Why do honeybees react the way they do to smoke?

By Geoff Tribe (after a discussion on smoke with Jenny Cullinan and Karin Sternberg)

Fire at Cape Point Nature Reserve

The 11km fire at the Cape of Good Hope Nature Reserve, March 2015

While viewing the aftermath of the recent fire that cut a swath of devastation through Cape Point Nature Reserve before it was extinguished, we wondered whether any of the wild swarms we were monitoring there had survived the blaze.

after the fire 1 after the fire 2, The question then arose as to what benefit the innate behaviour of imbibing honey in the presence of smoke could be? How did it contribute to the survival of a swarm whose nesting site is enveloped in fire?

The adage that it is better to work honeybees with a smoker and without protective clothing than vice versa is easily confirmed in practice. The earliest record of the use of smoke to control honeybees is from a tomb painting in Egypt from c. 1450 BC where a kneeling beekeeper is removing combs from the back of a stack of horizontal hives, while smoke was used to drive the bees off these combs towards the flight entrance at the front (Crane, 1999). The use of smoke to subdue honeybees has obviously been known for millennia, because a hieroglyph of the honeybee was part of the titular of the King of Egypt from 3100 BC, the honeybee being of importance in Egypt long before this time. But why should the bees behave the way that they do in the presence of smoke unless it had evolutionary benefits? Various materials may be used in a smoker to subdue honeybees, each beekeeper vouching for the merits of the material that he uses in calming the bees. These may include burlap sacks, dry pine needles and dry herbivore dung – all of which have the same effect on the bees despite what material is used as fuel. Historical records show that dagga was used by the Khoesan (Quena) to smoke the bees but without any comment on any added effects that it may have had on them!

Smoke consists of many compounds, and since all fuels are equally successful in subduing honeybees on whatever continent, they are possibly reacting only to one or more of the chemical components of smoke. These components are as yet unknown, and neither are the receptor sites on the bee where this recognition is facilitated. Another suggestion is that the receptors of the honeybees are reacting to a certain size of smoke particle.

What is not disputed is that smoke causes honeybees to imbibe honey and with their distended abdomens they are less inclined to sting. Honeybees have first to grip the target before the sting can be brought into play by bending the abdomen, a distended abdomen making this procedure more difficult. Smoke also disrupts the chemical communication of the bees. Bees protect their valuable honey stores from mammals by a concentrated mass attack on the intruder after it has been marked by a guard bee whose stinging apparatus is ripped out and the alarm pheromone is released in quantity. Hence the practise by beekeepers of smoking the area that has been stung to smother the odour and to confuse the bees. The Western honeybee (Apis mellifera) inhabits temperate areas which are subject to regular fires which are part of their ecology. Savannah grasslands and fynbos need fire to survive as such.

Honeybees however are unable to escape fire by flight as witnessed by entire apiaries being burnt to cinders and all swarms perishing. Honeybees only react to smoke if it wafts into their hive, which gives them little warning of impending disaster. Observation hive studies have shown that honeybees which abscond do so after consensus has been reached within the hive – which takes many days. By overcrowding an observation hive with bees it is possible to force the swarm to abscond. As the bees prepare to abscond, perhaps six dances indicating six potential nesting sites may occur initially. Over a few days this number is narrowed down to one dance and consensus has been reached. The swarm leaves the hive after the number of vibratory dances (an inhibitory dance where one worker places its head on another and shakes it by a dorso-ventral abdominal movement) (Fletcher, 1978) per unit time reaches zero. exposed nestThere would not be time to organise a mass evacuation of the hive in the event of a fire. In any case, smoke would disrupt all chemical signals involved in coordinating swarming and absconding. Exposed wild colonies would perish , the selection of secure nesting sites in fire-prone vegetation thus having primary survival value.

Nest under a rock

A wild nest deep under a rock

Laying queens would be unable to leave with a swarm escaping a fire because she would be too heavy to fly and, like the workers, would not be inclined to desert her brood. A swarm preparing to swarm or abscond first starves their queen and prevents her from laying eggs by not preparing cells for her in which to lay eggs. Absconding occurs soon after most young bees have emerged from their cells and the stores have been consumed – except that which is needed to establish the swarm at the new nesting site and is carried in the bees’ stomachs. This preparation again takes several days. Absconding usually takes place when conditions are adverse and there is an ‘awareness’ of this throughout the hive due to the ‘feedback’ mechanism in operation within the hive, where for example, the lack of incoming food causes a reduction or cessation in egg-laying.

Honeybees are therefore unable to escape by absconding in the face of the fire and exposed colonies would perish. The innate behaviour of honeybees to smoke must thus have become entrenched in those swarms which did survive the fire. The bees are therefore not trying to escape the fire by fleeing – so what is the value of imbibing honey? The value appears to be in the survival of those swarms which escaped the initial fire.

Natural fires in southern Africa in days of yore would burn over a vast area. For example, the fire that started in Riviersonderend in February 1869 that was driven by a hot Berg wind burnt the indigenous forest all the way from Swellendam to Humansdorp, driving elephants and other game into the sea for refuge. Swarms surviving such fires would emerge to find a ‘lunar’ landscape devoid of plants to sustain them and thus would be forced to abscond. Smoke causes bees to imbibe honey and retreat to the far recesses of the nest cavity. Bees will imbibe any honey flowing from melted combs and so function as living storage units. Under such conditions, to sting and die would cause the loss of this crucial honey store. Hence also a reason why they are then less inclined to sting. It appears that the innate behaviour of honeybees to imbibe honey in the presence of smoke has survival value in that it allows the eventual escape by migrating from a devastated landscape after the fire has passed. The honey stored in the bodies of the bees would allow the bees to survive the initial calamity and then to organise a migration out of the area and rebuild comb in the new nest site. This behaviour has become innate, which presupposes that the Western honeybees (and Apis species in general) have been subjected to fire for many millennia and have evolved to survive it.

The Heuningberg

The Heuningberg in the foreground dwarfed by the Groot Winterhoekberge behind them

A natural indication of the value in imbibing honey was recorded by O.F. Mentzel in 1787 at the Heuningberg (Fig. 4) where honey was bartered with the Khoesan. The translation records: “A few miles towards the east lie the Honig Bergen. These derive their peculiar name from the bees which find abundant nourishment in the many wild flowers and store their harvest in the crevices of the rocks. The Hottentots clamber up these mountains…to look for honey which they discover the sooner because the sun often makes it fluid, so that it runs onto the rocks, is again collected by the bees and carried back but in this way the storerooms are soon discovered”(Tribe, 1996). Temperatures in summer in this region exceed 40°C and the rock faces are blisteringly hot. Here the bees’ reaction is to heat and the melting of the comb but in the absence of smoke.

Do other organisms have innate reactions to fire? There appears to be none with the same innate response as that of honeybees. Various predatory birds are attracted to fire to capture escaping insects and rodents, but smoke is just a learned signal indicating a meal.

Anyone who has tried to remove a nest of the alien European wasp (Vespula germanica) which has invaded the fynbos and which forms massive multi-queen colonies because the African climate is so conducive to them, will rue the fact that they do not respond to smoke as do honeybees. The mass attack by these wasps soon results in the veil becoming completely covered in wasps with stingers extended and the odour from their stings becoming overpowering. With no response to smoke such as retreating from it, presumably because they nest in the soil and are not unduly affected by fires, they have no comparable reaction to that of honeybees. Even so, there is no honey stored in the wasps’ nests, their brood being fed progressively with regurgitated protein from soft bodied insects killed by their foragers.

The innate behaviour of honeybees to smoke has allowed people to effectively control them, making modern beekeeping including the movement of hives to pollinate crops possible.


Crane, E. 1999. Recent research on the world history of beekeeping. Bee World 80 (4): 174-186.

Fletcher, D.J.C. 1978. Vibration of queen cells by worker honeybees and its relation to the issue of swarms with virgin queens. Journal of Apicultural Research 17(1): 14-26.

Tribe, G.D. 1996. Heuningberg: past and present. South African Bee Journal 68(2): 39-47.

Images by Geoff Tribe, Jenny Cullinan and Karin Sternberg


4 thoughts on “Why do honeybees react the way they do to smoke?

  1. Aoife

    Helpful article, thank you. You didn’t say if the hives you were monitoring were ok…

    1. ujubee Post author

      Thank you for your interest! We are currently waiting for our post-fire permits to be approved to be able to enter the areas affected by the burn and investigate the nests. As soon as we do go in to these areas, we will post our findings in a new article. What we have observed from a distance, is that bees are still flying in the direction of the nests (we know that 3 nests were affected by the fire), and also judging by the bee activity on the Fire Asparagus plant, which was the first plant to grow and flower after the fire, there are bees around. Interestingly, the flowers of the Fire Asparagus have a honey scent. It seems that this plant is specifically targeting honeybees!

  2. 中心區

    It’s a pity you don’t have a donate button! I’d definitely donate
    to this excellent blog! I suppose for now i’ll settle for book-marking and adding your
    RSS feed to my Google account. I look forward to new
    updates and will share this site with my Facebook group.
    Chat soon!

    1. ujubee Post author

      Deprecated: Function create_function() is deprecated in /var/www/wp-content/plugins/wp-spamshield/wp-spamshield.php on line 1961

      Dear Elena,
      We do have a DONATE button; it should be located on the right hand side of the HOME page, BLOG page and ABOUT page on this website. We would very much appreciate any donations as we are currently self-funding this project and it is a challenge. We are so excited about the next articles that we will be publishing shortly, all based on our research, particularly in areas affected by fire and how our wild bee colonies have survived. The wild honeybees and little native bees are remarkable creatures and have adapted to this fynbos environment where burns are essential for the survival of this floral kingdom. Dr Geoff Tribe has also put together a fascinating article on rock art paintings of honeybee combs which I will be publishing soon. Please let us know if you have any problems donating or contact us on ujubeeconservation@gmail.com . Many thanks for sharing our blog on your facebook page and kindest regards, Karin and Jenny

Comments are closed.