Modelling selection, drift, dispersal and their interactions in the community assembly of Amazonian soil mites

Parasite-host network analysis provides insights into the evolution of two mistletoe lineages (Loranthaceae and Santalaceae)

Mistletoes are ecologically important parasitic plants, with > 1600 species from five lineages worldwide. Mistletoe lineages exhibit distinct patterns of species diversification and host specificity, however, the mechanisms underlying these differences are poorly understood. In this study, we analysed a comprehensive parasite-host network, including 280 host species from 60 families and 22 mistletoe species from two lineages (Santalaceae and Loranthaceae) in Xishuangbanna, located in a biodiversity hotspot of tropical Asia. We identified the factors that predict the infection strength of mistletoes. We also detected host specificity and the phylogenetic signal of mistletoes and their hosts. We found that this interaction network could be largely explained by a model based on the relative abundance of species. Host infection was positively correlated with diameter at breast height and tree coverage, but negatively correlated with wood density. Overall, closely related mistletoe species tend to interact more often with similar hosts. However, the two lineages showed a significantly different network pattern. Rates of host generality were higher in Loranthaceae than in Santalaceae, although neither lineage showed phylogenetic signal for host generality. This study demonstrates that the neutral interaction hypothesis provides suitable predictions of the mistletoe-host interaction network, and mistletoe species show significant phylogenetic signals for their hosts. Our findings also indicate that high species diversification in Loranthaceae may be explained by high rates of host generality and the evolutionary history shared by Loranthaceae species with diverse host plants in the tropics.

Hunger for sex: Abundant, heterogeneous resources select for sexual reproduction in the field

Major hypotheses on sex evolution predict that resource abundance and heterogeneity should either select for or against sexual reproduction. However, seldom have these predictions been explicitly tested in the field. Here, we investigated this question using soil oribatid mites, a diverse and abundant group of soil arthropods whose local communities can be dominated by either sexual or asexual species. First, we refined theoretical predictions by addressing how the effects of resource abundance, heterogeneity and abiotic conditions could modify each other. Then, we estimated the strength of selection for sexual species in local communities while controlling for phylogeny and neutral processes (ecological drift and dispersal), and tested its relation to resource and abiotic gradients. We show that sexual species tended to be favoured with increasing litter amount, a measure of basal resource abundance. Further, there was some evidence that this response occurred mainly under higher tree species richness, a measure of basal resource heterogeneity. This response to resources is unlikely to reflect niche partitioning between reproductive modes, as sexual and asexual species overlapped in trophic niche according to a comparative analysis using literature data on stable isotope ratios. Rather, these findings are consistent with the hypothesis that sex facilitates adaptation by breaking unfavourable genetic associations, an advantage that should increase with effective population size when many loci are under selection and, thus, with resource abundance.

Dispersal patterns of oribatid mites across habitats and seasons

Oribatid mites are tiny arthropods that are common in all soils of the world; however, they also occur in microhabitats above the soil such as lichens, mosses, on the bark of trees and in suspended soils. For understanding oribatid mite community structure, it is important to know whether they are dispersal limited. The aim of this study was to investigate the importance of oribatid mite dispersal using Malaise traps to exclude sole passive wind-dispersal. Oribatid mite communities were collected over a 3-year period from five habitat types (coniferous forests, deciduous forests, mixed forests, meadows, bog/heathlands sites) and three seasons (spring, summer, autumn) in Sweden. Mites entered traps either by walking or by phoresy, i.e., by being attached to flying insects. We hypothesized (1) that oribatid mite communities in the traps differ between habitats, indicating habitat-limited dispersal, and (2) that oribatid mite communities differ among seasons suggesting that dispersal varies due to changing environmental conditions such as moisture or resource availability. The majority of the collected species were not typically soil-living species but rather from habitats such as trees, lichens and mosses (e.g., Carabodes labyrinthicus, Cymbaeremaeus cymba, Diapterobates humeralis and Phauloppia lucorum) indicating that walking into the traps or entering them via phoresy are of greater importance for aboveground than for soil-living species. Overall, oribatid mite communities collected in the traps likely originated from the surrounding local habitat suggesting that long distance dispersal of oribatid mites is scarce. Significant differences among seasons indicate higher dispersal during warm and dry periods of the year. Notably, 16 species of oribatid mites collected in our study were sampled for the first time in Sweden. This study also demonstrates that Malaise traps are a meaningful tool to investigate spatial and temporal patterns of oribatid mite communities.