Principles of landscape ecology have been built on birds and plant species distribution, but the number of clues is now growing on below-ground organisms, whose dispersal may also be affected by above-ground landscape structure. For communities of microorganisms, the question remains if and how they answer to landscape structure, with or without time lag, and if some groups of microorganisms may react more than others. Here, we investigated if fungi or bacteria diversity is driven by the amount of forest cover in the current or the past landscape. We tested the Habitat Amount Hypothesis (HAH) on ancient forests of Cevennes national park, that were particularly fragmented 150 years ago, and are today surrounded by recent forests. As ancient forests are often more diverse in plant species, we hypothesized that the higher quantity of ancient forests in the landscape, the richer fungal and bacterial communities would be locally. More precisely, we expected that ectomycorrhizal fungi, and pathotrophic fungi, often indicators of mature forests, would be also more sensitive to forest history and therefore to the quantity of ancient forests than bacteria and saprotrophic fungi. We sampled 40 soil cores per 0.5 ha, pooled in 8 composite samples per plot in 27 landscapes and sequenced ITS and 16S marker by Illumina-Mi seq. To identify functional groups of fungi, we relied on their taxonomy and the use of public databases. Our results partly follow the HAH, as fungi richness was positively related with the quantity of ancient forests in the landscape and not by the focal patch size. Ectomycorrhizal and pathotrophic fungi were positively affected by the ancient forest cover, and so were saprotrophic ones, but not bacteria. Local factors also shaped the communities such as soil composition and elevation, confirming classical patterns in soil ecology. Interestingly, past landscape structure better explained fungi communities richness than contemporary landscape, suggesting a time lag in the response of communities to landscape modification and a potential extinction debt. Our results invite to consider below-ground communities in landscape studies and historical ecology, as their structure and functions might be intimately linked with soil and landscape history.

Soil samples were collected on April 2017 (average temperature 4.4 °C, average precipitation 17.6 mm from www infoclimat.fr). In each study site, soil cores were collected along two 60 m transect, spaced from 20 m : every 20 m a soil core was collected, and pooled with four soil cores collected at 3.5 m from the focal point in each cardinal direction. As a result, 40 soil cores pooled in 8 composite samples were collected per plot. The composite samples were then opened back to the lab the same day, and 15 g of soil was isolated in a clean tea-bag and dried with 50 g of silicagel (eventually more if the sample was not dry the day after). Soil DNA was extracted less than one month later with Macherey Nagel NucleoSpin® Soil kit (Ref 740780.250). Amplifications targeted ITS1 nuclear rDNA primers for fungi (Fwd: ITS5 GGAAGTAAAAGTCGTAACAAGG from Epp et al. 2012 and a modified version of Rev: 5.8S_Fungi CAAGAGATCCGTTGTTGAAAGTK, Taberlet et al. 2018), and 16S rDNA (V5-V6) for bacteria (Bact01 primers - Fwd: GGATTAGATACCCTGGTAGT and Rev: CACGACACGAGCTGACG (Fliegerova et al. 2014)). Each PCR was performed with a unique set of primers, characterized by a unique set of 8 nucleotide tags added to the universal primers, allowing to distinguish samples later in the analysis (as described in Taberlet et al. 2018). Negative controls without any PCR reactants were integrated (10% of our reactions) and all PCR reactions were replicated twice. PCR conditions followed Nagati et al. (2018). All PCR reactions were pooled to prepare Illumina-Miseq libraries, which were sent to GenoToul Core Facility platform for sequencing.