In a previous study of saproxylic beetle fauna on fresh artificially created high stumps of spruce and birch on 20 clear-cuts (whereof 10 in hotspots) we found only weak relations between beetle fauna and forest-landscape variables associated with hotspots for biodiversity. The stumps were distributed over a large part of south Sweden in an area covering 200 km from East to West and 50 km North to South. The result gave weak support for the strategy of concentrating measures to mitigate the decrease of dead-wood dependent insects to hotspots. We hypothesised that the relationships would be stronger when the wood was in a later successional stage, as hanbitat predictability increases with a longer turn-over time of beetle assemblages. I this study we therefore resampled the saproxylic beetle fauna on the same high stumps after ten years, and analysed the results together with the data from the previous study. This generated 51,390 saproxylic beetle specimens, belonging to 445 species. By analysing the species numbers and species composition with variance partitioning we found that the first years after the high stumps were created, the beetle fauna was mainly explained by regional factors (longitude and latitude) which explained species composition to 21%, 4% and 25% for all saproxylic, birch associated and spruce associated beetles respectively. At this time, forest-landscape variables describing the surrounding forest in buffers up to 5 km radius explained almost nothing. After ten years the pattern was the opposite, with surrounding forest variables explaining up to 27% of the variation, and longitude/latitude almost nothing. Our results suggest that hotspots exist also for species associated with trivial wood types and that successional stage of the dead wood might help explain at which scale saproxylic species respond. The result also suggests that species later in the succession are more indicative of where hotspots are situated than species early in the succession.