restricted access 12. Selection by Pollinators and Herbivores on Attraction and Defense
In lieu of an abstract, here is a brief excerpt of the content:

162 TW E LVE Selection by Pollinators and Herbivores on Attraction and Defense LYNN S. ADLER Interactions between plants, their herbivores, and their pollinators are thought to have led to the diversification of both plants and insects. Historically, studies of plant-herbivore and plant-pollinator interactions have occurred independently . Research at both micro- and macroevolutionary levels has focused on the evolution of plant resistance in the context of herbivory, and on floral traits in the context of pollination. For example, researchers have long recognized the role of plant secondary chemicals in herbivore feeding preferences (e.g., Dethier 1941). Fraenkel (1959) proposed that the “reason for existence” of plant secondary chemicals was to attract and deter herbivores rather than as products of plant metabolism, building on earlier work by Stahl (1888) and others. Ehrlich and Raven’s classic article (1964) on coevolution concluded that “the evolution of secondary plant substances and the stepwise evolutionary responses to these by phytophagous organisms have clearly been the dominant factors in the evolution of butterflies and . . . in the evolution of angiosperm subgroups” (p. 382). This publication and others in the 1970s (e.g., Feeny 1976; Rhoades and Cates 1976) led to a surge of interest in chemical defenses mediating plant-herbivore interactions. More recently, selection on plant resistance traits has been studied by manipulating herbivores as selective agents (e.g., Mauricio and Rausher 1997; Stinchcombe and Rausher 2001), and several phylogenetic studies interpret the diversi fication of plants and herbivores in the context of coevolution mediated by plant resistance traits (e.g., Farrell et al. 1991; Becerra 1997, 2003; Farrell and Mitter 1998; Cornell and Hawkins 2003). By contrast, the diversity of plant floral traits has been interpreted as the result of evolution due to their obvious role in attracting pollinators and promoting efficient pollination . Sprengel’s 1793 pioneering treatise (first translated to English in 1996) interpreted floral function in terms of relationships with pollinators and inspired much of the subsequent field of pollination ecology. Darwin also interpreted the floral morphological variation of heterostylous plants in terms of its role in promoting outcrossing (Darwin 1877). More recently, pollination biologists have suggested that the evolution of floral traits may be shaped by a diversity of pollinators, rather than a single pollinator or guild type (e.g., Herrera 1996; Waser et al. 1996). Recent theoretical models and manipulative studies continue to focus on the role of pollinators in shaping the evolution of floral traits (e.g., Dafni and Kevan 1997; Aigner 2001; Fenster et al. 2004) and on floral trait phylogenetic diversity in the context of pollinator attraction and efficiency (e.g., Jurgens 2004; Sargent and Otto 2004; Manning and Goldblatt 2005; Ree 2005). Clearly, herbivores have been a major selective force in the evolution of plant defense, and pollinators have been a major selective force in the evolution of attractive floral traits. However, a growing number of studies suggest that traits that deter herbivores may affect pollinator attraction, and traits that attract pollinators may affect herbivores. Herbivores and pollinators could exert selection on plant traits either through direct interactions (a pairwise relationship between plants and insects mediated by the trait), or via indirect interactions (insect selection on plant traits that is mediated by a third species) (Wootton 1994). Over half a century ago Grant (1950) recognized that floral morphology may have evolved to protect ovules from damage by some pollinators, such as birds and beetles, as well as to promote pollen transfer. More recent phylogenetic studies demonstrate that floral traits have evolved in response to selection from both pollinators and herbivores (e.g., Armbruster 1997; Armbruster et al. 1997; Pellmyr 2003). For example, resin-secreting floral glands that defended flowers of Dalechampia vines were subsequently co-opted as a reward for resin-collecting pollinating bees (Armbruster 1997). Thus, the evolution of plant traits may be shaped by simultaneous or sequential interactions with both pollinators and herbivores. Attractive and defensive traits can be genetically correlated via linkage or pleiotropy. Thus herbivore-imposed selection on resistance may drive the evolution of floral traits and vice versa. Furthermore, herbivores and pollinators themselves are not independent of each other. Several insect taxa include species that are pollinators as adults and herbivores as larvae. Traits that attract adult pollinators therefore have the potential to increase subsequent herbivory in some systems. The consequence of these interactions for plant fitness will depend on the level of specialization and on community context. For example, a pollinating herbivore may benefit...