We examined the consequences of habitat fragmentation on the assemblage of floral visitors and pollinators to male- and female-phase inflorescences of the understorey dominant palm Astrocaryum mexicanum at the Los Tuxtlas tropical rain forest. In six forest fragments ranging from 2 to 700 ha, we collected all floral visitors, pollinators and non-pollinators, to male- and female-phase inflorescences at the time of their greatest activity. We used multivariate and mixed-effects models to explore differences in guild composition between sexual phases of inflorescences and the effects of forest fragment size on several metrics of the assemblages of floral visitors. We detected 228 786 floral visitors, grouped into 57 species, across the six forest fragments. On average, abundance and species richness of floral visitors to female-phase inflorescences were higher than to male-phase ones. Forest fragmentation had no effect on species richness but negatively affected Shannon's diversity index. Overall, the most abundant species of floral visitors were predominantly found in inflorescences of plants from the large fragments. In contrast, most of the less common species were more abundant in the smallest fragments. The abundance of pollinators (those found on inflorescences of both phases and dusted with pollen that was carried to flower stigmas), and the ratio of pollinators to other floral visitors, increased with fragment size in both sexual phases of the inflorescences but these effects were significantly stronger on male-phase inflorescences than on female-phase inflorescences. These results show that tropical forest fragmentation correlates with changes in the composition of flower visitors to a dominant palm, with a reduction in the abundance of pollinators, but that such changes co-vary with the sexual phase of the plants.
Email your librarian or administrator to recommend adding this journal to your organisation's collection.
* Views captured on Cambridge Core between September 2016 - 28th May 2017. This data will be updated every 24 hours.