FLORAL BIOLOGY, MICROCLIMATE, AND POLLINATION BY ECTOTHERMIC BEES IN AN EARLY-BLOOMING HERB

Abiotic factors may constrain the functioning of species interactions such as plant-pollinator mutualisms. I investigated how thermal environment affects the interaction between the early-blooming daffodil, Narcissus longispathus (Amaryllidaceae) and its major bee pollinator (Andrena bicolor; Andrenidae), focusing simultaneously on plant and pollinator sides of the interaction. I studied fruit and seed set, flower duration, and the intrafloral thermal environment of N. longispathus, and the thermal biology, foraging behavior, and thermoregulatory ability of A. bicolor over a 6-yr period in southeastern Spain. N. longispathus flowers from February to April, when unsuitable weather often limits pollinator activity, yet most flowers are successfully pollinated in all years and sites. Fruit set was weakly pollen limited, but among flowers setting fruit the proportion of ovules developing into seeds was not. Individual flowers lasted for 17 d on average, remaining functional during this period. On sunny days, the air inside N. longispathus flowers was significantly warmer than outside. Mean temperature excess inside flowers was as high as 8 degrees C, and was positively related to solar irradiance. Within flowers, air temperature was highest around the anthers; this intrafloral gradient was consistent with variation among perianth parts in radiation transmittance. Andrena bicolor foraged in N. longispathus flowering patches only on sunny days with air temperature >12 degrees-13 degrees C, and foraging behavior and flower visitation rate were temperature dependent. Bees were able to fly at relatively low thoracic temperatures (T-th; range 22 degrees-31 degrees C) and this was essential for successfully foraging at N. longispathus. Under the range of irradiance and air temperature found at foraging sites, A. bicolor individuals inside flowers were able to reach T-th suitable for flight by passive means alone. Under laboratory conditions, A. bicolor was unable to raise or otherwise regulate T-th by physiological means, but free-flying individuals thermoregulated behaviorally. Basking was used to raise T-th and intrafloral microclimate, by influencing the proportion of foraging time devoted to basking, played an important role in thermoregulation. Flower visitation rate was positively related to the average temperature inside visited flowers, and the probability of basking immediately after one floral visit declined with increasing flower temperature. I conclude that the favorable microclimate within N. longispathus flowers, their long duration, and the thermal biology of A. bicolor, were critical elements in this early-season pollination system.