It's often stated that swarm cells are on the bottom of the frame and supersedure cells are built in the middle of the comb. This may not always be the case.

Emergency Queen Cells

Certainly, if a hive feels they are queenless they may try to make an emergency queen from eggs laid in regular worker comb. Emergency queen cells are structurally different than swarm or supercedure cells. You should be able to tell that the cell was made as an extension of a regular worker brood cell. We can see few examples of emergency cells in this photo:

Emergency queen cellsPhoto by: Shawn Caza / CC: Attribution-NonCommercial-ShareAlike License

Queen Cells Used During Supercedure and Swarming Look Identical

But a hive will often have proper queen cups ready at the bottom or middle of frames which may be used for either supersedure or swarming. So, rather than the appearance or location of the cell, one must think about the wider context of the hive to figure out what the bees are trying to do.

For example, the following picture shows a frame with a good number of bees, a queen cup with an egg on the comb near the middle-right of the frame, and the brood nest in the process of being back-filled with nectar.

backfilling the broodnestPhoto by: Shawn Caza / CC: Attribution-NonCommercial-ShareAlike License

Let Context be Your Guide

In the case above we are looking at a strong hive building up and things are getting congested between a queen that is laying well and increasing honey stores. It's best to assume the bees want to swarm. By the way, did you spot the current queen in the above photo? Click here for a closer look.

Other Beekeepers That Have Dispelled This Myth

Walt Wright has also suggested a larger quantity of queen cups/cells or queens in a wide variety of stages of development, with the first queen laid at the periphery and subsequent queens laid across the brood nest, is indicative of swarming. Whereas a supersedure tends to be just a few queens all laid at about the same time on the periphery of the brood nest.

The myths about supersedure cells run so deep in beekeeping literature that you still may not be convinced after having read both my post and the preceding link, so I will also offer you this link to David Cushman's perspective on the topic.


Honeybee Democracy by Thomas D. Seeley

Honeybee Democracy is one of my favourite bee books. The book summarizes Seeley's life long quest to understand the natural behaviour of this marvellous insect. Rather then talking about what we should do to our bees, the book focuses on what and how the bees, when left to their own devices, decide to do some of the things they do.

Seeley goes into the details of his ground breaking experiments which encompass most of what we currently know about feral hives, the communication processes involved during a swarm and the characteristics of a good bait hive. We also get some insight into Seeley's scientific process and the creative approaches used to uncover honeybee mysteries.

While I consider this a valuable read for all beekeepers, the stories describing the honeybee's sophisticated approaches to communication and decision making will also appeal to the non-beekeepers with a healthy sense of curiosity. You can find a copy on Amazon here.

The following video lecture gives a summary of the ideas discussed in the book. Watch with caution if you haven't read yet read the book. You might spoil some of the suspense.

Handy chart showing the relationship between the development time for new queens, swarming and hive population.

The important thing to remember about queen development is it can take about a month from an egg in a queen cell till that egg turns into a mated laying adult queen. I have personally seen new queens laying in the range of 25-32 days from queen egg being laid. Others say 28 days plus or minus five days.

During that time:

  • Population will decrease due to the swarming bees leaving
  • Then population will increase for three weeks as existing brood emerges
  • Then population will decrease for three weeks as brood from the new queen develops into adults.See the pdf version of the chart below.

swarm calender

The classic, and still the most definitive text on the subject is 'Swarming Its Control and Prevention' by L. E. Snelgrove.

Snelgrove's recommended approach to swarm prevention involves keeping lots of space in the brood box for the queen to lay. Once queen cells are observed, his strategy involves separating the flying bees from the house bees. To achieve this separation, he designed what is now called the Snelgrove board, a simple yet clever piece of equipment that allows you to divide bees within the same hive.

The result of using a Snelgrove board is basically an artificial swarm or split, however, his approach allows for a little more flexibility, particularly if increases are not desired and you plan on recombining the split later. On the downside, you do need to spend a little time opening and closing entrances a few different times over the course of a number of days. If you don't want to bother with Snelgrove's particular technique you will still benefit from reading his book as it is provides a comprehensive look at honeybee reproduction behaviour, and how beekeeper interventions can alter their behaviour.

Queen cells
Queen cells on a new foundationless comb. Via flickr

In addition to his own method, the 112 page book also provides the authors opinion on the pros and cons of other swarm control approaches.

Caution. Those tempted to avert swarming by the destruction of queen cells should pay special attention to the comments on pg. 57 of Snelgrove's book. He observed that after queen cells are destroyed one may find new capped queen cells as soon as four days later since the bees may use existing young worker larvae to replace the damaged cells.

This blog post on swarm prevention by Michael Bush is another quick and simple explanation of the behaviour leading up to a swarm and what you can do to prevent swarming.

Essentially, Mr.Bush is also all about keeping the brood box open and giving the bees enough space up to the point where you see queen cells. You need to keep an eye out for backfilling. This means checking if the size of the brood nest is diminishing as a result of the colony filling cells where brood has emerged with nectar. As an example, the following photo shows a large section of nectar in the middle of a brood frame

backfilling the broodnest

Keep in mind that you may see a similar behaviour during an intense flow. The bees may initially store nectar quickly in any available cell. They will then move it up above the brood shortly afterwards. So some nectar near brood isn't necessarily anything to worry about, but If they don't move it upwards, then you should expect queen cells are coming.

Tips. You may also want to consider if there's anything you can do to help the bees cure the nectar into honey faster (As moisture is evaporated from nectar it requires less storage space). Many will tell you to increase upper ventilation during a flow, while a few might tell you otherwise.

If the bees are backfilling during the reproductive swarm period, it is suggested to create space in the brood nest itself. Simply adding honey supers may not be enough as the bees may need more brood space rather than honey space.

Once you see queen cells or eggs in queen cups, decide if they are swarm rather than supercedure cells. As I keep hives in urban areas where a swarm can be particularly unwelcome by neighbours, I tend to error on the side of caution and always assume it's a swarm cell.

Egg in a queen cup

Caution. Once swarm cells are laid in, splitting, or artificial swarming, is the most effective recourse remaining to prevent a swarm. At this point it becomes extremely difficult to change the plans made by your colony by simply destroying cells and adding more space. The older the age of the developing queen the harder it will be to change the colonie's mind.

Here's a video of a Michael Bush giving a lecture on the subject:


A few years ago Dave Barr introduced me to the concept of Growing degree days(GDD), a calculation of accumulated heat that helps you to predict the blossoming of plants and trees with pretty good accuracy. It's no subsititute for physically looking around you and taking note of what is in bloom, but following the GDD might help predict the timing of flows and corresponding seasonal bee behaviour.

See wikipedia for a detailed explanation on how the calculation is made and flowering time of various plants.

Use this calculator to check the GDD for your area. All GDD references are given using base 10 throughout the rest of this post.

The spring has been excessively warm this year in Toronto. We jumped up to 25 C GDD in the last week. The average for this time of year is 0 C GDD and in 2010, which seemed like a warm early spring at the time, we were only at 3.5 C GDD. Normally this might be a bit early for silver/freeman maples, but just as you would expect from our current GDD, they are just finishing their bloom

freeman maple and native bee

and the native collettes have emerged in time to profit from them and mate:

What I have noticed over the past few years is that bees will start returning to the hive with full pollen baskets as the growing degree days rise just above 0 C.

full pollen basket on honey bee

One of the more detailed theories on bee behaviour in early spring I've been able to find is on Michael Bush's site. It attempts to relate certain changes in hive behaviour to bloom times. You may be in a different region than Michael, and things may progress at an unusual rate this year, but I think it will be interesting to keep an eye out for any of the correlations he mentions in the next month or two.


- If the brood nest starts contracting before the peak of the apple blossoms - they may be switching from buildup to swarm preparation. Look for backfilling, or patches of nectar surrounded by brood like in the photo below.

backfilling the broodnest

- When the Black locust blooms (140 C-160 C GDD) should be the start of the main flow and the interest in initiating a swarming should be reduced.

- Established colonies start making white wax shortly afterwards.

My hypothesis is, that for here in Ontario, around 70-100 C GDD might be a good time to look for the first signs of swarming preperation, with Queen eggs most likely appearing around 115-160 C GDD. However, I imagine with the large amount of stores the bees still have after what was a mild winter, they could run out of space even before this.

Update 2012: Short mild winter, I saw the first queen egg in one of my hives at 64.5 GDD. You can get a quick peek of it behind the bees:

The first swarm in my area was a few days later at 65.5 GDD or about three weeks after the dandelions first bloom, two weeks after the apples and at about the time the lilac trees were in bloom.

forsythia swarm

2013: I observed eggs and larvae in queen cups at 88.5 GDD. A little under a week after the apple blooms, two weeks after the dandelions and about the time the lillacs bloomed, but it's possible they were supercedure cells. It was a cool and wet spring, and I'm not sure there was very many reproductive swarms. I did find a feral colony that had moved in sometime before 95 GDD, but the first report of an actual swarm in the area did not come till around 265 GDD.

2014: Very, very cold winter, with a long cool spring. First report of a queen cell came from a friend around 50 GDD at the time the apple bloom started and two weeks after the dandelions. The first swarm report I heard reported came four weeks later at 252 GDD, about a week after the black locust bloom.

Last week we were called to collect a swarm.

Right away we noticed a few indications that led us to believe the swarm was queenless:

  1. There was no single cluster. Bees were on the ground in two different locations and there were a few up in a tree.

  2. The bees were spread out and fanning. While it was hot, it's likely they were searching for queen scent rather than trying to keep cool. Even once we got them into a nuc box they spread out over the available surface of the box rather than clustering.

  3. They became very interested in the bee gloves of my friend John and actually started running en masse towards him when he first started looking for a queen.(You can see this in the video. Unfortunetly it's a little obscured as some bees had also become very interested in my camera at the same time.) I don't expect the residual bee smells on the gloves and camera would have been nearly as interesting if they had a queen giving them the pheremones they were really after.

We collected the bees and joined them with one of our weaker colonies. The colony we joined it with was a split with an old queen removed from a hive that had supersedure cells. We placed the swarm on top in a seperate box with a sheet of newspaper inbetween. Five days later we opened the hive to discover: the bees had already eaten two holes through the paper, one dead queen being mobbed by bees and one healthy looking queen going about her business.

We couldn't believe it. We searched for other explanations. Is it possible a virgin queen drifted back to the wrong hive from the other half of the split? Seems unlikely. So maybe the swarm did have a queen afterall. Upon reviewing the video again, I noticed relatively few bees fanning at the begining as compared with the end when some bees were already placed in a nuc box. Perhaps that should have been seen as sign the bees were just trying to keep cool vs. actually wondering about the current location of their queen?

Lesson? Check and double check when there is any doubt about a queens existence.