Global distribution patterns of mycoheterotrophy

Aim Mycoheterotrophy is a mode of life where plants cheat the mycorrhizal symbiosis, receiving carbon via their fungal partners. Despite being widespread, mycoheterotrophic plants are locally rare, hampering the understanding of their global environmental drivers. Here, we explore global environmental preferences of mycoheterotrophy, and investigate environmental drivers of differential habitat preferences of mycoheterotrophic plants associated with arbuscular (AM) and ectomycorrhizal (EM) fungi. Location Global. Time period Current. Major taxa studied Mycoheterotrophic flowering plants. Methods We compiled the largest global dataset of epiparasitic mycoheterotrophic plant species occurrences and examined which environmental factors, including soil type, climate, vegetation type and distribution patterns of mycorrhizal autotrophic plants, relate to occurrence patterns of mycoheterotrophic plant species associated with AM and EM fungi. Results Mycoheterotrophic plant species avoid cold and highly seasonal climates and show a strong preference for forests. AM-associated mycoheterotrophs are predominantly found in broadleaved tropical evergreen forests whereas EM-associated mycoheterotrophs occur in temperate regions, mostly in broadleaved deciduous and evergreen needleleaved forests. The abundance of AM and EM autotrophic plants was a weaker predictor for mycoheterotrophs occurrences than forest type. Temperature and precipitation variables - but not edaphic factors - were the best predictors explaining the distribution patterns of mycoheterotrophs after accounting for the effects of forest type. For individual lineages, major differences in environmental preferences (often related to edaphic factors) occurred that were significantly associated with plant evolutionary relationships, indicating that these cheater plants have limited adaptive capabilities. Main conclusions The strong global geographical segregation of AM and EM mycoheterotrophs does not reflect the abundance of their potential autotrophic hosts, but seems to be driven by differential climate and habitat preferences. Our results highlight the non-trivial nature of mycorrhizal interactions, and indicate that identity of the partners is not enough to understand the underlying mechanisms promoting plant-fungal interactions in mycoheterotrophic plants.