Pollinator Competition as a Driver of Floral Divergence: An Experimental Test

Abstract

Optimal foraging models of floral divergence predict that competition between two different types of pollinators will result in partitioning, increased assortative mating, and divergence of two floral phenotypes. We tested these predictions in a tropical plant-pollinator system using sexes of purple-throated carib hummingbirds (Anthracothorax jugularis) as the pollinators, red and yellow inflorescence morphs of Heliconia caribaea as the plants, and fluorescent dyes as pollen analogs in an enclosed outdoor garden. When foraging alone, males exhibited a significant preference for the yellow morph of H. caribaea, whereas females exhibited no preference. In competition, males maintained their preference for the yellow morph and through aggression caused females to over-visit the red morph, resulting in resource partitioning. Competition significantly increased within-morph dye transfer (assortative mating) relative to non-competitive environments. Competition and partitioning of color morphs by sexes of purple-throated caribs also resulted in selection for floral divergence as measured by dye deposition on stigmas. Red and yellow morphs did not differ significantly in dye deposition in the competition trials, but differences in dye deposition and preferences for morphs when sexes of purple-throated caribs foraged alone implied fixation of one or the other color morph in the absence of competition. Competition also resulted in selection for divergence in corolla length, with the red morph experiencing directional selection for longer corollas and the yellow morph experiencing stabilizing selection on corolla length. Our results thus support predictions of foraging models of floral divergence and indicate that pollinator competition is a viable mechanism for divergence in floral traits of plants.

Motifs in bipartite ecological networks: uncovering indirect interactions

“…A similar situation is found in Fig. 2e, where A 1 and A 2 represent pollinators and P represents a plant species: exploitative and interference competition between pollinators can occur, as can indirect facilitation, where an increase in the abundance of A 1 can provide a beneficial effect to A 2 through P (Rathcke 1983, Temeles et al 2016. Similar ideas can be applied to other motifs where many specialists interact with a single generalist, such as motifs 4, 7, 8, 17, 18 and 44 ( Fig.…”

Sex differences in pollinator behavior: Patterns across species and consequences for the mutualism

“…Positive directional selection on corolla length has also been observed in the deeper-flowered, female-visited species H. bihai in the field (Temeles et al, 2013), and in greenhouse studies females select for longer corollas in H. caribaea (Temeles et al, 2016). Similar morphological and behaviourial differences between males and females have been observed in a number of other hummingbird species (Temeles, Miller, & Rifkin, 2010), raising the possibility that these species segregate floral resources by sex as well.…”

A hurricane alters pollinator relationships and natural selection on an introduced island plant

“…Flowering of H. caribaea on Dominica occurs from February to July, with peak flowering in April and May (Gowda & Kress, ). Heliconia caribaea is primarily visited and pollinated by male purple‐throated caribs which defend clumps of these plants as territories (Temeles & Kress, ; Temeles et al., ). Female purple‐throated caribs and less commonly green‐throated caribs ( A. holosericeus ) intrude on male territories to steal food and feed from undefended H. caribaea plants (Temeles & Kress, ).…”

“…We compared measurements of these birds to a sample of birds captured on Dominica from 2007 to 2016 and to a sample of birds captured in 2018 at the Tamarind Tree Hotel in Salisbury, Dominica. The first sample includes birds captured at various sites on the island during our previous studies (Temeles & Kress, ; Temeles et al., , ), and all measurements were made by the same person (EJT). The second sample consists of birds that were feeding from artificial food sources (sucrose solution in hummingbird feeders) provided by the hotel owners and staff immediately following Hurricane Maria and again were measured by EJT.…”