Relyea, R. A. 2001. The relationship between predation risk and anti-predator responses in larval anurans. Ecology 82:541-554.

Organisms that produce alternative, nondiscrete phenotypes in response to environmental conditions are expected to alter their phenotypes in relation to the degree of environmental change. This idea has been applied to the evolution of antipredator responses by prey, in which it has been hypothesized that prey should respond more strongly to predators that pose greater mortality risk. In a companion paper, I quantified predatorinduced behavioral and morphological responses in six species of larval anurans across five different predator environments and found that these responses were prey- and predatorspecific. In the present study, I addressed whether the responses were related to the level of predation risk posed by each of the predators. Within each prey species, I found that different predators posed different levels of predation risk; within each predator species, different prey species experienced different levels of risk. The differences in predation risk could be understood mechanistically after I quantified differences among predators in their ability to capture, handle, and consume prey and differences among prey in behavior and morphology. Using multivariate analyses, I found that predation risk had no significant effect on how a given prey responds to predators, although there were significant univariate behavioral effects; higher predation risk was related to greater decreases in activity and greater spatial avoidance. I also examined the relationship between risk and response across the six prey species within a predator treatment and found that higher predation risk across species leads to greater decreases in activity in the presence of Umbra and greater increases in tail depth in the presence of Anax. Thus, while previous studies have found relationships between predation risk and prey response when focusing on relatively few species, few predators, and a single trait, this more powerful test using 30 predator–prey combinations and nine traits suggests that the relationship is not well supported. This finding arises from the fact that larval anurans, as well as many other taxa, exhibit predator- and prey-specific behavioral and morphological changes in response to predator- and prey-specific risk.

能够响应环境条件产生可变、非离散表型的生物，其表型变化应与环境变化程度相关。该思路已被应用于猎物反捕食响应的演化研究，其中提出假说：猎物对致死风险更高的捕食者应表现出更强的响应。在一篇配套论文中，我针对5种不同捕食者环境下的6种无尾目幼体（larval anurans）的捕食者诱导型行为与形态响应进行了量化分析，结果发现这些响应具有猎物特异性与捕食者特异性。本研究旨在探究这些响应是否与各捕食者所带来的捕食风险水平相关。分析结果显示：在同一猎物物种中，不同捕食者带来的捕食风险水平存在差异；在同一捕食者物种中，不同猎物物种所承受的风险水平亦不相同。后续我通过量化各捕食者在捕获、处理与取食猎物方面的能力差异，以及各猎物在行为与形态上的差异，从机制层面阐释了捕食风险差异的成因。通过多变量分析，我发现捕食风险对特定猎物的捕食响应并无显著影响，但存在显著的单变量行为效应：更高的捕食风险与猎物活动水平的更大幅度下降以及更强的空间躲避行为相关。此外，我还针对单一捕食者处理组中的6种猎物物种，分析了风险与响应之间的关联，结果发现：在泥狗鱼属（Umbra）捕食者存在时，跨物种的更高捕食风险会导致猎物活动水平更大幅度下降；而在巨蜻属（Anax）捕食者存在时，跨物种的更高捕食风险会使猎物尾鳍深度出现更大幅度的增加。因此，尽管既往研究在仅关注少量物种、少数捕食者以及单一性状时，发现了捕食风险与猎物响应之间的关联，但本研究通过涵盖30种捕食者-猎物组合与9个性状的更严谨检验表明，该关联并未得到有力支持。这一结论的成因在于：无尾目幼体以及诸多其他类群，会针对捕食者特异性与猎物特异性的风险，表现出具有捕食者特异性与猎物特异性的行为与形态变化。