In Ka¨ ttilstad and Bja¨ rka-Sa¨by I took samples of
wood mould from all hollow oaks possible, whereas in
Sankt Anna, trees were randomly selected. Sampling
was impossible from trees with entrances that were too
narrow or too high from the ground (56 m) or if the
wood mould surface was too deep to reach. I sampled 8
l of wood mould from each tree, and if less was available
the tree was not included in the study. If there was
more than one hollow in a tree, one was randomly
selected for sampling. In Ka¨ ttilstad, 94 out of a total of
281 oaks with wood mould were surveyed, in Bja¨ rka-
Sa¨by 42 out of 97 and in Sankt Anna 19 out of 96. The
sampling was carried out in May and July 1996 in Ka¨ ttilstad,
in August 2000 in Bja¨ rka-Sa¨by and in July 1997
and June 1998 in Sankt Anna.