Detecting small environmental differences: Risk-response curves for predator-induced behavior and morphology. 2008.

Most organisms possess traits that are sensitive to changes in the environment (i.e. plastic traits) which results in the expression of environmentally-induced polymorphisms. While most phenotypically plastic traits have traditionally been treated as threshold switches between induced and uninduced states, there is growing evidence that many traits can respond in a continuous fashion. In this experiment we exposed larval anurans (wood frog tadpoles, Rana sylvatica) to an increasing gradient of predation risk to determine how organisms respond to small environmental changes. We manipulated predation risk in two ways: by altering the amount of prey consumed by a constant number of predators (Dytiscus sp.) and by altering the number of predators that consume a constant amount of prey. We then quantified the expression of predator-induced behavior, morphology, and mass to determine the level of risk that induced each trait, the level of risk that induced the maximal phenotypic response for each trait, whether the different traits exhibited a plateauing response, and whether increasing risk via increasing predator number or via increasing prey consumption induced similar phenotypic changes. We found that all of the traits exhibited fine-tuned, graded responses and most of them exhibited a plateauing response with increased predation risk, suggesting either a limit to plasticity or the reflection of high costs of the defensive phenotype. For many traits, a large proportion of the maximum induction occurred at low levels of risk, suggesting that the chemical cues of predation are effective at extremely low concentrations. In contrast to earlier work, we found that behavioral and morphological responses to increased predator number were simply a response to increased total prey consumption. These results have important implications for models of plasticity evolution, models of optimal phenotypic design, expectations for how organisms respond to fine-grained changes (i.e. within generation) in their environment, and impacts on ecological communities via trait-mediated indirect effects.绝大多数生物都拥有对环境变化敏感的表型可塑性性状（plastic traits），这类性状可表达出环境诱导的多态性。尽管传统研究大多将表型可塑性性状视为诱导态与非诱导态之间的阈值开关，但越来越多的证据表明，诸多性状可通过连续的方式做出响应。本实验将无尾类幼体——林蛙蝌蚪（*Rana sylvatica*）——暴露于梯度递增的捕食风险环境中，以探究生物如何响应细微的环境变化。我们通过两种方式操控捕食风险：一是维持捕食者（龙虱属*Dytiscus* sp.）数量恒定，改变其捕食的猎物总量；二是维持猎物总量恒定，改变捕食者的种群数量。随后，我们对捕食者诱导产生的行为、形态及体重性状的表达量进行了量化，旨在明确：诱导各性状产生响应的风险阈值、各性状达到最大表型响应的风险阈值、不同性状是否呈现平台期响应，以及通过增加捕食者数量或提升猎物消耗量这两种方式提升风险时，是否会引发相似的表型变化。研究结果显示，所有受试性状均呈现出精细调控的分级响应模式，且多数性状随捕食风险升高呈现平台期响应，这暗示可塑性存在上限，或是防御表型存在较高的代谢成本。对于多数性状而言，其最大诱导响应的大部分都在低风险水平下即可达成，这表明捕食者的化学信号在极低浓度下即可发挥作用。与此前的研究结论不同，我们发现，针对捕食者数量增加所产生的行为与形态响应，本质上仅是对总猎物消耗量提升的响应。上述研究结果对可塑性演化模型、最优表型设计模型、生物如何响应环境细粒度变化（即代内变化）的理论预期，以及通过性状介导的间接效应对生态群落产生的影响等领域，均具有重要的参考价值。