Background: Dioecious plants generally display sexual dimorphism in male and female floral traits, potentially attracting slightly different pollinator communities. The sharing of common floral visitors between male and female flowers and their timing of visits to both sexes is of critical importance to ensure plant's reproductive success. Palm inflorescences are visited by abundant and diverse insect communities, yet the temporal patterns of insect visits on both sexes remain poorly known.

Results: We report 59 morphospecies in the arthropod community, dominated by three beetle families: Staphylinidae, Nitidulidae and Curculionidae. Male inflorescences were more abundantly visited than female, but visitors of the later were taxonomically more diverse. Among the 16 pollinator candidates identified, 9 visited both inflorescence sexes synchronously at dusk /night whereas the others did so asynchronously during the day.

Conclusion: Our study provides new insights into the pollination mechanism of P. aequatorialis. We found evidence of differential pollinator attraction between floral sexes, which may be explained by the sexual dimorphism of both flowers. Synchronicity in dusk/night visits of both inflorescence sexes suggests a sexual synchronization of the signal used to attract pollinators.

We sampled flower-visiting insects of the tagua palm Phytelephas aequatorialis every 4 hours from the beginning (bud opening) to the end (significant fading in coloration and scent of rotting tissue) of the anthesis of these inflorescences by using interception traps. Interception traps were made of a transparent foil sheet (H = 57 cm and L = 26 cm), stretched out between two funnels (H = 32 cm and diameter = 26 cm), and held vertically in the vicinity of each focal flower (Fig. 1C). The bottom funnel was equipped with a collector so that flying insects colliding with the transparent foil fell into. The collector was filled with water and a wetting agent acting as a surfactant to prevent captured arthropods from escaping. Each trap was installed from the start of the flower anthesis i.e., when the peduncular bract (the bud integument covering the flower) opened, followed by either the quick elongation for the male inflorescence or the revealing of the tepals for the female inflorescence (Ervik et al. 1999). The collectors were retrieved and replaced by new ones every four hours. Because male flowers decay quickly, trapping was stopped within 24 to 48 hours after the beginning of the anthesis. Female anthesis may last several days so that trapping was stopped 28-52h after the opening. Insect fauna trapped in each collector was transferred into a plastic tube and conserved in 70° alcohol before identification and counting in the lab.