Prey Capture by Carnivorous Plants Worldwide 1923-2007

Available phylogenetic data illustrate that in all carnivorous lineages, the ancestral trap type is a sticky, flypaper-type trap (Ellison and Gotelli, 2001). In the Caryophyllales, pitfall traps (Nepenthes) and snap traps (Dionaea and Aldrovanda) are derived relative to the sticky pads of Drosera. Similarly, in the Lamiales, the sticky-leaved Pinguicula is ancestral to Genlisea with its eel (or lobster-pot) traps and Utricularia with its vacuum traps. In the Ericales, the Sarraceniaceae with its pitfall traps are derived relative to Roridula, another species with flypaper traps. Muller et al. (2004) hypothesed that carnivorous genera with rapidly evolving genomes (Genlisea and Utricularia) have more predictable and frequent captures of prey than do genera with more slowly evolving genomes; by extension it could be hypothesized that in general, carnivorous plants with more complex traps should have more predictable and frequent captures of prey than do those with relatively simple traps. Increases in predictability and frequency of prey capture could be achieved by evolving more elaborate mechanisms for attracting prey, by specializing on particular types of prey, or, as Darwin suggested, by specializing on particular (large) sizes of prey. In all cases, one would expect that prey actually captured would not be a random sample of the available prey. Furthermore, when multiple species of carnivorous plants co-occur, one would predict, again following Darwin that interspecific competition would lead to specialization on particular kinds of prey. Because the traps of carnivorous plants accumulate identifiable remains of prey, analysis of trap contents can provide an aggregate record of the prey that have been successfully "sampled" by the plant. Such samples could be used to begin to test the hypothesis that carnivorous plant genera differ in prey composition and to look for evidence of specialization in prey capture. Over the past 80 years, numerous ecologists have gathered data on prey contents of carnivorous plants of a number of species and genera from around the world, but these data have never been summarized or synthesized; this summary and synthesis is accomplished in Ellison et al. (2008). The accompanying data are in this file.现有系统发育数据显示，所有食肉植物（carnivorous plant）支系的祖先捕虫器均为粘性粘蝇纸型捕虫器（flypaper-type trap）（Ellison与Gotelli，2001）。在石竹目（Caryophyllales）中，瓶状捕虫器（pitfall trap，对应猪笼草属（Nepenthes））与触发式捕虫器（snap trap，对应捕蝇草属（Dionaea）和貉藻属（Aldrovanda））均由茅膏菜属（Drosera）的粘性捕虫叶演化而来。同样在唇形目（Lamiales）中，具粘性叶的捕虫堇属（Pinguicula）是两类衍生类群的祖先：一类是具鳗型（或龙虾笼型）捕虫器的螺旋狸藻属（Genlisea），另一类是具真空捕虫器的狸藻属（Utricularia）。在杜鹃花目（Ericales）中，具瓶状捕虫器的瓶子草科（Sarraceniaceae）由另一种具粘蝇纸型捕虫器的捕虫树属（Roridula）演化而来。Muller等（2004）提出假说：基因组快速演化的食肉植物属类（如螺旋狸藻属与狸藻属），相较于基因组演化较慢的类群，能更稳定可预测且更频繁地捕获猎物；由此可进一步推导，总体而言，具更复杂捕虫器的食肉植物，其猎物捕获的可预测性与频率应高于那些捕虫器相对简单的类群。猎物捕获的可预测性与频率的提升，可通过演化出更精巧的猎物吸引机制、特化捕食特定类型的猎物，或如达尔文所提出的，特化捕食特定（大型）体型的猎物来实现。在所有情形中，均可预期实际捕获的猎物并非可用猎物群落的随机样本。此外，当多种食肉植物共存时，同样可依据达尔文的理论预测：种间竞争将推动类群对特定猎物类型产生特化。由于食肉植物的捕虫器会留存可识别的猎物残骸，对捕虫器内容物的分析可获取该植物成功“采样”到的猎物的综合记录。此类样本可用于初步检验“食肉植物属类在猎物组成上存在差异”的假说，并寻找猎物捕获特化的相关证据。过去80年间，众多生态学家已收集了全球多个物种与属类的食肉植物的捕虫器内容物数据，但这些数据从未被汇总或整合；Ellison等（2008）完成了此项汇总与整合工作。本文件所附带的数据即为此类数据。