A no competition, competition

Unless you’re in an unseasonably warm part of the country, mid-April is usually early enough to put out your bait hives. This year, because of the unusually cold snap in the last week or so, it might still be a bit early. However, colonies are developing well and as soon as the weather properly warms up they will start thinking about swarming.

Regular readers, look away now

I’ve written a lot about bait hives in previous years. Anyone who assiduously follows this site and – unlike me 😉 – remembers what’s been written before can skip ahead to the next section.

But for those who need an aide memoire

The purpose of a bait hive is to attract a swarm that you (surely not?) or someone else has temporarily misplaced i.e. lost 1. When a colony swarms it settles in a temporary bivouac from which the scout bees fly to survey the area for a suitable new nest site.

The two stage process of swarming

The scout bees have very particular requirements.

They’re looking for cavities of about 40 litres volume with a small, clearly visible, south facing, entrance near the base.

Evolution and fussy house hunters

Bees have evolved to nest in trees – not quaint cartoon churches as shown above – and cavities in trees come in all shapes and sizes.

This is why they don’t care what shape the cavity is. However, cavities with small entrances are easier to defend, which is why they prefer them 2.

In addition, bees favour cavities situated more than 5 metres above ground level. Again, this makes evolutionary sense. It’s not just Winnie the Pooh that likes honey. The higher up a tree they nest, the less likely they would be detected by a bear 3 on the ground. And if the nest remains undetected (or unreachable by a climbing bear) there’s a chance the colony will thrive and reproduce (swarm) to pass on the ‘high altitude’ nest site preference gene.

And if you’ve evolved to nest in a tree cavity in a wood filled with other trees, it again makes evolutionary sense for the entrance to be clearly visible. If it wasn’t, bees on their orientation flights would inevitably get confused (and therefore lost).

Finally, bees have a strong preference for cavities that smell … of bees.

A cavity that’s already heavily propolised, or contains used drawn comb, offers distinct advantages to the incoming swarm. They will have less work to do and so more chance of building up before winter arrives.

The ideal bait hive

And you can reproduce these requirements by offering a used single brood box National hive with a solid floor and an entrance reducing block in place … facing south and situated well off the ground.

I discussed the evolutionary selection pressures that have shaped the preference for a 40 litre box rather than that convenient spare nuc box I’m repeatedly asked about a smaller box a few weeks ago.

In that post I also discussed why I ignore the preference for bait hives located 5 metres above the ground:

  • I want to be able to watch scout bee activity. This is tricky if they’re a long way off the ground 4.
  • It’s a lot safer retrieving a bait hive from a hive stand at knee level than it is when climbing a ladder. I usually move occupied bait hives late in the evening (when the bees are all in residence) and prefer not to do this balanced precariously on top of a ladder.
  • And – though not listed last time – my knee level bait hives are sufficiently successful I don’t need to increase their attractiveness. I don’t doubt they’d be more efficient located at altitude 5 but they work well enough that I don’t  feel the need to risk altitude sickness or a broken leg …

But, what I’ve not really discussed before is the location where bait hives should be sited and the importance of appreciating the ‘competitive‘ aspects of bait hives.

Natural competition

When you place a bait hive in the environment, whether it’s in your garden or the corner of a field or 5 metres up an oak tree 6, you are providing a potential nest site that will be judged in competition with other natural sites in the area.

And ‘the area’ is probably about 25 square kilometres.

If you struggle to visualize that then it’s the area covered by this circle centred on the roof of Fortnum & Mason’s, where there are some hives. London Zoo to Battersea Power Station and the Round Pond to Southwark Bridge … a large area 7.

Fortnum & Mason, 181 Piccadilly, London … scout bee range (in theory at least)

Scout bees survey over 3 km from their nest site, though swarms rarely relocate that far 8.

Why don’t they move ‘that far’?

Again, evolution may have selected bees that choose not to move away from the environment in which the swarming colony has flourished and built up strongly enough to be able to swarm.

Scout bees find nests, they don’t survey the available forage around those nests. So it makes sense to stay in the general area where forage is proven to be good enough (to allow swarming).

However, I suspect a compelling reason that swarms don’t move far from their original nest site is that there are plenty of alternative nest sites available.

Church towers 9, roof spaces, chimneys, tree cavities 10, compost bins, abandoned sheds etc.

Choices, choices

Think about the environment near your hives. Whether urban or rural, there are bound to be thousands of potential cavities within 3 km.

Some will be too small, some will be poorly defendable 11 and some will be unsuitable for other reasons.

But there are very likely to be some that are ideal, or pretty close to it.

Mature woodland and older man made environments are likely to have ample choices.

Occupied bait hive

Occupied bait hive …

And then there’s your lonely bait hive.

Chance in a million?

How can it possibly compete with all those natural cavities in the environment?

Bait hive ...

Bait hive …

The first thing to do is to ensure it adheres as close as is practically possible (and safely achievable) to the idealised requirements determined by Martin Lindauer, Thomas Seeley and others.

  • a 40 litre cavity = National brood box 
  • a small entrance of 10-15cm2 = entrance block, solid floor
  • south facing
  • shaded but in full view
  • over 5m above ground level 12
  • smelling of bees = one old, dark comb against the sidewall (no stores!)

Secondly, locate it within ~500 metres of your own apiary (to hopefully re-capture your own ‘lost’ swarms 13 ) or, more speculatively, anywhere in an environment in which there are other managed or feral colonies.

Which does not mean over the fence from another beekeeper’s apiary!

Be courteous … don’t poach 🙂

The density of bees throughout much of the UK is very high. Look at Beebase to see the numbers of apiaries within 10 km of your own. When I lived in Warwickshire it was ~180-220, in Fife it was ~35-40 14.

In the talks I’ve given on bait hives this winter – where I customise the presentation to the audience location – few areas with active BKAs have under 100 apiaries within 10 km of their teaching apiary.

In both Fife or Warwickshire I never failed to attract swarms to bait hives in my garden every single year … and in several years up to three swarms to a single bait hive location.

And, with one or two exceptions, these weren’t swarms I had lost 15.

The density of managed colonies in the UK means that a suitable bait hive just about anywhere stands a chance of being occupied.

So, that’s how to win the competition with the natural nest sites that are available.

No competition

But do not put out multiple bait hives in one area.

I have recently re-read an old paper by Thomas Seeley and Kirk Visscher on quorum sensing by scout bees. Quorum sensing is a term for a decision making process where enough bees agree on the same choice, rather than the majority.

Seeley and Visscher (2004) Quorum sensing experiment

Like many good experiments it has an elegant simplicity.

They reasoned that if you provided a bivouacked swarm with a choice of suitable nest sites it would reduce the numbers of scouts that favoured each nest site, and in doing so, would increase the time to reach a decision as to which was best.

And it does.

More potential nest sites leads to an increase in time taken to reach a decision

Unsurprisingly, with more nest box sites to choose between, the scout bees per box were reduced in number (top panel), dancing to advertise preferred nest sites was delayed (second panel), and piping – the ‘prepare for take-off’ signal (third panel) for the bivouacked swarm – was also delayed.

I’ll discuss how this favours a quorum sensing mechanism (and some other aspects of the study) if and when I get time in the future.

For the moment the key take home message is ‘more choice = slower decision making’ by the swarm.

And, if you delay the decision making, there’s a chance it’ll start raining, or the swarm will be collected by another beekeeper … or they’ll opt to move into the old tower of that quaint cartoon church.

One area?

I started the last subsection with the sentence ‘But don’t put out multiple bait hives in one area’.

What is one area?

I was being deliberately vague because I don’t know the answer.

Since I don’t know where the bees might come from 16, I don’t know what’s within range of the bivouacked swarm.

Widely separated bait hives (black) are likely to be within reach of more swarms than clustered bait hives (white)

In practical terms this means I space my bait hives at least 500 metres apart. Widely separated bait hives are likely to be within reach of more swarms than clustered bait hives.

More importantly, clustered bait hives are likely to lead to competition between scout bees from the same swarm, resulting in reduced scout bee attention..

Until recently I’ve not kept bees in my garden. I would always place a bait hive in the garden and one near my out apiaries. With permission, I’d locate them in other places as well.

Having moved, I now have much more space and have bees in the ‘garden’. When my bait hives go out 17 they will be placed in likely spots on opposite sides of our bit of scrubby wooded hillside 18 .

But what’s a likely spot?

Ley lines

And if you thought that last bit was slightly vague … brace yourself.

Over the years I’ve noticed that some bait hive locations are much more successful than others.

Under offer ...

Under offer …

My tiny courtyard garden in Fife was a magnet for swarms. I placed a bait hive in a warm corner of the garden on the day we moved in, and within 10 days a swarm had arrived.

Planting tray roof …

Every year, without fail, multiple swarms would occupy bait hives 19 in that corner of the garden. I even had two swarms competing for one bait hive in 2019.

A sheltered south-east facing hedgerow in Warwickshire was equally effective.

Bait hive

Smelling faintly of propolis and unmet promises

As was a south-west facing spot sheltered next to my greenhouse in a previous garden.

Des Res?

Des Res?

But other locations have been far less successful.

Of course, this is a positive reinforcement exercise. I’m more likely to site a bait hive in a location I’ve previously been successful in.

But what else might account for this differential success rate? And can it be exploited in the rational location of bait hives?

Is it, as some suggest, that bait hives work best when they are located at the intersection of ley lines? This, and the possibility of creating Varroa-resistant bees by exploiting geopathic stress lines, surely deserves a post of its own 20.

Call me sceptical

However, as a scientist – and knowing others have been more than a little sceptical about the existence of ley lines – I think there’s a more prosaic explanation.

Without exception, my most successful sites for bait hives have been well sheltered to the north, and – in most cases – to the north-east and north-west directions as well.

For example, the bait hive is situated on the south face of a wall running east-west (Under offer, above), or in a corner sheltered to the north and east (Planting tray roof, above), or facing south-east in a very dense hedgerow running north-east to south-west (Smelling faintly etc., above), or sheltered by surrounding walls or outhouses but with a clear entrance facing south-west (Des res?, above).

And … since I’ve been aware of this for at least five years, and probably subconsciously aware of it for much longer, that’s exactly the type of location I choose to site my bait hives.

Which is, of course, another example of positive reinforcement 🙂

However, it works for me. I choose sites that are well sheltered to the sides and back of the bait hive, and I try and orientate the bait hive to face south (ish).

At knee level 😉

Give it a try.



  1. Hence the alternative name, swarm trap.
  2. And why you should not use open mesh floors on bait hives.
  3. Or, presumably an early hominid hunter-gatherer … or a honey badger (which, I’ve just learned from Google, can also climb trees).
  4. So emphasising the selective pressure from bears, honey badgers and early hominids that have forced them to choose nest sites at altitude in the first place.
  5. Evolution is rarely wrong.
  6. Don’t, just don’t!
  7. And a useless comparison for those unfamiliar with London … sorry.
  8. I’ve discussed how far swarms disperse previously and it’s usually less than half a kilometre.
  9. Whether quaint or not.
  10. The enthusiasm with which old trees are felled – for ‘safety’ or to route a new railway – means that, ironically, tree cavities may be in short supply.
  11. Notwithstanding the absence of bears, honey badgers or early hominids.
  12. Don’t say I didn’t warn you …
  13. Again, surely not?
  14. Here on the remote west coast it’s just 1 … and that’s mine.
  15. As if!
  16. Insert standard quip here about not being from one of my hives …
  17. It’s still far too early here …
  18. With almost zero expectation of attracting a swarm as my queens are clipped, my swarm control is awesome and there are no other beekeepers within 5 miles.


    One of those three statements is not entirely accurate.

  19. Of all sorts of different types, so it wasn’t the hive per se, or even its orientation as I sometimes had them facing west instead of south.
  20. Don’t hold your breath.

17 thoughts on “A no competition, competition

    1. David Post author

      Hello Susan

      Usually the evening it arrives, or the following evening. If you do it soon after arrival the bees will not have re-oriented to their new location, so you can move it any distance. If you wait for 3-4 days the bees really need to be moved over 3 miles. I move them soon after arrival to a ‘quarantine’ apiary so they can be monitored for brood diseases and treated for Varroa, and allow me to set up a new bait hive in the same location.


  1. Archie McLellan

    No more than one hive in an area. Thanks David, I didn’t know that. But it makes a lot of sense.

    1. David Post author

      Hi Archie

      That’s my interpretation of the science … of course, someone will now come along and comment that they always put five bait hives in a row and always attract swarms to them.

      Not that that proves that it’s effective, or more effective, than just one in ‘an area’.

      I’ve got no real supporting evidence of my own – but have no reason to doubt Seeley and Visscher. However, with the bait hive in my garden I sometimes also found scouts examining stacked up supers or spare brood boxes. Anecdotally – and that’s all it is – when I blocked off access to the spare stacks of boxes (by aligning them properly, or making sure the entrance was sealed, or whatever) a swarm would arrive at the bait hive within a day or two.

      Not in the slightest bit scientific!


  2. Jo Horsley

    We don’t like to collect swarms, unless they are our own, but some sites have to have bait hives out or the land owners get fed up with being bothered by swarms, some get quite angry and never quite believe they are not ours but are coming from other apiaries nearby. One of our worst turned out to be about half a mile from 6 abandoned hives that shed swarms all summer and another is near a very good beekeeper who frequently goes abroad in the summer, we quite like getting his swarms as they are not bad bees. So the other criteria for successful swarm trapping is someone nearby who makes little or no effort to control swarming.

    1. David Post author

      Hi Jo

      “A very good beekeeper who frequently goes abroad in the summer” is an oxymoron, isn’t it?

      If you assume that the majority of bees in the UK are ‘managed’ (although some are mismanaged 😉 ) then the distribution of hives that might swarm is directly related to the distribution of people. Unless that is you make some assumptions that people in the North are more likely to keep bees, or people in Wales are more likely to let their colonies swarm … either of those statements might be correct, but I’m not aware of any data to support them.

      People in the UK

      The map shows the distribution of everyone in the UK, mapped by postcode. With the exception of the white areas I suspect that there are ample bees in range of the majority of beekeepers who want to try and attract a swarm to a bait hive.

      And, if you ask the BBKA how busy their ‘swarm line‘ is in April/May/June it will provide additional evidence that there’s no shortage of swarms 🙂

      And, you’re absolutely correct … some people get really fed up with swarms that appear on their property.


  3. Luke

    Hi David,

    Good to know about multiple bait hives in one location – I currently keep my bees in the city (along with several others) so it’d be interesting to try and map how many bait hives are out out in back yards and the like, and to see their success rate.

    On the topic of swarms, I’ve already had to split one of my colonies into a nucleus this year. Imagine my surprise to find charged queen cells ABOVE the excluder! There were no other eggs in the super (despite there being space to lay) and I found the queen in the brood box so I can only assume the bottom of her abdomen was slender enough to poke through the excluder and lay into prepared cups there; luckily I saw some larvae stuck to the top of the excluder and thought to check in the super otherwise I’d have probably lost a bunch of the little devils!


    1. David Post author

      Hi Luke

      I rather expect there are relatively few bait hives set out … at least in comparison to the number of swarms that is.

      Interesting comment on the QC’s above the excluder. I’ve had them stuck to the QE.


      And have several times had a queen sneak through – either due to bent wires on the QE, or inattentiveness on my part.


  4. Patsi Bucknall

    Hi David
    If you’ve missed the removal window and have to move the swarm away by at least 3 miles, how long do they need to remain in this location before you can move them back again closer to the original bait hive site? We have a bait box on our apiary site and would like to get any swarms we inherit (whether they’re our bees, feral or someone elses) ultimately set up back in that apiary.

    1. David Post author

      Hello Patsi

      I usually reckon an entire brood cycle to be on the safe side. However, the reality is that bees can be quite accommodating … if you make clear that the hive had been moved, for example by partially stuffing the entrance with grass for a few days so the bees have to force their way out, they will reorientate to the new hive location.

      If it’s a swarm of unknown provenance that’s arrived I’d want to be sure that there were no foulbroods arriving with it … that really means waiting until you’ve got brood in all stages from the new queen. That’s 2-3 weeks which should be sufficient for them to forget about the original site they arrived at.


  5. David Oliver

    Thanks for all the info David. I’ve currently got six bait hives dotted around my garden so I may have ‘over done it’. Reading your reply to Archie above, I wonder whether it might be effective to leave them until I see scouts at them all and then block off the entrances to all but one…or do you think I should just take five down now and leave one up? Unfortunately there are no other locations I can site them at the moment.

    I’ve been looking on the beebase website which I’ve just joined but can’t find where you can see how many beekeepers are local to you…..whereabouts is it?

    Thanks, Dave

    1. David Post author

      Hi Dave

      I’d be concerned that you’re going to ‘dilute’ the interest of the scout bees. If you wait for the scouts to arrive and then close them off they might be persuaded to favour a distant site instead. However, there’s no guarantees with any of this 😉

      Personally, I’d just place a single bait hive in what I considered the most promising position. I don’t think my blocking off access to stacks of empty boxes triggered the swarm relocating to my bait hive … I think it was a coincidence. Had they had no choice – and only found the bait hive – perhaps they’d have arrived there the day earlier?

      The NBU have just rehashed BeeBase to take into account the new Varroa monitoring regulations (and done a super job of formatting the output … er, NOT!). This screenshot shows one of my old apiaries (now vacant) with an arrow pointing to the number you’re looking for.


      It doesn’t give the location of other apiaries. It just shows the number within 10 km. You can therefore calculate the average apiary density … in this case area = ~314 square km, containing 257 apiaries = 0.8 apiaries per square km. Remember that scout bees can survey at least 3 km from the original site. Also remember that a single apiary may contain multiple hives … any or all of which might chuck out swarms.

      That apiary was in Warwickshire which has a high density (number per area, not intelligence 😉 ) of beekeepers. My Fife apiaries have ~35 other apiaries within 10 km, and I routinely captured swarms in bait hives there.


  6. Dave

    Thanks for such a detailed response David….I’ll have to find some other locations where I can spread out my bait hives!

    Given I don’t yet have any bees and therefore don’t have any hives listed on beebase, presumably I’ll not be able to access the part of the site that you’ve very helpfully screen shotted above?


    1. David Post author

      Hi Dave

      You can create an apiary for the site where you WILL have bees (e.g. your garden) and indicate the number of colonies as zero. I have had a couple of empty apiaries, such as a quarantine apiary, in the system like that. You’ll get the same info I displayed. Then, when you get the bees (or they arrive in a bait hive) you can update the entry.


  7. chris

    followed the advice and got a bait hive filled (national poly box with foundationless frames).

    Its made up of foundation less frames and a manky brood frame in a national hive and a solid floor.

    I will vape them when i get them back to home (4miles away) and put another empty on the same site.

    My question is what now ….

    should I move the entire box and leave the swarm in the big box to get on with life ?


    leave that box there and transfer them to a nuc and then transport the nuc back to home ?

    1. David Post author

      Hi Chris

      It all depends upon the size of the swarm and the availability of equipment. If it’s a small swarm you could move them to a nuc, but you’ll need to keep an eye on them to be sure they don’t outgrow the box. I tend to leave them in the box they ‘chose’ and therefore would set up a new bait hive after moving them.

      If the box is full of foundationless frames then interleave them with foundation or drawn comb. With a completely empty box they sometimes build the comb in the wrong place. Remember that the box must also be level!



Leave a Reply

Your email address will not be published. Required fields are marked *