Insects are pollen vectors for thousands of angiosperm species. Because flowers are the physical locations of pollen transfer, any physical and/or behavioral compatibility between flowers and pollinator biology may increase plant reproductive success. This is especially true for a plant species with a pollinium-pollination system, such Asclepias syriaca (common milkweed), which involves a morphological match between its flowers and pollinator body parts that transfer pollen. Previous research suggests that body size and leg size may both be important for pollen removal, where individuals with larger bodies may be able to carry larger pollen loads and individuals with intermediate leg sizes may be more likely to catch their legs on pollinaria and remove them from flowers. However, no study has directly linked these body parts to pollen movement. Therefore, we investigated the relationship between the body lengths (body size) and tibia widths (leg size) of A. syriaca's visitors and the amount of pollen they carry (number of corpusculi attached to pollinators) and efficiency of pollen transfer (proportion of pollinia removed that are deposited).
We captured and examined 352 insects over two seasons, 2014 and 2015. The insects were hymenopterans (bees and kin), lepidopterans (butterflies and moths), dipterans (flies) and coleopterans (beetles). The most abundant pollinator taxon was Apis mellifera, followed by bumble bees (Bombus spp.), and solitary bees in the Megachilidae family. Apis mellifera had more attached corpusculi than all other taxa. Across taxa, intermediate size similar to that of A. mellifera resulted in the highest number of attachments. More specifically, bumble bees, on average, had larger tibia than A. mellifera, but those with smaller tibia had more corpusculi attached to their legs. Megachilidae, on average, had smaller body lengths than A. mellifera, but those with longer bodies had more corpusculi attached to their bodies. Attachment to legs resulted in the highest transfer efficiency, but there was no relationship between transfer efficiency and body or leg sizes. This inconsistency may be the result of the confounding influence of pollinarium chaining or visitor behavior. Nonetheless, our data partially support both the body size and leg size hypotheses. The high number of pollinia attached to A. mellifera may reflect their anatomical similarity to native Apoidea that co-evolved with A. syriaca.