Counting and tracking varroa

Count, because by eye you see it too late

Varroa becomes visible on the bees once there are already many of them. Most of the population sits in sealed brood, out of sight: by the time infestation shows on adult bees, the colony is usually already in trouble.

Counting simply replaces an impression with a comparable number — comparable with the hive next to it, and above all with itself a fortnight earlier.

The four methods, and what they measure

• The sticky board under an open mesh floor measures natural mite fall: the mites that drop on their own, over a given number of days. It is non-invasive and does not require opening the hive — but the exact number of days in place must be recorded, otherwise the figure means nothing.
• The alcohol wash uses a sample of bees, usually taken from a brood frame with the queen set aside. It is the most faithful way to estimate an infestation rate, and the only one that kills the sample.
• The icing sugar roll works on the same sampling principle without killing the bees. Gentler, slightly less faithful: some mites stay attached.
• The CO2 method anaesthetises the sample long enough to dislodge the mites; like the sugar roll, it returns the bees to the colony.

One point changes everything: the sticky board gives a daily mite fall, the other three give an infestation rate as a percentage. Those two quantities are not comparable with each other. Pick a method and stay with it for the season, otherwise your tracking mostly measures your changes of method.

Interpreting a count

A single count mainly serves as an alarm. It is the series that informs: two measurements two to three weeks apart tell you which way things are going, which no single figure can do.

Reference values depend on the season, colony size and region, and they are revised. Rather than copying them out here, remember the logic. In spring the colony is small and tolerance is low: a natural fall of a few mites per day is enough to justify intervening. In summer the colony is more populous and the same figure weighs less — but the mite population grows far faster.

One order of magnitude often cited as a line not to cross is around 2000 mites in a colony in total (source: apisavoir.ch); that is the reference behind the projection Apivox displays. For seasonal natural-fall thresholds, refer to the current version of the varroa control concept published by apiservice.

A last point for reading results: during the brood season, the population models commonly used assume the mite population doubles on the order of once a month. A figure that looks «moderate» today can be critical four weeks later — which is the whole point of a projection.

Counting when it matters

Three moments make a count genuinely useful, without depending on a regional calendar: before deciding on a treatment, to know whether it is needed; during or just after the treatment, to check that it worked; and going into winter, because the late-summer mite load decides the quality of the winter bees.

The post-treatment count is the one most often skipped and the most rewarding: a treatment can fail, and not knowing costs a colony.

What a count does not tell you

• It only speaks for the hive you measured. Colonies in the same apiary diverge: always measuring the same «test» hive tells you about that hive, not about the apiary.
• It does not predict reinvasion. A properly treated colony can reload from neighbouring colonies, inside or outside your apiary.
• It does not replace clinical observation: deformed wings, patchy brood and sudden depopulation are visible to the eye and must override a reassuring number.