A Game of Drones on the Isles of Scilly
BeeCraft has carried news of the Game of Drones on the Isles of Scilly (October 2022) as well as an introduction to the natural laboratory that the islands offer (August 2021). The Game of Drones unexpectedly caught the public imagination and BBC Radio 4 and The Times have been following the story, so what can we say at the end of the pilot drone-marking project of 2022?
Do drones island hop and, if so, what effect might this have on the sustainability of honey bee colonies and beekeeping on the archipelago 25 miles off the tip of south-west England?
The 2022 Game of Drones on the Isles of Scilly has given intriguing if tentative insights about the behaviour of honey bee drones and some ideas about what that means for colony genetics and the risks of inbreeding or spread of disease and pathogens across an isolated group of islands. The main question posed was: do male honey bees, well known for their drifting between colonies, fly across water to bring their genes to a virgin queen on another island?
Origins of the Game of Drones
It started in 2017 when Nick Bentham-Green and Jo Widdicombe of BIBBA visited Scilly in search of a native dark honey bee. They encountered some tetchy bees of unknown provenance but thought they were varroa free and perhaps free of the two foulbrood diseases and other pathogens that are affecting bees on the mainland. The arrival of Jilly Halliday, a keen beekeeper, on Tresco in 2019 opened up new opportunities to investigate the Scillonian honey bees, so in 2021 Nick returned and with Jilly and the island beekeepers took samples of the islands’ drone pupae to try to understand their genetic ancestry and inter-relatedness (see BeeCraft July 2021 for details of that story). Drones were chosen because DNA analysis of them would provide relatively quick, low-cost results.
Before leaving the islands in 2021, the team wondered how they could investigate if the bees on each island were isolated or if their genes were mixing freely by island-hopping drones. The islands, 25 miles from the Cornish mainland, weren’t going to be visited by honey bees without human help, accidental or otherwise. But the islands were close enough to each other to present the possibility that bees might island hop. They offered a perfect natural laboratory and an idea, nicknamed the Game of Drones, was born: could drones be tracked to see if they were flying to other islands and their mating sites?
The honey bees of Scilly
Across the five inhabited islands that are known to have bees, there are up to 50 known colonies in 15 apiaries or clusters (see map on page 11) including at least three wild/feral colonies. In total, that number might be sufficient to provide genetic diversity that would not lead to inbreeding. Andrew Abrahams on the Scottish island of Colonsay reports no inbreeding problems with his 60 colonies, originating from 30 queens, that have been isolated for about 40 years. Because of the size of the individual islands and the limited forage, no individual island could reasonably be expected to maintain the number of colonies that would nullify the risk of inbreeding. However, if bees are mixing across all the islands, inbreeding would be less of a risk.
So, might the drones be island hopping and thereby helping to increase the genetic diversity?
The Game of Drones methodology
To discover if drones could be spreading their genes across the islands, a simple pilot experiment was devised. The aim was to mark as many drones as possible on the thorax with different colours depending on which island they were found – blue for Bryher, green for St Mary’s, purple for Tresco, orange for St Agnes and red for St Martin’s. The marking was almost identical in materials and method to traditional queen marking performed by many beekeepers. For logistical reasons, marking had to happen at the beginning of July but there were enough drones around to make the project feasible and the warm weather of summer 2022 on Scilly was ideal for honey bee flight. In any event, as a pilot project, it was going to reveal many of the issues that could be better addressed in a properly funded project in 2023.
More than 500 drones were marked across the five islands. This was perhaps fewer than hoped, but the populations of drones varied considerably from colony to colony – some had plenty, some had none – and this may have reflected the unusual weather of 2022. Drone population sizes are strongly influenced by the forage available (if there is a dearth, drones will be sacrificed) and, by the beginning of July, the dry conditions were severely affecting the available island forage. On some islands, such as St Martin’s, there are two distinct flows: pittosporum in spring, blackberry in summer but a significant dearth in between that could well affect drone numbers. (As beekeepers, members of the team soon realised how little attention we had previously paid to drones, but soon became much more attuned to drone numbers as a significant indicator of colony health and/or overall state.)
Andrew Lawson and Jilly Halliday marking drones – purple for Tresco
Tresco Abbey Gardens apiary
Ever the innovator, Matthew Elmes had devised a rapid drone-marking scheme with Jo Widdicombe of Cornwall. For colonies with many drones, the males were lured into a trap between two queen excluders so that they could be rapidly marked through the upper excluder. In other cases, the drones were marked as they moved across the comb or, if they were fleet of foot, lifted and marked.
The mainland members of the team weren’t on Scilly long enough to wait to check for island-hopping drones – that was left to the Scillonian beekeepers. Although, that later inspection of hives was not as regular as we might have hoped in the busy summer season (beekeepers must take advantage of the tourist income), there were no reports of island-hopping drones. However, a little drone-drifting was seen on Tresco in an apiary which had not been part of the drone-marking exercise.
DCAs on Scilly
Since drones are believed to fly mainly or only in the afternoons, marking was undertaken in the mornings, leaving the afternoons free to search for drone congregation areas (DCAs) that might offer evidence of island hopping.
In successful searches for DCAs in southern England, I have found that temperatures generally need to be 19oC or above (see BeeCraft, November 2016). However, with its frequent winds and breezes, the island temperatures don’t often exceed the low 20oCs, so I had low expectations of finding one.
Luckily, while we were there the temperatures were in the low 20oCs and, using a lure on the end of an extendable carbon fishing rod and/or a radio-controlled drone with a lure on board, we found four DCAs on three of the islands. We were unsuccessful on St Martin’s, although conditions seemed favourable, and on Bryher, where the wind chill may have inhibited the drones on an otherwise fine day.
Fittingly, a DCA on St Mary’s was found at a obsolete windmill called the Buzza Tower, which overlooks the harbour at Hugh Town. (Disappointingly, the name Buzza seems to have no connection to the sound of bees.) Two DCAs were also found at either end of the famed Abbey Gardens on Tresco.
The DCA on St Agnes proved a satisfying surprise. Lure aloft on the fishing rod and walking up the single path to the island’s lighthouse, a couple of drones drifted by. Obviously, there was a DCA somewhere nearby, but houses and bracken-filled fields prevented straying from the path to try to find the exact area. Then, we came across the house of Mike Hicks, the island beekeeper. Hoping to find him in, we went into his garden with the fishing rod lure slung casually over one shoulder. Buzzing was heard – the garden was on the edge of the DCA! Mike was off island but, with his son Aidan, we explored further and, using the radio-controlled drone, found the core of the DCA above the pittosporum shelterbelt hedges on the small farm. Mike and Aidan had indeed noticed some unexplained bee movements over the years, but never realised the farm hosted a DCA.
DCA hunting
Drone congregation areas, where drones gather in the afternoon in the hope of mating with a virgin queen, are one of the many mysteries of the honey bee world. In different environments, rules for DCAs locations vary – so much so that no one has yet established common criteria.
In the UK, it’s very unusual to find a DCA without a lure (a caged queen or a synthetic version of the 9–ODA to which drones are attracted). At a DCA they arrive usually within one minute, showing that they are already in the vicinity. In southern England, BeeCraft has carried several recent reports of DCA hunts (eg, November, 2016; August, 2017).
On the Isles of Scilly, we used a synthetic 9–ODA lure hoisted 3–4m above ground level and on occasion a radio-controlled drone taking the lure to a height of up to 30m above ground level. The topography of DCAs on Scilly closely matched that of DCA finds in other parts of southern England – hilltops, but on Scilly on land 20–30m above sea level rather than up to 200m often discovered elsewhere. (The highest point on any of the islands is 51m, on St Mary’s.)
The Isles of Scilly
Almost entirely owned by the Duchy of Cornwall (now William, Prince of Wales), the islands number about 80, five of which are inhabited and known to have colonies of honey bees. Located about 40km from Land’s End in Cornwall (too far for any swarm to venture, especially against the prevailing south-westerly wind), the shortest distance between any the five main islands varies from 300m to just under 2km. From the tip of St Agnes in the south to the tip of St Martin’s in the north west is just under 12km.
Mysteries
There are many surprises we uncovered in the Game of Drones. Two stand out.
Surprise 1: On the Tresco DCA which faced Bryher, one kilometre away across shallow water, we caught some of the drones in a net to see if any had coloured thoraxes. Curiously, only two of more than 100 drones temporarily trapped were marked – both purple, indicating their Tresco origin. The provenance and sheer number of the other drones remains a mystery and we suspected that there are wild colonies nearby that we didn’t know about. In the past few weeks Jilly Halliday has indeed discovered two previously unknown wild colonies on Tresco.
Students of the Five Islands Academy releasing drones that they had marked
Surprise 2: On the Friday that we discovered the DCA at the Buzza Tower in Hugh Town, St Mary’s, the drones were plentiful. On the Sunday, the team had been asked to remove a tetchy colony in a chimney, not far below the Buzza in the built-up area of the town. That done, we returned to the Buzza to test its DCA on the Tuesday afternoon that some of us were due to leave Scilly. The conditions and time were almost identical to the previous Friday but not a single drone came to the lure. It seems that on the Friday, all of the Buzza DCA drones had originated from the chimney, just 100m away due north. Now, with the nest relocated, there wasn’t a drone to be found at the Buzza. And yet, 100m due south of the Buzza, there were two colonies with (marked) drones in an allotment apiary. Sadly, as we had a boat to catch, we didn’t have time to search for another DCA that might have been used by the allotment apiary drones.
Speculation rather than conclusions
Fascinating though our pilot study on the natural laboratory of Scilly was, we are left with as many questions as answers, some of which we hope to address with funding and more rigorous methodology in 2023. Meanwhile, we can dare to speculate.
Although drones are known to drift from hive to hive, perhaps long-distance drifting is uncommon, especially if there is a sea to cross. Perhaps drones are a little lazy – or energy efficient. Perhaps they fly predominantly to a very local DCA (which could be very close to or even in the apiary in poor weather).
Even if a few drones do island hop, what is their chance of engaging in a successful mating? Given that recent research has indicated there may be as many as 11,000 drones per queen (unlikely to be so high on Scilly) and that we have long known that queens may mate with 16 or more drones, what chance is there for a few island-hopping drones to make a success of their lives by ending it through spread their genes?
However, and unfortunately, a single island-hopping drone could make an impact if it carried a pathogen, such as a virus or varroa mite.
Of course, it takes two to mate. Might the queens island hop? This seems unlikely given that research and casual observations suggest they leave the hive on mating flights lasting much less than 30 minutes. There is little research evidence about how far queens are prepared to fly, but perhaps new tracking technologies will one day shine a light on that.
Perhaps, as the preliminary DNA analysis suggests (see STOP PRESS), any inter-island genetic mixing is down to human transportation and possibly swarm movements rather than island hopping by drones.
Next season
We hope to return next season to investigate further and to refine the methodology. For example, with funding we could start earlier, make additional visits and help the local beekeepers in looking for marked drones since that is required exactly when the local beekeepers are working full tilt at income-generating jobs. We would aim to mark many more drones, and (perhaps with numbered discs) to show from which apiary or even colony they were first seen. We might also mark the queens distinctively so that if they do island hop in swarms we could have conclusive evidence of that.
Tailpiece
In mid-September, Jilly Halliday found a queenright colony in her Tresco apiary with a huge number of drones, one still with its purple ID marking, more than eight weeks after being marked (textbooks say that drones live an average of about seven weeks). Nectar was still coming in from the heather and belladonna.
On 8 October, the drones were finally thrown out of the hive.
Game of Drones teamÂ
Matthew Elmes and Josh Syrett of Pollenize, Kylie Carter beekeeper on St Mary’s and Jilly Halliday of Tresco (left to right in the photo), plus staff and students of the Five Islands Academy, Grace Twiston-Davies of Wildflower Collective (supporting the project from Falmouth), and of course other beekeepers of the Isles of Scilly: Ruth and Graham Eggins on Bryher, Mike and Aidan Hicks on St Agnes, Ben Gilett on St Martin’s, and Andrew Lawson and Richard Hobbs on Tresco. Lots of people and organisations on Scilly helped facilitate the project. The B4 Project provided significant funding and the St Mary’s Boatmen’s Association ensured that the team didn’t have to emulate the drones’ methods of island hopping.
Stop press: DNA results
Just before we went to press, preliminary results of the 2021 genetic analysis of 55 antennae from drone pupae in 11 apiaries across the five bee islands of Scilly was released by Beebytes, the honey bee genetics specialists at the Roslin Innovation Centre near Edinburgh.
Diagram explainer. The diagram below combines two types of analysis – the genetic heritage of the islands’ bees (pie diagrams) and their inter-relatedness (thickness of linking lines).
Lineage/ancestry. Each pie chart represents a single drone pupa and the outline colour (and position) the island on which it was sampled. The pie charts display the genetic ancestry, as indicated by the lineages. C-lineage (Carniolan and Italian origin, yellow in the pie charts) is strongest on St Agnes, while M-lineage (north-west European origin, black in the pie charts) is strongest on Tresco. O-lineage (Middle East origin, also found in ‘Buckfast’ bees, blue in the pie charts ) exists but at such a low percentage (<6%) that it is barely visible on the diagram pie charts.
Relatedness. The thickness of the lines joining the pie charts indicate the relatedness of the bees (the thicker, the closer the relatedness), so while drones on the same island may show a similar genetic lineage (ancestry) they may not be closely related and may therefore form separate clusters even on the same island.
The BeeBytes interactive diagram can be seen online at beebytes.org/scilly.
Initial thoughts
The results have only just become available and so need to be considered in depth, but early thoughts suggest that, with the marked exception of Tresco and Bryher, there seems to have been surprisingly little interchange of bees across the islands.
The similarities across Tresco and Bryher may arise from travelling swarms but we have also just discovered a split of one of the colonies imported to Tresco from (varroa-free) Isle of Man six or seven years ago was given to Bryher. Although the two islands are only 300m apart at their closest points and therefore easier for drones, queens and/or swarms to island hop between them, human intervention may be the key in explaining the bees’ close relatedness.
St Agnes stands out with its high C-lineage ancestry and this tallies with what is known about the importing of bees to the island (for the latest understanding of this, see the fascinating recollections of Mike Hicks, the St Agnes beekeeper in the separate panel on page 10).Â
Stephen Fleming, co-editor
Stephen Fleming is co-editor of BeeCraft, the international award-winning honey bee and beekeeping magazine. A beekeeper for more than 30 years, he is a researcher by training and obtained a PhD at the University of Reading. He has been interested in environmental issues since he was a child and beekeeping continues to open yet more doors that increase his understanding of the natural world.
