ABSTRACT
The species-area relationship (SAR) and its mechanisms regarding microbes are not as clear as those of plants and animals; this may result from the impact of sampling effects and the confusion between SAR and distance attenuation. We hypothesize that we can find more accurate microbial SAR curve, after removing these two factors. In this study, 27 leaves of three horticultural plants were selected as island models, and microbial biodiversity assessment was done with HTS (high-throughput sequencing). The separate and small systems using leaves as islands allow us to conduct a comprehensive survey of the microbial biodiversity of the leaves, without disturbance from sampling effects and distance attenuation effects. Interestingly, we did not find microbial SAR in those 27 leaves (also not found in evergreen trees Magnolia grandiflora and Eriobotrya japonica), but we did find significant microbial SAR in deciduous tree Ficus altissima. No significant differences were found between the different trees at the alpha diversity level of microbial biodiversity, but quite different on beta diversity. The results of beta diversity partition showed that F. altissima had the highest similarity of the microbial community among the leaves compared to those of M. grandiflora and E. japonica. Since leaf genesis in deciduous plants is more simultaneous than in evergreen plants; the result suggested that inconsistent historical background of leaf islands may mask microbial SAR. Thus, intensive sampling and consistent historical background are important for understanding microbial SAR.
