Aim This study aimed to clarify species boundaries as well as phylogenetic relationships among species of the Primulina eburnea complex, a rapidly diversifying group of edaphic specialists confined to karst habitat islands in southern China. We investigated patterns of genetic isolation and diversification in this complex. Location Limestone karsts in southern China. Methods We analysed data from three chloroplast DNA (cpDNA) regions and eight nuclear microsatellite loci, representing five closely related species in the P. eburnea complex sampled from 62 populations across its entire distributional range. We conducted a Bayesian phylogenetic analysis, constructed maximum parsimony haplotype networks and assessed population genetic diversity, population differentiation and geographical structure within and between species. Results Of the 60 identified haplotypes, eight were shared between populations and only one was shared between species. Both cpDNA and the nuclear markers revealed a high level of population genetic differentiation and a strong phylogeographical structure. Four main genetic clusters were identified according to their geographical proximity, although geographical structuring differed slightly between the cpDNA and the nuclear loci. The pattern of genetic structure of both the cpDNA and microsatellite genomes can be explained by patterns of isolation by distance (IBD), across both the distribution of the complex and the populations of individual species. Main conclusions This is the first study to investigate geographical isolation and speciation in a clade associated with karst habitat islands in southern China. Geographically structured population differentiation of the P. eburnea complex suggests that divergence is mainly driven by genetic drift, with little evidence of gene flow. Allopatric speciation is the main mode of diversification in the group. These results highlight the importance of geographical isolation in promoting population differentiation in karst habitat islands.

The genus Primulina is an emerging model system in studying the drivers and mechanisms of species diversification, for its high species richness and endemism, together with high degree of habitat specialization. In this study, we sequenced transcriptomes for eleven Primulina species across the phylogeny of the genus using the Illumina HiSeq 2000 platform. A total of 336 million clean reads were processed into 355 573 unigenes with a mean length of 1336 bp and an N50 value of 2191 bp after pooling and reassembling twelve individual pre-assembled unigene sets. Of these unigenes, 249 973 (70%) were successfully annotated and 256 601 (72%) were identified as coding sequences (CDSs). We identified a total of 38 279 simple sequence repeats (SSRs) and 367 123 single nucleotide polymorphisms (SNPs). Marker validation assay revealed that 354 (27.3%) of the 1296 SSR and 795 (39.6%) of the 2008 SNP loci showed successful genotyping performance and exhibited expected polymorphism profiles. We screened 834 putative single-copy nuclear genes and proved their high effectiveness in phylogeny construction and estimation of ancestral population parameters. We identified a total of 85 candidate orthologs under positive selection for 46 of the 66 species pairs. This study provided an efficient application of RNA-seq in development of genomic resources for a group of 'stone plants' from south China Karst regions, a biodiversity hot spot of the World. The assembled unigenes with annotations and the massive gene-associated molecular markers would help guide further molecular systematic, population genetic and ecological genomics studies in Primulina and its relatives.

The mutualism between fig plants and fig wasps has been recognized as one of the most specialized systems of symbiosis. Figs are pollinated by their highly specific pollinating fig wasps, and the pollinating fig wasps are raised within the syconia of figs.Recent studies indicated a difference between monoecious and dioecious figs in the dispersal range of pollinating wasps, which has potential consequences for gene flow. In this study, we detected the gene-flow pattern of the dioecious climbing fig, Ficus pumila L. var. pumila, at both local and regional scales. At the local scale, spatial autocorrelation analysis indicated strong genetic structure at short distances, a pattern of limited gene flow. This result was also supported by a high inbreeding coefficient (F IS = 0.287) and significant substructuring (F ST = 0.060; P \ 0.001). Further analysis indicated that the effective gene dispersal range was 1,211 m, and the relative contribution of seed dispersal was smaller than that of pollen dispersal. The inferred effective range of pollen dispersal ranged from 989 to 1,712 m, while the effective seed dispersal range was less than 989 m. Lack of longdistance dispersal agents may explain the limited seed dispersal. The high density of receptive fig trees was the most likely explanation for limited pollen dispersal, and the position of syconia and relatively low wind speed beneath the canopy may contribute to this phenomenon. At the regional scale, significant negative correlations (kinship coefficient F ij ranging from -0.038 to -0.071) existed in all comparisons between the studied population and other populations, and the assignment test grouped almost all individuals of the studied population into a distinct cluster. Asynchronous flowering on the regional scale, which provides a barrier for the pollinating wasps to fly from the studied population to the other populations, is probably responsible for the limited gene flow on the regional scale.

No abstract

Limestone karsts in southern China are characterized by high edaphic and topographic heterogeneity and host high levels of species richness and endemism. However, the evolutionary mechanisms for generating such biodiversity remain poorly understood. Here, we performed species delimitation, population genetic analyses, simulations of gene flow and analyses of floral morphological traits to infer the geographic history of speciation in a species complex of Primulina eburnea from limestone karsts of southern China. Using Bayesian species delimitation, we determined that there are seven distinct species that correspond well to the putative morphological species. Species tree reconstruction, Structure and Neighbor‐Net analyses all recovered four lineages in agreement with currently species geographic boundaries. High levels of genetic differentiation were observed both within and among species. Isolation–migration coalescent analysis provides evidence for significant but low gene flow among species. Approximate Bayesian computation (ABC) analysis supports a scenario of historical gene flow rather than recent contemporary gene flow for most species divergences. Finally, we found no evidence of divergent selection contributing to population differentiation of a suite of flower traits. These results support the prevalence of allopatric speciation and highlight the role of geographic isolation in the diversification process. At small geographic scales, limited hybridization occurred in the past between proximate populations but did not eliminate species boundaries. We conclude that limited gene flow might have been the predominant evolutionary force in promoting population differentiation and speciation.

Genome sizes vary widely across thetree of life and the evolutionary mechanism underlined remains largely unknown. Lynch and Conery (2003) proposed that evolution of genome complexity was driven mainly by nonadaptive stochastic forces and presented the observation that genome size was negatively correlated with effective population size (N e ) as a strong support for their hypothesis. Here, we analyzed the relation between N e and genome size for 10 diploid Oryza species that showed about fourfold genome size variation. Using sequences of more than 20 nuclear genes, we estimated N e for each species after correction for the effects of demography and heterogeneity of mutation rates among loci and species. Pairwise comparisons and correlation analyses did not detect a negative relationship between N e and genome size despite about 6.5-fold interspecies N e variation. By calculating phylogenetically independent contrasts (PICs) for N e , we repeated correlation analysis and did not find any correlation between N e and genome size. These observations suggest that the genome size variation in the Oryza species cannot be explained simply by the effect of effective population size. K E Y W O R D S :Effective population size, genome size, Oryza, polymorphism.

No abstract

No abstract