Antipredator Behavior Promotes Diversification of Feeding Strategies

Integrative and Comparative Biology Integrative and Comparative Biology, volume 52, number 1, pp. 53–63 doi:10.1093/icb/ics074 Society for Integrative and Comparative Biology SYMPOSIUM Antipredator Behavior Promotes Diversification of Feeding Strategies Cris C. Ledón-Rettig1 and David W. Pfennig Department of Biology, University of North Carolina, CB#3280, Coker Hall, Chapel Hill, NC 27599, USA 1 E-mail: Synopsis Animals often facultatively engage in less risky behavior when predators are present. Few studies, however, have investigated whether, or how, such predator-mediated behavior promotes diversification. Here, we ask whether tadpoles of the spadefoot toad Scaphiopus couchii have a diminished ability to utilize a potentially valuable resource–– anostracan fairy shrimp––because of behavioral responses to predation risk imposed by carnivorous tadpoles of the genus Spea. Observations of a congener of Sc. couchii that occurs in allopatry with Spea, coupled with an ancestral character state reconstruction, revealed that Sc. couchii’s ancestors likely consumed shrimp. By experimentally manipulating the presence of Spea carnivore-morph tadpoles in microcosms, we found that Sc. couchii reduce feeding and avoid areas where both Spea carnivores and shrimp occur. We hypothesize that the recurrent expression of such behaviors in sympatric populations of Sc. couchii led to the evolutionary fixation of a detritivorous feeding strategy, which is associated with a reduced risk of predation from Spea carnivores. Generally, predator-mediated behavior might play a key role in promoting diversification of feeding strategies. Introduction Predation is a ubiquitous—and potentially potent— agent of natural selection. Thus, most species have experienced prolonged and intense selection for adaptations that reduce the risk of being eaten (reviewed by Endler 1991). A common antipredation strategy among animals is to facultatively engage in less risky behavior when a predator is present (Skelly 1994; Peacor and Werner 2001). Although predator-mediated behavior has traditionally been regarded as having no long-term consequences, it can profoundly impact the population dynamics of prey (Werner and Peacor 2003; Schmitz et al. 2004; Miner et al. 2005; Preisser et al. 2005; Agrawal et al. 2007; Kishida et al. 2010) and even promote divergence between populations of prey (Edgell et al. 2009; Scoville and Pfrender 2010; Ingram et al. 2011). Specifically, traits associated with an antipredator behavior might diverge between populations experiencing different regimes of predators. In a population recurrently experiencing predation, traits associated with an antipredator response would be continually expressed and subject to the selective pressures of the predator environment. Consequently, traits associated with a predator-free environment would be subject to relaxed selection, which might result in the evolutionary loss of the ability to express such traits (reviewed by Lahti et al. 2009; Pfennig et al. 2010). Once this occurs, the formerly induced response is expressed constitutively and becomes ‘‘fixed’’ in the population (Edgell et al. 2009; Scoville and Pfrender 2010). Although populations experiencing high levels of predation might be expected to undergo such fixation, populations experiencing low levels would not (Scoville and Pfrender 2010). In this way, behavioral responses to predators (or any environmental cue) may actually precede, and even facilitate, genetically canalized change (Price et al. 2003; West-Eberhard 2003). Predator-mediated behavior might be especially important in promoting the diversification of feeding strategies, particularly when both predators and prey share common resources; i.e., when they belong Advanced Access publication May 17, 2012 ß The Author 2012. Published by Oxford University Press on behalf of the Society for Integrative and Comparative Biology. All rights reserved. For permissions please email: . From the symposium ‘‘The Impacts of Developmental Plasticity on Evolutionary Innovation and Diversification’’ presented at the annual meeting of the Society for Integrative and Comparative Biology, January 3–7, 2012 at Charleston, South Carolina. 54 tadpoles of Couch’s spadefoot toad, Scaphiopus couchii, which often co-occur with Spea tadpoles, generally avoid consuming shrimp, even if offered no alternative prey (Ledón-Rettig et al. 2008, 2009). Moreover, most Sc. couchii tadpoles grow poorly if limited to shrimp or to an otherwise highly proteinaceous diet (Buchholz and Hayes 2000; Ledón-Rettig et al. 2008, 2009). By contrast, as described below, another species of Scaphiopus that does not co-occur with Spea, Scaphiopus holbrookii, readily preys on shrimp and actually grows as well on shrimp as on detritus. Two types of observations from natural populations suggest that diminished shrimp-eating ability in Sc. couchii stems from selection imposed by Spea. First, Sc. couchii generally avoid breeding in the same shrimp-rich ponds inhabited by Spea (Cornejo 1985). Second, when they do breed in the same pond, Sc. couchii tadpoles generally remain in shallow water on the pond’s margin (D. Pfennig, personal observation). By contrast, the highest densities of shrimp and most carnivorous Spea tadpoles occur in deeper water at the pond’s center (Fig. 1) (Pomeroy 1981, 23; D. Pfennig, personal observation). These observations suggest that Spea might have actively excluded Sc. couchii from the shrimp resource. Spea tadpoles represent a real threat of predation to Sc. couchii (Pomeroy 1981; Cornejo 1985) and actually prefer Sc. couchii as prey over the tadpoles of other species (Pfennig 2000). We, therefore, specifically sought to test the hypothesis that a recurrent threat of predation by Spea caused Sc. couchii tadpoles to facultatively alter their behavior such that they indirectly avoided the shrimp resource and subsequently lost the ability to capitalize on this diet. We evaluated this hypothesis through an ancestral character state reconstruction and through a series of experiments. We began by using the reconstruction to determine whether Sc. couchii’s poor performance on shrimp (i.e., their avoidance of shrimp and their inability to assimilate or grow well on such a diet relative to one of detritus) (Ledón-Rettig et al. 2008, 2009; Buchholz and Hayes 2000) is evolutionarily derived. Next, using Sc. holbrookii (a congener of Sc. couchii that does not face predation from Spea), we experimentally evaluated whether avoidance of a diet of shrimp is derived in Scaphiopus. Finally, we performed an additional experiment to test whether predatory Spea influence Sc. couchii’s foraging decisions, such that they would likely not be able to access the shrimp resource. Our results suggest that avoidance of the shrimp diet is indeed derived in Scaphiopus and that the presence of Spea carnivores causes Sc. couchii tadpoles to avoid areas where both carnivores (...truncated)