• Plants use roots to access soil resources, so differences in root traits and their ecological consequences could be a mechanism of species coexistence and niche divergence. Current views of the evolution of root diversity are informed by large‐scale evolutionary analyses based on taxonomically coarse sampling and led to the ‘root trait phylogenetic conservatism hypothesis’. Here we test this hypothesised conservatism among closely related species, and whether root variation plays an ecological role.
• We collected root architectural traits for the species‐rich Cape rushes (Restionaceae) in the field and from herbaria. We used machine learning to interpolate missing data. Using model‐based clustering we classified root syndromes. We modelled the proportion of the syndromes along environmental gradients using assemblages and environmental data of 735 plots. We fitted trait evolutionary models to test for the conservatism hypothesis.
• We recognised five root syndromes. Responses to environmental gradients are syndrome specific and thus these represent ecomorphs. Trait evolutionary models reveal an evolutionary lability in these ecomorphs.
• This could present the mechanistic underpinning of the taxonomic radiation of this group which has been linked to repeated habitat shifts. Our results challenge the perspective of strong phylogenetic conservatism and root trait evolution may more generally drive diversification.

中文翻译：

---

超多样化海角植物丛中根部建筑生态形态的意外多样性和进化不稳定性。

• 植物利用根来获取土壤资源，因此根性状及其生态后果的差异可能是物种共存和生态位分化的机制。根系多样性进化的当前观点是通过基于分类学粗略抽样的大规模进化分析得出的，并导致了“根系特征系统发育保守性假说”。在这里，我们在密切相关的物种之间测试这种假设的保守性，以及根变异是否起生态作用。
• 我们从田间和草本植物中收集了物种丰富的开普敦草（Restionaceae）的根部建筑特征。我们使用机器学习来插值缺失的数据。使用基于模型的聚类，我们对根综合症进行了分类。我们使用735个样地的组合和环境数据对沿环境梯度的综合症比例进行了建模。我们拟合了特质进化模型，以检验保守主义的假设。
• 我们认识到五种根综合症。对环境梯度的响应是特定于综合征的，因此代表了生态形态。性状进化模型揭示了这些生态型的进化不稳定性。
• 这可能是该类生物分类辐射的机制基础，这与栖息地的反复移动有关。我们的结果挑战了强烈的系统进化保守性观点，根性状进化可能更普遍地推动了多样化。