EFFECT OF INFLORESCENCE-FEEDING INSECTS ON THE DEMOGRAPHY AND LIFETIME FITNESS OF A NATIVE PLANT

The importance of herbivorous insects for plant demography and fitness is still controversial. The fact that their effect has not been fully evaluated within the context of the complete plant life cycle contributes to this controversy. This study is the first to test directly the influence of flower and seed consumption by a guild of inflorescence-feeding insects on the demography and fitness of a native plant over its whole life cycle within its natural environment. First, we discuss the evidence required to assess the effect of inflorescence-feeding insects on sequential stages in the plant life cycle. Then, we present the detailed results of two experiments that quantify the effect of inflorescence-feeding insects on seeds, seedlings, juveniles, and subsequent flowering progeny of Cirsium canescens (Platte thistle), a native, monocarpic perennial species of Sandhills prairie, Nebraska. Exclusion of inflorescence-feeding insects by insecticide had four main effects. Total seed output increased (P < 0.0005 in both experiments), indicating that resources were not limiting seed production. Flower heads produced later in the season contributed to the seed pool when insects were reduced (P < 0.01), suggesting that insect feeding restricted the phenology of flowering and pollination. The density of seedlings increased around plants protected from inflorescence-feeding insects (P < 0.0005), showing that safe sites were not limiting. Finally, the increase in seedlings led to higher numbers of flowering adults (P < 0.009), demonstrating that inflorescence-feeding insects significantly reduced lifetime fitness. Neither individual compensation for seed loss, nor density-dependent compensation for increased plant population densities, was observed when insects were excluded. Thus, we conclude that the inflorescence-feeding insect herbivores limited seed production, seedling recruitment, plant density, and maternal fitness of this plant under natural conditions. This outcome challenges current theoretical models of the effect of insects on plant population dynamics. We predict that our result will be general for short-lived perennial plants with life histories in which persistence is tightly linked to regeneration from current seed, such as for fugitive perennials with transient seed banks.