Varroa treatment: measure first, treat second. Not the other way round.

I run an apiary large enough, and spread across enough sites, that I long ago stopped believing in treating "by feel". Every year I see the same thing among fellow beekeepers and among beginners: the honey comes off, someone digs out whatever was left from last season, drops it into the hives and considers the job done. Because "that is how I have always done it". The trouble is, varroa could not care less how you did it in the past. All it cares about is how many mites are sitting in a particular colony right here, right now, and you will not read that off from your armchair or your habits.

In this article I want to sort out three things: how to measure infestation, what we actually treat with today, and at what times. I start with diagnosis, because over the years I have learned that this is what decides the outcome, not the choice of product. Treating without measuring is shooting in the dark. Sometimes you hit, more often you waste the product, breed resistant mites, or, in the worst case, only realise the colony is beyond saving when you see bees with deformed wings. And by then it is over.

Why treating "blind" leads nowhere

Varroa is not a yes-or-no matter. Mites are present in practically every colony, and the question is never "are they there" but "how many". Over the years I have seen that two colonies standing side by side, on the same site, run identically, can carry wildly different mite loads. On top of that, every colony tolerates the parasite differently. What one colony shrugs off will flatten another. That is why there is no single magic threshold that fits everything, and anyone who sells you one is oversimplifying.

If you can see mites crawling across the floor with the naked eye, or bees with deformed wings, the situation is simple: you are late, and badly so. Visible symptoms are not a signal to "treat now", they are a signal that you have lost this round. The whole art is in reacting much earlier, and earlier only works when you measure regularly.

There is another side to this that gets talked about less: resistance pressure. Amitraz, which most apiaries run on, is not forever. The more often and the more thoughtlessly we pour it in, the faster mites become resistant to it, and in some regions of Europe and in the USA this is already a documented fact, described among others by Rinkevich and by researchers in France. Treating without a diagnosis is the shortest route to standing there in a few years with empty hands and a product that no longer works. Rotating substances and measuring the effect is not fussiness, it is the condition for having anything left to treat with a decade from now.

And here we come to the point I consider the most important in the whole puzzle. Diagnosis serves two purposes, not one. The first is obvious: knowing when a colony needs help. The second tends to be forgotten, and it matters just as much: knowing when that help is not needed, so you do not treat without reason. Because there is no varroa treatment without a cost to the bees. Every treatment, whether hard chemistry or acids, is to some degree a poisoning of the colony. Organic acids irritate the bees, can temporarily damage brood and weaken the colony, and in poor conditions can kill the queen. Studies on 65 percent formic acid have shown a temporary decline in brood after treatment, and queen losses following formic acid are well known from practice. Amitraz and thymol, in turn, leave residues in the wax. And the queen is the most sensitive point here, because she decides the colony's future, and she often dies not from varroa but from a treatment carried out without need or in the wrong weather. The conclusion is simple: if infestation is low, the best thing you can do for a colony is to leave it alone. That is why I measure before I reach for any product, so that I treat the colonies that genuinely need it and do not poison the ones that are coping.

How to measure infestation, concretely

Let me start with what you even count the drop on, because the choice of floor changes the whole job.

Open mesh floor versus solid floor

An open mesh floor, screened with a slide-out tray, simplifies the counting of the drop itself. Mites dropping off the bees fall through the mesh onto the tray below, where the bees can no longer reach them, so nobody carries them out, ants do not eat them and the result is clean. You slide in a tray, pull it after a few days and count. Convenient, especially when you are testing in series.

I myself, however, run on solid floors, with no mesh, and I have a concrete reason for it. A solid floor is not just a place where the drop collects, it is also a mirror of the colony. If the bees have a mess on it, mould, plenty of dead bees, dragged-out cappings or ejected brood, I can see at once that something is off, before I even open the nest. That information is worth as much to me as the drop itself, and it trains you to look at the colony as a whole rather than through the lens of a single number. The drop can still be read off a solid floor, you just slide in a tray or a sheet lightly smeared with petroleum jelly so the mites stick and nobody tidies them away. You only have to remember the correction: without a tray, some mites are carried out or dragged off by ants and wax moth, so the raw reading tends to be understated.

It is a matter of working style, not dogma. A mesh floor gives a cleaner drop count and less work in the counting, a solid floor gives a broader picture of the colony's state. Do whatever suits you, as long as it is deliberate and with a correction for what a given floor shows and what it does not.

Measurement methods

Natural drop on a tray. The simplest and cheapest thing you can do, and practically hands-off. You slide in a clean tray for three to seven days, on a solid floor lightly smeared with petroleum jelly so the mites stick, then you count and divide by the number of days for a daily drop. No killing of bees, no fuss. Excellent for watching the trend and, more importantly, for checking how well a treatment worked, because the drop after treatment is the most honest proof that the product did its job.

Powdered sugar method. You take about 300 bees, roughly half a cup, ideally nurse bees working on open brood, since those are the ones most crawling with mites. You tip them into a jar with mesh instead of a lid, add a spoon of powdered sugar, roll them, wait a moment and shake the contents out over a white plate of water. The sugar dissolves, the mites float up and you count them easily. The bees go back to the hive alive. More work than the tray, but you get a hard number on the spot.

Alcohol wash, or flotation. The same as the sugar method, only harsher, because those 300 bees will not be coming back. In return it is the most accurate of the home methods, because the alcohol strips off literally every mite, nothing hides. I use it where I want a certain result rather than an estimate, for instance when assessing colonies earmarked for breeding.

Drone brood inspection. You uncap a patch of sealed drone brood and see how many mites are sitting under the cappings. Drone brood is varroa's favourite quarters, so the scale is obvious at once. It is also a biotechnical method in its own right, but I will come back to that.

My everyday tool is the Varroa EasyCheck. It is a ready-made cup that does sugar, alcohol or carbon dioxide flotation in a convenient, repeatable form. In a larger apiary it is my main tool for a quick measurement: I take a sample, run the test and have the infestation percentage on the spot, without messing about with a jar and a plate. The principle is the same as the powdered sugar method or the alcohol wash, only the execution is faster and more repeatable from colony to colony, and in serial testing that makes an enormous difference.

Remember one thing: do not measure a single colony and project the result onto the whole apiary. Infestation varies between colonies even on one site. Test several colonies from each location, only then do you get a real picture.

When to treat, thresholds in numbers

There is no universal threshold, because colonies tolerate infestation differently. But there are concrete numbers I go by, and they are what tell me when there is a signal to act. The most sensible way is to measure infestation as a percentage, that is how many mites per 100 bees in the sample, because that figure is comparable regardless of colony strength.

On a sample of about 300 bees, by the sugar method or more accurately by an alcohol wash, the conversion is simple: divide the mite count by three and you have the percentage. In spring, when colonies are small, I already react at 1 to 2 mites per 100 bees. In the thick of the season the threshold is usually 2 to 3 percent, that is roughly 6 to 9 mites on the full sample of 300. The most important thing, though, is tightening the threshold in the second half of summer. As the winter bees start to emerge, I want to be down around 2 percent, and on sites with a high risk of drift from other apiaries even lower. This is not overkill but logic: these bees have to survive winter, so every mite now counts double. If in August the sample shows well above these figures, say a dozen or more mites per 300 bees, that is no longer a "signal to treat", it is a sign that you are late and the winter generation is already partly spoiled.

A note on method: the powdered sugar roll understates the result by roughly a third to a half compared with the alcohol wash, so if you count with sugar, multiply the reading by about one and a half to get closer to reality. The alcohol wash and the EasyCheck in alcohol mode give the value closest to the truth.

With the natural drop you look at the number of mites falling per day. In July a healthy picture is below 1 mite a day. Several or more a day in midsummer is already a red light, and a drop running into the teens per day means an infestation you should have moved on earlier. Just remember that the drop is an approximate measure, heavily dependent on colony strength and time of year, and on a solid floor without a tray it tends to be understated on top of that. For the treatment decision itself the alcohol wash or the EasyCheck is more reliable, while the drop is better for watching the trend and checking after a treatment.

After every treatment you measure again. If the drop is still high or the sample still shows more than 1 mite per 100 bees, you go in a second time. That is the whole point: you have hard numbers before and after, not a feeling that something got done.

What we treat with, an overview

I split the available arsenal into three baskets: hard chemistry (amitraz, flumethrin), organic acids and oils, and biotechnics. A good strategy is always a thought-out combination of the three, not clinging to one product for life.

Before I get to specifics, one thing that explains most of the rules below. In season, with brood in full swing, the bulk of the mite population, comfortably 70 to 85 percent, sits hidden and reproducing inside the sealed brood. On the adult bees you have only a minority at any one time. This is not my theory, it is the basic biology of the parasite, well described in the review by Rosenkranz and colleagues from 2010. That is why any product acting only on the bees, without penetrating the cappings, deals at most with that minority in a colony with brood. This is the key to understanding why some products require a broodless colony and others do not.

Formic acid, my mainstay for years

I will start with what I run on myself and have for several seasons, because it is my main product and I trust it completely. Formic acid has a quality nothing else available without a prescription has: it is the only one that penetrates the cappings and reaches the mites reproducing inside the brood. That changes the whole logic. I do not have to catch a broodless window, I do not have to cage queens, I do not have to juggle timings, because the acid reaches varroa even where the other products have no entry.

On top of that it is a natural product, it does not leave the problematic residues in wax and honey that hard chemistry does, and it is versatile, because it works in the thick of the season, with brood at maximum. For me it simply gives the best results, and that is why it became the mainstay rather than a mere add-on.

How I use it: my carrier is a ceramic dispenser, that is a plate or disc that releases the acid evenly by evaporation. And here is the most important thing, because with formic acid this is no place for improvisation. I take the dose straight from the plate manufacturer's instructions and stick to it to the letter, because it is the build of the particular dispenser that determines how much acid it holds and how fast it releases it. No averaged number of millilitres picked up online will replace the instructions for your own carrier, and with formic acid a mistake in the dose ends in a dead queen. In practice you use concentrated formic acid (85 percent), and manufacturers state that a single plate holds around 40 ml, with halves giving a gentler, more even evaporation that the bees tolerate better. I lay the plate on the frames, the acid releases gradually over a few days, and I repeat the treatment only when the drop after the first round shows it is needed, not on a fixed schedule. Instead of plates you can use a ready-made evaporator or an absorbent pad, but the rule is the same: you follow the carrier's instructions, not a number you overheard.

If you prefer everything set out step by step, the plate manufacturers provide ready instructions with doses and soaking method [insert link to the plate manufacturer's instructions here]. And always work in protective gear, because 85 percent formic acid is corrosive: acid-resistant gloves, goggles, a mask with an acid-vapour filter and good ventilation are not fussiness but basic safety.

One thing demands discipline: temperature. Formic acid is temperamental in heat and on hot days can harm the queen and drive the bees to fanning. So I plan the treatment with the forecast in hand and do not do it in the middle of a heatwave. Anyone who does not watch this will sooner or later lose a queen and blame the acid, when the real fault was their own haste.

In fairness I will add that the efficacy of formic acid in field trials is variable and depends heavily on temperature and method of application, and immediately after treatment a temporary decline in brood has been recorded, recovering within a few weeks. For me that is not an argument against, but a reason to use it in a series of doses and always verify the drop afterwards, rather than assuming a single application settles the matter.

Amitraz, the workhorse of the apiary

The most popular active substance here, mainly because it is effective, cheap and, in Poland, subsidised under the national beekeeping support programme. It comes in two forms, and that distinction is crucial.

Fumigation, that is Apiwarol and similar tablets you burn. You drop in a glowing tablet and the smoke does the work. Simple and quick, but let us be clear: amitraz as smoke acts only on the mites on the bees and does not get under the cappings. That is why Apiwarol only makes sense in a broodless colony. Fumigate a colony full of brood and you take down that minority sitting on the bees, while the majority calmly emerges from the cells and a week later your infestation is back. Effective fumigation means either a series of treatments spread across the brood cycle or, far better, aiming at a broodless window. And since in season the colony is breeding at full tilt, you have to create that broodless window yourself. This is where queen caging or the Scandinavian method come in, which I will get to shortly.

Strips, that is Biowar, Apivar, Apistrip. You hang two per colony, within the nest where the bees are densest, and leave them for several weeks: Biowar usually 6 to 8 weeks, Apivar 8, and up to 10 with brood present. This is not accidental but by design: the strip releases amitraz slowly across the whole multi-week period and keeps mite mortality elevated through a full brood cycle. Thanks to that it also catches the mites emerging gradually from the cells, and that is its advantage over smoke. It is why strips cope despite brood while fumigation does not. There is one drawback, but a serious one: you must remember to remove them on time. Strips left in the hive permanently are the textbook way to breed resistant varroa.

Also on strips is flumethrin (Bayvarol), a different substance but similar logic. I keep it mainly for rotation, so as not to hammer the mites with the same thing over and over.

Oxalic acid and the other acids

Oxalic acid is the classic of the broodless period. You apply it by trickling, that is dribbling a water-and-sugar solution over the bees in the seams, in a ratio of roughly 7.5 g of crystals to 100 g of sugar and 100 g of water, or by sublimation with a vaporiser. In a broodless colony efficacy can reach 90 to 95 percent, as the review by Rademacher and Harz confirms, and it costs pennies. Importantly, unlike hard chemistry, there is so far no evidence of varroa resistance to oxalic acid or of significant accumulation in wax, so it is a substance you can rely on long-term without spoiling your own future. But, like amitraz smoke, it acts only on the mites on the bees. If there is brood, a good part of the varroa is hidden under the cappings and the acid will not touch it. You can see this clearly in the numbers. In Gregorc's study the same oxalic treatments carried out in colonies with brood gave barely around 40 percent efficacy, and repeated in autumn, in the broodless period, close to 90 percent. Same substance, same application, and the difference comes solely from the presence of brood. That is why oxalic acid is a weapon for late autumn and winter, once the queen has stopped laying. Do not overuse it, because too high a dose or too frequent repetition weakens the bees. In practice once per generation, that is separately on the summer bees and separately on the winter bees, is entirely enough.

Oxybee and Api-Bioxal are ready-made oxalic acid preparations, for those who do not want to mix their own. VarroMed is a combination of oxalic and formic acid, a ready suspension for trickling, for use during low flight activity, so in spring and autumn.

Thymol in plates or gel (ApiLifeVar, Thymovar, Apiguard) is a thyme oil that works through vapour over a longer period. Like oxalic acid, it does not penetrate the cappings, so for it to work at all its concentration in the hive air must be maintained through a full brood cycle, in practice about a month. It is also strongly temperature-dependent, because thymol vapour is produced more abundantly the warmer it is. Field trials show a spread here from around 40 percent efficacy in the cold to 90 percent and more in warmer conditions, with ApiLifeVar coming out the most stable regardless of the weather. I treat thymol as a supplementary option, useful in September warmth, but weak as a sole line of defence, especially in a cooler climate.

Biotechnics, queen caging and the Scandinavian method

This is not eco for the sake of eco, but a real tool that lowers varroa pressure and cuts the amount of chemistry you have to pour into the apiary.

The simplest and very effective is cutting out drone brood. Varroa loves drone comb, so by systematically cutting out a frame of sealed drone brood you carry a good share of the mites out of the colony along with it. There is one condition here that is easy to forget: the drone comb must already have been laid up once and properly capped before you cut it. The whole mechanism rests on the fact that the mite enters the cell just before capping and gets shut inside with the larva, so only sealed drone brood holds the varroa in. Cut the comb too early, while still open, and you catch practically nothing, because either the parasite is not in there yet or it has time to move out. From experience it even looks as though, if you keep offering the bees carelessly managed drone comb, the varroa in time, oddly enough, starts to avoid it. That is why it pays to run this consistently: let the bees draw and lay up the frame, wait for it to cap, and only then cut. Field studies confirm that regular removal of sealed drone brood significantly slows the build-up of the mite population in season, so it is not folk lore but a real tool that limits pressure. You do it during inspections, and along the way you relieve the swarming impulse. Two birds with one stone.

A more serious tool is queen caging. You confine the queen in a cage or on a single comb for a couple of weeks so she stops laying. Once the last brood has emerged and all the varroa is out on the bees, you treat, most often with oxalic acid or amitraz fumigation, and hit the exposed mites with very high efficacy. The essence is that you artificially create a broodless period in midseason, and only then does a product that acts on the bees alone get full scope.

A second version of the same idea is the Scandinavian method, that is shaking bees onto foundation. You turn the colony into an artificial swarm: you shake all the bees off the combs into a box with foundation only, without a single frame of brood. All the brood, and with it the bulk of the mites trapped under the cappings, you remove from the colony. What is left is bees and foundation, so the colony is broodless from the outset and all the varroa is on the adult bees. Then you treat straight away, most often Apiwarol fumigation or an oxalic trickle, and knock the mites down without mercy, because nothing is hiding under the cappings any more.

You just need to know it is a hard method on the colony. The shaken bees are left with no stores and no nest, they have to rebuild comb from foundation practically from scratch, so it goes hand in hand with feeding. And here is the detail beginners trip over: you feed thin 1:1 syrup, which stimulates the bees to secrete wax and rebuild quickly, not thick 2:1 for winter stores. Good weather and an ongoing flow also help the colony get back on its feet. The efficacy against varroa is high precisely because the treatment lands on a colony that is completely broodless.

I will say plainly why I do not do any of this myself: with formic acid, queen caging and shaking onto foundation are simply unnecessary. Since formic gets under the cappings and kills the mites in the brood anyway, the whole circus of caging queens, shuffling combs, feeding and watching timings on every colony falls away. With a handful of hives it may be interesting work, but in a larger apiary it is logistics I deliberately avoid. If, however, someone runs on amitraz or oxalic acid as their base, queen caging and the Scandinavian method are strong tools worth having up your sleeve, because without a broodless period those products will never show their full potential.

Calendar, when to do what

Varroa is fought essentially all year round, only with different tools. Below is the rhythm that works for me.

Early spring, March and April. Monitoring the drop over two to three weeks. Not so much to treat as to know what level the colonies came out of winter at. If something looks bad, I react selectively, without waiting for the season.

Spring and early summer, May and June. Biotechnics rule here, that is cutting out drone brood during inspections. And no easing off on measurement, because this is the period when the mite population grows fastest and the moment is easiest to miss.

July, right after the lime flow. This is the moment you must not miss, and I consider it one of the most important in the whole beekeeping year. The instant you take off the lime honey and the last of the surplus honey, you go in with a treatment at once, no delay. The reason is hard: this is exactly when the winter bees are being raised, the ones meant to carry the colony through to spring. If they emerge weakened and loaded with the viruses varroa carries, no wintering will save them, however hard you feed. And an important correction for our times: the winter generation is no longer just July and August. With today's longer, warmer autumns queens lay for longer, so September works on the winter bee too, and sometimes even early October. That means the treatment window stretches further than the old calendars said, and you have to keep a hand on the pulse through all of it, not tick the job off after August. This is where I go in with formic acid on ceramic dispensers, as described above, because it also reaches the mites under the cappings and does not make me wait for a broodless window.

August and September. Closing off and checking after the July treatment. If monitoring shows something is still sitting there, I go in a second time so the last generation of winter bees emerges clean. With warm autumns I do not force the topic shut in August, but keep watching the drop and the sample into September, because the colony is still breeding and still producing winter bees. This is practically the last real moment to make it before the cluster forms.

Late autumn and early winter, November and December. The classic knockdown in the broodless period. Once the queen has stopped laying and all the mites are exposed on the bees, in goes oxalic acid by trickle or sublimation, or fumigation, and this is when efficacy can be at its highest of the whole year, because nothing is hiding under the cappings. Many beekeepers do this treatment and for them it works. But I will say plainly: I no longer do it myself. I get the main work done in summer and early autumn with formic acid, which reaches varroa in the brood as well, so I go into winter with colonies clean enough that a knockdown is unnecessary. On top of that there is a real problem with today's warm autumns: a genuinely broodless period arrives later and later and is often short or never quite closes, and oxalic on a colony that still has even a patch of brood will not work as it should anyway. If you rely on the oxalic knockdown, you have to wait patiently for a real absence of brood, not do a calendar treatment in November on a hunch.

After every treatment, slide in the tray and count the drop. That is your feedback. Without it you do not know whether the treatment worked or whether you just have the feeling of a duty done.

The most common mistakes

Over the years I have seen the same errors repeated over and over, among beginners and among long-standing beekeepers alike. Here are the ones that cost the most colonies.

Treating once a season, only in late autumn. This is mistake number one and the costliest. Someone waits with the treatment until November, does an oxalic in the broodless period, knocks varroa down beautifully, almost to zero, and is pleased with himself. Except the horse has bolted. The winter bee is made from July right through September, and with today's warm autumns even later. If it emerged on a high mite load, it is already damaged, drained by mites and loaded with viruses, and no autumn clean-up will undo that. You are killing mites on corpses that will not see spring anyway. For those who do the autumn knockdown, it is at most a finishing touch, never the only treatment of the year. The main work has to fall in summer and early autumn, before the winter generation emerges.

Fumigating or trickling oxalic on a colony full of brood. I come back to this because it is a classic. Since the bulk of the mites in season sit under the cappings, a single smoke or trickle takes down only that minority on the bees. A week later the varroa emerges from the cells and you are back at the start, only with the false sense that you treated. A bee-only product plus brood in the hive equals half an effect.

Running on one substance forever. Amitraz alone, year in year out, with no rotation, is a straight road to breeding resistant varroa on your own site. In parts of Europe amitraz is already faltering for exactly this reason. Interleave hard chemistry with acids and biotechnics, do not give the mites years of quiet selection under a single product.

Strips left in the hive too long or permanently. Hung and forgotten, strips release a weaker and weaker dose over time, and a sublethal concentration is the best resistance training the parasite could ask for. On top of that, amitraz accumulates in the wax. Strips go in for a set time and come out on schedule, full stop.

Treating during a flow or with surplus honey on. No product goes into a hive holding honey meant for people. Amitraz and thymol migrate into the honey and wax, and oxalic spoils the taste. You treat after the surplus honey is off, never during it.

Believing that more means better. A double dose of oxalic or formic acid will not kill varroa faster, but it will weaken the bees and often kill the queen. Acids hold to their ratios and doses not out of caution but because exceeding them harms the colony more than the mites. A chapter of its own is formic in the heat, which can drive a colony out of the hive or kill the queen. The weather forecast is part of the treatment, not an extra.

No check after treatment and blindness to autumn reinvasion. You did a treatment, ticked it off and did not check the drop. How do you know it worked? Maybe the product was weak, maybe the varroa is already resistant. The other thing: a colony cleaned in summer can rack the infestation back up in autumn, because drift from neighbouring, neglected apiaries can let mites in through the back door. That is why monitoring does not end with the July treatment. You keep a hand on the pulse right through late autumn, until the cluster forms.

Copying the calendar instead of looking at your own colonies. The dates in this article, and in any other, are a skeleton, not gospel. A season can be early or late, sites differ in varroa pressure, and individual colonies in infestation. The calendar tells you when to look more closely, but it is the measurement that tells you when to act.

In closing

The whole article boils down to one sentence: do not treat varroa because it happens to be July, treat because you measured the infestation and know you have to. You match the product to the situation, that is to the presence of brood, the weather and the level of infestation. You rotate substances so as not to breed resistant mites. And you always check the effect, rather than taking it on faith.

It is a bit more work than dropping in a tablet and forgetting about it. But this is exactly where the line runs between a beekeeper who knows what is happening in his hives and one who counts losses every spring and scratches his head over where he went wrong. After all these years I can say one thing: the mistake was almost always the same. He did not check.

Frequently asked questions

When should you treat varroa? The most important treatment is done in summer, right after the last honey is off, most often in July after the lime flow, and continued in August, and with warm autumns into September too, because that is how long the winter bee is now being made. Some beekeepers add an oxalic acid knockdown in the broodless period in late autumn, though not everyone needs it if the summer work was done properly. You always confirm the exact timing with a measurement of infestation, not with a date from the calendar alone.

How do you check the level of varroa infestation? Most simply by the drop on the floor tray, more accurately by the powdered sugar method or an alcohol wash on a sample of about 300 bees, and selectively by inspecting sealed drone brood. It is worth testing several colonies from each site, because infestation varies between hives.

What is the varroa treatment threshold? Infestation is measured as a percentage, that is mites per 100 bees in the sample (about 300 bees, sugar or alcohol wash). In spring you already react at 1 to 2 percent, in the thick of the season at 2 to 3 percent, and in the second half of summer, as the winter bees emerge, the threshold tightens to around 2 percent, because now every mite counts double. As a secondary signal, a natural drop clearly above 1 mite a day in July is also a warning.

What is the best way to treat varroa in summer, when there is brood in the colony? In the thick of the season formic acid works well, because it is the only one that penetrates the cappings and reaches the mites in the brood. Amitraz strips also work despite brood thanks to their multi-week exposure, whereas fumigation and oxalic acid only work well in a broodless colony.

Do you have to treat varroa if infestation is low? No. Every treatment is a burden on the bees, and on the queen especially, so with low infestation it is best to leave the colonies alone. This is exactly why diagnosis matters so much: it protects you both from treating too late and from poisoning colonies without need.

Why is July so crucial? Because the winter bee is made in July and August. If it emerges on a high mite load, it will be weakened and virus-burdened, and the autumn knockdown will no longer undo that. The summer treatment saves the generation that has to survive winter.

Sources and further reading

If you want to dig deeper than blogs, these are a good place to start. This is peer-reviewed material, not forum tales.

• Rosenkranz P., Aumeier P., Ziegelmann B. (2010). Biology and control of Varroa destructor. Journal of Invertebrate Pathology 103: S96-S119. The most cited review on the biology of the mite and the methods of controlling it.

• Rademacher E., Harz M. (2006). Oxalic acid for the control of varroosis in honey bee colonies - a review. Apidologie 37: 98-120. A review of oxalic acid efficacy, confirming around 90 percent in broodless colonies.

• Gregorc A. (2005). Efficacy of Oxalic Acid and Apiguard Against Varroa Mites in Honeybee Colonies. Acta Veterinaria Brno 74: 441-447. A study showing the difference in oxalic acid efficacy in colonies with and without brood.

• Rinkevich F. D. (2020). Detection of amitraz resistance and reduced treatment efficacy in the varroa mite, Varroa destructor, within commercial beekeeping operations. PLoS ONE 15: e0227264. Documentation of the growing resistance of varroa to amitraz in commercial operations.

• Ramsey S. D. et al. (2019). Varroa destructor feeds primarily on honey bee fat body tissue and not hemolymph. PNAS 116: 1792-1801. The paper that changed our understanding of what the mite feeds on and why it is so destructive.

• Gajda A., Beekeeping compendium on varroa (SGGW). The Polish source of the infestation thresholds and monitoring methodology referenced in the text.