It should be fairly straightforward, shouldn’t it? A colony makes several virgin queens, eventually whittles them down to one, and she flies off on several mating flights. She returns to the welcoming embrace of her workers, and begins the long process of laying hundreds or thousands of eggs every day during the summer, autumn, and spring. This is the strategy that honey bees have come up with, and they have been around a lot longer than humans, so it has proven to be successful.

Elusive Perfection

However, nature being what it is, not every queen’s story is quite so uneventful. As I frequently say, there is a lot of death and failure in nature. As beekeepers, many of us like to think that we should be achieving at least 90% ‘success’ in all that we do. Ideally, 90% or more of our colonies come out of winter strong and ready for explosive growth. Hopefully, by the power of skilful and observant beekeeping, 90% of our colonies won’t swarm, and if we make queens, 90% of grafts will produce viable cells, and 90% of our virgin queens will successfully mate. Well, for me, this is not the case. Every so often it is, but on average, it’s worse than that. On the plus side, my bee farming business pays for itself, including running a van and availing myself of the unique services of a beekeeping mole (my son, Alex). You don’t have to be perfect to be successful.

Mating Failures

I’m going to try to quote the odd study to give the veneer of legitimacy to much of what I say. However, the longer I keep bees, the more I suspect that there is a broad spectrum of honey bee behaviours, and researchers probably struggle to show more than a glimpse into the world of the bee. Just because something can’t be backed up by a study doesn’t mean it’s false, and just because a study says one thing, doesn’t mean it will always apply. That being said, some studies indicate that the open mating of virgin honey bee queens is successful about 70–80% of the time (20–30% loses).^[1][2]

As the ’80/20 Rule’ (Pareto principle) seems to apply to many things, let’s go with that, and say that 80% of open matings work out and 20% fail. That means that 20% end up hopelessly queenless. Without intervention, they become laying workers and eventually dwindle away, their demise possibly being hastened by robbing from other bees or wasps. I think something similar, but flipped, may apply to swarms; around 80% will die and only 20% survive^[3]. So, if five colonies swarm to produce ten colonies (five parent colonies plus five new ones), four of the parent colonies would be expected to survive along with one of the swarms, bringing us back to five colonies. This would imply a honey bee population in a steady state (neither growing nor declining).

Anyway, back to mating success. It all depends, doesn’t it. Under certain ideal conditions, mating success can approach 100%, and in diabolical conditions it can be zero. Averages are all very well, but they don’t help me in my specific location/conditions at a particular time. If I’m staring at the ruined combs of a colony full of laying workers, I don’t care what some study says – I care about how to reduce the chances of it happening again. What are the factors that make the success rate fall? Here are some, in probable order^[4] of influence:

Weather

This is beekeeping, so obviously ‘weather’ is going to rear its heads. I think of the weather as being akin to the Hydra of Greek mythology, but maybe I’m odd. Low temperatures, cloudy skies, and strong winds all play their part in reducing the likelihood of queens (and drones) taking mating flights. Evidence from work in Germany using RFID to track mating flights^[5] shows that, although more mating flights occur above 18°C than below, they still happen as low as 16°C. It seems that in cooler conditions queens have more frequent mating flights of a shorter duration, whereas in warm weather flight duration is longer, but fewer flights are made. Strong winds produce higher risks to the young queen, so she may not fly in such conditions, and may be more likely to fail to return.

Drones

People who spend their days instrumentally inseminating queens, and trying to control mating using isolated places and drone-donor colonies, tend to place great importance on both the number and quality of drones available. They are critical. Some might even say, more critical than the virgin queen herself. An average queen inseminated with lots of top quality drone semen will be a far better queen than a beautiful virgin that mates with too few drones, or drones with below-par sperm counts.

What we want is lots of healthy drones in the vicinity (within a 1 km radius) of our mating apiary. They should be well nourished, sexually mature, and from a different genetic line to the virgin queens. Varroa mite infestation, and associated viruses, reduce the fitness of drones. However, chemical varroacides can damage their sperm. Treatment should therefore be done before drone rearing takes place – maybe oxalic acid when broodless in winter is the answer. According to Coloss^[6] there should be eight drone-donor colonies for 50 virgins (a drone donor colony is a normal colony with 2 drone combs). You need to plan the grafting time to coincide with when drones have emerged and become sexually mature, which is two weeks after emergence.

So, make sure you have enough drones, and look after them!

Time of Year

This season (2025) has been unusual, with everything coming early, and I did my first grafts on 14th April. Normally, there is a window of opportunity that is best for making queens, running from when swarming starts to sometime in July. The reason is a combination of weather, drone availability, nutrition (pollen and nectar), and bright daylight conditions allowing mating flights even after 5pm, with several in a day. It is possible for queens to successfully mate late in the season, even in September, but most commercial queen producers stop in August. Late supersedure queens frequently turn out to be drone layers.^[7]

Apiary Density

This can be too low or too high. In low hive density areas, you may have to provide the majority of the colonies that produce the drones that your virgins mate with. This is great for being able to control breeding, but it is a rare situation in most of the UK. In areas with very high hive densities, there is no shortage of drones, but you have little control over which ones mate with your queens. However, there are other issues with numerous bees in a small area. Perhaps there is inadequate forage to support all of those colonies throughout the season, leading to underfed larvae and weaker drones. Moreover, the risks of diseases spreading are significantly increased, and viruses can damage queens, drones, and workers.

If I could wave a magic wand and have my choice, it would be for a low hive density. As it happens, my bees are typically in places where there are plenty of other hives. According to BeeBase, the ‘apiary density within 10 km’ ranges from 219 to 336 for my apiaries, which seems like a lot. However, the maps feature on the same website shows something more believable i.e. 57 to 149 apiaries and 148 to 269 colonies.

Location/Micro-climate

Some places do better than others as apiaries, and that includes mating apiaries. Perhaps it is to do with the topography, landmarks, degree of shelter or other factors. Some spots seem to enjoy a microclimate that is always a bit warmer and less windy than surrounding places.

There is also the matter of hive (or nuc) location and orientation within the apiary. Anything that can be done to reduce the chances of a returning queen ending up in the wrong hive must be positive. That includes differentiating hives using colour and patterns. Brother Adam’s mating hives were split into four compartments, each having an entrance at 90 degrees to its neighbour.

Wasps & Birds

Once wasps become a pest, and start bothering honey bee colonies, things get a bit tricky. They tend to find weak colonies, then hammer them relentlessly. Mating nucs are usually weak colonies. I combine my mini-plus boxes into double or trebles to ensure that they are strong enough to keep wasps at bay.

When queens disappear, it is easy to blame it on the countless swallows and martins that continually swoop upon insects throughout the summer. I think it happens, in some areas more than others. At one of my apiaries where I have had significant problems with missing queens, there are a great many swallows, and I think they are swallowing my bees. Apparently, in some locations, dragonflies are also partial to a slow flying virgin queen. The RFID study^[5] tracked 64 queens on mating flights and 11 of them failed to return, for whatever reason.

Drifting

We know that bees drift, especially when hives are set up in a long line. In such an arrangement, you will often find more bees (and honey) in the hives at each end of the row, and smaller colonies in the middle. We also know that queens can end up returning to the wrong hive, which usually results in their death. Given that any hive can swarm, and will therefore need a new virgin queen to become mated, it makes sense to try to help queens successfully return. Place hives in clusters rather than long rows, and differentiate their appearance in some way.

Research^[8] shows that honey bees have three spectral types of photoreceptors peaking in UV, blue and green parts of the spectrum. So, they can clearly differentiate between UV, blue, and green. We cannot see UV, whereas bees cannot see red (it looks like black to them). The three receptors have their peak sensitivities at the following wavelengths: 344 nm, 436 nm, and 556 nm (please click each wavelength to view the colour). From what I can tell, the research indicates that bees respond more strongly to the shorter wavelengths than longer, i.e. blue more than green.

In terms of shapes, bees can clearly identify parallel lines, alternating black-white-black-white etc. Furthermore, they seem good at spotting radial symmetry^[9], such as the shape of many flowers, or other radial shapes like stars. We can use this knowledge to mark hive entrances and lids so that the bees can clearly tell them apart.

Inspections

Most mating flights start with orientation flights around midday to 1pm, then the actual mating flights from 1pm to 4pm (or even later). Therefore, if you have a hive or nuc with a queen likely to be taking mating flights, don’t inspect it in the afternoon. The queen faces enough hazards without the beekeeper adding more.

If I place a ripe queen cell into a mating nuc, then I will leave it alone for three to four weeks. Ideally three, but occasionally, I get distracted. If the queen is not mated within three weeks of emergence, I will make sure the virgin is actually there, and give them another week. Subsequently, I assume something has gone wrong.

When I find queen cells in a hive, I try to leave just one, and leave them for four weeks. If I don’t have a mated queen by then, something is wrong. Leaving them longer normally results in laying workers or a drone laying queen. If I intervene with a frame of open brood and a caged queen before laying workers start, there is hope. I recently dropped a mated queen directly into a queenless and broodless colony – straight from a mating nuc into the colony (no cage or anything) – and they accepted her and are now doing fine. Laying workers are the end of the line – they get shaken out.

Notes

1. https://www.researchgate.net/publication/330125121_Balling_Behavior_of_Workers_Toward_Honey_Bee_Queens_Returning_from_Mating_Flights 30% of queens were ‘lost during mating flights’, although 7% didn’t even fly, so I’d adjust it to 25% losses. Some never returned, some drifted, and some were balled at the entrance.
2. https://pmc.ncbi.nlm.nih.gov/articles/PMC11201093/ mating success of honey bees was 78% on the mainland and 60% on island locations (stronger winds).
3. https://theapiarist.org/feral-facts-and-fallacies/ David Evans’ excellent article delves into the facts about feral bees.
4. Okay, I’m sure of the first two on the list, but after that, who knows?
5. Heidinger, I.M.M.; Meixner, M.D.; Berg, S.; Büchler, R. Observation of the Mating Behavior of Honey Bee (Apis mellifera L.) Queens Using Radio-Frequency Identification (RFID): Factors Influencing the Duration and Frequency of Nuptial Flights. Insects 2014, 5, 513-527. https://doi.org/10.3390/insects5030513
6. Büchler, R., Andonov, S., Bernstein, R., Bienefeld, K., Costa, C., Du, M., … Wilde, J. (2024). Standard methods for rearing and selection of Apis mellifera queens 2.0. Journal of Apicultural Research, 64(2), 555–611. https://doi.org/10.1080/00218839.2023.2295180
7. BBKA winter losses survey (sorry, the link has gone now) – not sure how much faith I put in that, but 38% of losses were ‘queen related’
8. Hempel de Ibarra N, Vorobyev M, Menzel R. Mechanisms, functions and ecology of colour vision in the honeybee. J Comp Physiol A Neuroethol Sens Neural Behav Physiol. 2014 Jun;200(6):411-33. doi: 10.1007/s00359-014-0915-1. Epub 2014 May 15. PMID: 24828676; PMCID: PMC4035557. https://pmc.ncbi.nlm.nih.gov/articles/PMC4035557/
9. Giurfa, M., Eichmann, B. & Menzel, R. Symmetry perception in an insect. Nature 382, 458–461 (1996). https://doi.org/10.1038/382458a0