Plant - insect interactions in a riparian grassland: a community approach to pollination

Abstract

This investigation into plant and insect interactions was carried out at a riparian grassland site in Shincliffe, Durham and commenced during the month of May and ended in early July. Vegetation surveys were carried out throughout the growing season to establish the species composition, plant and flower density, dominant plant species and various plant characteristics of the community as a whole. The plant found most frequently in flower during May was Stellaria holostea, whilst in June the plants most frequently in flower included S. holostea, Cruciata laevipes and Aegopodium podagraria. In July A. podagraria was once more the most frequently flowering plant. There was also a progressive increase in plant diversity, morphology and flower colour throughout the study period. Observation periods throughout the investigation enabled the overall insect and pollinator assemblage of the community to be identified. Insect abundance was found to be diumally heterogeneous throughout the study period. The only 'within' differences in ranked activity were identified for Diptera in June and 'others' in July. There were also significant differences in the absolute diurnal activity within the orders of Diptera, Coleoptera, Hymenoptera and 'others'. There were also quite striking and significant seasonal variations in the insect assemblage. Pollinator abundance was also found to be diurnally heterogeneous in June and July, whilst being homogeneous throughout May. In addition the dominant pollinators varied seasonally. There were no significant 'within' pollinator differences in ranked or absolute activity during any of the months. During all three months plant - pollinator interactions were looked at in some detail. Significant relationships were found to exist between the ranked number of flowers and the ranked number of pollinators per quadrat. In May there were also significant relationships between the ranked number of Diptera, Coleoptera and flowers per quadrat. In June there was a significant correlation between the ranked number of Hymenoptera and flowers per quadrat. There were no such findings for the month of July. Finally, during May there was a significant correlation between the ranked number of flowering stems and the ranked number of pollinator visits per quadrat. The preference and avoidance of certain plant characteristics such as colour and flower morphology were identified for some groups of pollinators. However, although some of these interactions were analysed, the precise nature of pollinator behaviour was difficult to prove. A closer look at individual plant species and their associated pollinator assemblages enabled generalisations to be made, but these associations were not apparent when the community was studied as a whole.