Tag Archives: swarms

Bee lining for Christmas

Bee hunting

Bee hunting

Following the Wild Bees† by Tom Seeley is an entertaining little book that would make an ideal Christmas present for a beekeeper. It describes the methods used to locate feral colonies (or any colonies actually) by bee hunting or bee lining, so called because you follow the line or direction they return to the colony from a nectar source you provide. It’s an ideal Christmas book for two main reasons; it’s a summer activity, so will remind the reader that balmy sunny days will – finally – replace the cold, dark days of winter and, secondly, it will allow the enthusiast the time to build the essential two-chambered ‘bee lining box’ which is used to trap, feed and mark the bees being ‘lined’.

I don’t intend to provide a précis of the method … you should buy and read the book for that. However, as a taster, you can visit the companion website to the book or watch a short video of Tom Seeley bee hunting …

Tom Seeley is a Professor in the Department of Neurobiology and Behaviour at Cornell University. He is a highly respected entomologist and, unlike many scientists, writes in an engaging and accessible manner. He explains complicated experiments in layman’s terms and makes parallels between his observations on honey bees and wider societal issues. Anyone who has read his book “Honeybee Democracy” will appreciate how simple and elegant his description of the science is.

His explanation of bee hunting is no less clear. Following the Wild Bees is really a ‘how to’ guide, rather than a popular science book, though each chapter does contain a separate section on the science behind the ‘how to’, together with lots of anecdotes. The book is subtitled “The craft and science of bee hunting”. If you’re not aware of feral colonies in your own area this book might help you find them … however, if you live in an area with lots of other beekeepers it will probably just help you find their apiaries (and you can also do that with Google maps).

Wild? They’re livid feral.

The most up-to-date review of feral colonies in the UK can probably be found in Catherine Thompson’s 2012 doctoral thesis (brace yourself … this links to PDF of the 173 page thesis!). Catherine surveyed a number of feral colonies in the UK and showed that, although there were limited but significant genetic differences between feral colonies and managed colonies, the feral colonies were no more ‘native’. Catherine also neatly demonstrates the limitations of studying wing veination (morphometry) as an indicator of genetic purity – it usually isn’t. Feral colonies are essentially relatively recent swarms lost by local beekeepers.

Why ‘relatively recent’?

High levels of DWV

High levels of DWV …

The feral bees Catherine studied had much higher levels of deformed wing virus (DWV), both indicative of – and as would be expected of – uncontrolled Varroa infestation. Therefore, whilst it might appear appealing to have colonies of wild bees in the local church tower they’re almost certainly riddled with DWV and Varroa. This presumably explains why so many of the feral colonies Catherine analysed died during the study period (2.5 years). The swarms lost by beekeepers (that occupy the church tower for example) quickly succumb to the detrimental effects of uncontrolled Varroa replication and the consequent transmission of viruses. Furthermore, through the activities of robbing and drifting that feral colony is likely to act as the generous donor of viruses and mites to the local managed beekeepers hives.

Perhaps not so appealing after all.


I recommend you read Following the Wild Bees. Do so sitting in front of a roaring log fire in mid-winter. Plan and build a ‘bee lining box’ (or buy one) and consider where you might go prospecting for ‘wild’ bees once the long summer days return.

But also plan to put out bait hives to catch swarms (yours or others) and clip your queens … every one ‘lost’ is an opportunity to establish a future source of Varroa and virus infestation.

Under offer ...

Under offer …


 ISBN-10 0691170266 … it’s worth shopping around for a copy as the prices vary widely (at the time of writing). WH Smiths had it for well under a tenner recently.

Bee shed inspections

A brief update on how things have been progressing in the bee shed. This is my first full season keeping colonies full-time within a shed or building though I’ve successfully overwintered mini-nucs in an unheated greenhouse in the past.

Under construction ...

Under construction …

When installed at the end of last season there was almost no need to open the hives, so it’s only this Spring that the pros and cons of the bee shed have begun to be properly understood.

The colonies are completely enclosed with simple tunnels leading to exits on the East/South East face of the shed. All the colonies are housed in standard National cedar boxes or poly nucs. Other than clear perspex insulated crownboards, there is no additional insulation and the shed is not heated. The shed is situated in open parkland with woodland and arable land nearby containing good forage and there is a permanent water supply nearby.

Colony development and Varroa loads

Colonies went through the winter in single National brood boxes, fed with fondant and treated with oxalic acid by vaporisation in September (before moving them to the shed) and in midwinter. The first inspection was conducted in late March. Colonies were building up well and were significantly stronger than colonies headed by sister queens in the same apiary or in my other apiary. Between late February and early May colonies dropped only 3-4 mites in total, with Varroa boards located within pull-out trays in the hive floor. I’m sure I missed a few mites, but doubt it was very many. We’ve recently uncapped a full frame of drone brood – each cell uncapped individually – and found no Varroa present. Mite levels are therefore reassuringly low – for reasons to be discussed in a future post – with no signs of DWV-related disease.

Varroa tray ...

Varroa tray …

Since mid-April colony development has been very good and they are now on double National brood boxes with 2-3 supers. A fourth super went onto one colony on the 25th of May and the stack now nearly reaches the shed roof. A four frame nuc has been split off one colony already to cool it down a little. Quite a bit of developing brood has also been harvested at weekly intervals for our research, usually by simply cutting a big slab out of the middle of a frame. This has probably also held the colonies back a bit and it’s only now I’m starting to plan for swarm prevention/control.


Inspections have been easier than expected. These colonies are headed by queens with reasonable genetics (Heinz queens – local mongrels of 57 varieties, reared by me in 2015). The bees are steady on the comb and tend not to fly up at you when the crownboard is lifted. They’re nothing particularly special, but would be considered reasonably placid and non-aggressive.

The colony is gently smoked from outside the shed (through the entrance tunnel) and a small amount is wafted under the crownboard or between the QE and the bottom super. After allowing them to settle the supers and crownboard are removed and placed outside on an overturned roof. The queen excluder and adherent bees are also left standing outside (unless it’s cold when the bees are shaken off into the open hive).

Inspecting the colony is straightforward. Any frames removed to make space are rested on the hive stand. Double brooded colonies are split into two, with one box stood aside on an eke on the roof of an adjacent hive roof. Inevitably, the queenless half of the split tends to get tetchy within a few minutes, so it’s best to deal with them first. When frames need to be shaken free of bees this can be done either over the open hive or, better still, directly into a gap between the frames. If done outside many of the nurse bees on the frame fail to get back to the hive (they’ve probably not been on orientation flights yet).

The smoker is usually stood just outside the shed door … if you keep it in the shed during inspections you can end up being kippered 😎

Flying bees

Perhaps surprisingly, even going through all 22 frames in a double colony, the shed does not fill with a maelstrom of flying bees. Undoubtedly this is partly because they’re reasonably calm colonies. Those that do fly rapidly find the window or open door and make their exit. When I first started doing inspections in the bee shed I’d manually help the stragglers outside after reassembling the hive. It turns out that there’s really no need … almost all the bees quickly vacate the shed by making a beeline ( 😉 ) for the bright lights of the windows or doors.

The great escape ...

The great escape …

Just how quickly the bees leave the shed was emphasised last Sunday when selecting larvae for grafting. I opened and inspected a double brooded colony, found a suitable frame with 24 hour larvae on it and placed it in a two frame nuc for protection. Within 5 minutes I could work without a veil (I react very badly to stings to the face so take particular care over this) without interruption from flying bees.

Weather and temperature

I’m sure that the temperature influences the behaviour of the colonies in the shed. They certainly forage – or perhaps collect water to use fondant or crystallised stores – at lower temperatures than those situated outside. When inspections are conducted on a cold day (say 10-11°C) they are even more steady than usual. However, those that do fly take longer to leave the shed and they can end up clustering in small, rather pathetic, little groups which then need to be scooped up on a hive tool and dropped into the colony. On cool days I don’t leave the supers or QE outside the shed as the bees would rapidly get chilled. Work commitments mean that inspections must be conducted on certain days, so I don’t have the luxury of simply waiting until it’s a bit warmer. Although the shed is unheated the temperature differential between the inside and outside is significant – perhaps 4-8°C – or more if the sun is shining on the window side of the shed. On a warm, sunny day the temperature inside the shed can easily reach the mid-20’s which makes inspections a hot and sweaty activity.

Needless to say, inspections on damp or wet days are much better than on colonies located outside. I avoid days when it’s raining hard, partly for my own comfort to avoid getting wet accessing the apiary, but also because I’d prefer not to force the bees to fly on a really wet day. However, on damp or drizzly days, inspections proceed as normal.

And the bad news is …

Almost everything I’ve written above is positive and my overall initial impression is that the bee shed offers very significant advantages for the sort of beekeeping I need to do. However, there are some drawbacks and design issues that either currently cause problems, or might in the future.

The first is that it’s too small. The shed is 12 x 8 feet and I should have got one at least half as long again. This is largely because it’s also used for equipment storage and has a small table for working on. With four hives I need storage for 8-12 supers, additional brood boxes and spare frames. If I was starting again, knowing what I know now, I’d get an 18 x 10 shed with the intention of housing at least 6 colonies and some additional nucs (by contrast mine will accommodate 4 full colonies and 2 nucs down the sunny side of the shed, with the possibility of 2-3 additional nucs at a squeeze). It’s not only equipment storage that takes up the room … you need considerable room to work as well, with space for turning, stacking and temporary placement of hive parts. Working in the bee shed encourages an efficiency of movement – or causes a lot of collisions – I’d not expected.

Essential storage ...

Essential storage …

Secondly the lighting is – at best – variable. On a sunny morning there’s ample light to see eggs and tiny larvae. However, as the colonies have grown, the added supers restrict the amount of light getting through the windows. On an overcast day, or late in the afternoon, the lighting is pretty hopeless – good enough to see queen cups/cells, good enough to locate the queen, but (particularly on dark frames) too dim to see eggs, small larvae or to check frames for signs of disease. It’s not unusual to have to carry frames outside to inspect them fully. I’m currently investigating 12V LED systems run from a solar panel-charged caravan battery. My only concern is that this might disorientate the bees and slow their exit from the shed during inspections.

Multiple supers ...

Multiple supers …

Thirdly, I should have spent more time designing the hive stands. I made them an inch or so too low which caused some problems with locating the hive entrances centrally in the T&G planks, but was not insurmountable. More problematically, as a consequence of the leg locations it’s difficult to keep the floor clear of hive debris that falls through the OMF. With the Varroa boards in place this isn’t an issue, but when they’re out – which I prefer if there’s a chance of the shed getting very warm – the debris needs to be regularly swept up to keep the shed clean. Some sort of removable debris trays would have been a good addition, but are not easy to fit retrospectively. However, the overall hive stand design – with the legs going through the suspended floor to avoid vibrations – works very well.

Finally, swarm control has yet to be tackled. My preferred simple method is doing a vertical split (or using a Snelgrove board that I’m experimenting with this year) but this requires an upper entrance which, obviously, cannot easily be arranged. One possibility is using the Demaree method of swarm control. Alternatively, it would be straightforward to remove the queen into a nuc and let the colony requeen. Currently I’m trying to postpone the inevitable by removal of some brood, ensuring they have enough space within the brood boxes which I swap (top to bottom, bottom to top) periodically, ensuring they have sufficient space in the supers and keeping a close eye on them. The queens are clipped. If they do swarm they’re likely to end up in a lump outside the hive entrance – the ground is flagged and so they should hopefully be relatively easy to scoop up.


Divide and conquer

Tom Seeley (of Honeybee Democracy fame) published an interesting paper in the journal PLoS One recently on “How honey bee colonies survive in the wild: testing the importance of small nests and swarming” – the paper is available as a PDF following this link (Loftus et al., 2016 PLoS One 11:e0150362).

Size matters

Using his normal elegant methodology Seeley formally tested the observed reduction in colony size and increased swarminess (is that a word?) of – feral or otherwise – colonies ‘selected’ to survive without Varroa treatment by simply abandoning them. The hypothesis – based on previous studies and an understanding of the biology of Varroa – was that colonies ‘forced’ to swarm by being confined in small hives would inevitably:

  • lose significant amount of Varroa through the act of swarming
  • experience a brood break so delaying Varroa replication while requeening
  • consequently survive better than large colonies in which pathogen levels inexorably increased to a level that would destroy the colony

Testing the hypothesis

He tested this by establishing adjacent apiaries (so they have the same microclimate) with either small (~40 litres … about the same as a National brood box) or large (~170 litre) volume hives and installing nucs in each which contained similar levels of brood, bees and Varroa. No Varroa control was performed. Those in the small hives were not managed to prevent swarming whereas those in the large hives were – with the caveat that the colony was kept together (i.e. queen cells were destroyed, brood frames were spread and ample supers were added). The study lasted two years, with regular monitoring of the colony strength, Varroa infestation level etc.

High levels of DWV

High levels of DWV …

To cut a long (but nevertheless interesting and worth reading) story short … the results support the original hypothesis. During the first year of the study the colonies developed in a broadly similar manner from transfer of the nuc to the large or small hive in June until the season’s end. However, by the following May the large hived colonies were almost twice as populous as those in the small boxes. This continued until August, with the average adult bee population in the small and large hives being ~10,000 and ~30,000 respectively. During this second season 10/12 small hives swarmed, whereas only 2/12 of the large hived colonies swarmed. In the latter mite levels dramatically increased to >6/100 adult bees (i.e. riddled with the little b’stards – my opinion, Seeley is too polite to comment). For comparison, the picture above has ~100 bees in it, with one visible Varroa, but has lots of overt deformed wing virus disease. In contrast, the small hived colonies – with the exception on one sampling point discussed later – had three to five times fewer mites than seen in the large hived colonies. By the second winter 10/12 large hived colonies had perished whereas only 4/12 small hived colonies had succumbed, and one of these was to a drone laying queen, not disease. Perhaps most tellingly, 7/12 large hived colonies had signs of overt deformed wing virus (DWV) disease – pathetic, tottering newly emerged workers with stunted abdomens and shrivelled wings – whereas none of those in small hives showed obvious disease.

Great … Varroa-tolerant colonies … where can I get some?

A small swarm

A small swarm

So, what does this mean in terms of practical beekeeping? Firstly, it suggests that it is possible to keep honey bee colonies without treatment or intervention. But – and it’s a biggy – the colonies will be too small to collect meaningful amounts of honey and will spend their time and energy swarming instead. 10,000 adult bees does not a colony make, as Aristotle didn’t say. Or at least not a colony that’s of any practical use for the honey-gathering goal of beekeeping. Ted Hooper (“Bees and honey“), and many others, have made the point that one big colony will gather more nectar than two smaller colonies. Secondly, these small colonies will chuck out loads of Varroa-riddled swarms. Seeley has previously demonstrated that swarming colonies lose ~35% of their Varroa load with the bees that leave the colony. Although this clearly benefits the original colony it potentially distributes Varroa-laden bees (and the smorgasbord of viral pathogens that are the real problem) to whichever local beekeeper finally hives them. This explains the need for prompt Varroa treatment of any swarms you might acquire.

On a more positive note this study clearly shows the benefit of a brood break in terms of restricting the replication and amplification of Varroa. Presumably this is primarily due to the 3+ week window with no sealed brood for Varroa to infest, though it may also mean that broodless colonies might get rid of Varroa at a faster rate with no brood present to distract them. It would be interesting to have compared mite levels immediately after swarming and in the subsequent weeks until the new queen starts laying. Randy Oliver has also discussed the benefits of a brood break during empirical development (and computer modelling) of his beekeeping methods for Varroa control. In his forthright manner he explains “Take home message: early splitting knocks the snot out of mite levels“.

Why discuss this if they’re no use for beekeeping … ?

There was one exception to the generally low mite levels in the small hived colonies and that was late summer in the second year when they all exhibited a large spike in Varroa numbers. This was attributed to robbing-out a collapsing, and soon to die-off completely, large hived colony in the adjacent apiary. The two study apiaries were in the same field. This emphasises the points made in earlier posts about the impact of drifting and robbing and the, at least theoretical benefits of, coordinated Varroa control. Of course, ~2 mites per 100 adult bees in the small hived colonies is not really a low number at all. Assuming a colony size of 10,000 adults with 80% of the mites in capped cells the total Varroa load would be ~1000 in the colony, the threshold level above which the NBU consider treatment is required to avoid loss of the colony.

Divide and conquer

The Varroa loss achieved by swarming, coupled with the break in brood rearing, help the colony conquer – or more correctly tolerate – Varroa levels that otherwise rapidly increase and destroy a colony. However, this is neither a practical or acceptable solution to the Varroa problem. ‘Beekeepers’ (an oxymoron surely?) that allow their colonies to swarm indiscriminately both reduce their chance of getting a good honey crop and impose their – potentially Varroa-ridden – swarms on the neighbourhood. This is irresponsible. In contrast, beekeepers who carefully monitor their colonies and use an effective combination of integrated pest management – for example, including an enforced brood break during the ‘June gap’, or a vertical split, perhaps – benefit from large, healthy, honey-laden§ colonies which overwinter better.

§ at least in the good years 😉




I’ve been dabbling with BEEHAVE, a computer simulation of a honeybee colony. It’s not beekeeping, but it’s about as close as you can get in the middle of winter. BEEHAVE was developed by Matthias Becher in the University of Exeter and the paper that describes the model is published and Open Access [PDF]. The model includes a wealth of user-modifiable variables such as forage availability, climate, beekeeping activities and pathogens, and outputs information on colony size, speed of development, age structure, honey stores etc. The BEEHAVE simulation is implemented in the open source language NetLogo and is freely available. The parameters that influence colony development – egg laying rate, drone/worker ratios, forage (nectar and pollen) availability, mite replication rate etc. are all based on measured and published data (or logically extrapolated from this if they don’t exist) so that the in silico performance is a fair reflection of what might be expected in the field.

If you can, do … if you can’t, simulate it 🙂

I’m interested in the rational and effective use of miticides to control Varroa-mediated transmission of DWV (and other viruses) in the hive. Using BEEHAVE and a standardised set of conditions allows predictions to be made of how effective a particular Varroa control might be. For example, here’s a simple question we can try and answer:

How important is a midwinter mite treatment if you’ve treated earlier in the year?

Using BEEHAVE set to all the default conditions and ‘priming’ the colony with just 20 mites on the 1st of January it’s possible to see what happens if no treatments are applied over one or more years. It’s then possible to repeat the predictions with the inclusion of a Varroa treatment. For the purpose of this brief introduction to BEEHAVE I’ve used a miticide which is applied and active for a total of 28 days and which kills 95% of phoretic mites. This might broadly reflect Apiguard treatment (2 x 14 days) or vaporised oxalic acid (OA; 3 treatments at 5 day intervals, but documented to kill mites for up to one month). I’ve additionally looked at the application of a single treatment with oxalic acid in midwinter, again killing 95% of phoretic mites, the sort of effect that OA trickling might achieve if there’s no brood present.

No treatment … they’re doomed

No treatment

No treatment

BEEHAVE modelling is based on a series of underlying probabilities (e.g. likelihood of a developing pupa to become mite associated, likelihood of that being a drone or worker pupa) so doesn’t produce the same results every time it is run¹. For example, the graph above shows adult bee numbers (left axis, blue lines) in an untreated colony for three simulations of up to five years each (horizontal axis), together with the associated mite number (right axis, red lines). Mite number build up strongly as new brood is reared each spring, with mite numbers peaking at ~24,000 in the fourth summer. In the third and fourth winters mite number per bee range from 2-4. The default conditions of 20 mites, coupled with a minimum viable colony size of 4000 bees, results in one colony succumbing in the fourth winter and the two remaining dying in the fifth winter (bee numbers drop to zero). Real studies – with untreated hives in the field – have shown similar outcomes (Martin, 1998 [PDF]) though colonies tend to die between winters 2 and 3, presumably because the input mite populations are higher². In all subsequent graphs the data plotted is the average of three simulations.

One treatment … better than nothing

It’s worth remembering at this point that the advice from the National Bee Unit is that mite numbers in the colony should be maintained below 1000 (Managing Varroa [PDF]). To try and achieve this we need to investigate the influence of applying miticides in the simulation – in mid-June (left graph), mid-September (middle) or late December (right). I appreciate mid-June is very early in the season, but it emphasises an important point.

That’s a bit better 🙂 These plots show the averages of adult bee and mite numbers (using the format shown above, blue for bees, red for mites). None of the in silico colonies expired during the simulation though the mite numbers are dangerously high irrespective of the treatment during the mid/late summer months. Note that range of the scale on the right hand (mite numbers) axis differs in each graph. Treatment in mid-June (left) delays the summer exponential rise in mite numbers and, in terms of overall impact on mite numbers (and consequent adult bee losses) is measurably better than only treating in midwinter (right). Of the conditions tested, mid-September (centre) is clearly the best … Varroa levels are reduced at the same time as the colony starts to contract, leaving the remaining mites less opportunity to reproduce. Maximum colony size remains about the same year on year and Varroa numbers never reach more than one third of those seen in either mid-summer or midwinter treatments. However, not everything is rosy … Varroa levels are dangerously high from the third summer on, and levels are increasing each winter. Remember that these simulations were started with just 20 mites in the colony².

Do your colonies have only ~20 mites in them this winter?

Two treatments … a double whammy

Two optimal treatments

Two optimal treatments

It’s only when you combine early autumn and midwinter treatments that mite numbers are really well controlled. Under the highly optimised conditions – both treatments were set to be 95% effective against phoretic mites – Varroa numbers remain below the NBU recommended maximum of 1000 for the duration of the simulation. Clearly the combination of the mid-September slaughter of phoretic mites, coupled with a midwinter mopping up – when there’s little or no brood present – provides really tight control of Varroa levels. However, the importance of this is perhaps even more apparent when you consider the consequences of a sub-optimal mid-September treatment.

The graph on the left shows the consequences of using a miticide that achieves only 85% efficacy … perhaps reflecting Apiguard usage when the ambient temperature is too low for the thymol to be spread throughout the colony. Under these conditions mite numbers rapidly get out of control. Compare that with the graph on the right which includes an additional midwinter treatment where mite numbers are far better controlled … though only to about the same level as is seen with a 95% knockdown of mites in mid-September (centre graph in the ‘one treatment only’ section, above).

And the answer is …

Occupied bait hive

Occupied bait hive …

Although the majority of miticides are broadly similar in their maximum published efficacy, I suspect that they are often used in a way or under conditions that do not routinely achieve these maxima. For example, the 30 year average September temperature in England is just below 13°C, much lower than the temperatures in which Apiguard efficacy reached the reported maximum of 99%, and lower than the Vita-recommended minimum temperature (15°C). Therefore, the answer to the original question (which was How important is a midwinter treatment if you’ve treated earlier in the year?) is … if there’s any chance the late summer/early autumn treatment was sub-optimal then a midwinter treatment is very important to prevent Varroa levels building up in the colony, resulting in the spread of virulent strains of DWV and other viruses. The other broad conclusion is that miticides are much more effective – in terms of impact against the total mite population – when brood levels are low or absent. That’s why brood breaks coupled with miticide treatments e.g. applying vaporised oxalic acid to a recently hived swarm or one that has moved in to a bait hive, are a very powerful combination to reduce the impact of mites, and the viruses they transmit, on the colony.

There are additional considerations which influence the choice and timing of miticide treatments. In a future post I’ll address the timing of the autumn treatment and the critical development of the overwintering bees that get the queen and the colony through to the following Spring.

¹BEEHAVE provides the ability to model colony development based upon measured and measurable parameters within a honeybee colony. Of course, in the real world a host of factors influence our bees – climate, forage availability, bad beekeeping, good beekeeping, integrated pest management, swarming, queen longevity etc. These are all variable within BEEHAVE but have been left unaltered from the defaults for the purpose of this post in which only the timing and efficacy of miticide treatment was altered. All the data for this post were generated using the rather verbosely numbered BEEHAVE_BeeMapp2015 version.

²Mite levels were deliberately started at a very low level to emphasise how quickly they build up if not controlled. Running the simulations with a higher mite input simply shifts all the graphs to the right e.g. increasing input mites to 200 (not an unreasonable number for many midwinter colonies) with no treatment, results in the virtual colony dying in early December of the third year, with mite levels having reached ~5300 in the first summer and ~19000 in the second.

This is the second in a series of related posts about Varroa control. The first was on drifting in honeybees. I’ve created a separate page that lists these and other posts on the how, why and when of Varroa treatment.

2015 in retrospect

The winter solstice seems like a good time to look back over the 2015 beekeeping year. With the day length about to start increasing, what went right and what went wrong? Back in March I wrote that my plans for the year were different from the usual OSR – swarming – queen rearing – summer flow – harvest – Varroa treatment – feed-’em-up and forget ’em routine as I was moving to Scotland in the middle of the season. Some of these things happened, though perhaps less than in a usual year.

Mid-season memories

Mid-season memories …

Spring – better late than never

Cloak board ...

Cloake board …

The OSR yielded poorly as the spring was cold and late. I didn’t even look inside a colony until mid-April. Colonies were only getting strong as the OSR flowers went over meaning that most of it was missed. The weather was unseasonably cold, with mid-May being 2-3ºC cooler than average. Queen rearing started in the third week of May and although grafting went well, queen mating was really hit and miss, with low temperatures and lots of rain lasting through May and June. On a more positive note, I used a Cloake board for the first time and was pleased with the results (I’ll write about this sometime in 2016 after using it a bit more). I didn’t use any mini-nucs this year as I didn’t want the hassle of dealing with them mid-season when moving North. Instead, I did all of my queen mating in 2-5 frame nucs, often produced as circle splits from the cell-raising colonies. This worked well … and considering the lousy weather was probably a lot less effort than using mini-nucs which would have required constant attention and lots of feeding. Using poly-nucs I could prime them with a frame of brood and a frame of stores and adhering bees, dummy them down and leave 3 frames of foundation (or wherever possible, drawn comb) ready to be used on the other side of the dummy board. Once the queen was mated the colony would build up well and if – as often happened this season – the queen failed to get mated or was lost (drowned?) during mating flights it was easy to unite the queenless unit with a queenright one, so not wasting any resources.

Go forth and multiply

Split board

Split board …

Beginners often find the coordination of colonies for queen rearing, and the apparent difficulty of grafting (it isn’t), a daunting prospect. When I’ve been involved in teaching queen rearing it’s clear that the relatively small scale approach I use (queenright cell raiser, grafting and – usually – mini-nucs) is often still too involved for the very small numbers of queens most beekeepers with just a couple of hives want. It was therefore interesting to raise a few queens using vertical splits, simply by dividing a strong colony vertically and letting the bees do all the work of selecting the best larvae, raising the queen and getting her mated. It has the advantage of needing almost no additional equipment and only requires a single manipulation of the hive (and even that can probably be simplified). Having documented the process this season I’ve got a few additional things I’d like to try in 2016 to make it even easier and to allow better stock selection. After that it will be incorporated into queen rearing talks and training.

Changes in Varroa treatment

The big change in Varroa treatment in the UK was the licensing of Api-Bioxal. Whether or not you consider the 50-fold or more cost of VMD-approved oxalic acid (OA) over the generic powder is justified is really a separate issue. Oxalic acid is an effective miticide and, if administered appropriately, is very well tolerated by the colony. Despite the eyewatering markup, Api-Bioxal is significantly less expensive than all other approved miticides. For the small scale beekeeper it’s probably only 20% the cost of the – often ineffective – Apistan, or either Apiguard or MAQS. Under certain circumstances – resistant mites, low temperatures or the potential for queen loss – there are compelling reasons why OA is preferable to these treatments. If we hadn’t been using OA for years the online forums would be full of beekeepers praising the aggressive pricing strategy of Chemicals Liaf s.p.a in undercutting the competition. Of course, if we hadn’t been using generic OA for years Api-Bioxal would probably be priced similarly to Apiguard 🙁

Sublimox in use

Sublimox in use …

I’ve used OA sublimation throughout 2015 and been extremely impressed with how effective it has been. Mite drops in colonies treated early in the season remained low, but increased significantly in adjacent colonies that were not treated. I treated all swarms caught or attracted to bait hives. Some were casts and there were no problems with the queen getting out and mated (though the numbers of these were small, so statistically irrelevant). Late season treatment of colonies with brood also seems to have worked well. Mite drops were low to non-existent in most colonies being monitored through late autumn. Colonies get mildly agitated during treatment with a few bees flying about under the perspex crownboard (you can see a couple in the image above … this was a busy colony) and a few more rapidly exiting the hive after the entrance block is removed. But that’s it. The colony settles within a very short time. I’ve seen no loss of brood, no obvious interruption of laying by the queen and no long-term detrimental effects. Sublimation or vaporisation of OA can – with the correct equipment – be achieved without opening the hive. I expect to use this approach almost exclusively in the future.

Moving bees

Moving colonies from the Midlands to Fife was very straightforward. Insect netting was an inexpensive alternative to building large numbers of travel screens. It’s the same stuff as Thorne’s sell for harvesting propolis so I’ve got enough now to go into large scale propolis production 😉 The colonies all settled in their temporary apiaries well and I even managed a few supers of honey during the latter part of the season.

Small hive beetle reappeared in Southern Italy shortly after the honey harvest was completed there. Che sorpresa. This was disappointing but not unexpected (and actually predicted by some epidemiologists). As I write these notes the beetle had been found in 29 Calabrian apiaries between mid-september and early December. It’s notable that there’s now a defeatist attitude by some contributors to the online forums (when not if the beetle arrives here) and – since not everyone are what they seem on the interweb – there are some playing down the likely impact of the beetles’ arrival (and hence the demand to ban imports) because they have a vested interest in selling early season queens or nucs, either shipped in or headed by imported queens. I don’t think there’s any sensible disagreement that we would be better off – from a beekeeping perspective – without the beetle, it’s just that banning imports of bees to the UK (admittedly only a partial solution) is likely to cause problems for many beekeepers, not just those with direct commercial interests. I remain convinced that, with suitable training and a little effort, UK beekeeping could be far less dependent on imports … and so less at risk from the pathogens, like small hive beetle. Or of course a host of un-tested for viruses, that are imported with them.

And on a brighter note …

Bee shed ...

Bee shed …

The new development in the latter part of the year was the setting up of a bee shed to house a few colonies for research. This is now more or less completed and the bees installed. It will be interesting to see how the colonies come through the winter and build up in spring. The apiary has colonies headed by sister queens both in and outside the bee shed so I’ll be able to make some very unscientific comparisons of performance. The only problem I’ve so far encountered with the shed was during the winter mite treatment by oxalic acid vaporisation. In the open apiary the small amount of vapour that escapes the sealed hive drifts away on the breeze. In the shed it builds up into a dense acidic hazy smoke that forced me to make a rapid exit. I was wearing all-encompassing goggles and a safety mask so suffered no ill effects but I’ll need an alternative strategy for the future.

Due to work commitments, house, office and lab moves, things were a lot quieter on the DIY front this year. The Correx roofs have been excellent – the oldest were built over a year ago and are looking as good (or as bad, depending on your viewpoint) as they did then. They’ve doubled up as trays to carry dripping supers back from the apiary and I’ll be making more to cover stacks of stored equipment in the future. Correx offcuts were pressed into service as floors on bait hives, all of which were successful.

With well-fed colonies, low mite counts, secure apiaries and lots of plans for 2016 it’s time to make another batch of honey fudge, to nervously (it’s got hints of an industrial cleaning solution) try a glass of mead and to finish labelling jarred honey for friends and family.

Happy Christmas

Lomond Hills and OSR

Lomond Hills and OSR

Miticide cost effectiveness

There goes a few pence ...

There goes a few pence …

My recent comments on the cost of Api-Bioxal prompted me to look in a little more detail at the cost of miticides routinely available to beekeepers. The figures quoted below are the best prices listed by one of three leading beekeeping suppliers in the UK (E.M. Thorne, Maisemore’s and C. Wynne Jones – there are lots of other suppliers, but I’ve used these three and been satisfied with their service). I made the following assumptions: the beekeeper is purchasing sufficient to treat three single-brooded full colonies for three years (i.e. something with a reasonable shelf-life) with as little left over as possible. Costs per colony treatment were calculated for 9 colonies (3 x 3 years) only … any ‘spare’ can therefore be considered as free. This means that for Apiguard, available in packs of ten trays (5 colony treatments) or a 3kg tub (30 colonies), the cost is calculated per colony from two packs of 10 trays as a full course of treatment for one colony requires two trays. Obviously, buying in bulk – for example through a co-operative purchasing scheme in your beekeeping association – should reduce these costs significantly. No postage costs were included.

Apiguard – two boxes of 10 trays (C. Wynne Jones) = £41 = £4.55/colony

Apistan – two packs of 10 strips (C. Wynne Jones) = £41 = £4.55/colony

MAQS – one 10 dose tub (all suppliers) = £57.60 = £6.40/colony

Api-Bioxal – one 35g sachet (C. Wynne Jones) = £8.20 = £0.91/colony

Oxalic acid (OA) crystals – one 300g tub (Maisemore’s) = £4.32 = £0.48/colony

Note that this simplistic comparison hides a number details.

  1. These various treatments should be broadly similar in their efficacy (see below) in reducing the mite population, but must be used according to the manufacturers instructions for maximum efficiency. Under optimal conditions all quote at least 90% reduction in mite levels. However, Apistan (and Bayvarol, not listed) is pyrethroid-based and resistant mite populations are very widespread. In the presence of totally or partially resistant mites, Apistan will be of little or no benefit. Interestingly, Apistan resistance (which, like resistance to pyrethroids in other species, is due to a single amino acid substitution, so readily selected) appears to be detrimental to the mite in the absence of selection. This means that it may be possible to use Apistan effectively every 3-5 years as part of an integrated pest management as long as other beekeepers in the area follow the same regime. During the years Apistan is not used the pyrethroid-resistant mites should reduce in number, so restoring the efficacy of the treatment. I’m not aware that this idea has been properly tested, but it might be worth investigating.
  2. Only the first four treatments are approved for use in the UK by the VMD.
  3. Both the oxalic acid-containing treatments – Api-Bioxal and OA crystals – require preparation before use, or specialised equipment for delivery. OA vaporisation (sublimation) also necessitates both care and personal protection equipment to prevent exposure to the chemical which is a lung irritant. The costs indicated do not include these additional requirements.
  4. The treatments are not equivalent or necessarily interchangeable. For example, a) only MAQS should be used when honey supers are present, b) Apiguard is moved around the hive by active bees, so treatment is recommended when average daytime temperatures are above 15ºC , and c) there are reports on discussion forums of repeated OA vaporisation treatment – 3 at 5 day intervals – for colonies with brood present. The costs indicated above assume a single treatment (in midwinter or of a swarm/shook swarm in the case of OA) with any of the listed compounds.
  5. Finally, the ‘excess’ amount spare after treating the colonies over three years differs significantly. The first four have sufficient left over for one further treatment. The OA crystals will have enough left over for a further 190 colonies … and buying a 300g tub is probably about the most expensive way to purchase OA per gram 🙂

Bang for your buck

As indicated above, all of the Varroa treatments listed should give 90+% knockdown in mite numbers if used properly. This means following the manufacturers’ instructions – in terms of dose, time and duration of application. A key point to remember is that the mite is only susceptible when outside the capped cell and that 80% or more of the Varroa in a colony at any one time will be inside capped cells if there is brood present. For this reason, it is preferable to treat during natural (or induced e.g. a shook swarm) broodless periods. It has even been suggested that the midwinter OA treatment should be preceded by destruction of any brood present. Although this makes sense, I can understand why some beekeepers might be reluctant to open a colony to destroy brood in the middle of winter. There have been numerous reviews of individual and comparative efficacy of the various Varroa treatments – for example this well-referenced article on mite treatment in New Zealand from 2008. If used properly there’s little to choose between them in terms of efficacy, so the choice should be made on the grounds of suitability, convenience and cost.

‘Suitability’ is a bit of a catch-all, but requires you broadly understand how and when the treatment works – for example, Apistan is a pyrethroid so works well against sensitive mites, but is pretty-much useless against resistant populations, and resistance is widespread in the UK. ‘Convenience’ is generally high in the ready-prepared commercial treatments – it takes seconds to insert a tray of Apiguard – and much lower if the compound has to be prepared or you have to get dolled up in protective gear. In this regard, the absence of a pre-mixed liquid version of Api-Bioxal is a disappointment. Thorne’s still supply (at the time of writing) Trickle 2, a very convenient pre-mixed 3.2% w/v OA treatment for mid-winter trickling, but for how much longer? Similarly, the gloves, mask, goggles and power needed to treat a colony by OA sublimation makes it far from convenient for a single treatment.

Closing thought …

1 lb jar of honey

1 lb jar of honey …

Despite the great differences between the cost/treatment/colony it’s worth noting that even the most expensive is not a lot more than the price of a 1 lb jar of top quality local honey … just like the stuff your bees produce 😉 So, in the overall scheme of things, Varroa treatment is relatively inexpensive and very important to maintain colony health and to reduce overwintering colony losses.

See also Managing Varroa (PDF) published by the Animal and Plant Health Agency

Late arrivals

Stacked boxes

Stacked boxes …

I’m moving house in a couple of weeks and so stacked unused ‘bee equipment’ in a pile on the patio for packing. Some of the supers contained drawn comb from previous years, some of the broods were empty and some contained prepared foundationless frames. I thought I’d taken care to align everything reasonably well to ensure they were ‘beetight‘ when I finished up late on Thursday evening. However, I’d misaligned a chest high stack in the middle and unknowingly left a finger-width crack which allowed some scouts to decide it was a desirable site. Sometime mid-morning on Friday – when I was in the office – a good sized swarm arrived. I hadn’t noticed any scouts checking out the location. I originally thought it was robbers cleaning out honey from the supers, but a quick peek under the roof (there was no crownboard on the stack) showed they were busy drawing comb. Going by the numbers of bees present it looked like a prime swarm, but you can’t be sure unless you find the laying queen.

They couldn’t have chosen a much less suitable (for me … it obviously suited them 😉 ) stack to set up home in. The bottom three boxes were empty broods, topped with three supers, two of which were part filled with drawn drone foundation. Inevitably the spacing of the frames in the supers was all over the place. Removing the roof gently showed they were already building brace comb, attached to the roof and/or the frames. The bees were accessing the stack somewhere in the middle, on the face against the wall. What a mess.

Rearranging the hive

Rearranging the hive …

I fired up the smoker and got kitted up. It was relatively easy to split the stack and put a temporary floor below the supers (with the entrance facing the wall) and put a crownboard in place. The colony were agitated but not aggressive. There were far too many bees to try and find the queen. It was a hot day and there was a whirling maelstrom of bees. I was concerned that the queen – if she was mated – would start laying up the drawn drone foundation in the supers. By evening the stack was quietly humming away, with all the bees inside, so I moved them a few feet away to a purpose-built stand (the ubiquitous milk crate) … swarms can be relocated within 24-48 hours of arrival during which time the “3 foot, 3 mile rule” can be safely ignored.


Blackberry …

Early on Saturday morning I put a new floor and brood box filled with frames on the stand, then added a clearer board and put the two supers full of bees on top. The hope was that many of the foragers would move down into the brood box, leaving the queen and attendants above the clearer. I peeked through the perspex crownboard on Sunday morning and the number of bees in the supers was much reduced. A quick inspection located a very dark unmarked laying queen in the supers. One wing was pretty tatty so she might be quite old. To my surprise the bees had re-engineered a big patch of the drawn drone comb in the super frame to make worker-sized cells and that was the area she’d laid up. In addition, they’d also piled in a surprisingly large amount of nectar – presumably from blackberry which is just developing well at the moment. I rearranged the brood box, moving the queen on the laid-up super frame into the bottom box, then shook the remaining bees off the super frames and closed the colony up.

Ready for OA treatment ...

Ready for OA treatment …

Finally, late on Sunday evening I treated the colony by oxalic acid vaporisation. With no sealed brood in the hive it’s a perfect time to reduce the phoretic mite numbers by at least 95%. Since I have no idea about the provenance of the swarm – other than being sure its not from one of my colonies, all of which have marked and/or clipped queens – this gives at least some peace of mind that a range of unpleasant diseases aren’t being introduced to the apiary with the bees, or the mites they’re carrying. I’ll check the Varroa drop over the next few days and monitor the quality of sealed brood before deciding what to do with them. However, I suspect they’ll either be requeened or given away to an association member still wanting bees, or quite possibly both as I unite other colonies in preparation for moving.

The faint sniffing is my hay fever … I’m not testing the OA vapour. The latter is a significant lung irritant and I’m wearing safety goggles and a mask for personal protection. I’ll post something separately on the Sublimox vaporiser later in the season.

Note Unlike an earlier swarm only about ten mites dropped after OA vaporisation within the first 24 hours which is very reassuring. Some claim that only healthy colonies swarm and, although there is some truth in this (i.e. only strong healthy colonies build up sufficiently to swarm), it doesn’t mean the swarm won’t have a high phoretic mite load. Since, by definition, swarms are brood-free it’s an ideal time to treat them.

Queen clipping – why?

I sometimes have colonies in my (very) small suburban garden … it’s great to be able to watch the bees before leaving for the lab or to observe them early in the season bringing in pollen from the crocuses. It’s also a convenient staging post between my out apiaries and a whole lot easier than carrying heavy boxes around through waist-high field margins. However, I’m aware that my neighbours may not share my enthusiasm for bees. I therefore do my utmost to only keep well-behaved colonies in the garden by selecting for docility as a priority when queen rearing. In addition, I make sure any queens heading colonies in the garden are clipped. Queen clipping is the trimming of one wing, preventing the queen from flying any distance should the colony swarm. In the absence of a queen, a prime swarm leaving the hive will either return to the hive or will cluster with the queen a very short distance from the hive.

Clipped queen ...

Clipped queen …

A colony in the garden swarmed on open queen cells (QC) last Sunday afternoon. The colony had chosen to ignore the super, so filled the brood box with nectar (I suspect I’d added the super too late and the colony had already started to think about swarming). Consequently the colony ran out of space. The QC’s were about 3-4 days old and unsealed. There’d been none present at the previous inspection (remember that colonies usually swarm once the first QC’s are sealed). The colony was half-way through a vertical split (to be described in the future) with the original queen in the top box and the newly emerged virgin in the bottom box. I’d been away and arrived home to find the top box swarming and the air filled with bees. With an unclipped queen they would usually settle in a nearby tree or bush and then send out scouts to find more desirable accommodation.

I might have been fortunate enough to catch this, but they might have settled somewhere inconvenient like the chimney or on the kids trampoline in the garden next door. However, because the queen was clipped, she couldn’t fly and the bees just milled about for 15 minutes … a fantastic sight and sound. Eventually they returned to the hive … but to the bottom box. Shortly after they’d settled I found the queen and a small retinue of workers on the ground about a metre from the hive entrance (see photo above). I quickly went through the top box, shaking bees off the frames and knocked off all the QC’s. I also swapped out a couple of nectar-filled frames for drawn comb. I then ran the queen back into the entrance. With luck the reduced density of bees and increased space to lay will discourage them from swarming again*.

A queen with a clipped wing generally swarms later than an unclipped queen, potentially giving you a few extra days between inspections. However, as the example above shows, you can’t rely on this so seven day intervals between inspections are still recommended. Had I not found the queen she would have probably crawled back to the hive stand, climbed up the leg and ended up under the open mesh floor. Although this is not ideal, it provides another opportunity to recapture her and it’s far preferable to losing the bees altogether or bothering your neighbours with swarms.

Summer storm ...

Summer storm …

Although the weather was wonderful when the colony swarmed, it rapidly changed later in the afternoon when we were treated to downpour of biblical proportions … any swarm caught out in the torrential rain and hail would have probably fared very badly.

Time to close the hive up ...

Time to close the hive up …

The image above (the densest cloud formed a wide band from the North East to the South West, almost directly above three of my apiaries) is a composite of three images stitched together in Photoshop. I was desperately trying to get through the last few hive inspections but had to abandon things and seek shelter in the car. The rain and hail didn’t last long, but what it lacked in duration it more than compensated for in volume (both sound and fluid ounces).

Perhaps surprisingly, in the 30 minutes or so before the heavens opened the bees were remarkably well behaved.

* Update on checking six days later (today) the blue marked and clipped queen is back and laying again in the top box. It looks like she’s been getting a lot of attention as the blue paint has almost disappeared. There are no signs of any more queen cells but they’re still not taking much notice of the super. Unfortunately, they are showing signs of robbing another colony in the garden, so I’ll shortly be moving them to another apiary.

In the meantime, I prepared a stack of boxes in preparation of moving house and – within 24 hours – another swarm moved in. I’d missed a finger-wide gap in the stack and the bees occupied a chest-high pile of broods and supers. These look like another generous donation from a neighbour … thank you.

Swarm care and treatment

As I left my out apiary last Friday evening I gently tapped a bait hive buried in the grass and nettles and was met with a healthy buzz and a few inquisitive worker bees … it had obviously been occupied by a swarm in the last day or two, despite me not having seen any scout bees investigating it.

Super poly bait hive

Super poly bait hive …

I checked the hive on Sunday and found a small swarm covering about 3-4 frames with a dark, unmarked laying queen. She’s definitely not from my colonies as all mine in that apiary are marked and accounted for. At the back of the hive was an unoccupied wasps nest, beautifully constructed from paper-thin chewed wood pulp. This is the second bait hive with squatters this year.


Waspkeeping …

Although there’s a reasonable flow from something at the moment (early blackberry?) I gave them a couple of pints of syrup on Monday and Wednesday evening to help them draw comb on the foundationless frames that fill the box. It’s not advisable to feed syrup immediately in case the honey stores brought with the swarm are contaminated with foul broods – by delaying feeding (or not feeding at all) the bees use their stores to draw wax. I also treated the colony – which has no sealed brood yet of course – with oxalic acid vapour from a Sublimox vaporiser (shown below being used early in the season on a full colony) bought a few months ago from Icko Apiculture.

Sublimox vaporiser

Sublimox vaporiser …

The phoretic mites on a swarm carry an unpleasant payload of viruses including deformed wing virus. It’s therefore good practice to keep the swarm in isolation until it’s known to be healthy, and to treat appropriately for mites as soon as possible. I also treated the churchyard swarm caught last Thursday with OA vapour despite not yet being sure whether the queen is mated or not – if she is then it’s best to treat before the colony have sealed brood, if she isn’t then OA vaporisation is sufficiently ‘gentle’ that I don’t expect the treatment to interrupt her from getting out and mated in the current good weather. By treating with OA vapour late in the evening I wouldn’t interrupt a mating flight and could be pretty sure that most of the bees – and therefore most of the mites – were ‘at home’.


Gotcha! …

It’s good practice to keep records on where swarms were found, hived and how they were subsequently treated.


Churchyard swarm

Quiet churchyard

Quiet churchyard

While away on ‘bee health’ business for the day in York I received a call around midday that there was a swarm settling in a small tree in a local churchyard. The combination of the words “small tree” and “within arms reach” is always reassuring, so I promised to have a look when I got back. Inevitably I was delayed and it was nearly 9pm by the time I turned up in the churchyard. I fully expected it to have been collected by another beekeeper, or to have disappeared to a bait hive or even the church tower … but it hadn’t. The swarm was quite small (I suspect it may be a cast) and tightly clustered – exactly as described, in a small tree easily reachable without steps. Excellent.

As the final peals of the bells died out I dropped the swarm into an eight frame poly nuc box, gently lowered a full complement of foundationless frames on top, replaced the perspex cover sheet and roof and waited while the few stragglers entered the box. It was lovely sitting in the gathering gloom listening to the fanning bees at the entrance – indicating the queen was in the box – as the evening ebbed away. It was too dark for any photographs unfortunately. Shortly after 10pm the nuc box was installed in my garden on a levelled stand – to allow the bees to draw the foundationless frames out vertically – and they were busy making orientation flights when I checked at 6am the following day.

Although the bees looked perfectly healthy I’ll keep them away from my main apiaries until I see some brood and can check them thoroughly. In the meantime, and before they have sealed brood, I’ll treat them with oxalic acid vapour to minimise the phoretic mite numbers. To help them draw out new foundation I’ll give them a few pints of thin syrup (I’m still using up some old fondant left over from last winter) if the current nectar flow dries up – the rape is gone and the blackberry is just starting, but my other colonies are bringing something in. Finally, I’ll keep a close eye on their temper and general behaviour and, if unacceptable, will either requeen them or unite them with another colony.