Background Mycorrhizal strategies are very effective in enhancing plant acquisition of poorly-mobile nutrients, particularly phosphorus (P) from infertile soil. However, on very old and severely P-impoverished soils, a carboxylate-releasing and P-mobilising cluster-root strategy is more effective at acquiring this growthlimiting resource. Carboxylates are released during a period of only a few days from ephemeral cluster roots. Despite the cluster-root strategy being superior for P acquisition in such environments, these species coexist with a wide range of mycorrhizal species, raising questions about the mechanisms contributing to their coexistence. Scope We surmise that the coexistence of mycorrhizal and non-mycorrhizal strategies is primarily accounted for by a combination of belowground mechanisms, namely (i) facilitation of P acquisition by mycorrhizal plants from neighbouring cluster-rooted plants, and (ii) interactions between roots, pathogens and mycorrhizal fungi, which enhance the plants' defence against pathogens. Facilitation of nutrient acquisition by clusterrooted plants involves carboxylate exudation, making more P available for both themselves and their mycorrhizal neighbours. Belowground nutrient exchanges between carboxylate-exuding plants and mycorrhizal N 2 -fixing plants appear likely, but require further experimental testing to determine their nutritional and ecological relevance. Anatomical studies of roots of clusterrooted Proteaceae species show that they do not form a complete suberised exodermis. Conclusions The absence of an exodermis may well be important to rapidly release carboxylates, but likely lowers root structural defences against pathogens, Plant Soil (2018) particularly oomycetes. Conversely, roots of mycorrhizal plants may not be as effective at acquiring P when P availability is very low, but they are better defended against pathogens, and this superior defence likely involves mycorrhizal fungi. Taken together, we are beginning to understand how an exceptionally large number of plant species and P-acquisition strategies coexist on the most severely P-impoverished soils.

Summary The abundance of nitrogen (N)‐fixing plants in ecosystems where phosphorus (P) limits plant productivity poses a paradox because N fixation entails a high P cost. One explanation for this paradox is that the N‐fixing strategy allows greater root phosphatase activity to enhance P acquisition from organic sources, but evidence to support this contention is limited. We measured root phosphomonoesterase (PME) activity of 10 N‐fixing species, including rhizobial legumes and actinorhizal Allocasuarina species, and eight non‐N‐fixing species across a retrogressive soil chronosequence showing a clear shift from N to P limitation of plant growth and representing a strong natural gradient in P availability. Legumes showed greater root PME activity than non‐legumes, with the difference between these two groups increasing markedly as soil P availability declined. By contrast, root PME activity of actinorhizal species was always lower than that of co‐occurring legumes and not different from non‐N‐fixing plants. The difference in root PME activity between legumes and actinorhizal plants was not reflected in a greater or similar reliance on N fixation for N acquisition by actinorhizal species compared to co‐occurring legumes. Synthesis. Our results support the idea that N‐fixing legumes show high root phosphatase activity, especially at low soil P availability, but suggest that this is a phylogenetically conserved trait rather than one directly linked to their N‐fixation capacity.

Accepted ArticleThis article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Running headline: Soil-borne pathogens and plant diversity Accepted ArticleThis article is protected by copyright. All rights reserved. Summary1. Soil-borne pathogens can contribute to the maintenance of local plant diversity by reducing differences in competitive ability between co-occurring plant species. It has been hypothesised that efficient phosphorus (P) acquisition by plants in Pimpoverished ecosystems might trade-off against resistance to root pathogens. This could help explain high plant diversity in severely nutrient-impoverished ecosystems.However, empirical evidence of such a trade-off remains scarce.2. In hyperdiverse shrublands in south-western Australia, non-mycorrhizal cluster-rooted Proteaceae are very efficient at acquiring P. However, Proteaceae co-occur with many other plant species using other P-acquisition strategies, such as ectomycorrhizal (ECM) associations.3. In a glasshouse experiment, we grew Proteaceae and ECM plant species from hyperdiverse shrublands alone and in competition with each other, and in the presence or absence of native soil-borne pathogens (Phytophthora spp.). We hypothesised that native Phytophthora species are more detrimental to Proteaceae than co-occurring ECM plants, due to a trade-off between highly efficient P acquisition and pathogen defence, and that this equalises differences in competitive ability between these two plant groups.4. When seedlings were grown alone, biomass of non-mycorrhizal plants was reduced in the presence of Phytopthora, while ECM species were unaffected by this pathogen.When non-mycorrhizal and ECM species were planted together, ECM plants grew better in the presence of Phytophthora than in its absence, because Phytophthora reduced the growth of the non-mycorrhizal competitors.5. Growth of ECM plants was positively correlated with percent root colonisation by ECM fungi, but this was only significant when ECM plants were grown in the presence of Phytophthora.Synthesis. Our study shows that native soil-borne pathogens equalised differences in competitive ability between seedlings of contrasting nutrient-acquisition strategies, thus supporting the hypothesis proposing a trade-off between highly efficient P acquisition and resistance against root pathogens. We found that non-mycorrhizal cluster-rooted species may be the most efficient at acquiring the growth-limiting Accepted ArticleThis article is protected by copyright. All rights reserved.resource, but that co-occurring ECM species are better defended against root pathogens. Our results suggest that native soil-borne pathogens and ECM contribute to the maintenance of the plant hyperdiversity in severely P-impoverished ecosystems.

Fungi are highly diverse organisms, which provide multiple ecosystem services.However, compared with charismatic animals and plants, the distribution patterns and conservation needs of fungi have been little explored. Here, we examined endemicity patterns, global change vulnerability and conservation priority areas for functional groups of soil fungi based on six global surveys using a high-resolution, long-read metabarcoding approach. We found that the endemicity of all fungi and most functional groups peaks in tropical habitats, including Amazonia, Yucatan, West-Central Africa, Sri Lanka, and New Caledonia, with a negligible island effect compared with plants and animals. We also found that fungi are predominantly vulnerable to drought, heat and land-cover change, particularly in dry tropical regions with high human population density. Fungal conservation areas of highest priority include herbaceous wetlands, tropical forests, and woodlands. We stress that more attention should be focused on the conservation of fungi, especially root symbiotic arbuscular mycorrhizal and ectomycorrhizal fungi in tropical regions as well as unicellular early-diverging groups and macrofungi in general. Given the low overlap between the endemicity of fungi and macroorganisms, but high conservation needs in both groups, detailed analyses on distribution and conservation requirements are warranted for other microorganisms and soil organisms.

Fine root endophytes (FRE) were traditionally considered a morphotype of arbuscular mycorrhizal fungi (AMF), but recent genetic studies demonstrate that FRE belong within the subphylum Mucoromycotina, rather than in the subphylum Glomeromycotina with the AMF.These findings prompt enquiry into the fundamental ecology of FRE and AMF. We sampled FRE and AMF in roots of Trifolium subterraneum from 58 sites across temperate southern Australia. We investigated the environmental drivers of composition, richness, and colonisation of FRE and AMF by using structural equation modelling and canonical correspondence analyses. Root colonisation by FRE increased with increasing temperature and rainfall; but decreased with increasing phosphorus. Root colonisation by AMF increased with increasing soil organic carbon but decreased with increasing phosphorus. Richness of FRE decreased with increasing temperature and soil pH. Richness of AMF increased with increasing temperature and rainfall, but decreased with increasing soil aluminium and pH. Aluminium, soil pH, and rainfall were, in decreasing order, the strongest drivers of community composition of FRE; they were also important drivers of community composition of AMF, along with temperature, and in decreasing order: rainfall, aluminium, temperature, and soil pH. Thus FRE and AMF showed the same responses to some (e.g., soil P, soil pH) and different responses to other (e.g., temperature) key environmental factors. Overall, our data are evidence for niche differentiation among these co-occurring mycorrhizal associates.