In lieu of an abstract, here is a brief excerpt of the content:

Understanding the physiological and behavioral mechanisms underlying host-plant specialization in holometabolous species, which undergo complete development with a pupal stage, presents a particular challenge in that the process of host-plant selection is generally carried out by the adult stage, whereas host-plant utilization is more the province of the larval stage (Thompson 1988a, 1988b). Thus, within a species, critical chemical, physical, or visual cues for host-plant identi- fication may differ over the course of the life cycle. An organizing principle for the study of host-range evolution is the preference-performance hypothesis (Jaenike 1978). According to this hypothesis, ovipositing females should maximize their fitness by selecting plants on which offspring survival will be high; in other words, over a range of potential host plants, adult female preference should be correlated with larval performance . Evidence in support of this hypothesis has been mixed (Thompson 1988a, 1988b; Scheirs and deBruyn 2002). Poor correlations have been attributable in some cases to opposing evolutionary forces that include conflicting selection pressures exerted by trophic factors such as predation and parasitism (Camara 1997; Heisswolf et al. 2005); ecological factors such as relative rarity of most-suitable hosts (Rausher 1980), recent invasion of less-suitable hosts into an otherwise coevolved community (Wiklund 1975), or thermal constraints on voltinism patterns (Scriber 1996, 2002); phenotypic plasticity of host-plant selection (Mercader and Scriber 2005); or parent-offspring conflict in the case of species in which adults are also phytophagous (Roitberg and Mangel 1993). Although positive correlations are frequently found in other taxa (viz., Craig and Itami, this volume), such correlations are exceptional in lepidopterans. The dichotomy between adult behavioral preference and larval physiological performance in Lepidoptera is key to longstanding debates over the evolution of specialization in general. The timing and nature of adaptation to plant chemistry throughout the life cycle are central to these debates. Almost 60 years ago, Dethier (1948) suggested that “the first barrier to be overcome in the insect-plant relationship is a behavioral one. The insect must sense and discriminate before nutritional and toxic factors become operative” (p. 98). Thus, Dethier argued for the primacy of adult preference , or detection and response to kairomonal cues, in host-plant shifts. In contrast, Ehrlich and Raven (1964) reasoned that “after the restriction of certain groups of insects to a narrow range of food plants, the formerly repellent substances of these plants might . . . become chemical attractants ” (p. 602), arguing for the primacy of shared allomonal phytochemistry and larval detoxification in initiating host shifts (with the evolution of kairomonal responses following ). Oviposition “mistakes,” in which females lay eggs on host plants that do not support larval growth (e.g., Berenbaum 1981b), are cited as evidence for the importance of behavioral cues in initiating host shifts; colonization of novel host plants that share the same range of defense compounds is claimed as evidence of the primacy of larval performance in the process (see Berenbaum 1990 for a review). In view of the fact that the physiological and behavioral changes necessary to effect a host shift arise by random mutation, primacy is in a sense irrelevant; host shifts are completed when the necessary behavioral and physiological traits are in place, irrespective of the order in which they occur (Berenbaum 1990). Complicating the resolution of this issue are recent experimental studies indicating that preference and performance traits involved in host shifts may not be entirely under heritable genetic control and are instead influenced by epigenetic and even nongenetic factors. Evidence suggests that larval feeding experience in at least some taxa may influence adult oviposition preference either via “chemical legacy” (retained chemical signals) or via some form of retained memory (Barron 2001; Rietdorf and Steidle 2002; Akhtar and Isman 2003; but see van Emden et al. 1996). No real resolution of such debates can take place until 3 O N E Chemical Mediation of Host-Plant Specialization: The Papilionid Paradigm MAY R. BERENBAUM AND PAUL P. FEENY the biochemical and genetic bases of preference and performance are elucidated for a broad array of species. How Lepidopterans Prefer A priori prediction of which chemicals influence oviposition preference and larval performance and whether linkages in chemical perception or processing between developmental stages exist is hampered by the fact that, within the Lepidoptera , caterpillars and adults experience very different chemical environments, with very different attendant behavioral and physiological consequences. Adult females, with relatively few exceptions, search for host plants while flying; complementing visual cues for identifying...

Share