae sp06 Tillandsia sp. sp25 Gaiadendron punctatum sp07 Gaultheria reticulata sp26 Tristerix longebracteatus sp08 Macleania rupestris sp27 Passiflora cumbalensis sp09 Pernettya prostrata sp28 Bejaria resinosa sp10 Stenomesson aurantiacum sp29 Berberis sp. sp11 Axinaeae pauciflora sp30 Vaccinium floribundum sp12 Oreocallis grandiflora sp31 Brugmansia sp. sp13 Asteraceae sp32 Gaultheria erecta sp14 Rubus sp. sp33 Gaultheria glomerata sp15 Chuquiragua jussieui sp34 Nasa sp. sp16 Vallea stipularis sp35 Tillandsia buserii sp17 Fuchsia sp. sp36 Tillandsia complanata sp18 Centropogon sp. sp37 Tillandsia stenoura sp19 Disterigma alaternoides sp38 Mutisia alata Appendix table 2. General information about the number of samples and pollen grains found on the network. TOTAL Locality Aguarongo El Gullán La Tranca Number of (species) 31 22 18 20 Number of samples 1 454 441 433 580 Number of pollen grain 191 058 70 570 68 233 52 255 Con-specific pollen grain 185 717 69 002 66 700 50 015 Average number of Con-specific pollen grains per species 237 159 340 96.2 Hetero-specific pollen grain 5341 1 568 1 533 2 240 Average number of Hetero-specific pollen grains per species 3.11 1.83 2.15 4.95 Appendix table 3. Results of the Permanova analysis assessing differences in community composition. The model indicates significant differences in community composition among groups. df sumofSqs R2 F Pr Model 2 0.92426 0.43247 2.2861 0.003* Residual 6 1.21291 0.56753 Total 8 2.13717 1 Appendix table 4. Linear model estimates for CP/HP ratio and node degree in Node degree IN Estimate Std. Error Degree freedom t-value P R2m R2c CP/HP ratio - 0.09992 0.03019 114 -3.31 0.00125 0.088 0.088 Appendix table 5. Linear model estimates for floral traits and node degree in values. Node degree in Estimate Std. Error Degree freedom t-value P R2m R2c Floral Abundance 0.110 0.018 109 6.03 <0.001 0.252 0.252 Opening corolla -0.05157 0.05108 102 -1.01 0.315 0.01 0.01 Floral tube length 0.0081 0.051 102 0.16 0.874 0.0002 0.0002 Stamen exsertion 0.02771 0.05489 34.18 0.505 0.617 0.007 0.015 Nectar concentration -0.04651 0.16 90 -0.290 0.772 0.0009 0.0009 Stigma exsertion 0.01837 0.03020 59 0.608 0.545 0.006 0.006 Nectar production 0.02995 0.03092 99 0.969 0.335 0.010 0.010 Appendix table 6. Linear model estimates for floral traits and node degree out values. math_shortcuts Node degree out Estimate Std. Error Degree freedom t-value P R2m R2c Floral Abundance -0.03227 0.03422 75 - 0.943 0.349 0.012 0.012 Opening corolla 0.08435 0.06738 92 1.252 0.214 0.017 0.017 Floral tube length 0.087 0.064 92 1.362 0.176 0.019 0.019 Stamen exsertion 0.007 0.068 40 0.11 0.913 0.0003 0.0003 Nectar concentration -0.4012 0.2110 87 -1.901 0.061 0.040 0.040 Stigma exsertion -0.01830 0.04222 64 -0.433 0.666 0.003 0.003 Nectar production 0.10332 0.04363 87 2.368 0.020 0.061 0.061 Appendices table 7. Node degree in for species found at each site and sampling period. Site/Repetition Species Aguarongo El Gullán La Tranca 1 2 3 1 2 3 1 2 3 Asteraceae 0 0 0 0 0 0 0 0 Axinaea merianiae 6 3 4 Axinaea pauciflora 1 2 0 Barnadesia arborea 7 2 5 0 0 0 2 0 Bejaria resinosa 2 2 Berberis sp. 7 Bomarea uncifolia 0 0 1 1 1 0 2 2 Brachyotum confertum 6 4 3 0 6 6 0 3 Brugmansia sp. 0 Castilleja sp. 2 0 Centropogon sp. 0 Chuquiraga jussieui 5 5 0 5 0 4 Disterigma alaternoides 2 2 4 Disterigma empetrifolium 2 Ericaceae 0 0 0 0 0 0 0 0 0 Fuchsia sp. 9 7 5 0 0 2 2 3 Gaiadendron punctatum 0 3 4 Gaultheria erecta 1 2 Gaultheria glomerata 2 3 7 4 Gaultheria reticulata 0 3 4 7 5 6 9 Macleania rupestris 2 5 7 4 9 0 Mutisia alata 3 Nasa sp. 0 Oreocallis grandiflora 1 3 2 6 6 8 3 2 10 Passiflora cumbalensis 2 3 5 Pernettya prostrata 4 2 2 2 4 Rubus sp. 3 2 2 1 3 0 Salvia corrugata 1 5 3 0 2 6 2 Sin identificar 0 0 0 0 0 0 0 Stenomesson aurantiacum 5 5 1 1 4 4 Tillandsia buseri 1 Tillandsia complanata 4 4 Tillandsia sp. 0 0 0 0 0 0 Tillandsia stenoura 3 2 0 Tristerix longebracteatus 0 Vaccinium floribundum 1 8 4 2 3 7 5 9 Vallea stipularis 0 7 7 0 0 Viola arguta 1 7 3 2 2 2 5 7 3 Appendices table 8. Node degree out for species found at each site and sampling period. math_shortcuts Site/Repetition Species Aguarongo El Gullán La Tranca 1 2 3 1 2 3 1 2 3 Asteraceae 4 3 3 4 5 2 8 7 Axinaea merianiae 1 9 0 Axinaea pauciflora 3 7 1 Barnadesia arborea 3 2 3 1 1 2 0 1 Bejaria resinosa 0 0 Berberis sp. 1 Bomarea uncifolia 6 8 2 4 0 2 6 4 Brachyotum confertum 8 9 11 3 9 12 6 13 Brugmansia sp. 2 Castilleja sp. 0 2 Centropogon sp. 2 Chuquiraga jussieui 1 0 1 1 1 0 Disterigma alaternoides 0 1 0 Disterigma empetrifolium 0 Ericaceae 1 6 3 4 3 5 3 8 5 Fuchsia sp. 3 5 3 3 1 0 2 1 Gaiadendron punctatum 1 1 1 Gaultheria erecta 0 0 Gaultheria glomerata 0 0 0 0 Gaultheria reticulata 6 0 2 0 4 2 1 Macleania rupestris 0 1 0 1 1 1 Mutisia alata 0 Nasa sp. 2 Oreocallis grandiflora 4 1 1 5 4 10 1 3 4 Passiflora cumbalensis 0 1 5 Pernettya prostrata 2 0 0 3 0 Rubus sp. 0 2 0 1 4 3 Salvia corrugata 0 1 6 1 0 1 3 Sin identificar 3 1 3 3 1 6 3 Stenomesson aurantiacum 1 6 0 0 3 1 Tillandsia buseri 0 Tillandsia complanata 1 0 Tillandsia sp. 6 9 4 1 2 4 Tillandsia stenoura 0 0 3 Tristerix longebracteatus 1 Vaccinium floribundum 0 0 0 0 0 0 0 0 Vallea stipularis 1 6 1 4 2 Viola arguta 2 2 5 2 2 5 7 2 7 Appendix figure 1. Heatmap of correlation coefficients among floral traits. The color and size of the circles indicate the strength and direction of the correlation (red = positive, blue = negative). Strong correlations are observed between stigma and stamen exertion, and between floral tube length and stigma-stamen distance. Appendix figure 2 . Beta-diversity among sampling sites and periods. NMDS ordination based on Bray-Curtis dissimilarities, showing variation in community composition across three sites (Aguarongo, El Gullán and La Tranca) and three sampling periods. Each point represents the composition of a community based on data of the number of flowers produced by each species, with colors indicating site and shapes indicating sampling period. Information & Authors Information Version history V1 Version 1 03 September 2025 Peer review timeline Published Ecology and Evolution Version of Record 30 Sep 2025 Published Copyright This work is licensed under a Non Exclusive No Reuse License. Collection Ecology and Evolution Keywords community ecology comparative description plants statistical terrestrial Authors Affiliations Ann Frías-Romero 0009-0003-5456-408X [email protected] Universidad del Azuay View all articles by this author Boris Tinoco 0000-0003-1054-8945 Universidad del Azuay View all articles by this author Bryan Rojas Universidad del Azuay View all articles by this author Ariana Vélez Universidad del Azuay View all articles by this author Samara Zeas-Bermeo Universidad del Azuay View all articles by this author Catherine Graham 0000-0001-9267-7948 Swiss Federal Institute for Forest Snow and Landscape Research WSL View all articles by this author Metrics & Citations Metrics Article Usage 230 views 201 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Ann Frías-Romero, Boris Tinoco, Bryan Rojas, et al. Floral roles in hummingbirds-mediated indirect plant interactions in Tropical Andean communities.. Authorea . 03 September 2025. DOI: https://doi.org/10.22541/au.175692134.44757608/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.175692134.44757608/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'a02846239eb3c13d',t:'MTc3OTkxOTAzMg=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();