Many developing countries harbour large numbers of species that face little‐understood, gradual changes in their environment, including chronic anthropogenic disturbance (CAD), a high frequency but low intensity form of disturbance. These countries also lack the resources to study each species, so conservation practices have been generalized, assuming that complete taxonomic groups may be managed in the same ways. This approach could be justified if closely related species respond similarly to threats. We assessed if related species respond similarly to CAD, whether this affects phylogenetic diversity and if such patterns occur in different systems. Species’ identity and abundance was recorded in 59 sites differing in CAD intensity in a semi‐arid grassland and a tropical dry forest in Mexico. Disturbance response indices were calculated for each species. Nested analysis of variance coupled with null models were applied to determine if generalizations within taxa are justified and which taxonomic level explained the most variation in disturbance response among species. We obtained phylogenetic trees using molecular data for the grassland and published data for the tropical forest. Phylogenetic signal was measured with Pagel's λ. Community mean phylogenetic distance was regressed on CAD. Higher taxonomic levels explained more variation than expected by chance, indicating that related species respond similarly to CAD. However, only species in the same genus behaved similarly enough to make generalizations reliable. This is the result of an underlying, if modest, phylogenetic signal in CAD responses. Mean phylogenetic distance decreased with CAD in the grassland but not the tropical dry forest. This suggests that CAD was a stronger environmental filter in the grassland but weak in the tropical dry forest, where it was less intense. Policy implications. Species inform us how their relatives respond to chronic anthropogenic disturbance (CAD), supporting the idea that generalizations in management are possible. However, this procedure is seemingly reliable only within genera, which explain over 80% of the accumulated variance in species’ CAD responses, and not within orders or families. CAD may reduce phylogenetic diversity, perhaps leading to ecosystem function loss.

Background: Plant richness world-records have been reported to occur either in tropical rainforests, or, at smaller spatial scales, in chronically disturbed grasslands. The tight relationship between scale and record richness suggests that there is an unyielding limit to diversity. Hypotheses: If such limit exists, current records should hardly be beaten. Chronic disturbance in grasslands allows richness to approach the limit. Studied species: All vascular plants at the study site. Study site and years of study: A natural, semiarid grassland at Concepción Buenavista, Oaxaca, Mexico, in 2012 and 2014. Methods: At 21 sites we recorded species richness in randomly sampled 0.1 × 0.1 m quadrats, and measured chronic disturbance caused by different agents. We estimated the maximum richness at each site through maximum likelihood. At one site we searched for high richness quadrats. Results: The study site shares the current world record of plant richness at the 0.1 × 0.1 m scale with 25 species. The estimated maximum diversity at each site decreased with land degradation, but had a maximum at intermediate-high levels of livestock activity. Conclusions: Our results support the idea of a hard limit to species richness, and thus theories of plant coexistence that envisage such limit. Our study site shares attributes with other record-holding grasslands, such as severe resource limitations and chronic disturbance by ungulates, suggesting that these conditions promote high species richness. We hope that, by reporting this diversity record, we also help to change the notion that Mexican grasslands are secondary communities having a negligible biological value. Keywords: Chronic anthropogenic disturbance, semiarid grassland, species richness, coexistence, intermediate disturbance hypothesis. Un récord mundial de diversidad en un pastizal de Oaxaca, México ResumenAntecedentes: Los récords mundiales de riqueza de especies de plantas se han reportado en selvas perennifolias o, en escalas pequeñas, en pastizales con disturbio crónico. La relación estrecha entre escala y riqueza récord sugiere la existencia de un límite inviolable para la diversidad. Hipótesis: Si tal límite existe, los récords actuales difícilmente serán rotos. El disturbio crónico permite que la riqueza de los pastizales se acerque al límite. Especies en estudio: Las plantas vasculares en el sitio de estudio. Sitio de estudio y fechas: Un pastizal natural semiárido, Concepción Buenavista, Oaxaca, 2012 y 2014. Métodos: En 21 sitios registramos la riqueza de especies en cuadros de 0.1 × 0.1 m y medimos el disturbio crónico causado por distintos agentes. Estimamos la diversidad máxima de cada sitio por verosimilitud máxima. En un sitio buscamos los cuadros con mayor riqueza. Resultados: Nuestro sitio comparte el récord de riqueza en la escala de 0.1 × 0.1 m con 25 especies. La riqueza máxima estimada se redujo con la degradación del suelo, pero alcanzó un máximo en un disturbio medio-alto por ganado. Conclusiones: Nuestros resultados apoyan la idea de que e...

Plant-soil feedback (PSF) occurs when plants change the biota and physicochemical properties of the soil, and these changes affect future survival or growth of plants. PSF depends on several factors such as plant functional attributes (e.g., life cycle or photosynthetic metabolism) and the environment. PSF often turn positive under dry conditions because soil biota confers drought tolerance. Conspecifics and close relatives share pathogens and consume similar resources, exerting negative PSF on each other. These ideas have mostly been tested under controlled conditions, while field studies remain scarce. To reevaluate these findings in nature, we analyzed plant-soil feedbacks over a drought-stress gradient in a phosphorus-limited semiarid grassland. We planted seedlings of 17 species in plots where community composition had been monitored for six years. To determine PSF intensity, we measured how seedling longevity was affected by previous occupancy of conspecifics and heterospecifics. The previous occupancy-survival relationship (OSR) was used as a proxy for PSF. Evidence for OSRs was found in one-third of the species pairs, with inconclusive evidence for the rest suggesting weak feedbacks. This is in line with the expectation that PSFs in the field are weaker than under controlled conditions. As expected, positive PSFs were more frequent as drought stress increased. The strongest OSRs were caused in dry plots by C 4 perennial grasses, which had very positive OSRs on several C 3 annual forbs, but negative effects on each other. Well-documented differences between these two functional groups may explain this result: C 3 plants are more sensitive to drought, and thus may be favored by tolerance-conferring microbiota; in contrast, water-efficient C 4 perennial grasses compete for phosphorus strongly, perhaps driving strong negative PSFs between them. Finally, close relatives had more negative OSRs on each other than on distant relatives as expected, although only in dry plots. This pattern was mostly due to the negative effects of closely related C 4 grasses under dry conditions, and their positive effects on distantly related dicots. Our results highlight the importance of plant traits and of the environmental context in determining the direction and strength of PSFs under field conditions.

Aims Diversity in communities is determined by species’ ability to coexist with each other and to overcome environmental stress that may act as an environmental filter. Niche differentiation (ND) results in stronger intra- than interspecific competition and promotes coexistence. Because stress affects interactions, the strength of ND may change along stress gradients. A greater diversity of plant growth forms has been observed in stressful habitats, such as deserts and alpine regions, suggesting greater ND when stress is strong. We tested the hypothesis that niche differences and environmental filters become stronger with stress. Methods In a semiarid grassland in southern Mexico, we sowed six annual species in the field along a hydric stress gradient. Plants were grown alone (without interactions), with conspecific neighbors (intraspecific interactions) or with heterospecific neighbors (interspecific interactions). We analyzed how the ratio of intra- to interspecific competition changed along the gradient to assess how water availability determines the strength of ND. We also determined if hydric stress represented an environmental filter. Important Findings We observed stronger intra- than interspecific competition, especially where hydric stress was greater. Thus, we found ND in at least some portion of the gradient for all but one species. Some species were hindered by stress, but others were favored by it perhaps because it eliminates soil pathogens. Although strong ND was slightly more frequent with stress, our species sample was small and there were exceptions to the general pattern, so further research is needed to establish if this is a widespread phenomenon in nature.

Clustering of species with similar niches or traits occurs in communities, but the mechanisms behind this pattern are still unclear. In the emergent neutrality model, species with similar niches and competitive ability self‐organise into clusters. In the hidden‐niche model, unaccounted‐for niche differences stabilise coexistence within clusters. Finally, clustering may occur through alliances of species that facilitate each other. We tested these hypotheses using population‐growth models that consider interspecific interactions parameterised for 35 species using field data. We simulated the expected community dynamics under different species‐interaction scenarios. Interspecific competition was weaker within rather than between clusters, suggesting that differences in unmeasured niche axes stabilise coexistence within clusters. Direct facilitation did not drive clustering. In contrast, indirect facilitation seemingly promoted species alliances in clusters whose members suppressed common competitors in other clusters. Such alliances have been overlooked in the literature on clustering, but may arise easily when within cluster competition is weak.