For the Love of Bees and Beekeeping

September 2015

Evolution and Double-Queen Management

by Keith Delaplane


After a break in August to talk about CCD politics, I’m happy to pick up the thread again on our evolutionary history of the honey bee with a view toward clues to good management. In July’s installment we stumbled upon one of those clues. Readers will recall that we talked about the evolutionary unit of selection which I defined as “an entity at any level of biological hierarchy that can be acted upon by natural selection.” I quoted Richard Lewontin who states it more specifically that “any entities in nature that have variation, reproduction, and heritability may evolve,” and that the principle “can be applied equally to genes, organisms, populations, [or] species.1” As strange as it may sound, I think this points us toward the biological basis for an old beekeeping trick for increasing honey yields – the double-queen hive.

So how do we get from the evolutionary unit of selection to really big honey crops? To begin, it’s important to understand the unit of selection because it’s not always clear at what level natural selection is acting as we ascend the ladder of biological hierarchy. In the case of honey bee colonies, we can ask whether natural selection is acting at the level of the worker, the queen, or the whole colony. In the case of organisms such as ourselves, we can ask the same thing about our genes, our cells, or our whole bodies. The reason this is important is because if natural selection acts simultaneously at different levels then that sets the stage for evolutionary conflicts, even within the same colony or organism. An example of such a conflict in the case of the honey bee colony is the fight-to-the-death practiced by rival virgin queens2. This behavior is good for the winner but not for the colony which loses a valuable back-up queen. An example in organisms like you and me is cancer cells which rebel from the rest of the genome, evade the other cells’ defenses, and proliferate at the expense of the organism3. So whether it’s a selfish queen or a rogue cancer cell we see that the concept of a unit of selection helps us explain the existence of biological phenomena that are at odds with the whole to which they belong. The most famous reductionist interpretation of this is the “selfish gene” hypothesis made popular by Richard Dawkins who states that it is the gene that is the fundamental “replicator” in biology and that all subsequent levels of biological organization are machinery coopted by the selfish gene to propagate itself4. The gene doesn’t necessarily “care” about the machinery as long as the gene is replicated.

Understanding that it is units of selection upon which evolution acts all the way up the biological hierarchy helps us appreciate that the products of evolution at any level and point in time are not perfection but rather optimization with other units of selection. The trajectory of life has been for units of selection to group together into higher-level units of selection – from genes to chromosomes to cells to multicellular organisms to families to societies to superorganisms, each emerging level integrated and internally stable. This stability can only happen if the evolutionary conflicts in lower subsumed levels are suppressed or absent. This grouping of formerly independent units of selection into higher-level integrated units can be thought of as the evolution of individuality – each “individual” reproductively competent and contingent upon control of the internal conflicts in levels below it5. My June column covered one of the most important ways lower-level conflict is handled in the honey bee colony – coercing workers to forego egg laying and help their mother produce siblings.

Having said all this, I want to return to that example of lower-level conflict in honey bees that remains unresolved – rival sister queens fighting to the death. Now it is obvious that the colony is the loser in this situation because two queens is arguably better than one when it comes to colony size, defense, foraging strength, and genetic variation, to speak nothing of the safety net of having an extra queen. The ants and termites saw the sense in this long ago, and their mature colonies can harbor numerous queens and their respective progenies. If it is true that evolution is optimization not perfection, then we can confidently say that there is a real cost to the colony when queens fight to the death. I propose to my readers that we get insight into that cost when we observe the difference in honey yields between double-queen (DQ) and ordinary single-queen (SQ) colonies.

Double- (or even multiple-) queen management is nothing new and boils down to the manipulation of a colony to sustain more than one queen. The idea gained momentum after research by C.L. Farrar in the 1930s showed that honey yields increase disproportionately as a colony’s worker population increases6. In other words, one colony of 60,000 bees will produce more honey than the sum of two colonies each with only 30,000 bees. It’s no overstatement that ...