Integrating more biological and ecological realism into studies of multitrophic interactions
Integrating more biological and ecological realism into studies of multitrophic interactions
复制标题
将更多的生物学和生态现实主义融入多营养相互作用的研究中
DOI：
--
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
J. Ellers
中科院分区：
文献类型：
--
作者：
J. Harvey;M. Malcicka;J. Ellers
植物与昆虫食草动物相互作用的研究为我们目前对群落生态学模式和过程的理解奠定了基础。然而，我们相信，一些最持久的假设已经在研究中得到检验，这些研究的结论可能是根据各种标准的缺陷得出的。在这份意见文件中，我们强调了我们认为犯了根本性错误的几个领域。以植物-草食动物-寄生生物相互作用的研究为例，我们重点关注五个主要问题，并为未来的研究提供了框架。首先，我们认为，这一理论大部分是基于使用在简单农业生态系统中生长的人工选择作物进行的研究，这些作物代表了在更复杂的自然系统中生长的自然选择植物的较差替代品。其次，我们表明非自然物种组合也经常用于检验多营养相互作用中的一系列假设。这些通常是基于易于获取和/或饲养或分子测序而不是基于生物学或生态现实性来选择的。第三，我们认为一些研究反映了对所研究物种的重要生物学和生态特征的了解不足。然后，我们建议植物与昆虫相互作用的研究需要一种基于生活史的方法，结合生态现实主义，以便更好地理解自然界的模式和过程。第四，我们认为许多研究使用生物体的单一基因型——对于昆虫来说尤其如此——忽略了局部效应，从而忽略了进化“热点”的概念（Thompson，2005）。虽然植物的不同种群/基因型在实证文献中更为突出，但昆虫中往往缺乏这一重要方面。最后，我们强调统计显着性并不一定转化为生态显着性，尽管这两个参数通常是相关的。本文的主要目的是强调多营养相互作用的实证文献中常见的重要缺失要素或错误。我们的批评并不是对他人工作的控诉（其中一些包括我们自己的工作！），而是指出我们需要谨慎解释许多发现。它还呼吁对那些经常“净化”生态以产生统计显着性的研究持更加批评的态度。
Studies of plant–insect herbivore interactions have provided foundations for our current understanding of patterns and processes in community ecology. However, we believe that some of the most enduring hypotheses have been tested in studies where the conclusions may be derived on the basis of flaws in various criteria. In this opinion paper, we highlight several areas in which we feel that fundamental errors have been made. Drawing upon studies of plant–herbivore–parasitoid interactions as examples, we focus on five main issues of concern and provide a framework for future studies. First, we argue that much of this theory has been based on studies using artificially selected crops growing in simple agroecosystems that represent poor proxies for naturally selected plants growing in more complex natural systems. Secondly, we show that unnatural species combinations are also often used to test a range of hypotheses in multitrophic interactions. These are often chosen on the basis of easy accessibility and/or rearing or molecular sequencing rather than on biological or ecological realism. Thirdly, we suggest that some studies reflect a poor understanding of important biological and ecological traits of the species studied. We then suggest that research on plant–insect interactions needs a life history-based approach incorporating ecological realism in order to better understand patterns and processes in nature. Fourthly, we argue that many studies use single genotypes of the organisms – this is especially true with the insects – which ignores local effects and thus the concept of evolutionary ‘hot spots’ (Thompson, 2005). Whilst different populations/genotypes of plants figure more prominently in the empirical literature, this important aspect in the insects is often lacking. Finally, we stress that statistical significance does not necessarily translate into ecological significance, although these two parameters are often linked. The main aim of the paper is to highlight important missing elements or mistakes that commonly arise in the empirical literature on multitrophic interactions. Our criticisms are not meant as an indictment of others’ work (some of this includes our own!) but to point out that we need to interpret many findings with caution. It is also a plea to be more critical of research that often ‘sanitises’ ecology in order to generate statistically significant