J Plant Ecol ›› 2017, Vol. 10 ›› Issue (4): 634-648.DOI: 10.1093/jpe/rtw068

• Research Articles • Previous Articles     Next Articles

The role of the pollination niche in community assembly of Erica species in a biodiversity hotspot

Roderick W. Bouman1,2,3, Sandy-Lynn Steenhuisen4 and Timotheüs van der Niet1,2,5,*   

  1. 1 Naturalis Biodiversity Center, Understanding Evolution Group, Darwinweg 4, 2333 CR Leiden, The Netherlands; 2 Leiden University, Sylviusweg 72, 2333 BE Leiden, The Netherlands; 3 Hortus Botanicus Leiden, 5e Binnenvestgracht 8, 2311 VH Leiden, The Netherlands; 4 Department of Biological Sciences, University of Cape Town, Private Bag X3, Rondebosch 7701, South Africa; 5 School of Life Sciences, University of KwaZulu-Natal, Private Bag X01, Scottsville 3209, South Africa
  • Received:2015-09-30 Accepted:2016-06-19 Published:2017-07-24
  • Contact: van der Niet, Timotheus

Abstract: Aims Closely related species frequently co-occur in plant communities. The role of niche partitioning for allowing their coexistence has been called into question, but most studies on community ecology neglect pollination as a niche component, and thus potentially underestimate niche partitioning. Here, we investigate the importance of the pollination niche for determining coexistence in a biodiversity hotspot. We specifically test whether co-flowering insect-pollinated species of Erica, the most species-rich plant genus of the South African Cape Floristic Region, are characterized by different functional floral traits and different specialized pollination systems.
Methods Communities of co-flowering Erica species were studied at three sites in the fynbos biome in the Western Cape of South Africa. We focused on five species with flowers that conform to an insect pollination syndrome. We assessed whether floral traits important for pollinator attraction and mechanical fit overlap by comparison of scent (gas chromatography coupled with mass spectrometry), colour (spectrophotometry and models of bee vision), nectar volume and concentration and corolla morphology. Pollinator partitioning was quantified by comparison of floral visitor profiles and pollinator importance, calculated as the product of visitation rate, number of flowers probed per visit and pollen load for each functional visitor group among species within communities. Potential effects of competition were assessed by comparing visitation rates both directly and using anther ring disturbance as a proxy for lifetime visitation.
Important findings Floral phenotypes differed among species within communities in all functional traits. Visitor assemblages were diverse, including ants, beetles, flies, honeybees, moths and solitary bees. However, the Cape honeybee, Apis mellifera capensis, was the main pollinator of all but one Erica species. Insect visitation rates were significantly lower for E. imbricata and E. calycina relative to congeners, but high rates of anther ring disturbance, which differed significantly in only two out of seven comparisons, suggest that long-term visitation rates are mostly similar among species that share pollinators. This study shows that coexistence of co-flowering insect-pollinated Erica species is not determined by specialization for different pollination systems. Pollinator sharing might rather lead to facilitation among co-flowering species, and floral constancy mediated by extensive interspecific floral trait variation might mitigate its potentially negative effects.

Key words: Cape Floristic Region, coexistence, floral traits, heaths, pollination niche

摘要:
Aims Closely related species frequently co-occur in plant communities. The role of niche partitioning for allowing their coexistence has been called into question, but most studies on community ecology neglect pollination as a niche component, and thus potentially underestimate niche partitioning. Here, we investigate the importance of the pollination niche for determining coexistence in a biodiversity hotspot. We specifically test whether co-flowering insect-pollinated species of Erica, the most species-rich plant genus of the South African Cape Floristic Region, are characterized by different functional floral traits and different specialized pollination systems.
Methods Communities of co-flowering Erica species were studied at three sites in the fynbos biome in the Western Cape of South Africa. We focused on five species with flowers that conform to an insect pollination syndrome. We assessed whether floral traits important for pollinator attraction and mechanical fit overlap by comparison of scent (gas chromatography coupled with mass spectrometry), colour (spectrophotometry and models of bee vision), nectar volume and concentration and corolla morphology. Pollinator partitioning was quantified by comparison of floral visitor profiles and pollinator importance, calculated as the product of visitation rate, number of flowers probed per visit and pollen load for each functional visitor group among species within communities. Potential effects of competition were assessed by comparing visitation rates both directly and using anther ring disturbance as a proxy for lifetime visitation.
Important findings Floral phenotypes differed among species within communities in all functional traits. Visitor assemblages were diverse, including ants, beetles, flies, honeybees, moths and solitary bees. However, the Cape honeybee, Apis mellifera capensis, was the main pollinator of all but one Erica species. Insect visitation rates were significantly lower for E. imbricata and E. calycina relative to congeners, but high rates of anther ring disturbance, which differed significantly in only two out of seven comparisons, suggest that long-term visitation rates are mostly similar among species that share pollinators. This study shows that coexistence of co-flowering insect-pollinated Erica species is not determined by specialization for different pollination systems. Pollinator sharing might rather lead to facilitation among co-flowering species, and floral constancy mediated by extensive interspecific floral trait variation might mitigate its potentially negative effects.