Soil lichens have species-specific effects on the seedling emergence of three gypsophile plant species

Despite advancements in our knowledge on the importance of biological soil crusts (BSC) in arid and semiarid environments, little is known on the effects of specific crust organisms on vascular plant seed emergence. We conducted a growth chamber experiment to evaluate the effects of two BSC-forming lichens (Squamarina lentigera and Diploschistes diacapsis) on the emergence of three specialist vascular plants from gypsum Mediterranean habitats (Campanula fastigiata, Reseda stricta and Helianthemum squamatum). The design of the experiment included three disturbance levels (intact, disturbed and removed) and two sterilization levels (live vs. autoclaved) of BSC-forming lichens. Specifically, we evaluated three hypotheses: (i) lichens will reduce the emergence of vascular plants, (ii) the magnitude of the negative effect of lichens on emergence will be species-specific, and (iii) the two species of lichens evaluated will exert differential effects on seed emergence. Emergence in the presence of lichens was lower than that obtained in tests conducted under optimum lab conditions (reduction of 50%). The effects of crust disturbance and sterilization on seed emergence were species-specific, and also varied with the lichen species serving as substrate. When germinated with Squamarina, a disturbance×sterilization interaction was found in Campanula. Disturbance increased and reduced the emergence of this species when the lichen was dead and alive, respectively. When Helianthemum was sown on Squamarina and Diploschistes, disturbance reduced and increased, respectively, the emergence of this species. The emergence of Reseda was not affected by any treatment when sown on Diploschistes, but disturbance reduced it when shown on Squamarina. Our results highlight the importance of species-specific interactions between vascular plants and BSC-forming soil lichens. The potential mechanisms underlying these interactions are discussed.