Counting Varroa

It’s that time of the season again. With the exception of readers in the Southern Hemisphere, Colonsay, the Isle of Man or a few favoured locations in the Highlands of Scotland, miticide treatments should be on to reduce Varroa levels.

For reasons explained elsewhere, it’s important that this is done before the winter bees are exposed to the smorgasbord of viruses that Varroa trasmits when it feeds.

It’s not sufficient to just treat. You also need to have some idea that the treatment is reducing the numbers of Varroa in the colony.

Counting by numbers

It has been determined that only 10-20% of mites in a colony are phoretic i.e. attached to emerged workers. The majority of treatments (MAQS is the current exception) only target these mites. Therefore, treatments are usually applied over a period of several weeks to ensure that mites newly emerged from capped cells are also exposed.

There are a couple of obvious ways to determine the mite load before and after treatment. These include:

  • conducting an alcohol wash test, or a sugar-roll equivalent, of workers to quantify the phoretic mites.
  • uncap a known amount of worker brood (drone brood is almost certainly absent from colonies this late in the season) to quantify mite infestation.

However, both are pretty intrusive and – with the exception of the sugar-roll – involve the sacrifice of bees or brood, so perhaps not ideal at this stage of the season. However these are the most accurate way of measuring things.

Counting the corpses

Out, damn'd mite ...

Out, damn’d mite …

Alternatively, and this is what most beekeepers do, apply the treatment and count the mite drop.

To count the mites you need some way of catching the mites. Open mesh floors (OMF) can easily be fitted with a sheet of closely-fitting (most usefully white) Correx onto which the mites drop. Restrict the access of ants and other creepy crawlies to the tray or they may steal some of the corpses. Check these on a regular basis during treatment and you have a simple way of determining whether the treatment is working.

The treatment may be working, but has it been effective?

The scores are on the floors

If you count thousands of dropped mites and that number doesn’t diminish during treatment i.e. the drop per day early and late in treatment is broadly similar, then the treatment is working, but it’s not effective or finished as there are loads of mites still left.

What you need to observe is a reduction in mite drop when comparing early and late counts.

Depending upon the treatment, the first days’ drop isn’t necessarily indicative of whether the miticide is working (or of the phoretic mite load of the colony). It may take a day or two for the treatment to achieve maximum kill. Vaporised oxalic acid often gives a better drop after 24-48 hours, and continues to work over about 5 days.

As indicated in the footnote, the numbers of brood emerging per day will expose ‘new’ mites to the miticide, increasing the count. If emerging brood levels vary, so will the mite drop … but also remember that the efficacy of the miticide also varies over time.

What you’re looking for is a hugely reduced count of mites dropped per day at the end of the full treatment period when compared with the beginning.

I usually carefully monitor the first week or two and the last week. Simples.

Objective vs. subjective counting

Easy counting ...

Easy counting …

Some beekeepers count each and every mite that appears on the trays. Others just look for ‘lots’ at the beginning and ‘almost none’ at the end. I consider >50/day is ‘lots’ and only count smaller numbers.

The less frequently you count the more difficult it is to discriminate dead mites from all the other detritus that accumulates on the trays. The cell cappings, the pollen that’s being dropped, the wax scales and various other bits of bee, all make spotting the mites more tricky.

The larger the area you’re counting the more likely it is to either double-count or miss mites. Make life a bit easier by ruling a simple grid onto the tray and counting square by square.

Scrape the tray clean after counting the mites … if you leave the tray dirty you’ll end up double counting and struggling to spot mites that are knee-deep in the crud that’s fallen through the OMF.

Don’t try this at home

Varroa are a pretty regular size and shape. And colour for that matter. At least adult mites are. This raises the possibility – though perhaps only to those with a tendency towards geekiness – to try and count mites automagically§.

Rather than stand around the apiary squinting through myopic eyes at tiny reddish ovals you could simply photograph the tray and then process the image later.

Been there, done that … or at least tried to.

Fiji ...

Fiji …

There’s a freely-available, well-supported, image analysis package called ImageJ (also distributed sometimes as the auto-referential Fiji … Fiji is just ImageJ). It’s possible to count objects using ImageJ having set criteria that define them.

As an exercise in near-futility I’ve attempted to do this for Varroa. You first need to ensure the Varroa are of a standardised size and shade by scaling the image appropriately and correcting the colour. This can be done by using a photographers grey card of a known size, placed to the side of the Varroa tray. You then use this as a reference to scale the image and define the white balance.

Finally, you define the size, roundness and shade of a Varroa and process the image in ImageJß. It counts the mites and provides an overlay with each identified mite numbered. You’re then able to check whether it’s missed any.

It does.

Consistently variable

This is the point I’ve got stuck at … the accuracy is all over the place but it’s clearly not impossible. Problems include:

  • It overlooks mites lying on their ‘edges’, perhaps propped up on a speck of pollen or fragment of wax. Better colour definition and a wider range of ‘ovality’ might sort this out.
  • It misses mites lying immediately next to another mite – these look like 8 or ∞ rather than a simple solid oval. I’ve no clear solution to this other than counting lower densities of mites.
  • It ignores some mites that appear as ‘doughnuts’ because of reflection from the shiny carapace. Don’t use flash for the photography.
  • It counts some ovalish, reddish lumps of pollen that are about the right size as mites. D’oh!

At best the accuracy is above 80%, but it’s variable. The lack of consistency is the major issue. If it was always 80% it would be perfectly acceptable and a very fast way to record mite numbers. At worst – usually when the tray is messy and mite numbers are relatively low – it’s well below 50%.

This is an intriguing beekeeping-related task for long winter nights. If you’re a geek. My ambition is to take a quick smartphone photo, scrape the Correx tray clean and then (automagically!) do the counting at home with a cup of tea and piece of cake.

I’ll keep persevering … particularly with the tea and cake 😉

† It’s currently Spring in the Souther Hemisphere, so the wrong time to treat. The remaining locations (and Australia) have no Varroa so have no need to treat. Lucky blighters.

‡ This is a gross oversimplification. Obviously, a broodless colony will only have phoretic mites. Swarms that issue from colonies take 35% of the mites with them, leaving 65% on the remaining bees (or capped in cells). The actual number of phoretic mites likely depends upon the prior history of egg laying by the queen. It also is probably influenced by the overall level of mites in the colony (or ratio of uncapped brood to mites perhaps). I’m not sure if anyone has modelled this successfully, though it might be possible to do this with BEEHAVE.

§ Automagically is pretty obviously a concatenation of automatic and magic. It is usually defined as “(especially in relation to the operation of a computer process) automatically and in a way that seems ingenious, inexplicable, or magical”. Interestingly, the term was first used in the 1940’s, well before the advent of computers.

ß Once I’ve got this working better I’ll provide some instructions … in the meantime the menus that you need to use are Analyse … Set Measurement and Analyse … Count Particles. Image scaling needs to be done first in ImageJ. Currently I do the white balance in Adobe Lightroom (which is overkill, but convenient as all my images go through this software).


Integrity …



… make sure your stored broods and supers have it. For that matter, your hives need it as well.

Early autumn is when wasps appear in droves. They gatecrash picnics and pester our bees as they search for carbohydrates. This year, after a pretty poor summer, they’re a bit later that usual … at least in this part of the UK. Queen wasps have now stopped rearing brood – which requires a protein-rich diet for the colony – and they’ve now switch to carb-bingeing.

My apiaries are away from my house, so the first thing I become aware of is the increased numbers of wasps investigating the stacked up piles of broods and supers. ‘Wet’ supers containing residual honey after extraction are very attractive to the wasps. They look for any structural weakness in the stack … a poorly fitting roof, a warped crownboard or gaps at the joints in the boxes.

If any of these provide access there will be a never-ending stream of wasps making return trips from their nest site – which may be up to a mile away – and the source of the honey. When you walk past the boxes you can sometimes hear the rustling or scratching of the wasps as they look for additional routes out once they’re laden with honey.

They can even chew through poly boxes where there’s a hint of  structural weakness to gain entry or subsequent exit from the pile. In the picture (above) the corner of the upper Swienty poly super – at the point where the hive tool had slightly indented the relatively soft polystyrene – was mercilessly attacked, leading to a neat chewed hole large enough to give the wasps access.

Structural integrity

Maintain neat stacks of equipment, securely sealed at the base and the top. Try and avoid having an open mesh floor at the bottom of the stack, even if the entrance is sealed up. The scent of honey will still attract wasps from far and wide … and once they’re there they are very persistent.

I use sheets of spare Correx or solid split boards with all the entrances closed up tightly on the bottom of the stack. If you don’t have anything suitable a sheet of thick polythene forms a reasonably impenetrable barrier – to scent and wasps.

If there are gaps, and there probably will be as equipment ages and warps or gets bashed about, seal up the gaps as soon as you notice them. Waterproof gaffer tape is as good as anything for this as a temporary fix.


It’s not only the integrity of empty stacks of equipment that is being tested at this time of the season. Wasps will show similar levels of interest in colonies of all sizes – from double-brood monsters bulging with bees to mini-nucs containing only a cupful of workers and a queen.

Life’s a lot easier for the colony if they only have a single entrance to defend. Check the joints and junctions between boxes and seal any up where wasps might gain access. This includes the – entirely unnecessary in my view – ventilation holes in the roof. Unless these have well-fitted mesh there’s a chance the jaspers can get in and from there to the honey supers if your crownboard also has another – entirely unnecessary – ventilation hole.

Kewl floor entrance slot

Kewl floor entrance slot

Life’s easier still for the colony if the entrance is small and/or easy to defend. The kewl floors I favour, with an “L-shaped” entrance tunnel, provide ample opportunity for the guard bees to challenge any inquisitive wasps – at the entrance per se, or where the tunnel opens into the brood box. I don’t think I’ve ever had to provide additional protection to full-sized colonies using one of these floors.

Unbalanced or nearly-colonies

Small colonies, recently created colonies or colonies that are otherwise weakened are a different matter altogether. In all of these the colony is either understrength or lacking a full complement of worker types.

Reduced entrance ...

Reduced entrance …

Or both.

For example, newly created nucs containing a late-mated queen and a couple of frames of brood and adhering bees, may well not be up to defending themselves properly against the determined attention of wasps (which is the only sort of attention wasps are capable of).

Mini-nucs or queenless colonies are particularly susceptible and regularly succumb to attacks by wasps.

Maximise the opportunities these colonies have to defend themselves by restricting the entrance to a very narrow gap. The gaping maw of Thorne’s Everynucs definitely needs reduction, either by closing off 80% of it with grass stuffed into the gap (in an emergency) or using foam or Varroa mesh offcuts to achieve the same end.

Don’t be sloppy

Slopping large quantities of sugar-rich syrup about the apiary is a sure-fire way to attract the striped hordes. Make sure feeders are watertight (syrup-tight) and don’t leak. Pour carefully and mop up any spills that do occur.

I almost exclusively use fondant for late-season feeding. Spills are non-existent. The block is cut in half with a spade (or breadknife) and placed inverted on top of the colony. It takes seconds and works well. As an aside, you can feed fondant and treat with Apiguard at the same time – I’ve been asked about this recently.

Don’t leave offcuts of brace comb lying about in the apiary. Tidy up after you, don’t conduct unnecessary inspections or leave unattended frames propped up against the hive stand while you slowly go though a colony.

All of the above help the colonies avoid the attention of wasps … they also reduce the likelihood of colonies being robbed out by other bees, a topic I’ve mentioned before and will discuss in the future as it has very significant implications for disease transmission.

† Jasper is a slang name for wasps I was aware of as a boy in North London. The origin of the term appears unknown, but QI has an interesting discussion on it. It’s variously ascribed to Dorset, Lincolnshire (East Midlands), Yorkshire and Glasgow. It’s also not clear the derivation of the word … ‘jasper’ sounds like the genus Vespa which our common wasps belong to. Alternatively, there’s a striped mineral called Jasper which (potentially, though not to me) resembles the striped abdomen of wasps.

That’s all folks

That's all Folks

That’s all Folks

It’s late August and the end of my least successful beekeeping year ever. That sounds very negative, so perhaps it should be qualified. It’s the end of my least successful beekeeping year in terms of honey production.

However, in terms of the satisfaction I’ve got from my beekeeping, it’s been a pretty good year. Let’s examine these two things separately, dealing with the bad news first.

Tell ’em about the honey, mummy

My production colonies only generated about 25lb each of Spring honey. Some of this was clearly oil seed rape (OSR) as there were fields just about in range, but much of it was essentially mixed hedgerow and tree nectar, and none the worse for that. This was all extracted in late May or early June and is now stored, set, in buckets. Later in the year, once the temperature drops, I’ll prepare soft set honey for sale or distribution to friends and family.

25lb is firmly at the bottom end of the averages over the last few years though – in fairness – It’s only my second Fife Spring, so I don’t have much recently to compare it with. Colonies were doing well when I first inspected them, but in some cases that wasn’t until early May. The active beekeeping season is only 4-5 months long here (latitude 56.3° N).

June started well, with clear weather and high temperatures.

And then it started to rain. And continued for almost the entire month.

Lime can yield well in July

Lime can yield well in July …

None of my full-size colonies needed feeding, but most reduced their brood rearing. July nectar flows were poor. The lime yielded a small amount of very high quality honey, but for whatever reason – poor weather, colonies not strong enough, patchy flows – pretty-much nothing else. The summer honey was extracted in mid-August and is already disappearing fast.

I didn’t take any colonies to the heather as I was abroad for a chunk of July when I’d need to be preparing and shifting them to the moors. And, in all likelihood, they probably weren’t strong enough anyway.

And that was it … like last year, all over much sooner than expected.

There’s some balsam in central Fife along the River Eden that might give some late-season nectar and there’s ivy (but that is some way off flowering yet) but I usually let the bees keep anything they collect once the summer honey is extracted.

Flowering ivy

Flowering ivy

And the good news is

Beekeeping isn’t all about honey. There’s also tremendous satisfaction to be gained from working with the colonies, improving your stock and feeling that – although perhaps not in complete control – you’ve got a pretty good grasp of what’s happening and how things are going.

In this regard, 2017 was a success.

I know I lost one swarm (actually a cast from the queenless half of a split). I got a call to say that the apiary was thick with bees but they’d long gone by the time I extricated myself from meetings and got home. In itself this wasn’t a success. However, I learned my lesson and managed to hive a second cast that issued from the same colony a day or two later. I also had success with my bait hives.

With a couple of exceptions my vertical splits went well, with the resultant queens both laying well and heading well-behaved colonies. The couple that didn’t work developed into (drone) laying workers and were dealt with successfully by uniting.

In retrospect, considering the weather in early/mid-June I’m astounded any queens managed to get out and mate. By late July colonies headed by these newly mated queens were developing well, with frame after frame of brood exhibiting a pretty respectable laying pattern.

That'll do nicely

That’ll do nicely …

Throughout the season I had a pretty good idea what was happening in most of my colonies. There were no big surprises … “Oops, a virgin queen, where did she come from?”, or “Grrrr … no queen, no eggs and no swarm cells, I’m stumped”.

Colonies behaved in a thoroughly predictable manner. Strong ones were caught before they swarmed, split and were merged back to a double brood box. Nucs developed pretty well, though they needed close attention and some emergency feeding through June. No drama, no panic.

The end of the summer season, other than the truly woeful honey yield, has left me with a good number of nicely behaved and generally very strong colonies. As always there’s one exception, but I’ll unite that weakling late this week if things haven’t picked up.

All the gear, no some idea

Split board ...

Split board …

Gradually equipment standardisation is starting to pay dividends. I ran out of almost nothing (I certainly didn’t run out of supers 🙁 ) and managed to mix’n’match as needed to leave colonies secure, watertight and with the proper bee space when needed. Homemade split boards ended up being pressed into service as floors and it’s clear I’ll have to make some additional kewl floors this winter.

Bamboo-strengthened foundationless frames were a great success. Furthermore, I prepared a second batch mid-season and never got round to using them, so have plenty to start the season next year. Result! However, it’s sobering to realise that one of the reasons they weren’t used was that the nectar flow simply wasn’t strong enough to get them drawn properly.

Finally, whilst we’re on the subject of equipment, I’ve used about half a dozen Abelo poly hives this year in addition to the usual Swienty boxes with homemade floors and roofs. First impressions of the Abelo boxes are pretty positive and I’ll write something up later in the year on them.

Season’s end … or the start of the new season?

Late summer and autumn is an important time in the beekeeping year. Some even consider it the start of the next season, as success in the subsequent year is very dependent upon the preparation in the preceding autumn.

Feed'n'treat ...

Feed’n’treat …

All my colonies are scarfing down large quantities of fondant at the moment. They’ll all get another few kilograms as the autumn progresses. Unless there’s good reason to, it’s unlikely any colonies will be inspected again until Spring.

Varroa treatment is ongoing and the mite drop from most colonies is reassuringly low. I count the mites from each colony over a two week period. Over the first 5 days, some dropped just single figures …

All colonies are coordinately treated to maximise decimation of the mite population at a time when bees have a tendency to drift more and/or rob adjacent colonies – both being well-documented routes by which Varroa can be transmitted between hives. I’ve also helped a neighbouring beekeeper (with colonies within range of my own apiary) by loaning out my Sublimox so that, together, the mite population at a landscape-scale is reduced.

This is simple common sense. I don’t want my (nearly) mite-free colonies infested from neighbouring apiaries and I also don’t want the colonies I do have with appreciable mite levels (~50+ after 5 days treatment) to infest others.


It’s far too soon for much serious thought about 2018. However, I already know there are going to be some major changes to my beekeeping. The local Council have just announced that they will shortly (Spring next year) build a new road literally through the middle of my bee shed and apiary … finding a new location and getting things rebuilt is my major focus at the moment.

And finally … it’s harvest time and raining again …

Mainly dry ...

Mainly dry …

† Tell ’em about the honey, mummy was a catchphrase from a TV advert for Sugar Puffs breakfast cereal. The advert aired from 1976 to ’85 and featured the Honey Monster and Henry McGee (from the Benny Hill show).

Henry is the one on the right.

They don’t make advertising like that any longer. For obvious reasons.

‡ Scarf is American slang meaning to ‘eat voraciously’. It’s probably a bastardisation of the word scoff. Scarf has other meanings and I strongly suggest you don’t look these up.


That's All Folks

That’s All Folks

The phrase That’s all folks dates back to 1930 when it was used on the closing screen of a Warner Bros. Looney Tune cartoon.

Over the years many different characters used this line on both Looney Tunes and Merrie Melodies cartoons. Mel Blanc (1908-’89), the actor who voiced (stuttered) the most famous version … Th-th-th-that’s all folks! has the engraving That’s All Folks on his gravestone.

There’s a 1949 Merrie Melodies cartoon called The Bee-Deviled Bruin with the Three Bears, a colony of bees and a shortage of honey for breakfast. Typical slapstick ensues. It ends with That’s all folks”.

Right here, right now

In February 2016 I posted an article on When to treat colonies with miticides. It was read by subscribers, generated a bit of discussion – in particular on Apiguard use – and then disappeared into the howling wilderness that is the interwebs …

Google hides the searches that drives most of the website traffic to this site. However, it has been found … as is clear from the access stats (below), it is now being accessed extensively.

This reflects the change in the seasons as beekeepers turn their thoughts from harvesting honey to protecting their colonies from the ravages of Varroa and the viruses it transmits.

'When to treat' stats ...

‘When to treat’ stats …

It’s worth reiterating it here, though this won’t be news to regular readers, Varroa itself is probably not the problem. The problem is the smorgasbord of viruses that Varroa transmits when feeding on the haemolymph (blood) of honey bee pupae.

Benign and virulent

Most important of these viruses is deformed wing virus (DWV). This virus has a sort of Jekyll and Hyde personality. It’s probably present in all honey bees, transmitted between bees while larvae are being reared and during trophylaxis (the regurgitation of liquid food between bees). Under these conditions, and in the absence of Varroa the virus is probably benign.

DWV symptoms

DWV symptoms

Of course, it’s difficult to test whether it is really benign as there probably aren’t any bees that lack DWV. Even bees that have never been exposed to Varroa, such as the black bees on Colonsay, have DWV. Let’s assume that, even if not benign, it has minimal detrimental effect on the bees.

Varroa changes the route by which DWV is transmitted. Instead of being orally transferred – a route the bees have probably evolved to cope with – Varroa bypasses any defence mechanisms by ‘injecting’ DWV directly into the blood. Under these conditions DWV reproduces rampantly – for reasons that have yet to be determined – causing the ‘deformed wing’ symptoms most beekeepers are familiar with.

Symptomatic adult, or recently emerged, bees can contribute little or nothing to the colony.

Live fast, die young

But there’s more bad news. Asymptomatic adult bees with high levels of DWV have a shorter lifespan and die prematurely. This is perhaps not an issue during the heady days of summer when the turnover of worker bees is at its height – the queen is laying well, perhaps 1500-2000 eggs per day, the colony is bulging and individual workers “live fast and die young” after about 6 weeks.

Formally, I don’t think it’s been shown that mid-season workers have a shorter life span when they have high levels of DWV. What is known, and what is much more important, is that winter bees die prematurely if their DWV levels are high.

Winter bees are the ones with high levels of fat in their bodies. These are the bees that get the colony through the winter. Some might live for 5-6 months in the UK, and it’s known they can live for up to 9 months. If these bees die early, in the absence of any significant brood rearing, the colony dwindles and dies.

Game over.

Preventing the inevitable

Varroa transmits DWV and results in high levels of DWV. High levels of DWV in winter bees shortens their lifespan and results in colony losses. How can you prevent the inevitable?

In the absence of ways to directly control DWV levels (these are in development but you’re then tackling the symptom, not the cause) the only way to do this is to prevent the transmission of DWV to the winter bees by Varroa in the first place.

And you do this by applying effective miticides early enough that the winter bees are protected from exposure to Varroa, and the viruses it transmits.

How early is early?

I discussed this in the earlier article and am working on a more nuanced version at the moment. Essentially – and I’m writing this in mid-August – the answer is now or very soon.

In the definitive publication demonstrating the premature death of winter bees by DWV, Peter Neumann and colleagues detected a measurable reduction in longevity as early as November in the colonies they studied. These bees were age-marked and had emerged 50 days earlier. The eggs had therefore been laid in the first week of September … and been capped (together with any Varroa) as pupae in mid-September.

Therefore, treatments to reduce Varroa should be completed by mid-September to protect the winter bees. Since many treatments take ~4 weeks the time to treat is right now.


There’s climatic variation between parts of the UK and Bern, where the study by Peter Neumann was conducted. There’s also seasonal variation year on year.

In the balmy South the dates will be later than in the frigid North. In cool years with an early autumn they will be earlier, whereas an Indian summer will delay the need to treat.

Treatments are generally incompatible with a honey flow. If you take your bees to the heather you have to balance collecting a late heather crop with protecting the bees from the ravages of Varroa.

it’s not possible, or wise, to be dogmatic about precise dates … other than to say that miticide treatment is generally required earlier than you think to protect the winter bees from DWV.

Better late than never?

Well, yes, but the damage may well have been done.

The annual survey of beekeepers in Scotland regularly includes significant numbers using Apiguard in October. Close, but no cigar (actually, not even close) … it’s probably too late to reduce Varroa levels in a meaningful way, and it’s unlikely to be effective anyway as Apiguard needs average temperatures around 15°C.

All of the comments above on the timing of treatment make the assumption that the treatment is effective – the right dose, the right duration, the absence of resistance etc.

Finally, it’s worth noting that starting treatment in mid/late August does not reduce Varroa levels to the lowest achievable levels. Treating later in the year does this, because more mites are phoretic and ‘reachable’ by the treatment. To reduce mite levels to the minimum you also have to also treat midwinter … something for another post.

† Live fast, die young was the title of a biography of the actor James Dean by James Gilmore. It’s a popular phrase, being used for a movie and several song titles. The extended version Live fast, die young and have a good looking corpse, often wrongly attributed to James Dean, actually came from the 1947 book Knock on Any Door by Willard Motley.

‡ Close but no cigar is a mid-20 phrase from the USA. It dates back to the time when fairground stalls gave out cigars as prizes.


Right here, right now is a song released in April ’99 by Fatboy Slim (Norman Cook) from the album You’ve come a long way, baby‘. If you appreciate evolution you’ll enjoy the video …

… but you’ll need to like beat/dance music to appreciate the track.

Extrapolating Asian Hornets

Extrapolation in mathematical terms means “the extension of a graph, curve, or range of values by inferring unknown values from trends in the known data”. There’s a rather poor scientific joke which involves ‘extrapolating the line’ from a single data point.

Here’s the same joke from the incomparable

Extrapolating (with thanks to

… which brings us neatly to the Asian hornet (Vespa velutina). This invasive species predates honey bees and other pollinators and is a threat to beekeeping for two reasons:

  • by killing honey bees (which comprise > 50% of the diet) it can destroy colonies
  • by hunting at hive entrances it prevents bees from foraging, so reducing honey yields
Vespa velutina

Vespa velutina …

As many will be aware, at those who members of a beekeeping association or internet-savvy, an Asian hornet nest was discovered in Tetbury, near Bristol, in September 2016. Although the nest was destroyed (and other nests were not found) there remains the concern that the Asian hornet may be established here. If that’s the case, or if when it arrives again, how far and how fast will it spread in the UK?

A recent publication by Matt Keeling and colleagues has attempted to to address this. You’ll appreciate the relevance of the cartoon above … August 2016, no hornet nests, September 2016, one hornet nest. How many can we expect in the future?

A long way from home

Asian hornet distribution ...

Asian hornet distribution …

Unsurprisingly (the clue is in the name) the Asian hornet comes from Asia. The native distribution includes tropical South East Asia where it exists in a number of different colour morphs or sub-species. However, it spreads – naturally on its own and with the inadvertent help of man – and is now established in Korea, Japan and North West Europe.

The Asian Hornet appeared in South West France in 2004. It was thought that overwintering queens were imported with a shipment of pottery from China (it is the V. velutina nigrithorax subspecies – see map for native distribution). Since 2004 the Asian hornet has spread widely across France, Spain, Belgium, Portugal and Italy. You can view a dynamic Google map showing location and time of detection here.

It’s a short hop across the Channel from Northern France for an insect that flies strongly and disperses widely, so there was every expectation that the Asian hornet would appear in the UK in due course. Remember that bluetongue virus of sheep was introduced to the UK in 2006 by tiny Culicoides midges wafted across the Channel on the prevailing South Easterly winds.

Brexit or not, the Asian hornet can easily get to Southern England.

Closer to home

Asian hornets were first found in Tetbury, Gloucestershire, in September 2016 and subsequently in North Somerset. A nest was found in Tetbury and destroyed. Haplotype analysis demonstrated that, unsurprisingly, the hornets were related to those in Northern France. Worryingly, this analysis also demonstrated that the hornets were related, but no closer than at the grandparent level.

This suggests one of two scenarios. Either there were two separate incursions from France by hornets that were closely related or the Asian hornet was established in the UK before 2016 and the Somerset and Gloucestershire finds represent progeny (two generations later) of this initial invasion.

There have been no further reported Asian hornet finds on the mainland (as of early August 2017 and remembering that absence of evidence doesn’t mean evidence of absence) though additional nests have been found on Jersey and it’s possible that the hornet is established in the Channel Islands.

How far, how fast?

Whether or not the Tetbury hornet incursion has been contained there remains the question about how far and fast will the Asian hornet spread in the UK.

This is the sort of thing that can be modelled by epidemiologists. Taking account of a variety of factors such as the numbers of queens produced per nest, the environmental suitability for the hornet, the dispersal distances and – importantly for a species originating in tropical regions – the latitude (and hence temperature).

Much of the real numbers for these various factors come from studies of the spread of the Asian hornet in France. For example, the mean dispersal distance of new queens is about 18 miles i.e. Tetbury to central London in about 5 generations.

Predicted spread of the Asian hornet in UK …

In the figure above A-D are the early years (<5), E and F are the predictions for 10 and 20 years hence. Remember, this is an inexact science as the predictions are being made from a very small founder population. G shows the confidence limits which, at 90% (the palest shading) are very broad.

Is Scotland safe?

As a beekeeper in Scotland the potential to spread North is a concern … those of you living further South can simply jump ahead to the conclusions.

A key feature of the prediction Keeling and colleagues made was that increasing latitudes (and possibly altitude going by the pale shading of North Wales and the Pennines, though this could also be lack of suitable nesting environments – the Asian hornet prefers oak trees apparently) will restrict breeding success and hence spread due to reduced temperatures.

They modelled this by assuming reproductive success/spread (actually queens/nest) was 100% in Andernos-les-Bains, France (where much of the data used for the modelling was derived) and 0% in the North of England. This mirrors the situation for the European hornet which is much less common in Northern England. Assuming a linear decrease in queens per nest over this distance they reproductive success in Tetbury is only 38% that in Andernos-les-Bains. 

With these assumptions, Scotland is safe. I should add that I’m talking about natural distribution and spread. A single Asian hornet was discovered at a “retail warehouse in the central belt of Scotland” in March 2017. Although there’s no further official news about this discovery it seems likely it was imported ‘on the back of a lorry’ (and the SASA/NBU people will be able to tell where from after genetic analysis) to one of the giant distribution warehouses in Motherwell or Livingston. If the assumptions made about reproductive success at increasing latitudes are correct, this type of introduction is unlikely to result in the Asian hornet getting established in Scotland.

All of the modelling discussed so far assumes no intervention by FERA or by protective and vigilant beekeepers. The impact of nest destruction will be discussed in the future … but don’t get your hopes up 😉


Asian hornet nest

Asian hornet nest …

With only one (or two?) incursions and no further reports in 2017 it’s possible that the Asian hornet is yet to become established in mainland UK. Once here – and I think it’s pretty certain it will get here and become established – the speed with which it spreads may look like the model proposed here. If the underlying assumptions on reproductive success, dispersal distance etc. are representative of the situation in the UK then – without effective intervention – we can expect it to be widespread within a couple of decades.

However, a feature of all epidemiological modelling is that they are subject to revision and reinterpretation … inevitably leading to further publications.

† Don’t mistake the Asian or Yellow-legged hornet (Vespa velutina) with the Asian giant hornet (Vespa bloodyenormousandterrifying). The latter has a wingspan of 7.5cm and its proper scientific name is Vespa mandarinia. A number of newspaper articles in autumn 2016 made this mistake. Many also chose to use images of the European hornet (Vespa crabro) or other large stingy-looking insects … none of which are likely to help the public correctly identify the invasive Asian hornet.

Vespa mandarinia

Vespa mandarinia …

Peaceful easy feeling

The 6-8 week period between late June and harvesting the summer honey is a quiet period in the beekeeping calendar. At least, it is in mine. My colonies aren’t going to the heather, so there’s nothing to prepare for that. Swarm control is complete and many colonies are now headed by new queens, so the chance of swarming is minimal. The spring honey – what little there was of it on account of the incessant rain – was extracted in late June. It’s now easy going until the summer honey is taken off and the colonies are prepared for winter.

Inspect, or just observe?

The 7 day cycle of inspections that are so important as the season builds up become much less critical. If there’s a new mated, laying queen in a box with ample space, sufficient supers and enough stores (for adverse periods of weather) there’s actually little to be achieved by rummaging through the box on a weekly basis.

Instead, I generally just observe things from the outside. If pollen is being taken in by foragers, if there are good numbers of bees on orientation flights during warm mornings and if the hive is reassuringly heavy, then there’s probably no need to inspect weekly. A peek through a perspex crownboard can give a pretty good idea of how much space the colony has and whether they’re fully utilising the super. With experience, hefting the hive (gently lifting the back an inch or two and judging the weight) is a good indication of whether they might need an additional super. And that’s it … I generally leave these strong, healthy colonies to simply get on with things during July and into August.

But inspect when appropriate

Of course, some hives will need checking. For example, any hives that are clearly below-strength for an unknown reason should be carefully checked for signs of disease. Varroa levels can be readily, albeit pretty inaccurately, determined by putting a Correx Varroa tray below the open mesh floor and the colony should be inspected for obvious signs of deformed wing virus (DWV) symptoms.

High levels of DWV

High levels of DWV …

If there’s any doubt about the health of the colony consult a good book on the subject (Ted Hooper’s Bees and Honey is a reasonable start though some of the more comprehensively illustrated newer books might be better), ask your mentor or an experienced local association beekeeper and contact the local bee inspector if necessary.

Chronic Bee Paralysis Virus (CBPV) is a high-season problem for big, strong colonies. Sick bees exhibit characteristic shaking or shivering symptoms, look oily or greasy and accumulate in a large smelly pile below the hive entrance. A very distressing sight. I’ll be discussing CBPV in more detail over the next few months as it appears to be an increasing problem.

Queen problems

The other colonies I keep a close eye on are those with known or potential queen problems. These include colonies where the queen may not have mated, or those in which the queen appears to have got mated but the colony shows signs of early supercedure, suggesting that all is not right.

Hopalong Cassidy ...

Hopalong Cassidy …

The queen in the (rather poor) photograph above has a paralysed left rear leg. She’s a 2017 queen and emerged in early/mid June during a period of very poor weather. I found her as a skittish virgin very soon after emergence (quite possibly the day of emergence) then left the colony to get on with things. She was mated by the first week in July. Eggs were present but I didn’t see her in the colony. However, she wasn’t laying particularly well, either in terms of number or pattern.

Since I was disappearing to Malaysia on business for 10 days in late July I thoroughly inspected the colony before leaving. I discovered her hobbling around the frame, clearly with very severely impaired abilities. There was very little open or sealed brood in the colony. In the several minutes I watched her she didn’t lay any eggs despite checking lots of cells that looked polished and ready to me (but I accept she’s probably a better judge of a suitable cell than I am). She clearly could lay, and you can see an egg at the tip of her abdomen. I suspect that, although her walking wasn’t grossly impaired, she was unable to properly ‘reverse’ into the cell.

Not a bee ...

Not a bee …

Don’t delay, act today

Mid-July, a strongish colony with almost no brood, a crippled queen and no means of checking things for a fortnight meant that prompt action was needed. I removed the queen and united the colony over the top of another strong colony. The alternative was to wait and see if the colony disposed of her, or tried to supercede her. Either would have imposed a delay of about a month after my return, there were limited numbers of larvae for the colony to start from, a rapidly ageing worker population and little chance of the colony building up strongly through the autumn to overwinter successfully. This was a case of using them or potentially losing them.

I’ve no idea how the queen came to have a gammy leg. I’d not seen her since she’d been mated. One possibility is that two queens emerged at or near the same time, duelled in the hive leaving one dead and the other partially crippled. Although damaged, the queen still managed to leave the hive to mate successfully, but then struggled to lay properly.

We’ll never know.

Late evening

Finally, if you’ve not visited your apiary late on a warm, calm summer evening then you really should. Strong colonies can be heard from some distance away, a sort of low humming sound. There’s the heady smell of warm honey in the air as they evaporate off water from stored nectar in preparation for capping stores off for the winter ahead. Highly recommended.

† Gammy meaning (especially of a leg) unable to function normally because of injury or chronic pain … in contrast to the fictional cowboy Hopalong Cassidy used to label the image. Hopalong Cassidy had a wooden leg.


Peaceful easy feeling was the title of a song by the Eagles released in 1972 on their debut album (Eagles). The band, or what’s left of them after the recent death of Glenn Frey, continue to play live with four concerts last month.

Don Henley has just turned 70 and should really Take it Easy 😉

Size matters

Anyone reading the will be aware that there are a number of contributors there that enthusiastically recommend the treatment of colonies with vaporised (or, perhaps more accurately, sublimated) oxalic acid to reduce Varroa levels.

There goes a few pence ...

There goes a few pence …

Although vaporised oxalic acid (OA) has been used by some for many years, the speed with which it has recently been embraced by many UK beekeepers (at least those that contribute to discussion forums and, perhaps to a lesser extent, those I speak to in associations over the winter) probably reflects two or three things:

  • an awareness of just how effective oxalic acid is as a treatment
  • the increased availability of commercial oxalic acid vaporisers (or Heath Robinson-like plans to build-your-own)
  • the huge price-differential between oxalic acid and most other treatments

There are almost as many homegrown or imported vaporisers as there are treatment regimes to hammer down the mite levels. Of course, there’s the contentious point that oxalic acid is not approved by the VMD (Veterinary Medicines Directorate), despite having been in routine use for decades. Api-Bioxal is, but is probably unsuitable for sublimation due to the inert (as far as Varroa are concerned) additives it contains. Api-Bioxal can be vaporised but leaves a caramelised residue in the vaporiser pan that is hard to clean.

Out, damn'd mite ...

Out, damn’d mite …

‘Vaping’ is also popular in the US. Randy Oliver has covered it extensively on his site and it’s also regularly discussed on Beesource. OxaVap make/supply a vaporiser that appears very similar to the Sublimox I use. The OxaVap model has a useful temperature display that I would find much easier to read than the red/green diodes on the Sublimox … I’m red/green colourblind.

Active and passive vaporisers

The Sublimox and OxaVap vaporisers are ‘active’ … they blow out a dense cloud of OA-containing vapour through a relatively narrow diameter nozzle (the video below uses water to demonstrate this process). This provides advantages both in terms of ease and speed of delivery. These vaporisers simply need a 7mm hole drilled through the sidewall of the floor (see photo at the top of the page), or through an eke placed over the colony. The OA-containing vapour is ‘squirted’ in, permeates all corners of the hive within seconds and you can then move on to the next hive. The vaporiser doesn’t need cooling between treatments and the dose administered is tightly controlled.

Big Daddy

However, OA dosage isn’t critical. It has been shown to be well-tolerated by bees in studies from groups in the UK and Germany. If the dose isn’t critical and speed really is important then perhaps consider the vmVaporizer. At $3600 it’s about ten times the price of a Sublimox.

vmVaporizer ...

vmVaporizer …

The manufacturers claim you can treat 300 hives an hour with one of these … one every 12 seconds. For comparison, the Sublimox takes 20-30 seconds per hive. However, what takes the time is sealing the hive, moving the generator about, unsealing the hive etc. so you’d need a team of (well protected) helpers and some closely spaced hives to achieve a similar rate. The vmVaporizer is mains (110V) powered so would also need a generator or inverter.

The video above demonstrates the vmVaporizer in action. It produces copious amounts of oxalic acid vapour, albeit less ‘forcefully’ than the Sublimox. It seems the only way to control how much is delivered is by changing the duration the hive is exposed for.

Undoubtedly this is overkill for the majority of readers of this site, but it’s interesting to see what the commercial beekeeping community are using (much like browsing the decapping or bottling machines in the Swienty catalogue). There’s at least one satisfied UK-based beekeeper quoted on the vmVaporizer site so … Mark, if you happen to read this I’d be interested in how well the machine works and whether you can achieve the quoted hive treatment every 12 seconds?

And, does it work with Api-Bioxal?



Travel screens

Stressed bees trapped inside a hive generate large amounts of heat which can result in the comb melting and collapsing. If you are moving colonies any distance, for example between distant apiaries (or even nearby apiaries if it’s a warm day), it is really important to provide ample ventilation to the hive.

Abelo hives in transit ...

Abelo hives in transit …

The usual way to do this is to use travel screens. These consist of a wooden-framed metal mesh which is used in place of a crownboard. With a sealed entrance, an open mesh floor and a travel screen, the hive is securely and safely closed up for transport, but air still can circulate freely. On long journeys or particularly hot days (which can be avoided by moving bees at night) water can be poured or sprayed through the mesh travel screen to further aid cooling of the colony.

Moving day

Late evening in the apiary

Late evening in the apiary

I often end up moving colonies in midsummer, either distributing splits to new apiaries or taking my strongest colonies to good areas of forage to exploit the summer flow. Colonies are strapped up during the day and moved either very late in the evening or very early the following morning. This timing ensures that all the flyers are back before sealing the hive and that colonies are moved during the coolest part of the day. In June in Fife this means after 10pm or before 4am if the weather is good.

Nevertheless, despite my moving hives annually, the travel screens would only be in use for a few hours a year. The rest of the time they are simply in the way.

Be flexible

For this reason I don’t use wooden-framed travel screens. I’m not prepared to pay the £15-17 the commercial suppliers charge for something that will languish on top of a stack of supers in storage for almost the entire year.

Instead of framed travel screens I use fibreglass insect mesh (search eBay for that phrase and choose the cheapest – I bet they’re all the same). This is very strong and hardwearing, relatively inexpensive and available by the roll or in a range of sizes. It’s easily cut with strong scissors and you can therefore prepare customs sizes for full hives or nucs. I cut the mesh a few centimetres oversize to allow it to be securely attached to the hive.

I bought mine from eBay a few years ago. Current prices work out at about £1.25 a hive depending on the quantity ordered. It’s usually cut for you from a 1.2 metre wide roll and sold by the metre. Each metre gives you four National-sized travel screens and some offcuts.

Straps or staples

Colonies in transit should be securely strapped – both together and to stop the hive moving in the event of an accident. For full hives I use straps and an eke to hold the mesh in place.

Travel screen mesh and eke

Travel screen mesh and eke …

For nucs, or when I run out of ekes, I staple the mesh in place to keep it secure.

Mesh and staples

The mesh is similar to the stuff sold by Thorne’s to collect propolis (at over £3 a sheet). If left in place for a long period the bees will start to seal the holes up. If you want the propolis put the mesh in a bag in a freezer and scrunch it up to release the frozen propolis.

Foam blocks

While we’re on the subject of transporting hives it’s worth mentioning that the frames must be aligned with the direction of travel. If they’re not, there’s a danger that bees will be crushed as the frames move with the acceleration and deceleration of the car.

You also need to prevent the frames from moving laterally. In a National box, unless heavily propolised, there’s usually space for 11 frames and a dummy board. However, that still leaves a bit of additional space.

On a long journey or on rough roads these can shiggle apart. This causes two problems. Firstly there’s a chance that bees will be crushed as the frames swing from side to side. Secondly, since you won’t be opening the top of the hives immediately upon arrival (the bees need to reorientate rather than have their roof unceremoniously whipped off after the long journey) there’s a chance the bees will build brace comb in the gaps.

Foam block ...

Foam block …

I use closed cell foam blocks wedged tightly into the gap between the dummy board and the side wall – one on each side in line with the side bars of the frames. This keeps everything stable and the blocks can be removed when you first inspect the colony after moving. With nucs, where there’s less to move about, or with very inflexible dummy boards, you can get away with a single block wedged tightly half way along the top bar.

Nuc prepared for transport

Nuc prepared for transport …

If you use Langstroth-sized nucs with integral feeders (like the Thorne’s Everynuc) to allow National frames to fit then it helps to also nail a cross bar over the feeder to stop the frames shifting backwards and forwards during transit.


Apis mellifera aquaticus

Early June 2017 ...

Early June 2017 …

June in Fife was the wettest year on record. It started in a blaze of glory but very quickly turned exceedingly damp. The photo above was taken on the 7th of June. One of my apiaries is in the trees at the back of the picture. Six queens emerged on the 2nd or 3rd of June to be faced with a week-long deluge. The picture was taken on the first dry morning … by the afternoon it was raining again, so delaying their ability to get out and mate (hence prompting the recent post).

And so it continued …

Early July 2017 ...

Early July 2017 …

Here’s the same view on the 1st of July. Almost unchanged … ankle deep water en route to the apiary, the burn in flood and some splits and nucs now being fed fondant to prevent them starving.

A beautiful morning though 😉

Retrospective weather reports

Of course, you shouldn’t really worry about weather that’s been and gone, though comparisons year on year can be interesting. At the very least, knowing that the June monthly rainfall in Eastern Scotland was 223% of the 1961-99 average, I’ll have an excuse why queens took so long to mate and why the June gap was more pronounced than usual. Global warming means summers are getting wetter anyway, but even if you make the comparison with the more recent 1981-2010 average we still got 206% of the June monthly total.

The Met Office publishes retrospective summaries nationally and by region. These include time series graphs of rainfall and temperature since 1910 showing how the climate is getting warmer and wetter. If you prefer, you can also view the data projected on a map, showing the marked discrepancies between the regions.

June 2017 rainfall anomaly from 1981-2010

June 2017 rainfall anomaly cf. 1981-2010 …

Parts of the Midlands and Lewis and Harris were drier than the June long-term average, but Northern England and Central, Southern and Eastern Scotland were very much wetter.

It would be interesting to compare the year-by-year climate changes with the annual cycle of forage plants used by bees. Natural forage, rather than OSR where there is strain variation of flowering time, would be the things to record. As I write this (first week of July) the lime is flowering well and the bees are hammering it. The rosebay willow herb has just started.

Rosebay willow herb

Rosebay willow herb

Prospective weather forecasts

Bees are influenced by the weather and so is beekeeping. If the forecast is for lousy weather for a fortnight it might be a good idea to postpone queen rearing and to check colonies have sufficient stores. If rain is forecast all day Saturday then inspections might have to be postponed until Sunday.

If you have a bee shed you can inspect when it’s raining. The bees tolerate the hive being opened much better than if it were out in the open. Obviously, all the bees will be in residence, but their temper is usually better. They exit the shed through the window vents and rapidly re-enter the hive through the entrance.

I don’t think there’s much to choose between the various online weather forecast sites in terms of accuracy, particularly for predictions over 3+ days. They’re all as good or as bad as each other. I cautiously use the BBC site, largely because they have an easy-to-read app for my phone.

Do I need an umbrella?

For shorter-term predictions (hours rather than days) I’ve been using Dark Sky. This can usefully – and reasonably accurately – predict that it will start raining in 30 minutes and continue for an hour, after which it will be dry until 6pm.

The forecast in your area might be different 😉

Dark Sky via web browser

Dark Sky via web browser

There’s a well designed app for iOS and Android as well that has neat graphics showing just how wet you’re likely to get, how long the rain will last and which direction the clouds will come from.

Dark Sky on iOS

Dark Sky on iOS

It’s far from perfect, but it’s reasonably good. It might make the difference between getting to the apiary as the rain starts as opposed to having a nice cuppa and then setting off in an hour or two.

Rain stopped play

I’ve posted recently on delays to queen mating caused by the poor weather in June. I’ve now completed inspections of all the splits. Despite both keeping calm and having patience I was disappointed to discover that the last two checked had developed laying workers. Clearly the queen was either lost on her mating flight or – more likely (see the pictures above) – drowned.

I’ve previously posted how I deal with laying workers – I shake the colony out and allow those that can fly to return to a new hive on the original site containing a single frame of eggs and open brood. If they start to draw queen cells in 2-3 days I reckon the colony is saveable and either let them get on with it, or otherwise somehow make them queenright.

One of the laying worker colonies behaved in a textbook manner. A couple of days after shaking them out there were queen cells present. I knocked these back and united the with a spare nuc colony containing a laying queen.

Lime can yield well in July

Lime can yield well in July

The second colony behaved very strangely. I didn’t manage to inspect them until a week after shaking them out. There were no queen cells. Nor was there any evidence of laying worker activity in the frames of drawn comb I’d provided them with. Instead, they’d filled the brood box with nectar from the nearby lime trees. Weird. I united them with a queenright colony and I’ll check how they progress over the next week or two.

Apis mellifera aquaticus

My colonies are usually headed by dark local mongrel queens. My queen rearing records show that some are descended from native black bees (Apis mellifera mellifera) from islands off the West coast of Scotland, albeit several generations ago. These bees are renowned for their hardiness, ability to forage in poor weather and general suitability to the climate of Scotland.

Nevertheless, without further natural selection and evolution they will have still needed water wings, a snorkel and flippers to get mated last month 😉

Not waving but drowning


Carl Linnaeus

Carl Linnaeus

The taxonomic scheme ‘developed’ by Carl Linnaeus (1707 – 1778) is a rank-based classification approach actually dates back to Plato. In it, organisms are divided into kingdoms (Animals), classes (Insecta), order (Hymenoptera), family (Hymenoptera), genera (Apis) and species (mellifera).

The subspecies is indicated by a further name appended to the end of the species name e.g. Apis mellifera capensis (Cape Honey bees), Apis mellifera mellifera (Black bees)

Apis mellifera aquaticus doesn’t really exist, but might evolve if it remains this wet 😉

Honey and hay fever

300 jars of honey

300 jars of honey

I’m conflicted. As a beekeeper I appreciate offsetting the cost of indulging my hobby from honey sales. In a good year I get much more honey than I could ever give away to friends and family. Despite making some of my own equipment, there are the costs of purchasing (yet more) boxes, miticides, extraction equipment and winter feed. There’s also an ever-growing wishlist of things that, whilst not essential, would be very welcome. Abelo’s heated honey creamer looks very nice 😉  Bottling, labelling and then selling honey – either from the door or from local shops – provides a few quid to help … a sort of self-perpetuating process in which I transfer all that summer effort by the bees into the coffers of Thorne’s and C. Wynne Jones.

However, I regularly get asked for local honey to ‘prevent the symptoms of hay fever’. Emails or phone calls go something like this:

“My son/daughter/husband/wife suffers really badly from hay fever and I read that locally produced honey could help her symptoms” … followed by a request to confirm that what they’ve read is correct and could I sell them some honey.

As a scientist I can’t do the former and so usually fail to achieve the latter. No way to run a business perhaps, but honesty is the best policy.


Bless you

Bless you

Hay fever is an allergic reaction to pollen in the air. About 20% of the population have, or will develop, hay fever. I never had it as a child, but in my 30’s developed a strong reaction to some grass pollens that still makes a fortnight or so in mid/late June pretty miserable. Hive inspections with bad hay fever are really miserable.

Symptoms are characteristic – itchy eyes, sneezing and a runny nose (where does all that stuff come from?!). Anti-histamines, either prescription or over-the-counter, help prevent the allergic reaction from occurring. Usually this is sufficient to make the symptoms bearable.

Severe hay fever symptoms, where anti-histamines or corticosteroids are insufficient, can be treated by immunotherapy. Over several months, the patient is exposed sub-cutaneously or orally, to low and increasing doses of the allergen (the compound that causes the allergy) to help develop immunity. Full desensitisation takes about three years.

Honey contains pollen

Honey contains small amounts of pollen. The presence of the pollen forms the basis for lots of tricky questions in the BBKA examinations and is a feature used by food standards to discriminate between flavoured sugar syrup and real honey.

This is probably where the ‘honey prevents hay fever” stories originate. It’s this small amount of pollen that is supposed to stimulate the immune system of hay fever sufferers. A sort of DIY desensitisation course using toast or porridge to help deliver the allergen. Tasty 😉

All this seems pretty logical and straightforward. Honey contains pollen. Low doses of pollen are used to stimulate immunity that, in turn, stops hay fever from developing. Local honey prevents hay fever … I must get this printed on my labels to boost sales further.

Don’t let the facts get in the way of a good theory

Unfortunately, there are a couple of irritating facts that scupper this nice little theory. The first is  a sort of error of omission, the second is the absence of evidence supporting the theory (or, more accurately, the evidence that the theory is wrong).

Honey certainly contains pollen. At least, real honey doesMelissopalynologists – those who study the pollen in honey – can identify the genus of plants that the bees have been visiting and so may be able to deduce the geographic origin of the honey.

The key part of that last sentence is “that the bees have been visiting”. The vast majority of pollens in honey are from the flowers and trees that they visit to gather nectar. These pollens are usually large and sticky so they adhere to the passing bee and are then transferred to another plant when the bee moves on.

What’s missing are any significant quantities of pollens from wind-pollinated plants such as grasses. Studies have shown that almost all pollens that cause allergies such as hay fever are from these wind-pollinated species. It’s logical that these pollens are largely absent … since the flowers, grasses and trees that produce them are anemophilous (wind-pollinated) they don’t need to generate nectar to attract bees, so the bees don’t visit. So there’s little or none of this type of pollen in honey.

No bees legs ...

No bees legs …

Testing, testing …

So that’s the error of omission. What about scientific support, or otherwise, for the theory that local honey prevents hay fever? After all, this must be an easy (and tasty) experiment to do. Feed a group of people honey and compare their hay fever symptoms with a group fed synthetic honey (or perhaps imported pseudo-honey sold from a supermarket near you).

Researchers in Connecticut did this experiment in 2002. They published their results in a snappily-titled paper “Effect of ingestion of honey on symptoms of rhinoconjunctivitis” published in the Annals of Allergy, Asthma and Immunology.

Rhinoconjunctivitis, or perhaps more correctly, allergic rhinoconjunctivitis, is the symptoms of hay fever – the itchy eyes, sneezing and runny nose. Three groups of a dozen hay fever sufferers, pre-screened for reactivity to common wind-borne allergens, were randomly assigned to receive local ‘raw‘ honey, filtered non-local honey and honey-flavoured syrup (the placebo group). They took one tablespoon of honey, or substitute, a day and recorded their hay fever symptoms. The abstract of the paper neatly summarises the results:

Neither honey group experienced relief from their symptoms in excess of that seen in the placebo group.

… leading the authors to conclude that:

This study does not confirm the widely held belief that honey relieves the symptoms of allergic rhinoconjunctivitis.

Absence of evidence does not mean evidence of absence

So, this study does not confirm (prove) that honey prevents hay fever. What about the opposite? Can we use it as evidence that honey does not prevent hay fever symptoms?

1934 Loch Ness haox

1934 Loch Ness hoax

Tricky … as the skeptic James Randi asserted, you can’t prove a negative. I can’t prove that the Loch Ness Monster doesn’t exist. However, in the absence of convincing evidence that it does exist, I can be reasonably sure that Nessie is a 6th Century tale, embellished in the 19th Century and blatantly exploited by the 21st Century tourist industry.

Of course, lake monsters are ‘found’ worldwide, which isn’t evidence that any of them actually exist 😉

We’re getting into the messy intersection of science and philosophy here. I think it’s sufficient to say that there’s no scientific evidence that honey prevents hay fever. The Connecticut experiment was a properly controlled random study. To my mind (as a scientist) this is much more compelling evidence than any amount of anecdotal stories to the contrary.

An abbreviated version of which is what I tell potential customers who want me to confirm that buying my local honey will help alleviate their hay fever symptoms. Essentially, it won’t.

Sure, they might not get hay fever after eating my honey, but that’s almost certainly a coincidence. It’s a coincidence I’m happy to live with, but not one I’m happy to promote as a reason to buy my local honey.

Why buy local honey?

I don’t think it’s necessary to cite dubious medical benefits when encouraging people to buy local honey.

Why claim something that’s probably not true?

Far better to claim the things that are true, some of which are also clearly demonstrable:

  • It’s local, from the hedges and fields within 3 miles of the apiary. It wasn’t imported by the tonne from a location or locations unknown.
  • It’s a very high quality product – clearly to claim this you need to ensure it looks wonderful and that there are no legs or antennae lurking in the jar.
  • It hasn’t been excessively heated before jarring – all the goodness is still present, including pollen, just not the sort of pollen that will prevent hay fever.
  • The honey hasn’t been micro-filtered, pasteurised or tampered with in any way.
  • It varies during the season as the forage changes – a jar of spring OSR honey is very different  in flavour from a jar of mid-summer floral (hedgerow) honey. It’s a wonderful edible snapshot of the changing seasons.
  • Buying it supports a local cottage industry.
  • It tastes fantastic – clearly demonstrable.

The ‘taste test’ is usually the deciding factor. A couple of tester jars – clearly labelled – a limitless supply of plastic coffee stirrers and a discard pot will allow customers ample opportunity to ‘try before they buy’.

Which they surely will … 🙂

∑ Honesty is the best policy is an idiom dating back to the late 16th Century when Sir Edwin Sandys, a founder of the Virginia Company and one of the first settlers in America, stated “Our grosse conceipts, who think honestie the best policie”.

A corruption of the saying by Mark Twain “Don’t let the facts get in the way of a good story”. 

† Jean Emberlin (2009). “Grass, tree, and weed pollen”. In Kay et al. The Scientific Basis of Allergy. Allergy and Allergic Diseases. 1:942-962. John Wiley & Sons. ISBN 9781444300925

‡ This isn’t xenophobia. The UK is a net importer of honey. 95% of the honey eaten in the UK is imported – 50% of the 34,000 tonnes imported in 2012 came from China. Most honey on the supermarket shelves contains some rather vague term like Produce of EU and non-EU countries. You don’t know where it came from, and probably nor does the supermarket. There have been bans on imported honey due to it being not honey (just doctored corn syrup), or being contaminated with antibiotics.