4.2.2. Where to sample?

The difficulty with adult bees is that the virus titre can be influenced by the age and the tasks of the bee. It is therefore advisable to sample as much as possible the same age/task group throughout the experiment, in order to minimize the influence of such effects on the data (Van der Steen et al., 2012). In practical terms, this means either sampling from the brood chamber (mostly young nurse bees), the honey supers (medium-age bees) or at the entrance (older foraging bees). The choice of age-class for sampling is less important than consistently sampling the same age-class. See also the BEEBOOK paper on statistical methods (Pirk et al., 2013).