Differential pollen placement on an Old World nectar bat increases pollination efficiency

Abstract

Background and aims: Plant species that share pollinators are potentially subject to non-adaptive interspecific pollen transfer, resulting in reduced reproductive success. Mechanisms that increase pollination efficiency between conspecific individuals are therefore highly beneficial. Many nocturnally flowering plant species in Thailand are pollinated by the nectar bat Eonycteris spelaea (Pteropodidae). This study tested the hypothesis that plant species within a community reduce interspecific pollen movement by placing pollen on different areas of the bat's body.

Methods: Using flight cage trials, pollen transfer by E. spelaea was compared between conspecific versus heterospecific flowers across four bat-pollinated plant genera. Pollen from four locations on the bat's body was also quantified to determine if pollen placement varies by plant species.

Key results: It was found that E. spelaea transfers significantly more pollen between conspecific than heterospecific flowers, and that diverse floral designs produce significantly different patterns of pollen deposition on E. spelaea.

Conclusions: In the Old World tropics, differential pollen placement is a mechanism that reduces competition among bat-pollinated plant species sharing a common pollinator.

Keywords: Ceiba pentandra; Chiropterophily; Durio zibethinus; Musa acuminata; Parkia speciosa; Parkia timoriana.; Pteropodidae; Thailand; bat pollination; floral morphology; interspecific pollen transfer; pollen placement.

Fig. 1.

Eonycteris spelaea, a common nectar bat and important pollinator in Thailand.

Fig. 2.

Floral designs of plant species pollinated by Eonycteris spelaea in Thailand. (A) Ceiba pentandra. Five anthers with yellow pollen surround a white stigma. (B) Musa acuminata. A red bract curls back to reveal two rows of white flowers. (C) Parkia speciosa. Drops of nectar produced by flowers at the neck of the inflorescence (inset) accumulate above pollen-bearing flowers at the base. (D) Durio zibethinus. Many stamens surround a pale yellow stigma in the centre. Scale bars = 1 cm.

Fig. 3.

The average amount of pollen with 95 % confidence intervals (using back-transformed values) transferred by Eonycteris spelaea to female (A) Ceiba pentandra, (B) Durio zibethinus and (C) Musa acuminata flowers by male flowers of different plant species. Significantly more pollen was transferred between conspecific flowers (closed circles) than heterospecific flowers (open circles) (GLM pairwise contrasts with sequential Bonferroni correction, P < 0·001). Floral combinations (n = 8 trials per floral combination) along the x-axis are listed with male flower species preceding female flower species (e.g. ‘Musa–Ceiba’ means pollen transferred from male Musa acuminata flowers to female Ceiba pentandra flowers). Parkia inflorescences (P. speciosa and P. timoriana) were only used as pollen donors (male flowers). Durio–Ceiba and Durio–Parkia combinations could not be tested because their flowering phenologies did not overlap.

Fig. 4.

Back-transformed model estimates (with 95 % confidence intervals) of the number of pollen grains placed on different areas of Eonycteris spelaea: the crown of the head, chest, face and ventral side of the wings. The bat-pollinated plant species tested were (A) Ceiba pentandra (n = 16 bats), (B) Musa acuminata (n = 24 bats), (C) Durio zibethinus (n = 16 bats) and (D) Parkia speciosa and P. timoriana (n = 16 bats). Patterns of pollen deposition differed significantly among plant species (GLMM pairwise contrasts, P < 0·001), except between C. pentandra and D. zibethinus (P > 0·9). Within each graph, means with different letters are significantly different (GLMM pairwise contrasts with sequential Bonferroni correction, P < 0·05).