4.2. Measuring colony infestation rate

Three methods to estimate colony infestation have been designed (Ritter, 1981; De Jong et al., 1982). Acaricides can be used to kill all mites in a colony. Mites will fall to the bottom of the hive and can be counted (Branco et al., 2006). Without the use of acaricides, the natural mortality can be quantified from the bottom of the hive to determine the population size of the live mites. Alternatively, the infestation rates of adults and brood can be estimated from adult and brood samples. When the first two methods are used, ants must be prevented access to bottom boards. Their scavenging habit will result in the disappearance of dead mites before they can be counted and will thus bias the results (Dainat et al., 2011). Such a protection can be obtained by preventing access to the whole hive or to the bottom board. Hive protection can be achieved by using a stand with feet smeared with grease or resting in containers containing a liquid over which ants cannot walk (water or oil). Here it is important to regularly verify that dirt does not accumulate in the container or on the grease, allowing ants to reach the hive. Blades of grass can also form bridges and should be cut in the surrounding of the hives. Alternatively, the varroa board itself can be protected against ants. This is achieved by covering the board with sticky material (e.g. Vaseline, glue, absorbent paper impregnated with vegetable oil). Such ‘sticky boards’ can be purchased or homemade.

All three methods (using acaricides, monitoring natural mite fall and assessing infestation levels) were found to provide comparable results (Branco et al., 2006). For the adult infestation rate estimate, the sample size in relation to the level of precision required by the experimenter has been determined by Lee et al. (2010a). Their study provides methods with different workloads permitting to achieve several levels of precision. We present here the method with optimal time and effort investment ratio that is necessary to reach the precision necessary to researchers. Since researchers are mostly interested in the infestation rates of particular colonies rather than of whole apiaries, we do not describe the latter method here, but refer to Lee et al. (2010a) for the number of colonies to sample from in order to obtain a representative figure at apiary level.

The methods based on mite fall or on evaluating infestation rates from adult or brood samples are only reliable for colonies with medium to high infestation rate. The methods show imprecision when colonies have less than 3,000 brood cells, when the brood infestation rate is <2 % (unless very large samples are taken, see the BEEBOOK paper on statistics (Pirk et al., 2013)) or when the colony is collapsing (due to decreased amount of brood) (Branco et al., 2006; Lee et al

., 2010a). In these cases, the acaricide treatment can be used. Using synthetic acaricides to estimate parasite population size in the host is reliable provided a product with high efficiency (> 95 %, taking possible resistance by the mite into account) is used. However, it is destructive and can only be used for a quantification /diagnostic purpose. The mites being killed by the treatment and the hive being contaminated with acaricide residues, the treated colony cannot be used as source or host of mites.

4.2.5. Sub-sampling mites to count on a bottom board