1. Some plants respond to both neighbours and nutrient levels below‐ground independently, while others respond only to nutrients. These responses have important ecological implications, but debate about the appropriate control treatments used in experiments has led many to debate results.
2. When plants ignore neighbours, and only respond to nutrient levels this is sometimes called an ideal‐free distribution (IFD). When individual plants respond to neighbours by increasing root production to pre‐empt the uptake of neighbours, and do so at the expense of reproduction, this has become known as a tragedy of the commons (TOC). Here, I describe how the history of competition experiments has largely been a history of experiments that only control two of three critical variables: (a) total nutrients per plant; (b) soil nutrient concentration and (c) pot volume. These confounding effects are the source of debate about results.
3. In one study, split‐root pea plants were grown with and without a below‐ground neighbour at many pot volumes with equal total nutrients per‐plant. The researchers explicitly argued that plant growth responses would not be affected by the concentration of nutrients in soil, and therefore they explicitly confounded soil nutrient concentration with neighbour presence. Here, I reanalyse a subset of their data, controlling for soil nutrient concentration, but confounding pot volume with neighbour presence.
4. When I controlled for soil nutrient concentration, the presence of a below‐ground neighbour had no effects on plant size, or allocation, and the effects of volume also disappeared. The reanalysis suggests that the behaviour of peas in this study was consistent with an IFD model, and that the previously reported volume and neighbour responses may have been due to the confounding of nutrient concentration.
5. The results of the study in question depend very much on whether one does analyses that control for pot volume or whether one does analyses that control for soil nutrient concentration. Ultimately, these analyses lead to two different interpretations of the data. New experimental designs are emerging that fully control pot volume and nutrients across competition treatment, and I discuss possible paths forward in this ongoing debate.

中文翻译：

---

解释植物根系对养分，邻域和盆栽量的反应取决于研究人员的假设

1. 一些植物对邻居和地下养分水平均独立做出反应，而另一些植物仅对养分做出响应。这些反应具有重要的生态意义，但有关实验中使用的适当控制方法的争论已导致许多人辩论结果。
2. 当植物忽略邻居而仅对营养物水平做出反应时，这有时被称为无理想分布（IFD）。当个别植物通过增加根系产量来阻止邻居的吸收而对邻居做出反应，而这样做却以繁殖为代价时，这就是公地悲剧（TOC）。在这里，我描述了竞争实验的历史在很大程度上是只控制三个关键变量中的两个的实验历史：（a）每株植物的总养分；（b）土壤养分浓度和（c）盆容。这些令人困惑的影响是对结果争论的根源。
3. 在一项研究中，在许多盆中种植有或没有地下邻居的开根豌豆植物，每株植物的总养分相等。研究人员明确指出，植物生长反应不会受到土壤中养分浓度的影响，因此，他们明确地将土壤养分浓度与邻居的存在混淆了。在这里，我重新分析了一部分数据，控制了土壤养分浓度，但将盆栽体积与邻居的存在混淆了。
4. 当我控制土壤养分浓度时，地下邻居的存在对植物大小或分配没有影响，而体积的影响也消失了。重新分析表明，该研究中豌豆的行为与IFD模型一致，并且先前报道的数量和邻居反应可能是由于营养物浓度的混杂所致。
5. 所研究的结果在很大程度上取决于是否分析了控制盆栽量的方法，还是分析了土壤养分浓度的方法。最终，这些分析导致对数据的两种不同解释。新的实验设计正在出现，它们可以在整个比赛过程中完全控制锅的体积和养分，我将在这场持续的辩论中讨论可能的发展道路。