The first 30 aspens with a minimum inter-tree distance of 5 m within a transect were selected, working from the transect centerline and outwards to the edge (Fig. 2). Only trees that with certainty had been retained at final harvest, i.e. not such that possibly had regenerated after harvest, were selected. If two or more trees were at the same distance from the centerline, a dice determined Vemurafenib cost the selection. Diameter at

breast height and presence of all lichen species on the stem from the base and up to 2 m were recorded. The inventory was carried out in the summer and autumn of 2009. Taxa in the genera Caloplaca, Rinodina, and Xylographa were not determined to the species level and for 20 taxa the species identification was uncertain ( see Appendix). Species difficult to determine in the field were collected for identification under light microscope and with spot tests using chemical reagents and UV-light. All Bryoria species and all Lepraria species except L. lobificans and L. jackii were treated collectively. Small specimens of the genera Cladonia and Usnea were treated as Cladonia spp. and Usnea spp. respectively.

Micarea prasina might include M. byssacea and M. micrococca. Collema occultatum var. occultatum and C. occultatum var. populinum were treated as separate taxa since they differ in morphology, ecology and distribution. The nomenclature follows Santesson et al. (2004) except for Bacidia rosellizans that follows Ekman (2009), Casein kinase 1 Caloplaca pyracea that follow Arup

(2009), Biatora high throughput screening compounds globulosa and B. pallens that follows Printzen and Otte (2005) and the genus Stictis that follows Wedin et al. (2006). Information for each species regarding aspen-dependency (if a species’ main substrate is aspen, or in some cases aspen and Salix spp., the species was classified as aspen-dependent), dispersal mode, photobiont, growth form, and categorization as red-listed or signal species (indicator species for sites with high a conservation value; Nitare 2000) were recorded using Moberg and Holmåasen, 1982, Krog et al., 1994, Wirth, 1995, Foucard, 2001, Wedin et al., 2006 and Gärdenfors, 2010 and F. Jonsson’s expertise. A generalized linear mixed model (GLMM) was used to analyze differences in species richness between trees that had been exposed for 0–4 years and 10–16 years, with stand and tree as random factors. Stand was considered as a random factor since we assumed that trees on the same clearcut or in the same young forest were more similar. Tree was considered as a random factor to capture unexplained variation that caused over-dispersion. Using observation-level random effects is a recommended way to deal with overdispersed Poisson GLMM (Breslow, 1990).