Irrational Varroa control

About a month ago I wrote a post on rational Varroa control. I define this as choosing a miticide appropriate for your colony 1 and environment, administering it properly and at the right time so providing the maximum benefit to your bees. The long term goal of rational Varroa control is reduced colony losses due to mite-transmitted viruses.

Primarily this means reduced overwintering losses due to Deformed wing virus.

I regularly give talks on this topic. In these I provide a few brief examples of the misuse of miticides. Some of these examples originate – anonymised – in questions I’ve received in previous talks or by email. Some are from surveys of miticide usage.

Winter is coming – miticide treatments and fondant on

A few (a very few) are from direct personal experience 🙁

I’m aware that these examples qualify as ‘misuse’ because I’ve read up about the active ingredient in miticides and have a reasonable understanding of how, when and why they work.

Misuse could probably be broken down into three divisions:

  • Incorrect usage that does not reduce the efficacy of the treatment. 
  • Using the miticide in a manner that significantly reduces its efficacy, but otherwise does little or no harm.
  • Incorrect usage that has no impact upon the mite population and/or that significantly harms the colony and/or spoils the honey.

After a short 2 preamble, I’ll outline some of the more glaring examples. Keep these in mind when planning your mite control strategy and you should see improved winter survival, better spring build up, stronger summer colonies … and more honey 🙂

What can I use?

The Veterinary Medicines Directorate maintain a list of miticides approved for use on honey bees (and other animals). It’s a large database and I find the easiest way to see the current product list – which changes quite regularly as things are added and removed – is to use the search facility.

VMD database search … bees, budgies and bearded dragons

Simply select ‘Bees’ from species list (they’re between ‘Bearded dragons’ and ‘Budgies’!) and then hit the big Run search button at the bottom.

There are currently 31 approved products though this contains a number of duplicates as Great Britain and Northern Ireland are listed as separate territories.

In my view, the two most important useful columns in the returned table are the ‘Active substance‘ and the ‘Aligned product‘. 

The active substance is the chemical in the miticide that is responsible for killing mites. 

Look down the list. There are relatively few different active substances, present alone or in combination.

I currently count just five.

And then there were five … VMD approved miticides classified by active ingredient

Thymol, amitraz, pyrethroids, formic acid and oxalic acid.

I’m ignoring the minor components like eucalyptus oil that, individually, do not have high levels of miticidal activity.

Flumethrin and tau-fluvalinate (the active ingredients of Bayvarol and Apistan respectively) are both pyrethroids and have the same mode of action (and, more importantly, resistance to one usually confers resistance to the other).

What about management methods to control mites?

The VMD database lists the miticidal products approved for treating honey bees. 

They don’t list management techniques like drone brood uncapping or small cell foundation, or the application of non-toxic compounds like dusting with icing sugar.

There’s nothing to stop you using these approaches.

However, first conduct some sort of cost-benefit analysis to determine if they are worthwhile.

Is the disruption to the colony, or the potential tainting of honey stores, justified by the reduction of Varroa numbers?

For example, Randy Oliver has done some analysis of the impact of sugar dusting and shows that – at best and with weekly applications – it might be able to hold Varroa levels steady. You can download his Excel calculations and play with some of the assumptions.

If you increase the percentage of mites capped in cells from 60% (which is probably rather conservative) and decrease the percentage of mites dislodged by sugar dusting from 40% (which is probably rather aggressive) then mite replication rapidly outruns the control method applied.

That’s a questionable ‘benefit’ in return for the disruption of blowing 120 g of icing sugar into the hive every 7 days 3 … but there’s nothing to stop you using it as a control method.

Though it might not do much controlling … 🙁

I choose miticides that kill at least 90% of mites when used properly. I prefer to use miticides once or twice during the season, rather than dabbling every other week or month.

What can’t or shouldn’t I use?

You cannot use things like fenpyroximate, spirotetramat or spirodiclofen.


These are used for mite or tick treatment of other animals or plants 4. They have been shown to be effective against Varroa (though at high levels they may also kill bees) but are not approved for use. 

Remember also that VMD approval involves both the compound and the mode by which it is administered.

Amitraz is approved for use as the active ingredient in Apivar strips. However, in Great Britain and Northern Ireland you cannot fumigate colonies with amitraz … which, as Apiwarol, is a popular treatment method in Poland.

There are some (perhaps surprising) ‘restrictions’ in terms of approved usage. Api-Bioxal can be used to trickle treat twice per season, but only used once for vaporisation.


This is one of the many oddities buried in the depths of the VMD database.

Api-Bioxal is not approved for spray administration, but Oxuvar is. Both have the same active ingredient. 

No wonder beekeepers find this confusing … 🙁

Read the instructions and other documentation

Miticides approved by the VMD come with documentation. This takes the form of the instructions written (often in really tiny print) on the packet. The other thing written on the packet is the ‘use by’ date. 

Apivar instructions – duration of treatment

Keep the packaging 🙂

Read the instructions for use.

It may seem like an obvious thing to suggest but at least a third of the questions I get asked are answered in the documentation.

Part of the problem is the wording that’s often used or the apparent (and sometimes real) contradictions between the instructions provided for two miticides that have the same active ingredient and formulation 5

Api-Bioxal additional documentation

For more legible documentation you could refer again to the VMD database of approved miticides. If you following the Aligned product’ link you will have access to several additional pieces of information, most useful of which are the Summary of product characteristics and the Product literature. The latter is a copy of the literature in a very easy-to-read format, without any fancy colours or tiny fonts.

Keep records

You should keep records of when you treat and what you treat with, including batch numbers. I simply keep the packet the miticide was supplied in. I know when I treated as my copious 6 notes record the dates … and allow me to work out when the period of treatment is finished.

Apivar (but the batch number is on the back of the packet)

Even easier … just take a photograph of the empty packet, but make sure you include the batch number in the shot.

Examples of miticide misuse

It’s not possible to provide a comprehensive set of examples of miticide misuse (or Irrational Varroa control) … at least not in ~2000 words 😉 – after all, there might be one or two ways to use a miticide properly, but thousands of ways it could be used improperly.

Here is a non-exclusive and far-from-comprehensive list of miticide misuse or bad practice. 

Don’t do this at home … or in the out apiary 😉

Too little

Use the correct dose. Using less than the recommended dose ensures that some mites will survive. This may lead to the development of mite resistance.

Some beekeepers have been known to add a single strip of Apivar to colonies in preparation for going to the heather moors.

This is bad practice.

Even worse … some have been found leaving the strip in the hive while they were at the heather. Is that forgetfulness or just reckless? The honey will be tainted  and the long-term exposure to low levels of amitraz may contribute to the development of resistance 7.

Too late

Remember that the goal of the late summer treatment is to prevent the developing winter bees from being exposed to Varroa and the viruses that the mite transmits.

If you treat too late in the season you may well kill lots of mites, but the winter bees will already have been exposed.

Mite levels (solid lines) when treated July to November and timing of winter bee production.

In the graph above, treating in mid-October will kill significantly more mites than treating in mid-August.

But that’s not really the goal.

An October treatment will kill more mites because they’ve been breeding like rabbits throughout September.

And, what have they been reproducing on? Your developing winter bee pupae 🙁

Winter bee production is not an all or nothing event. The colony does not switch from producing summer bees to winter bees on a particular date. As late summer segues into early autumn an increasing proportion of the developing brood will be winter bees.

It’s your responsibility to ensure that enough of them are protected from Varroa so that they can lead a long and protective life, getting the colony through until February or March next year.

Too much

Many miticides are reasonably well tolerated by bees. Nevertheless, overdosing is also to be avoided. If you read the product characteristics for Api-Bioxal for example it states that:

Significantly higher bee mortality was observed in hives that received double (by sublimation) or triple (by trickling) dosages of product. In addition, when overdosed, the over-wintering  capacity of colonies was diminished and there may be detrimental effects on colony development in the future 8.

Remember that nucleus colonies are smaller than full sized colonies. It’s not unusual for a  beekeeper to administer a full dose to what is essentially a half-sized colony. 

In the case of Amitraz, ‘overdosing’ is well tolerated and two strips in a nucleus colony is unlikely to do the bees any harm.

However, the same cannot be said of MAQS. The product characteristics for MAQS specifically state that it should not be used for colonies with less than 6 frames of brood.

Too high

The formic acid-containing MAQS is poorly tolerated by the colony at high ambient temperatures. The literature suggests it should not be used when the peak daily temperature might exceed 29.5°C.

I’ve never used MAQS. I’m told by other beekeepers that they’ve had problems with queen losses at temperatures as ‘low’ as 25°C.

Remember, when you add the MAQS strips they need to be in the hive for 7 days. You therefore need to check the forecast for the week ahead before starting treatment.

Too low

Alison Gray and Magnus Peterson (both at Strathclyde University) have conducted surveys of Scottish beekeepers for about the last 15 years, summaries of which appear annually in The Scottish Beekeeper. Results are also collated into the COLOSS analysis. 

The surveys have evolved over the years, but have included questions on the type and timing of treatment.

I read some of these carefully when I returned to live in Scotland (in 2015). I was interested to see what other beekeepers were using. One thing that surprised me was the amount and timing of thymol (Apiguard) treatment.

For example, in the 2014 report (PDF), ~25% of treatments used were thymol, with 60% of these being applied in September, October and November.

A quick check of the Apiguard Product characteristics turns up the following statement:

Do not use the product when the maximum daily temperature expected during the treatment is lower than 15°C or when the colony activity is very low or when temperature is above 40°C.

I don’t know anywhere in Scotland in where the maximum daily temperature exceeds 15°C for four weeks (the duration of treatment) between September and November. 

I checked a personal weather station a couple of miles from my apiaries in Fife … in October 2013 the average temperature was 11.5°C, but the maximum failed to reach 15°C on 10 days through the month.

Inevitably, the efficacy of treatment would be reduced.

West coast temperatures 26/7/21 to 26/8/21 … 7 days fail to reach 15°C making Apiguard a very poor choice here.

You need good long range weather forecasting skills. Prior experience of what might be expected can really help your planning in these circumstances. 

Too long

Do not leave Apivar strips in the colony longer than stated in the instructions. 

How long is that?

Again, from the Product characteristics documentation:

If brood is not present or at its lowest level, the strips can be removed after 6 weeks of treatment. If brood is present, leave the strips in place for 10 weeks and remove the strips at the end of treatment.

Strips added to a colony with a young queen (and therefore laying well, and late into the year) on the day this article appears 9 should probably be removed in the first week of November.

Used and removed Apivar strips

That’s not a great time of year to be lifting roofs and prising up crownboards.

However, the alternative is worse. If you leave the strips in situ you ensure that any surviving mites (and there will be some) are continuously exposed to a low level of amitraz … perfect conditions to help select for resistance 🙁

Too short

This is much the same as too little (see above). If you remove the miticide before the correct period of time has elapsed then some mites will escape treatment.

No miticide is 100% effective. However, why store up problems for the future by unnecessarily reducing the efficacy of the treatment?

Remember … the only good mite is a dead mite

Too old

Miticides are not inexpensive. It is therefore very tempting to save and re-use opened packets. 

For example, beekeepers with just a couple of hives still have to purchase a packet of Apivar 10 for an eye-watering £31. It must be very tempting to tuck the unused portion away in the shed for the next time.

And, when they do, how many do a before and after count of phoretic mites to confirm that the treatment worked?

I suspect very few. 

If they did they might be in for a bit of a disappointment.

Amitraz, the active ingredient in Apivar strips, is quite unstable and rapidly degrades. I’ve used strips from previously opened packets and observed that they were significantly less effective 11.

What I should have done was read the Product characteristics documentation where it clearly states:

Shelf life after first opening the immediate packaging: use immediately and discard
any unused product.

Similar problems (short or non-existent shelf life) apply to some of the oxalic acid-containing solutions, in this case because they degrade to product hydroxymethylfurfural which is toxic to bees at high concentrations (see the notes on this at the end of the post of preparing Api-Bioxal for trickle treating).

One and two hive owners should coordinate their purchases and treatment of colonies to avoid wasting money.

Coordinated treatment has additional benefits, which neatly takes me to the final topic … 

Too few

OK, these titles are getting a bit contrived now, but this is the last one 😉

Treat all the colonies in the apiary simultaneously. I’ve written extensively about drifting and robbing. Both activities redistribute adult bees and the mites piggybacking on them around the apiary (or the wider environment).

If you treat just one colony it will soon 12 become reinfested with mites from neighbouring colonies.

If you don’t treat one colony and it develops high mite levels in autumn, perhaps due to a late surge in brood rearing, it will shed mites (hitching a ride on young bees going on orientation flights) to your adjacent treated colonies.

A final note on Apistan and pyrethroid-containing miticides

Resistance to Apistan is widespread. It’s so widespread that the National Bee Unit apparently stopped keeping records a decade or so ago.

Apistan is a very effective miticide … against mites that are sensitive. 

If you intend to use Apistan (or any of the approved pyrethroid-containing miticides) I would strongly suggest determining the level of infestation before treatment, and confirming a 90+% reduction in mite levels after treatment.

Unfortunately, just counting the dead mites on the Varroa tray is not evidence that the treatment has been effective.

If you find 1000 dead mites on the tray it just tells you is that you have 1000 mites less in the colony.

There may still be 9000 left if the treatment was only 10% effective … 

That’s not going to end well 🙁



  1. Or colonies …
  2. Which, now I’ve written it, appears to be anything but short.
  3. Where does all that sugar go? Some will be undoubtedly eaten, but some may end up – directly or indirectly – in the honey supers.
  4. Though not necessarily in the UK. For example, spirodiclofen is licensed for use against scale insects and mites of fruit trees in many EU countries.
  5. See the comments above on Api-Bioxal and Oxuvar. There is a Scottish Varroa Working Party organised by Luis Molero, the Chief Bee Inspector for Scotland, to try and get rid of some of these contradictions and to provide better advice on mite management for beekeepers.
  6. Oh how we laughed!
  7. So, probably reckless …
  8. Unfortunately, these product characteristics do not give details of how much mortality was observed, or links to the primary data.


    It’s also worth noting that some of these comments contradict studies on vaporised oxalic acid usage, for example by Thomas Radetzki, Medhat Nasr, Heinz Kaemmerer or Francis Ratnieks.

  9. 27th August 2021.
  10. 10 strips, sufficient for 5 hives.
  11. So ineffective I had to buy some more and retreat … bingo!.
  12. In days – we’ve done this experiment.

37 thoughts on “Irrational Varroa control

  1. Lindsay

    Great article, we’re new to beekeeping this year and got our 5 frame Nuc in July. I intend going into winter on brood and a half simply because the colony has grown so much since tranferring into a hive 7 weeks ago. I have put in an inspection board twice, the last time being last week and on both occasions I have not seen a single varroa, so my question is this do I really need to treat for varroa just now. Can I leave off treatment and maybe do an oxalic dribble in December? TIA

    1. David Post author

      Hello Lindsay

      It’s a free world … you can do what you want 😉

      Remember that a Varroa tray is a relatively poor way to monitor mite numbers. Did the mites fall down and crawl away? Were they eaten by visiting ants? Or blown away in the wind? Perhaps the colony is busily rearing brood, so mites have to spend a minimal period ‘surfing’ around on the colony until they find a pupa to infest (so are less likely to have been dropped)?

      I’d try uncapping some drone brood and see if there are any mites present. Alternatively, do an alcohol wash (or sugar dusting) on a cup full of bees which is a lot more representative than natural mite fall.

      Your winter OA treatment must be applied when the colony is broodless … if you live somewhere warm and the colony might rear brood all winter then not treating now and missing some mites in the winter may leave you with problems next year.

      Most beginners underestimate the devastating consequences of unmanaged Varroa infestations.


      Good luck

    2. Paul Lindstrom

      David, do I read the end comment “If you find 1000 dead mites on the tray it just tells you is that you have 1000 mites less in the colony. There may still be 9000 left if the treatment was only 10% effective…” to say that you/we should perform a “sugar roll” test as a compliment to counting the mite mite drop now and then to compare the results? I haven’t done sugar roll tests yet, but it sounds quite harmless for the bees, and might be useful.

      1. David Post author

        Hello Paul

        That comment was specifically directed to those who use Apistan to (probably fail to!) control Varroa infestation. Of course, it’s good practice to confirm that whatever miticide you use actually worked, by showing a reduction of the levels before treatment, after treatment. However, a certain level of pragmatism is needed … if reasonably regular mite monitoring (by natural drop, uncapping drones or sugar roll) through the season has given a idea of the overall level of infestation, and if the mite drop during treatment is about what is expected, then there’s probably no need to do further testing late in the season.

        I don’t think I’ve ever tested mite levels after my autumn treatment. I have an idea of what was present beforehand and I use effective miticides in a manner that ensures they achieve the maximum killing. Yes, there’s a possibility that the treatment might have failed because the mites are now suddenly all amitraz resistant … but it’s highly unlikely. I reason that the added disruption to the colony late into the autumn is more likely to cause harm.


  2. Elaine Robinson

    Hi David, good summary & reminder.
    What are your thoughts on treatment rotation? Suspect many of us stick with a treatment we like especially when choices are few (e.g temperature precludes Thymol if you live in the North; fear of queen impacts late in the season for MAQs; API-bioxal is only effective on phoretic mites; resistance issues and solubility in wax for pyrethroids), so that leaves Amitraz!
    Have the consequences been measured of using it each season, assuming the strips are removed as per the manufacturer instructions?

    1. David Post author

      Hello Elaine

      By alternating amitraz/Apivar in summer and oxalic acid in winter you are effectively rotating treatments. Any potentially amitraz-resistant mites lingering after the summer treatment should get ‘whacked’ by the OA in winter. The treatments have completely different modes of action and resistance to one will not confer resistance to the other (and resistance to OA is difficult to envisage … and, frankly, the published studies on amitraz are a bit weird).

      Based on summer/winter treatments, I don’t see a need to use different treatments in alternate summers (or winters).

      If I only treated once per season I might have to re-think this.

      As I indicate above, I’m puzzled by the published study on Apivar/amitraz resistance. There are increasing anecdotal (by which I mean not backed up by some hard science) reports of resistance being widespread, particularly from commercials, but I’ve not heard this from amateurs. My experience with Apivar is limited to the last five years or so. However, it remains very effective in my hands. Over this period – including some of our scientific studies – I’ve probably treated ~75-100 colonies per year, some with astronomically high mite loads 🙁 I’ve seen colonies drop ~3000 mites in a week in July (and if that number were phoretic, imagine the number in the capped cells!) and overwinter really well, with very low mite levels through the autumn.


      1. Elaine Robinson

        Thanks David that’s v reassuring as it’s exactly my treatment regime – Amitraz (Apivar / Apitraz) late summer, removing Nov, followed by Apibioxal sublimation Dec after a cold spell. Mites have been reassuringly low this season but I’m not letting my guard down!

        1. David Post author

          I don’t even wait until December if the colony is definitely broodless. The OA treatment is to “mop up” mites that escaped the Apivar and that have subsequently reproduced (or not). Once the colony is broodless all those mites should be accessible to the OA. Since the colony is broodless the mites will no longer be reproducing, so the numbers cannot increase (until brood rearing starts again). Whilst there could be an argument to delay treatment until the end of a broodless period – on the basis that there will be a natural loss of mites throughout this period through grooming etc. – it’s a more difficult thing to judge, so (of course 😉 ) I don’t bother. I take the easy route. Once the colony is broodless I treat with OA.

          In some years with an early end to the summer nectar, early Apivar treatment and rapid onset of a cooler autumn, this has been as early (too many early’s in this sentence) as late October.

          Mites have been low this season with us as well. We (or rather the guys in the lab) harvested hundreds of drone pupae for experiments and only very rarely found mites. Drone infestation rates would be a fraction of a fraction of a percent. I suspect this was because the spring was cold and long, delaying the onset of colony expansion and therefore holding back the mite population.


  3. Jack

    Hi David,
    Both recent varroa posts very interesting for me coming to the end of my first season. My miticide of choice this year is Apivar and you mention that it is well tolerated by the bees and 2 strips in a nuc is unlikely to harm them. A question related to that comment: My bees are on 14×12 nationals so I am wondering if the proscribed 2 strips per brood box would still be sufficient if the dosage has been optimised for standard frames, and am considering a 3rd strip to account for the extra bees in my boxes. Or better simply to leave 2 strips for longer (up to 10 weeks)? In other words, higher dosage or longer dosage time? Any thoughts?

    1. David Post author

      Hi Jack

      If you read the Apivar instructions (at least on the VMD website) you’ll see it’s not been tested in double brood boxes. However, I’ve used two strips successfully, and have also doubled up – with 2 strips per box – with no ill effects. Except to the mites.

      If you’re leaving the strips in for a full ten weeks it’s worth scraping any wax/propolis off them half way through the treatment. This reactivates them and gives better killing in the latter part of the course of treatment.


  4. Julia

    Oooh! One more active substance for your list: beta acids (such as those found in hop guard). I’ve found them far more effective then thymol products, tho, not strong enough to provide a solid “end of the summer” mite wipe-out. (Sad story omitted.)


    1. David Post author

      Hi Julia

      As far as I’m aware approval of Hopguard in Great Britain/Northern Ireland was never granted as Vita withdrew their licensing application. I commented previously that: “There appear to have been problems with toxicity and the risk/benefit analysis. If you read the European Medicines Agency report it appears that there were some reasonably significant shortcomings in the data submitted for assessment or the way the data had been obtained.”

      Good to know that it works for you. Interesting that you consider it better than Thymol products. I used Apiguard for several years (before I moved) and was happy with the protection it afforded the colonies. You’re in the Pacific North West (I think?) and so probably have a cooler, damper climate like we have in Scotland. I wouldn’t be confident using Apiguard here.

      There are lots of active substances … but relatively few that are approved for use 🙁


  5. Eric Beaumont

    Hello, David.

    I recall the book ‘Varroa – still a problem in the 21st Century?’ which stated that the efficacy of pyrethroids would return after a break of use for three or so years.

    The explanation given was that resistance uses varroa energy and so reduces mating efficiency, and that varroa soon abandons its adaptions when the chemical is no longer used.

    At this point it is once again vulnerable to the same chemical, which suggests that rotation of product is necessary to maintain effectiveness.

    Does this remain a valid explanation on which to plan the use of treatments which would otherwise lead to resistance?


    1. David Post author

      Hello Eric

      I’ve not read the book but I have written about Varroa resistance to pyrethroids. The evidence that pyrethroid resistance may disappear is based upon sequencing the gene encoding the target for tau-fluvalinate, a voltage gated sodium channel (VGSC).

      The detail is in that last link … 40/40 Apistan-treated mites contained the relevant mutation conferring resistance whereas only 25/239 Apistan-untreated mites (or more correctly, mites from historically untreated colonies) contained the mutation.

      I was a speaker at the Welsh BKA conference in 2017. In a Q&A session (with Wally Shaw and others) I made the point that the sale and use of Apistan should be banned for four years out of every five (or perhaps I said ‘allowed only every 3 years’ or something similar). When use was allowed it would then be exceptionally effective.

      But beekeepers don’t work like that 🙁

      Some buy near to out-of-date stock in the sales and then hoard it for a year or two. They use packets opened the previous season. They ignore the wording on the Apistan instructions which recommends “conduct(ing) appropriate testing (eg. Vita/NBU test or Beltsville test) to determine whether resistant mites are present prior to treating the colony”. They don’t like being told what they can and cannot do. Look at the brouhaha that followed the requirement to register Varroa infestation.

      Not all beekeepers of course, but sufficient to probably make any meaningful control of fluvalinates worthless.

      It’s been known that Apistan is pretty much ineffective for a long time (a decade or more, due to the widespread distribution of resistance) but the suppliers keep selling it and beekeepers keep buying it. I was in Thorne’s a week ago collecting an order. There were several packets of Apistan piled up with other pre-orders for beekeepers in the Fife and Midlothian region 🙁

      Finally – rant nearly over (!) – it’s worth making clear that I’m not aware of studies that have formally tested the detrimental impact of the VGCS mutation. How long does treatment need to be avoided to result in loss of the mutation in the population (the original paper was a bit vague on that point)? Do tau-fluvalinate residues in commercial foundation result in continued selection for resistance (assumed, but I’m not sure formally proven either)? Is the mutation lost faster for example in TBH owners or those who use foundationless frames?

      All of these would be valid and valuable studies … because, when it works, it works exceptionally well.


  6. Roy Haynes

    Hello David
    With regard to MAQS, you say that the product characteristics specifically state that it should not be used for colonies with less than 6 frames of brood. What it ACTUALLY says is that the CLUSTER should cover a minimum of 6 frames (approximately 10,000 bees). That is a significant difference. It also says that the colony should be disturbed as little as possible during application. What you don’t want to do is carry out a full inspection (to determine how much brood there is) and then apply the strips immediately. From bitter experience, this is likely to lead to queen loss. Best to leave a few days between the last inspection and the application of MAQS strips.

    What the instructions don’t seem to mention is that the colony is likely to ‘beard out’ on the front of the hive for the first two or three days of treatment. Therefore when checking the weather forecast you need to pay attention not just to temperature, but to likelihood of rainfall.

    In defence of MAQS, it is only a week’s treatment so you don’t have to guess what the the temperatures might be in 3 or 4 weeks time. The instructions also state that it is not essential to remove the strips after the 7 day period as the bees will dispose of the spent strips. Formic acid is not lipophilic and so will not leave residues in the comb.

    Best wishes
    Roy Haynes

    1. David Post author

      Thanks for the correction Roy … I’ll amend the text. It does mean that MAQS isn’t really suitable for many nucs (which are typically 5 frames).

      Both the instructions (Product literature) and the Summary of Product Characteristics available on the VMD database mention bearding … on pages 11 and 4 respectively “Colonies are expected to expand the cluster as part of controlling vapour concentration during the first 3 days of treatment. Bearding behaviour may be observed”.

      I’ve not checked whether the Product literature is exactly the same as the instruction distributed with the product because I don’t have access to the latter, but it usually is.

      I don’t think MAQS needs defending … it has a lot going for it. No residues as you say, short window of treatment, organic, compatible with the presence of honey supers and – the biggie – active against mites under cappings.


  7. Adam Darling

    Hello David,
    I have used MAQS a few times but the results are too variable – sometimes the bees tolerate it ok, othertimes bees crawl out of the hive to die in their hundreds on the grass outside or I lose a queen. And I have never used it in warm weather above 20C.
    The instructions, I recall, say to put the strips between boxes in a double brood colony. If the queen is above, there’s no means of getting any fresh air – I wonder if that could contribute to queen mortality? (And what do they die of?). And if you use MAQS when wasps are about and make a gap between boxes, do wasps get in perhaps? And how effective is it when there’s a gap? Are we reducing it’s efficiency too much?
    With putting the strips between brood (at 35C), the temperature is considerably higher that the ‘maximum’ outside temperature reccomended, so the strips are definitiely going to fume more compared to putting them on top of a single-box colony where the cluster 3 or 4 inches below the top of the frames, and under a thin plywood crown board where the temperature will be a lot closer to the outside ambient, so the treatment will be less harsh.

    I like the idea of “one week and you’re done” but the treatment seems too much of a gamble and I don’t want to play russian roulette with queen mortality over a few seasons to confirm the best way of using it with my hives and in my part of the world. It’s only for colonies that I don’t mind losing the queen in. (And that’s not many!).

    1. David Post author

      Hi Adam

      “Russian roulette with queen mortality” isn’t really a ringing endorsement 😉

      Interesting comments. I’d also thought about the ‘strips IN the cluster’ (i.e. at the interface of two brood boxes) being at ~35°C as well but had reasoned that this temperature drops off pretty fast once you reach the edge of brood being reared. Stores are at a lower temperature for example. My reading of the instructions doesn’t suggest you need to provide headspace around strips between brood boxes.

      One of the more disconcerting parts of the instructions are those on supersedure “Do not destroy queen cells that may be observed prior
      to, or post treatment. Supercedure, even if thought to be set in motion by treatment, is a natural process, and should be allowed to proceed for the health of the colony. Verify queen-right one month after treatment.”

      I’m not convinced that supersedure ‘induced by treatment’ is a natural process (and would argue whether this is supersedure or replacements being reared under the emergency response). I also think that ‘enforced supersedure’ is probably pretty risky at this time of the season. Drone numbers are plummeting and the chances of getting a new queen properly mated must be less good than earlier in the season.


  8. idse

    thanks David for these very instructive articles. Not too far south from you, I have got into a winter system of a partially filled super below the brood, to provide stores and reduce wind-draught reaching and cooling the cluster.
    IF I then treat the brood with apivar strips for 6 or 10 weeks this autumn, does that render those super frames un-usable for honey extraction next summer (having been moved back to the top in the spring).
    instructions so far found say:
    1-do not extract honey from the brood chamber (fine – understood)
    2-do not harvest honey when treatment is in place (fine – I am looking at a 6-9 month delay)
    3- do not recycle brood frames as honey frames (does that actually cover the super frames in the winter setup described above??)

    or is this a very silly question, with no demonstrable value in placing the super below the brood box in the first place….?

    1. David Post author

      Hello Idse

      I can’t comment on the last point “no demonstrable value in placing the super below the brood box in the first place” as I’ve not done this enough to make any meaningful comparisons. Perhaps 10% of my colonies go into the winter with a nadired super.

      However, I have written previously about Amitraz residue in wax. I don’t reuse supers that have been present when treatment was on the hives. It’s probably an overabundance of caution, but I’d prefer to be as certain as possible that the honey I produce hasn’t come from frames containing miticide residues.

      For reasons I don’t have time to relate, I have one colony that was treated with two nearly-sealed supers in place. Both those supers got marked with a ‘X’ and will be nadired on a couple of colonies over the winter. In the spring I’ll melt out the super frames before the colony expands down into them.

      An alternative approach, still giving you the potential benefits of the nadired super overwinter, would be to:

      • remove all supers before treatment
      • treat and feed
      • remove Apivar strips
      • add nadired super half full of sealed stores

      It takes 30 seconds to do the last of these and causes almost no disruption to the colony. It gives you the additional stores and the draught-proofing (has anyone actually measured this?! … the OMF almost certainly acts as a pretty good baffle I’d have thought) but leaves the super unexposed to the miticide.


      1. idse

        Thank you very much. Incredibly prompt answer. Gracious of you to not refer me simply to your amitraz-residue-in-wax article (which I clearly had not found).
        Very clear information.
        I had thought of the ‘remove super – treat – replace super’ option, but not sure how long after strip treatment the bees still carry around, and potentially spread, amitraz or degradation products: how long after treatment do drug levels and spread become insignificant?
        Miticide in bought wax foundation is both fascinating and scary – every day is a school day.

        1. David Post author

          Hello Idse

          I’m going to be writing about this later in the year. We know something about the half-life of amitraz breakdown products … but I don’t think there are any studies of spread within the hive. Miticide residues in commercial foundation is fascinating and scary an unpleasant reality 🙁

          I would do the ‘remove super – treat – replace super’ option without worrying too much.


  9. Vince Poulin

    David – back from the “North” and completed Alcohol washes on 4-hives. Results are good but I’m hesitant to rely fully on them which means taking a precautionary approach regardless of results. Hive 1 – 2 alcohol samples having an average 1.6% infestation (1.1% and 2.4%, N=444). . No mites in a July 31 sample of 286 bees for this hive. Hive 2, single sample N= 377 with 3.7% infestation. This is the highest result. This hive had the most mites on July 31, at 0.7% infestation, N=288 bees. Hive 3. Very good at 0.5% for a sample of N=409 bees. A 0 result on July 31. Hive 4. Pretty good at 1.3% infestation, N-303. This hive had mites on July 31 at 0.7% (N=289). Science says all hives have mites. Hive 2 exceeds 3% infestation which is the recommended threshold for treatment (3 bees/100bees). That leaves 3 hives with mites levels of 0.5-1.6%. Those are well below recommended treatment levels but – you have commented within a single apiary if you treat one – treat them all. The argument being due to drift and otherwise mite transfers. With respect to drift I haven’t seen it. I have 3 different strains of bees in the apiary. Carnolian, Buckfast and Italian. They are all distinctive in appearance. To date I have not seen say an Italian attempt to enter the other hives. Nor a Carnolian – (distinctively black) try it either . Nor have I noticed an odd strain in a box while doing inspections. Hive 2 needs treatment of certain and flowing a precautionary approach you will almost certainly suggest treating all the hives. I don’t like the idea of having any mites in the hives. We’ve chatted about this. However at 0.5%-1.6% do I treat all of them? I must say I do like moving frames around and having Apivar in the hives means this becomes limited in the future. That said, I lost all hives last year to mites – it would be a small price to pay.

    1. David Post author

      Hi Vince

      Welcome back. Drifting is low level but usually present unless your hives are well separated. I seem to remember the figure of 1% of bees on orientation flights over a three day period. Clearly with brood rearing now reducing there will be less orientation flights and less chance of drifting.

      However, in its place – and particularly prevalent at this time of the season – you might have robbing. Seeley’s work suggests that this is a more significant source of mite transfer. If colonies are similar strength (and strong) you might well get away with only treating the one with high mite levels. Perhaps consider robbing screens as well? They’re not something I have much (well, any!) experience of, but I understand the principle and they’re cheap and easy to build. You don’t want healthy strong colonies robbing weak, mite infested colonies. The latter should have screens fitted.

      The other thing to consider is when your season ends. It’s all over here on the east coast, but there’s another 1-2 weeks to go on the west (though likely curtailed by rubbish weather next week 🙁 ). As soon as I take the honey supers off I treat everything, whatever the mite levels. If you’ve only got a fortnight or so to go (and you’re not interested in the late season nectars, or happy to leave it for the bees) you could hold off treating any hives, and then treat them all.

      With 1.5% infestation (phoretic mite level) a colony at this time of the season probably contains several hundred mites not under cappings. Assuming 10-20% of mites are phoretic (and let’s be generous and assume the latter because the laying rate of the Q is reducing) you’ve probably got 2000+ mites in the hive. Once it gets cold all of these will be phoretic … but by that time a lot of your winter bees will have been feasted on by mites late in the season. Where will these mites go once the Q starts rearing brood after midwinter?

      These numbers are arm waving guesstimates, they depend upon the accuracy of your washes, the size of your colonies and – most questionable – my maths (!). However, they give you a scale of the problem.


      1. vince poulin

        Again, thanks greatly David for the guidance. It’s a done deal. I was lucky. A local supplier split a 50-pack of Apivar down to 10 which helped me set up the treatments. I began treatment in Hive 4 which had the highest infestation level – 3.7% (alcohol wash). This hive had a disturbing drone uncapping mite density of 20.7% which from last year’s data is in that threshold-collapse range. So hard to say I caught it in time. I will treat all the hives. They all have mites but are in better shape with alcohol tests showing 0.5, 1.3 and 1.6% infestation levels. Your “math” on mites of up to 2,000 in a hive at 1.6% tells us how risky it is to have any number of mites over even 1% in a hive. These little suckers are nasty predators that hide easily in the shadow of thousands of summer bees. I have three mid-season splits to test this week once these 4-hives are completed. I’m prepared to treat them if needed. What makes treatments challenging is shuffling brood boxes. Warre hives employ small volume boxes which end up stacking quite high. My wintering goal is to manage them as a single Lang deep. This means bringing the stack down to 2-brood boxes and 3 if that is not easily managed. Each Warre box is similar in volume to a 5-frame Lang (NUC), thus 2 of my brood boxes = a single Lang deep. Not sure how this works out with your Nationals. Maybe 1.5 National volume???). You had comments on drifting and robbing. I do place Wasp Excluders on all the hives. I think these may be similar to what you refer to as a Robber Screen. Wasps are at their peak at the moment. An interesting observation – wasps have shown greatest interest in Hive 4. This is the hive with the highest infestation level (3.6%). I’m guessing they can sense the hive has a high level of mites and also a reason to think I may be too late with the treatment for this hive. I gave thought to doing a quick OA treatment so as to hit phoetic mites hard before setting the hive up with Apivar but did not follow through on it. It seems like a strategy that could have value where a particular hive has been hit hard early on by mites. It makes sense Apivar needs some number of days before it is effective.

        1. David Post author

          Hi Vince

          Two quick comments … wasps will always target a weaker colony rather than a strong one. It’s possible that your hive with high mite levels is already perceived as being weaker by the wasps. I doubt they can directly sense the mite level.

          In my experience Apivar works pretty quickly on phoretic mites. We have done studies on very heavily infested colonies (thousands of phoretic mites) and shown that mite drop is maximal on days one and two after treatment, but pretty much flatlines after 6-7 days. If you’re seeing an extended period of drop from an Apivar treated hive it’s almost certainly mites emerging from freshly uncapped brood.


          1. vince poulin

            David – you are “spot-on” with expected mite drop after Apivar treatment. Dual treatments using OA is not required. I have begun treatments on 5 hives with 4-hives now providing data. Unfortunately 1-hive had a “leaky” mite board. Bees filled the “drop-chamber” and literally swept the board clean of any mites or debris. So – results in hand for 3 hives. Hive 1: 1.3% infestation (alcohol wash test result) Daily drop (3-counts) since treatment – 96, 38, 50. Hive 2: 3.7% infestation (had the mite board problem) – only 5, 7 and 19 could be counted. HIve 3: 0.5% infestation has seen 19, 19, and 17 mites. Hive 4: 1.6% infestation has had only 2-days post treatment. Mite counts 105 and 110 respectively. Your above comment of 100-200 phoretic mites in a hive at say 1.6% infestation may well be “spot-on”. Hive 1 (1.3%) is up to 184 mites in the drop (3-days post T) and Hive 4 (1.6%) is at 215 (2-days post T). These results suggest several other things, 1. Treatment timing is late but I had no other chance of treating earlier. 2. Hives with 1.3% and 1.6% alcohol wash test results support well your recommendation to treat all hives within the apiary when one theoretically needs treatment. I did have the one test result at 3.7% which is the one hive that exceeded the common treatment recommendation threshold of 3%. With mite drops of 184 and 215 recorded for Hives 1 and 4 we can see that the 3% threshold could mess up a lot of bee keepers. I have only done a few alcohol wash tests – most of my previous monitoring was with drone uncapping. But – for one hive (Hive 4) I took 2 samples. The first sample result was 1.1% (not bad) but the second came in at 2.3% – nearly 2X higher. These were samples of N=277 bees and N=167 bees respectively. Average was 1.6% infestation. All said – we have no assurance mites are spread out within a hive randomly. I did scoop samples from active brood frames where one can expect more mites given just days away from them dropping inside cells. You can also see a quite large variation in drone uncapping results. I’ve seen many segments of comb where mite densities are much higher than other spots. All this is to say these tests still very valuable. They are not perfect but the best we can do. Nothing wrong with working with a good dose of humility – if you find mites don’t let them fool you. Of course I’m not out of the woods. These are significant drops and as you have pointed only a small proportion of mites in the colony – 10X could already be present capped brood. I hope not! But for certain a possibiliey. I unfortunately had to leave at a critical time period and would have treated much earlier if I could. It is helpful knowing a bunch of phoretic mites have been eliminated and we have begun a 42-day treatment regime. Most of the hives have large amounts of eggs and pre-capped larva. This should help. Several had less suggesting the queens may only have begun thinking about producing winter bees. We may get lucky!

          2. David Post author

            Hi Vince

            I’m not sure how your climate compares to ours … to accompany your comprehensive mite records it would be worth making a note of the temperature etc as the season draws to a close. You may find, after some more years of recording the same sort of information, that a pattern emerges that will inform good or bad practice going forwards. I’ve got very interested in the influence of local weather on hive activity and am dabbling with some experiments (low science, not high science!) now, with some more planned for the winter and for the next queen rearing season. I already keep rough and ready notes on the change of the seasons – migrant bird arrival/leaving dates, particular forage availability etc. Here on the west coast it’s all rather new this year, but on the east coast I know the season started very late and ended at about the average time. That means I can probably expect below-average mite infestation levels (though I’m not formally monitoring them) and a broodless period sometime toward early November, unless we have a crazy warm ‘Indian summer’.

            And there’s not much chance of that this far north 🙁

            I think we can be pretty certain that mite distribution is far from random in the hive. As you suggest, they’re preferentially on the young bees occupying the centre of the brood nest and involved in brood management and cell preparation duties.


  10. Rory Fields

    Hi David, immensely helpful blog as ever – have you experimented or heard much about the long-term efficacy of Varromed? Obviously it’s a combination of Formic and Oxalic acid but I’m no scientist or academic and have found it impossible to find corroboration of how well it works – it seems poorly represented (whilst even ‘bee-gyms’ get a mention) in most guides to IPM and yet if it’s a very effective product, it’s simplicity itself to administer and not vastly expensive if you refrigerate and have >40 hives. Thanks for your time, Rory

    1. David Post author

      Hello Rory

      I’ve no experience with Varromed. The VMD database is very coy about the product and, like you, I’ve not managed to find much about efficacy – there’s an overview here, but don’t expect much detail. You can find the Product Information on the EMA website here. One of the striking things is the recommended treatment regime; 3-5 treatments can be applied in autumn alone (with winter or spring treatments suggested as well). Considering the reported damage oxalic acid does to open brood this is surprising. I’m going to be writing about some of the apparent internal contradictions (why is Api-Bioxal not approved for multiple repeat treatments by trickling?) in a future post … no wonder beekeepers find things confusing 🙁


  11. vince poulin

    David – these threads are now over 2-months old but a comment you made when talking about a 1.6% alcohol test was – possibly 100-200 phoretic mites at large in the hive. That was actually spot on as I noted in a followup post. In that hive on Day 1 and 2 of start of treatment it dropped 215 mites. Further on you estimated those numbers suggest up to 2000+ mites in the hive. “These numbers are arm waving guesstimates, they depend upon the accuracy of your washes, the size of your colonies and – most questionable – my maths (!). However, they give you a scale of the problem.” That quote is yours. I’m 69-days post start of treatment and have now a total mite drop of N=2223. I’d say those waving arms were on the mark. Now – my counts are not absolute. Bees getting into the mite board chamber often “wing-swept” areas clean. I think mites in those drop areas likely lost to the count. Mites on bees that die away from the hive, daily counts sometimes obscured by debris – especially paper from hive swap, and once a long period of time where counts not possible. But for sure – they are in the “order-of-magnitude” for assessment purposes. I just recently read an excellent article from University of Florida on mites and their life history:, (James Ellis, et. al.). Of great interest were the life stages from egg to adult mite inside the cell. And of particular importance was learning of 3-molt stages. My latest counts included what I called “juvenile” mites. They were white to translucent mites. I intuitively thought them to be “juveniles” even when placing a few under the scope. With this read I then took fresh samples and had a really good look and indeed they were not mites, but the 1 of 3 molts sheds. Why I’ve seen most of those now is likely due to fewer winter bees emerging. There is little brood left in any of the hives and only 2 have shown either cappings or recent DWV bees. I also found interesting male mites are smaller in size, yellow to tan in colour. In the past I would have claimed them to be “juvenile” mites due to my lack of understanding. All this is to say – maybe dust off one of your old posts and see how much clarity you have provide visitors on just these topics. A few images – etc., good detail on how all this shakes out from a life-history stand point. Just a though!

    1. David Post author

      Hi Vince

      Finally got round to responding (apologies!). When I count mites I only count the mated females that are either phoretic (on workers identified during an alcohol wash) or that fall through the OMF during treatment. The main reason for this is that my eyesight is probably not good enough to routinely count anything smaller! However, these are also the only ones that are important. The male cannot live outside the capped cell I don’t think (I’d need to check Jamie’s article to be sure).

      Reassuring (and not a little surprising) that my maths was in the right ballpark. I suppose by the law of averages I had to be about right sometime?!

      Have a look at the Varroa Model I discuss in the recent post … it allows you to input wash counts and determine when best to treat. I think (I’ve yet to explore it fully) that you can also explore a range of other interventions, including enforced brood breaks and treatment.


  12. Malcolm Wilkie

    I have a question. Can one sublimate with oxalic acid a box of bees that has a new queen that has not yet come into lay?

    1. David Post author

      Hello Malcolm

      You can.

      But I guess that’s not really the Q.

      I suspect you mean “Can I sublimate without damaging the queen or her prospects of mating (or starting to lay if she’s already mated)?” I’m not sure I have a definitive answer to that question. However, I always OA treat swarms that arrive in bait hives before the Q starts laying – both prime swarms and casts – and I don’t remember an instance when the Q has failed. I’ve also stated elsewhere that I’d prefer to risk losing the Q than to have a mite-infested colony.

      It’s worth noting that if you’re going to do this you must a) avoid any chance of interrupting a mating flight, and b) ensure as many bees are in the box as possible. I therefore treat late in the evening.


  13. Sharon Campbell

    Hi David
    I have just been reading your blog from last year. If you are still answering I have the following question. I am treating my hive with Apivar strips and having been doing a mite count every few days since 3rd September when I began treatment. The mite drop was heaviest straight after treatment (70 per day) and it dropped to 20 per day after 3 weeks. The number of mites has now gone up to 50 per day on the last 2 counts. (now 4 weeks since the treatment started) Is this normal? I am on a brood and a half on a WBC and there is still a lot of pollen going in. The temperature are around 17C during the day and lots of late summer forage available so I presume the Queen is actively laying.
    I live in the south of England UK

    1. David Post author

      Hello Sharon

      The high initial drop is the phoretic mites being killed off. After that the mite drop reflects the mites released from emerging brood. This will vary depending upon the laying rate if the queen 21 days earlier (and on a few other factors, but the laying rate is the major one). You can expect the laying rate to vary a bit, so don’t worry. Apivar needs 6-10 weeks, effectively at least 2 complete brood cycles, and you should see the mite drop significantly less towards the end of the treatment period. If (as I suspect) you have lots of brood in the hive, treat for the full 10 week period.


      PS Comments are turned off on posts after 2 years (often because I’ve updated the topic)


Leave a Reply

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