1985). In the earwig trees, crawling insects were excluded with
the same sticky barrier as in the banded trees but earwigs were
added to the canopy every 1–2 weeks. As the aim was to achieve
the same number of earwigs in the earwig trees as in the control
trees, the number of individuals in both was recorded beforehand
(see below) to estimate the number of earwigs that needed to be
added to the earwig trees. The added earwigs were collected from
non-experimental citrus trees on the same day that earwig abundance
was recorded and promptly added to earwig trees.
The treatments were established in March 2009 and in February
2010. The crawling insects could reach the canopy via tall weeds,
overlapping branches between adjacent trees, branches in contact
with the ground, or by crossing the sticky barrier using dead insects
as stepping stones. To avoid so, we inspected the trees on a
bi-weekly basis and, if any problem was detected, it was immediately
solved by trimming tall weeds, pruning branches or by
adding glue to the sticky barrier.
2.2. Earwig abundance
An artificial refuge was placed on two branches of each sampled
tree. These refuges, similar to those described by Suckling et al.
(2006), consisted of plastic shell vials (7.6 cm length; 1.2 cm diameter)
covered with garden hosing with caps on one side to create
darkness. The refuges were tied to branches with tape and checked
every 1–2 weeks (from April to October) for earwig abundance.
Although these refuges do not contain all the earwigs present in
the canopy, they shelter most of them (likely most of 75%; Piñol
and Espadaler, unpublished data).