Department of Evolution and Ecology, One Shields Avenue, University of California, Davis, California 95616, USA.
Ecology. 2011 May;92(5):1073-82. doi: 10.1890/10-1239.1.
Three recent meta-analyses of protective plant-ant mutualisms report a surprisingly weak relationship between herbivore protection and measured demographic benefits to ant-plants, suggesting high tolerance for herbivory, substantial costs of ant-mediated defense, and/or benefits that are realized episodically rather than continuously. Experimental manipulations of protective ant-plant associations typically last for less than a year, yet virtually all specialized myrmecophytes are long-lived perennials for which the costs and benefits of maintaining ant symbionts could accrue at different rates over the host's lifetime. To complement long-term monitoring studies, we experimentally excluded each of four ant symbionts from their long-lived myrmecophyte host trees (Acacia drepanolobium) for 4.5 years. Ant species varied in their effectiveness against herbivores and in their effects on intermediate-term growth and reproduction, but the level of herbivore protection provided was a poor predictor of the net impact they had on host trees. Removal of the three Crematogaster species resulted in cumulative gains in host tree growth and/or reproduction over the course of the experiment, despite the fact that two of those species significantly reduce chronic herbivore damage. In contrast, although T. penzigi is a relatively poor defender, the low cost of maintaining this ant symbiont apparently eliminated negative impacts on overall tree growth and reproduction, resulting in enhanced allocation to new branch growth by the final census. Acacia drepanolobium is evidently highly tolerant of herbivory by insects and small browsers, and the costs of maintaining Crematogaster colonies exceeded the benefits received during the study. No experimental trees were killed by elephants, but elephant damage was uniquely associated with reduced tree growth, and at least one ant species (C. mimosae) strongly deterred elephant browsing. We hypothesize that rare but catastrophic damage by elephants may be more important than chronic herbivory in maintaining the costly myrmecophyte habit in this system.
三篇近期关于植物与蚂蚁共生关系的元分析报告表明，植物与蚂蚁共生的保护作用与对蚂蚁植物的实测种群增长效益之间的关系出人意料地微弱，这表明蚂蚁植物对食草动物具有高耐受性、蚂蚁介导的防御具有显著成本，以及/或者防御效益是间歇性而非连续性的。对保护型蚂蚁-植物共生关系的实验操纵通常持续不到一年，但实际上所有的专性蚁栖植物都是长寿的多年生植物，对于这些植物，维持蚂蚁共生体的成本和效益可能会在宿主的一生中以不同的速度累积。为了补充长期监测研究，我们用 4.5 年的时间从其长寿的蚁栖植物宿主树（金合欢属的 A. drepanolobium）中实验性地排除了四种蚂蚁共生体中的每一种。蚂蚁物种在防治食草动物的效果以及对中期生长和繁殖的影响方面存在差异，但它们提供的食草动物保护水平并不能很好地预测它们对宿主树的净影响。尽管其中两种蚂蚁物种显著减少了慢性食草动物的损害，但移除三种 Crematogaster 物种在实验过程中导致宿主树的生长和/或繁殖呈累积性增长。相比之下，尽管 T. penzigi 是一种相对较差的防御者，但维持这种蚂蚁共生体的成本很低，显然消除了对树木整体生长和繁殖的负面影响，导致在最终普查时新分支生长的分配增加。金合欢属的 A. drepanolobium 显然对昆虫和小型食草动物的食草性具有很高的耐受性，维持 Crematogaster 蚁群的成本超过了研究期间获得的收益。没有实验树被大象杀死，但大象的损害与树木生长减少独特相关，至少有一种蚂蚁物种（C. mimosae）强烈阻止大象食草。我们假设，大象偶尔发生但灾难性的损害可能比慢性食草性对维持该系统中昂贵的蚁栖植物习性更为重要。
