Sophie and the Walrus

An imaginary argument about breeding better bees when you can’t control the drones

Sophie is imaginary. She is a composite of arguments I have heard and read over the years, and she does not speak for BIBBA, although she may well be a member. The Walrus is, of course, me. Every figure either of them quotes comes from the sources listed at the end. The article that Sophie and Walrus refer to will be published after this on my blog, and it’s called “Breeding Better Bees When You Can’t Control the Drones”.

Round 1: What are we actually arguing about?

Sophie: I’ve read your article. You open by saying success comes down to the weather and the quality of the bees. Then you spend several thousand words ignoring the bees best suited to our weather.

Walrus: I’m not ignoring them. The bees I use are extremely well suited to our weather. My breeder queens are the product of many years of careful selection, either here in the UK or in Northern Europe and Scandinavia, where the weather is no kinder than ours. My goal is honey for my customers, produced efficiently and profitably. Oh, and the pleasure and peace that come from being a part of the lives of bees. I’m not trying to transform the local gene pool.

Sophie: That’s a very polite way of saying you’re not interested in the future of British bees.

Walrus: It’s an honest way of saying I have a business to run. Your aims and mine are not the same.

Sophie: Fine. But a bee adapted to our climate copes better with our weather. That’s the whole point of local adaptation.

Walrus: Then we agree because mine are adapted to it. Decades of selection through cold springs, wet summers and long winters is adaptation, which is what breeders have done to produce my breeder queens. By almost every measure I can think of, they beat the local bees here.

Sophie: Adapted to a climate isn’t the same as adapted to a place. Büchler’s trial compared strains from all over Europe, and in each region the local ones outlived the incomers. And “almost every measure” is a big claim. Where’s your evidence?

Walrus: My honey crop, calm bees and fewer swarms. You are welcome to visit and see for yourself.

Sophie: That’s still practitioner experience.

Walrus: So is yours. We’ll come to that later.

Sophie: Even so, your clever bees can’t make the sun shine.

Walrus: Nor can yours. No amount of local adaptation makes a colony massively productive during six weeks of rain or a July drought. Being locally adapted is not some magical shield against starvation during a prolonged dearth.

Sophie: Thrift in a bad summer is still worth something. A colony that doesn’t starve in the June gap is a colony you don’t have to feed.

Walrus: Agreed, it’s worth something. But feeding at the right time can make a huge difference; it’s part of beekeeping in my opinion.

Round 2: The man on the peninsula

Sophie: You hold up Jo Widdicombe as your example, and then you don’t follow what he does.

Walrus: I admire what Jo does. He’s on the Rame Peninsula, fairly isolated from other beekeepers, with plenty of near-native bees around. If I lived there, I might well do the same. He says, “Once you have a more uniform strain, it is easier to breed true”, and I don’t disagree.

Sophie: So why not build that uniform strain where you are?

Walrus: Because I’m on the Cheshire plain, surrounded by beekeepers keeping every kind of bee you can imagine. Jo has a more or less closed population and I don’t.

Sophie: And your evidence that the local bee is worse?

Walrus: Years of handling them. My average “local bee” is more defensive, drips off the comb, and wants to swarm even when given space.

Sophie: Your “local bee” in Cheshire is a mongrel. You’re comparing mongrels with bought-in breeder lines, not with locally adapted near-natives. Of course the mongrels lose.

Walrus: Indeed. The surrounding bees are mongrels, and a line selected for decades beats them hands down. I’d rather start from the best bees I can get than from whatever happens to be here, unless what’s here is special.

Sophie: Or you could start from the best of the locals and improve them.

Walrus: In an area with this much gene flow in and out, that would take years of lower honey crops to get to somewhere worse than where I am already.

Round 3: Purity, predictability and performance

Sophie: Your article separates purity, predictability, and performance.

Walrus: I was going to call them the three Ps, but I decided against it.

Sophie: Why?

Walrus: Urinals. Don’t ask.

Sophie: … The point is that the three are linked. Purity gives you predictability, and predictability makes performance repeatable.

Walrus: Purity is one route to predictability. Long selection inside a closed population is another. And performance is what all honey producers are looking for.

Sophie: You and Jo agree on selection, though. In his words, “by repeated selection, standards can be raised”.

Walrus: We do. Bad-tempered colonies shouldn’t put out drones near where you’re mating queens. Where we differ is that he also culls anything that doesn’t look native. I cull for temper, not colour.

Sophie: Colour is a quick field test.

Walrus: It’s a quick test of colour, nothing more. I’m interested in traits.

Round 4: Genetics, as explained by a walrus

Walrus: Can I do my genetics bit? I warn you, it’s like quantum physics to me. I’m fine up to a point, then my head explodes.

Sophie: Go on.

Walrus: Traits aren’t inherited, genes are. Gentleness depends on lots of genes plus the environment. With each new generation, the genes are reshuffled like a pack of cards, so the trait has to be rebuilt from a new hand. The additive part is passed on, but the lucky combinations get broken up. That’s why the daughters of a brilliant queen drift back towards the average.

Sophie: Which is precisely why you need a uniform population. If everyone in the pool carries similar variants, the reshuffle changes little. That’s what breeding true means.

Walrus: Agreed. And I can’t get a closed population in Cheshire.

Sophie: Queens shuffle more than most. Beye and colleagues found recombination ten times higher than in other higher eukaryotes.

Walrus: I had to look up eukaryotes.

Sophie: And drones don’t reshuffle at all. A drone’s sperm are identical copies of his single set of genes, a random half of his mother’s. So selecting drones really means selecting the colonies that produce them.

Walrus: Which is the bit I can’t really control, although I try to influence it.

Round 5: How far do queens fly?

Sophie: Your own source, Jensen in the Peak District, states that 90% of matings occur within 7.5 km, half within 2.5 km, and the furthest was 15 km. In the most isolated part of Edale, 60% of matings were to Edale drones.

Walrus: Meaning 40% weren’t. And each queen had an effective mating frequency of 17.2. Your neighbours’ bees are in the mix, whether you like it or not.

Sophie: That study was on black bees, not far from you, and its authors concluded the Hope Valley was suitable for long-term black bee conservation.

Walrus: Edale is a semi-isolated valley. I’m on an open plain. I treat those distances as a rough guide. It’s going to vary by location.

Sophie: You can’t have it both ways. If the valley is irrelevant, so are the distances you use to show that your neighbours’ drones get a big say.

Walrus: A rough guide is what I said. And apiary vicinity mating happens too, so, as usual with bees, there’s no rule that always applies.

Round 6: Who is to blame for the temper?

Sophie: Brascamp’s Austrian study looked at 14,650 colonies and showed that worker-effect heritability was larger than queen-effect heritability for every trait.

Walrus: So the old saying that temper comes “from the drone side” is only half right. Drones supply half the worker genes, and the queen supplies the other half.

Sophie: Half of the bigger effect. That makes the drones you can’t control rather important, doesn’t it?

Walrus: It does. That’s why my daughter queens are not all the same. Most are excellent, but maybe ten percent are snappier than I’d like.

Sophie: And your remedy?

Walrus: We rarely use smoke, but if they’re grumpy, we fire up the smoker and that almost always calms them down.

Sophie: That’s not breeding. That’s sedation.

Walrus: That’s beekeeping.

Round 7: Show me the numbers

Walrus: Here’s the breeder’s equation: response to selection equals heritability times the selection differential. The article goes into all that.

Sophie: Go on, impress me.

Walrus: Gentleness, on the Austrian 1 to 4 scale, has a combined heritability of 0.37. Say my apiary averages 3.2, and I choose breeders whose colonies score 3.8. If I control both sides, the daughters improve by 0.37 × 0.6, about 0.2, so they score roughly 3.4. If I control only the queen side, it’s half that, so roughly 3.3.

Sophie: That’s feeble.

Walrus: It adds up, year-on-year, though progress slows as the variation shrinks.

Sophie: And swarming? Your table shows 0.06 for the Austrians and 0.34 for the Italians. Which is it?

Walrus: The Austrian figure is a combined heritability. The worker-effect figure on its own was 0.36. The two sets of data are from different bees, using different statistical models, so compare within a column, not across a row.

Sophie: That’s convenient.

Walrus: It’s accurate. In my experience temper is easier to shift than swarming anyway, and swarm prevention done in good time matters more than any heritability figure. Although, I did try a few of Paul Horton’s bees once (bred by Ivan Nielsen) and they wouldn’t swarm, no matter how cramped for space they were. Crazy.

Round 8: Pepper-pot brood

Sophie: Continual selection narrows the gene pool. You admit you’ve seen pepper-pot brood.

Walrus: Once, in 14 years. If a queen mates with a relative carrying the same version of the sex-determining gene (csd), some of her fertilised eggs become diploid drones instead of workers, and the workers remove them.

Sophie: With seventeen or so mates, one related drone would barely show. You’d need very few matings for it to be visible.

Walrus: Exactly my guess, and it is only a guess. I think she mated close to home with very few drones, and one was a close relative. And naturally, it happened to be one I inspected while my local beekeeping association was visiting.

Sophie: The gods have a sense of humour. But if your drone colonies are headed by sisters from one breeder, and your virgins are their sisters too, you’re raising the odds yourself.

Walrus: Which is why I change the line of breeder queen every season.

Round 9: Experience versus evidence

Sophie: BIBBA’s committee statement says locally adapted bees “have long been known by experienced beekeepers to consistently perform better”.

Walrus: That’s practitioner experience rather than measured data.

Sophie: So is half your article.

Walrus: True. Practitioner experience counts for a lot in beekeeping, so you got me there.

Sophie: And there is measured data. Büchler and colleagues: 597 colonies, 16 genotypes, 20 apiaries across Europe. Colonies with local queens survived 83 days longer.

Walrus: They were not treated for varroa. Bee farmers treat and feed. It tells us little about a managed honey operation.

Sophie: It tells you local bees are more resilient when things go wrong.

Walrus: When things go wrong in my apiaries, I fix them. And I wouldn’t call 83 days long term.

Sophie: It’s a relative difference under stress, not a lifespan.

Walrus: Then show me the same trial with treated colonies, and we’ll see the difference, if any.

Round 10: Crosses and crossness

Sophie: BIBBA also says random mating creates “arbitrary crosses that often result in aggressive behaviour”. And Jo warns, “Crosses can give enhanced vigour in the first generation, but this is not maintained.”

Walrus: That F2 aggression applies to crosses between distinct types of bee (pure subspecies), such as a native queen mated to Italian drones. Near me most bees are already mongrels. Crossing one set of mongrels with another isn’t the same thing.

Sophie: Yet your colonies get worse when they requeen themselves. That’s the cross breaking down.

Walrus: It’s a steady slide back towards the local average, which is just what the reshuffle predicts. It’s not a collapse. And it’s why I keep buying good mothers.

Round 11: The black bee lives

Sophie: Native bees aren’t extinct. Hassett and colleagues found 97.8% of 412 bees sampled across 24 Irish counties were pure Apis mellifera mellifera.

Walrus: That’s in Ireland, not Cheshire.

Sophie: Buswell and colleagues (2024) say 17 out of 70 colonies from Britain and Ireland showed no significant introgression. Colonsay and parts of South West England still harbour pure colonies. On Colonsay and Oronsay it’s an offence to keep any other honey bee.

Walrus: And 53 out of 70 did show introgression, even though they were sampled from beekeepers keeping non-imported stock. Hybridisation is widespread. The study doesn’t tell us what share of all our bees are pure, but it will certainly be lower than that.

Sophie: It tells you they survive. That’s a start.

Round 12: Where is BIBBA heading?

Sophie: BIBBA’s aims include breeding and rearing locally adapted bees “as a stepping-stone towards the restoration of native and near-native bees”.

Walrus: And near-native is never defined.

Sophie: Nobody is demanding that every hive be pure black.

Walrus: Good. If near-native means a dark, locally selected bee with some mixed ancestry, that’s one thing. If it means nearly pure Amm across all of lowland England, I think it’s almost unachievable.

Sophie: Nobody has proposed that either.

Walrus: Well, if “restoration of native and near-native bees” applies to the whole country, I have doubts.

Round 13: Selecting from what we have

Walrus: Selection works on heritable variation, and nothing in genetics says a population has to be pure before selection can work.

Sophie: And every figure you quoted comes from pure-subspecies programmes with pedigree records. You couldn’t find a single published study showing progress in open-mated, mixed stock.

Walrus: I said that myself.

Sophie: So the evidence supports our model, not yours.

Walrus: Your model works in rare, isolated places. For most beekeepers in England there’s no controlled, stable population to select within. It’s theoretical rather than practical.

Sophie: It’s theory because people like you won’t join in.

Walrus: Or because the geography won’t let them. But also, what gives one group of beekeepers the right to dictate to the whole country what bees they can keep?

Sophie: Nobody’s dictating. We campaign. You’re free to ignore us, and clearly you do.

Round 14: Flooding the neighbourhood

Walrus: My method is to buy instrumentally inseminated breeder queens from strongly selected lines, lately from Sweet Apiaries. I also select my own standout queens. Then I graft, open-mate the virgins in an area I flood with my drones, and requeen anything defensive.

Sophie: Your drone flooding doesn’t work. Sprenger and colleagues put drone colonies on an island in Lake Neusiedl, about 1.8 km from the mainland and joined to it by a dam. On average only 63% of the offspring carried the island drones’ cordovan marker, and the authors concluded the island wasn’t suitable for a mating station.

Walrus: I admit my flooding is probably ineffective, but maybe it’s better than nothing.

Sophie: And with half of matings within 2.5 km, your drones get a big say in your neighbours’ queens.

Walrus: They do. My drones go into the local mix, and I don’t pretend my contribution is zero. That’s one reason I requeen anything defensive, and my bees are good bees.

Sophie: So, you’re flooding your neighbours with bought-in genetics.

Walrus: Into a neighbourhood already full of mixed bees. When you say “bought in” it’s not as if they are from another continent, and as I said before, they are well adapted to my conditions.

Round 15: The treadmill

Sophie: You buy at least one expensive breeder queen every season. That’s a treadmill, not a breeding programme.

Walrus: It’s a business cost, and I’m fine with it.

Sophie: You depend on a handful of breeders. What if supply dries up, prices rise, or the rules on moving queens tighten?

Walrus: Then I’d need a plan B. I would select from my best queens and those of other beekeepers that I respect and trust. I might even move in that direction anyway.

Sophie: Our approach builds a population that keeps its quality without constant top-ups.

Walrus: After how many years? Switching to that approach would cost me productivity for several seasons, maybe longer. It’s a risk.

Sophie: Sustainability isn’t only about next season.

Walrus: My approach is sustainable for as long as I keep doing it.

Sophie: That’s the definition of dependence.

Walrus: Look at most farming; people rely on others all the time for various inputs (seed, fertiliser, pest control, breeding material, etc.). It’s how the industry works.

Round 16: The New World Carniolan

Sophie: You hold up Susan Cobey’s New World Carniolan as proof that selection from mixed stock works.

Walrus: Yes, that seems to be the case. It began in 1981 from queens of mixed US and Canadian backgrounds, run as a closed population. Each year 30 to 60 breeders are chosen, their drones pooled for instrumental insemination, and 200 to 250 colonies tested, with the top third kept. Ruttner called them “the most Carniolan like bees” he had seen in North America.

Sophie: Every mating is controlled, and after about 20 generations they still had to bring in Carnica semen from Austria, Germany, and Slovenia because selection narrows the gene pool. That’s nothing like your apiary.

Walrus: Very true. Sue’s approach needs control of queens and drones, and a broad, genetically diverse founding base. If I were twenty years younger, I’d learn instrumental insemination. At 63, it seems a bit late. Jolanta and her team at Denrosa Apiaries went to visit and learn from Sue Cobey, so I’ll be interested to hear how it has gone.

Sophie: So, the best example of your approach requires exactly the control that local breeding groups try to build.

Walrus: That’s why I don’t pretend to be doing what Sue does. Where I can’t control the drones, I can still choose the mothers. Furthermore, the best example of control involves a syringe, not an island or peninsula.

Round 17: Truce?

Sophie: So what would you tell a new beekeeper?

Walrus: If you live where native or near-native bees make up most of the population, join the local queen breeding group and get involved. Your drones matter to everyone within at least 7.5 km, so don’t swim against the tide.

Sophie: I’ll drink to that.

Walrus: Be aware that what many people think are Amm or nearly Amm are actually nothing of the sort. Many dark bees contain a good dose of Carniolan genetics. A Plymouth study of Cornish dark bees found most were mixed with Carniolan genes, and only between 4 and 15 of 43 came out pure, depending on how strictly you define it.

Sophie: But in some places they are pure, like in Ireland.

Walrus: If you live somewhere like me, with very mixed-up bees, rear your own queens rather than buying in mated ones of unknown origin. Graft from a UK-bred breeder queen, from your best colonies, or both, and keep records so you can track progress.

Sophie: And requeen the nasty ones.

Walrus: Yes, so their drones leave the breeding population.

Sophie: We agree on more than I expected.

Walrus: We both want calm, productive bees that don’t swarm the moment the sun comes out. We just disagree about where to start and how to achieve that goal.

Sources

• Jo Widdicombe, talk to the Scottish Native Honey Bee Society
• Beye et al. 2006, Genome Research
• Jensen et al. 2005, Conservation Genetics
• Brascamp et al. 2016, Apidologie
• Andonov et al. 2019, BMC Genetics
• BIBBA committee statement on imports
• BIBBA aims
• Büchler et al. 2014, Journal of Apicultural Research
• Hassett et al. 2018
• Buswell et al. 2024, Journal of Apicultural Research
• The Bee Keeping (Colonsay and Oronsay) Order 2013, SSI 2013/279
• Sprenger et al. 2025, Archives Animal Breeding
• Developing a breeding program with Sue Cobey, Beekeeping Today podcast
• Ellis et al. 2018, Journal of Insect Conservation