Breeding honey bees to resist varroa and disease

Four things called 'resistance'

Most confusion comes from sliding down the ladder without saying so: a stock is selected on an assay score and sold as if the validated phenotype followed. Sometimes it does. The trials below show that it cannot be assumed.

The traits breeders select

• Hygienic behaviour — workers detect, uncap and remove dead or diseased brood. It is usually scored by freezing a patch of brood and measuring how much the colony has cleaned out a day or two later.
• Varroa-sensitive hygiene — a more specific form, in which workers remove pupae in cells where a mite is reproducing, interrupting the mite's breeding.
• Grooming — workers remove mites from their own bodies or from nestmates, sometimes damaging them.
• Suppressed mite reproduction — a lower proportion of mites in the colony producing offspring, which may result from the behaviours above.

The Asian honey bee, varroa's original host, does several of these routinely, and populations of the western honey bee that have survived untreated in several parts of the world show combinations of them. A 2020 systematic review brought together every study of populations bred or naturally selected for survival with the mite, and the genomic work trying to find the genes behind them.

What the trials show

A 2026 Canadian trial is the kind of evidence the ladder asks for. Colonies from stock selected for hygienic behaviour and varroa-sensitive hygiene — either by field tests or by antennal protein markers that predict the behaviour — were compared with unselected domestic stock and imported commercial stock, in three regions, with and without mite treatment. The selected stocks were at least as productive, kept varroa populations significantly lower at some places and times without treatment, and showed no biologically meaningful cost. Their larger autumn populations ate more food over winter, which the authors traced to colony size rather than to a cost of resistance.

A 2018 survey of 649 colonies in eastern Australia is the useful counterweight. Seventeen per cent met the hygienic selection threshold, and colonies in a breeding programme were almost six times more likely to be highly hygienic — the selection was working on the trait. But hygiene did not predict whether a colony had chalkbrood, or how bad it was. A trait selected because it helps against some brood diseases did not help against this one, in that population.

And traits compete. A one-season study in Korea found that non-hygienic colonies with strong grooming, and colonies with intermediate traits, had lower mortality than hygienic colonies with weak grooming — though their mite counts did not differ significantly. One season in one country is not a verdict, but it is a reminder that breeding for one behaviour is a bet on that behaviour.