Give us a Break

Mr & Mrs W taking a break in Argyll
Mr & Mrs W taking a break in Argyll

Somebody asked me yesterday, “what makes a good beekeeper?” I don’t know if I’m qualified to answer that, but what I said was, “If you can keep your bees alive from year to year, you’re doing pretty well.” That’s probably not the best answer, but experienced beekeepers who keep on top of disease and nutrition issues are going to have far fewer losses than novice beekeepers. 

A standard measure of success is “average honey yield per hive”, but that one can be heavily skewed by weather conditions. Also, beekeepers may have a high yield per hive that lives, but they may conveniently ignore dead or small colonies from the calculation. I suppose each beekeeper will decide what works for them, and this is how it should be.

Reducing Winter Losses

I’m very much still learning, and each year I hope that list of mistakes will get shorter. I seem to be running at 20% losses per year, which I would like to get down to 10%. If disease and nutrition are taken care of, then much is down to the quality of the queen; poor performers need to be culled and replaced with something better.

From what I see, the central pressures on honey bee survivability are three things; the weather, varroa mites and the beekeeper. We cannot control the weather. Beekeepers kill a lot of colonies through inexperience, mistakes, neglect or stressing the bees too much. It’s up to us to continue to learn and to improve. That leaves varroa mites. I go on about them a lot, but that’s because they are a big deal. 

Going Treatment Free?

There is a growing body of evidence that if we all stopped treating our bees, eventually the survivors would be able to tolerate the mites without our help. However, over 90% of honeybees would die in the process. The survivors will be those with certain genetic and behavioural traits which help the bees to deal with mites. It seems likely that some of these traits involve frequent swarming and widely spaced colonies, plus hygienic behaviour involving uncapping and re-capping brood cells. 

For many years beekeepers have used honey bees as their livestock. We have selectively bred for traits which are beneficial to us. Lots of honey, low swarming tendency and gentle disposition are the things I look for. What are the chances that the post-apocalypse survivor bees will have those traits? If 90% of managed colonies die in the first couple of years that will be an end to commercial beekeepers, which will have massive knock-on effects. Suddenly the number of pollinators will plunge, with who knows what consequences for the rest of agriculture and the natural world.

I’m a Treater

I try to keep my bees healthy and strong, which means treating for varroa mites at certain times. My idea is to try to rotate treatments so that the parasites do not develop resistance. I prefer to use naturally occurring substances such as oxalic acid and thymol, but sometimes I use a synthetic miticide (Amitraz). The whole thing is very tricky because I don’t want to treat when the honey supers are on, but I don’t want the mite load to get so high that I get a honey crop only to see my bees drop dead. Also, oxalic acid is most useful when there is little or no sealed brood in the hive (wintertime or with a new swarm). I also need the winter bees, which are made in September/October in my area, to be nearly mite-free so that the colony can get through winter successfully.

Hive Population Dynamics

The complex problem of hive population dynamics, which includes bees, brood and mites, can be studied using a mathematical model. Randy Oliver spent a year developing a model which is based on hard data from beehives plus copious research papers. Obviously, you can’t do beekeeping based on a spreadsheet; you have to be with the bees and make decisions based on what you see. However, I find it helpful to use Randy’s model to try out different scenarios to see what should happen theoretically, on average. If something looks horrible in the model, I won’t risk it in real life.

Walrus and Randy Oliver
Walrus and Randy Oliver

Recently I have been running two different varroa control methods through the hive model. One is the regime I followed last year. The other option involves giving the bees a brood break in July by physically taking brood frames away. This latter idea was advocated by Ralph Buchler at the 2019 National Honey Show. I have tweaked the model colony details to match my bees in my area and have assumed some mite drift into the hives in the autumn (250 mites). In reality, all colonies are different, but this exercise was to compare 2 strategies, so running them both through the same model makes sense.

My Hives Assuming no Treatment

Chart 1: No brood break, no treatment
Chart 1: No brood break, no treatment

In all scenarios I’m starting the year with 40 mites in the colony and in late summer/autumn there is an influx of 250 mites from drift/robbing. The yellow bars represent bees, orange is sealed worker brood, and the black line shows the percentage of varroa infestation in that worker brood. Chart 1 shows one of my typical hives over the year, assuming no mite treatment. What happens is that by the end of summer, when winter bees are about to be made, the mite population really ramps up and those winter bees will be sick. The hive will die out in the winter.

Chart 2: Amitraz in August, OAV in Nov
Chart 2: Amitraz in August, OAV in Nov

Chart 2 shows what I did last year. I took the honey off quite early then treated with Amitraz plastic strips for 6 weeks from mid-August. In late November, I used Oxalic acid vapour (sublimox) to kill the majority of the mites, which would mostly be phoretic, as there is little or no brood at that time. During the time that winter bees are made (Sep/Oct), the infestation level is quite low (4%-8%), so hopefully, they were healthy.

The problems with the method I used last year (Chart 2) are:

  • Varroa infestation was high at the time of Amitraz treatment; there is little margin for error; if I’d left it much later to get more honey the colony could have succumbed
  • The brood in November is highly infested (30%+), so those bees will struggle
  • I don’t want to rely heavily on Amitraz every year because it’s a synthetic miticide and resistance can become a problem. Thymol is likely to be slightly less effective and may put the queen off laying for a week or two, which I don’t want.

Brood Break or Shook Swarm

Chart 3 is an interesting one. The idea is to remove all brood frames from the hive apart from one frame of eggs/larvae. This “bait comb” is removed once the brood is sealed. Apparently doing this can remove 95% of the mites. The brood combs can either be destroyed, which sounds insane to me, or used to make up new hives or nucs. I’ll talk about them later.

Chart 3: Brood Break in July, no treatment
Chart 3: Brood Break in July, no treatment

Going back to Chart 3 what seems to happen with no treatment apart from the brood break is that the hive dies out in the following spring, but maybe not. It looks like the winter bees are going to be OK, but there will be quite a few (800 – 1200) mites attached to these bees over the winter. They weaken the winter bees and also jump into the new brood laid in spring. I think with just doing this mechanical method of mite control that the bees will die in their second year. They last longer than the scenario in Chart 1 (no summer brood break).

One potential consequence of taking away all brood in July and around 10,000 bees is that the honey crop suffers. In my case, the flow is usually tapering off in August anyway, so it may not be so bad. Also, with no brood to look after the remaining bees will possibly be storing more of what they collect rather than using it to feed larvae. 

Chart 4: Brood break in July then OAV late Nov
Chart 4: Brood break in July then OAV late Nov

Chart 4 is what happens when Oxalic acid vapour is used in late November/early December. It seems to result in a colony that survives; a lot will depend on there being no brood at the time of treatment.

Which One is Better?

The main differences between what I did last year (Chart 2) and this new brood break plus OAV method (Chart 4) are:

  • brood infestation in August is virtually zero after the brood break, whereas in Chart 2 it’s up around 20% – 25%
  • there are fewer bees in August in the brood break scenario, which could impact the honey crop if there is a flow in August
  • the mechanical removal of varroa in July replaces a chemical treatment in August, which seems like a good thing
  • honey supers can stay on much longer with the brood break if desired, to catch some willowherb and balsam honey

Making Increase

Hives made up of the mite-infested brood frames plus bees would need to be treated with Amitraz straight away (Chart 5). They would also need a new queen; I’d probably pop a virgin in there from my queen rearing operation (sounds grand but it isn’t). The large quantity of emerged nurse bees would be ideal for drawing out foundation into comb. Alternatively, I could let them make emergency cells and knock them down, then use the queenless hive as a cell builder. It would be quite late in the season for that, but there should be drones about into early September. 

Chart 5: New Hive from Removed Brood Frames
Chart 5: New Hive from Removed Brood Frames

In Chart 5, I assumed that the hive was made up with 8 frames of brood plus bees, treated with Amitraz immediately, given a virgin queen and then zapped with OAV in late November. You could argue that the OAV is over the top, but it’s incredible how many mites can drift in during the autumn; better safe than sorry in my view.


Based on these results, I can’t say that I am filled with unbridled enthusiasm about the “brood break/shook swarm in July” idea. It is around about the time when I would make up nucs to over-winter anyway, so I might give it a go on some hives. I need to see if the outcomes in real-life match those in my fancy spreadsheet model.

2 thoughts on “Give us a Break

  1. Hi Steve. What do you think about Samuel Ramsey’s deductions that as varroa are never phoretic and are truly parasitic all their lives creating a brood break when varroa levels are not low is pointless. The now broodless varroa feed on the fat body of the nurse bees who, depleted, cannot feed the brood that does eventually need feeding leading to dwindling of the colony.

    • Hi Dani! Yes, fair point, except you could even argue they are not a parasite but a parasitoid. In my model I did the brood break in July when the mite count would be high, so it would be beneficial then (at least from a mite reduction standpoint). To me, it’s all about ensuring the winter bees are healthy and then a final OAV treatment when broodless in winter means they start off the following year with a fighting chance.

What do you think?

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: