Abstract
Species extinctions undermine ecosystem functioning, with the loss of a small subset of functionally important species having a disproportionate impact. However, little is known about the effects of species loss on plant-pollinator interactions. We addressed this issue in a field experiment by removing the plant species with the highest visitation frequency, then measuring the impact of plant removal on flower visitation, pollinator effectiveness and insect foraging in several sites. Our results show that total visitation decreased exponentially after removing 1-4 most visited plants, suggesting that these plants could benefit co-occurring ones by maintaining high flower visitor abundances. Although we found large variation among plant species, the redistribution of the pollinator guild affected mostly the other plants with high visitor richness. Also, the plant traits mediated the effect of removal on flower visitation; while visitation of plants which had smaller inflorescences and more sugar per flower increased after removal, flower visitors did not switch between flower shapes and visitation decreased mostly in plants visited by many morpho-species of flower visitors. Together, these results suggest that the potential adaptive foraging was constrained by flower traits. Moreover, pollinator effectiveness fluctuated but was not directly linked to changes of flower visitation. In conclusion, it seems that the loss of generalist plants alters plant-pollinator interactions by decreasing pollinator abundance with implications for pollination and insect foraging. Therefore, generalist plants have high conservation value because they sustain the complex pattern of plant-pollinator interactions.