By Geoff Tribe and Ujubee
Occasionally when a swarm issues from a nest and hangs from a branch a little way off, a few drones may be seen hovering around the swarm, possibly attracted to the pheromone secreted by the queen. Drones are however not known to accompany a reproductive, absconding or migrating swarm, although drones have been recorded departing from clustered swarms around a tethered queen (Burgett, 1974). Yet this appears to have happened in a swarm which had arrived unseen in a natural nest under a rock no more than 30 days previously where multiple drones were immediately present – which would mean that drone comb would have had to have been constructed simultaneously with that of worker comb and brood rearing immediately begun.
But the duration of the life stage of a drone is 28 days and it is only sexually mature after 36 days on which it takes its first mating flight. Drones do not take part in any of the functions within a nest; their sole function is to mate with a queen, which happens in the air in drone congregation areas (DCA) a considerable distance from the nest. A virgin queen will mate with up to 40 drones and some sperm from each drone will be stored in her spermatheca. Thus over a period of time the nest will contain workers of various patrilines (paternal lines). Inbreeding, the mating of a queen with a related drone, is detrimental to a colony and DCAs containing hundreds or thousands of drones ensure that the chance of mating with a related drone is extremely remote.

Honeybee drones attracted to a strip of cloth attached to fishing gut and a helium filled balloon which had been lowered to near ground level within a drone congregation area in Pretoria.
How the drones find these DCAs is not fully known, but wind direction, landmarks and pheromones all appear to play a role. Both queens and drones find these DCAs independently and so respond to the same cues. Drones leave the nest independently on optimal days for flight, although at peak flying times the rush of drones to the exit and away may appear like a stampede. Flight usually begins at 12h00 and ends at 16h00 in which an average of three flights of varying duration by a single drone may take place. As the period for flight nears, drones appear to become ‘restless’ and begin combing their eyes with their forelegs, especially at the nest entrance just before take-off. In between flights the drones rush to open honey cells to replenish their energy – which takes 3 minutes on average.
Drone comb is built in an established colony when there is a surplus of pollen, and is built after that of the requirements in worker brood has been met. The drones tend to congregate in a remote part of the nest away from the brood nest when inactive and take no part in any of the functions within the nest. They never appear to follow the foraging or swarming dances of the workers or are located in the area where the dances take place. The interaction of workers with drones is limited largely to allowing them access to uncapped honey and of expelling them from the hive during times of dearth, mainly due to a shortage of pollen. The drones are either forcefully expelled from the nest by workers or are denied entrance to the nest on return from mating flights.
Swarms that issue from a nest may be reproductive or absconding (migrating) swarms, the latter induced by disturbance (predation, manipulation) or due to a dearth of resources (Winston et al. 1979). Congestion in the nest of a rapidly expanding population of bees results in the issuing of reproductive swarms where the old queen departs with young bees (Hogg 2006).
During 2016 a wild honeybee colony that had constructed its nest under a boulder and enclosed the opening with a propolis wall absconded. This nest site remained empty for several months and was inspected at irregular intervals. On 28/6/2016 the nest was again inspected and found to be empty. The propolis wall was analyzed and removed as part of a research project, but on 29/7/2016 it was found to be inhabited by a new swarm which had immediately begun rebuilding any damaged comb. The most amazing discovery was that many drones were departing on mating flights and could only have accompanied the swarm on its migration because drone brood needs 28 days to develop from egg to adult and is only sexually mature after 36 days and is only produced after sufficient worker brood is present in the colony. It is unlikely that the drone presence could be the result of drift because of the numbers involved, the lack of honey stores and the nearest colony being located a considerable distance away.
A swarm about to abscond sends out flight experienced bees which locate suitable nesting sites. On return to the nest they dance to indicate the direction and distance to the proposed nesting site. There may be dozens of dancers indicating a dozen or more sites, but over time only one dance will predominate (Seeley & Buhrman 1999). This means that all the scout bees have visited all the potential sites prior to the swarm issuing and have agreed by consensus on which site is the most suitable. Dancing scouts can be recognised in having no pollen on their legs and they do not stop at intervals to offer nectar to those following the dance.

Small swarm of Apis mellifera capensis hanging from a strand of barbed-wire during the flowering of canola near Caledon during which many swarms trek.
At the beginning of absconding, thousands of bees pour out of the nest and mill about in a swirling mass which begins after a period of time to coalesce into a swarm which alights on a branch or another such structure where they remain for several hours or even a day or two before they finally depart for the selected nesting site. Yet, the dance indicating their destination continues on the outside of the swarm of hanging bees, using the mass of bees as the substrate. Sometimes a few drones are seen with such swarms but, being so close to the original nest, they are usually regarded as having wandered on return to the nest by being attracted to the queen pheromone. Somehow the swarm manages to keep together despite individual bees flying in circles around the queen as they move rapidly through the air in a straight line while at the same time orienting towards a predetermined destination.
Drones however never appear to follow dances and therefore cannot be expected to understand them. Yet a large number presumably arrived at this new site and successfully migrated with the swarm. This could have been accomplished by one of three ways or a combination of all three – pheromones, sound (vibration) and sight.
A swirling swarm of bees on the move creates a distinct sound which is easily audible as it passes by; it can be followed by eye and although it moves at a pace, it could be followed by a runner if over level terrain; and both the queen and workers are releasing their pheromones as they fly to keep the bees together in the swarm. It has been established that in the case of reproductive swarms at least that there is a continual flight of worker bees between the old hive and the new site while swarming is in progress which steer them to the new site which may be several kilometres away (Wenner 1992; Seeley & Buhrman 1999). Hundreds of scout bees thus generate an aerial pathway by establishing an odour trail of Nasonov pheromone as they fly back and forth through the moving swarm (Wenner 1992; Schmidt et al. 1993). As explained by Lindauer (1955): While the swarm cloud is proceeding gradually along one sees a few hundred bees shooting ahead through the crowd in the direction towards the nesting place, then flying back at the margin of the swarm cloud, again pushing forward rapidly, and so on, until the goal is reached. Bees already at the site release Nasonov pheromone which serves to further attract the swarm and serves as a ‘settling’ or ‘orienting’ pheromone (Schmidt et al. 1993; Schmidt 1994; Wenner 1992). Because of the pivotal role played by pheromones, migrating swarms invariably issue only when weather conditions are favourable, especially the absence of strong winds.
What is also amazing is that swarming bees would tolerate so many drones which cannot forage for themselves and so have to rely on the workers to feed them. On arrival at the nest site, initial comb building relies on the honey stored in the honey-stomachs of the migrating workers which is of crucial importance for establishing a new colony. Perhaps drones are only able to accompany swarms in times of plenty of forage – drones regarded here as of value to the species even though they will not mate with a related queen from their own colony i.e. the workers tolerate/rear their drones for the benefit of another colony.
Drones cannot live on their own and rely on the workers to provide them with sustenance and to heat the nest. An absconding swarm will therefore have severe repercussions for any drones in the nest unless they could depart with the swarm. Bees abscond due to various causes including pests, disease and disturbance, but the main reason is usually a dearth of nectar and pollen. Ordinarily the latter would mean few if any drones in the nest. For drones to leave with the swarm would presuppose that the drones were aware of this and, despite limited interaction with workers, would have recognised the cues as the nest started its relatively lengthy preparations to swarm. This would involve a drastic decrease in brood-rearing, few if any larvae present, all sealed worker brood having emerged and no stored pollen, nectar or honey (Winston et al. 1979). Brood rearing would have ceased, the queen would have been fed less lavishly to enable her to fly, the last capped brood would be emerging, and the honey stores would be depleted. Could distinctive sounds that are produced during the last few days before the swarm issues from its present colony, ‘a buzzing like that of an army setting out on the march’ (Wenner 1992) be recognized by drones as to what it portends? In the case of reproductive after-swarms the peculiar humming noise (Esch 1967) within the hive could also be augmented by the piping of queens. So the signs would be there but are drones able to recognise them?
When travelling with the swarm, do drones remain on the periphery or immerse themselves within the body of the swarm? Drones are strong flyers and able to fly as fast as or even faster than worker bees. The workers are aware of where they are going because they reach consensus before they depart to the new site they have selected, but the queen who had not taken part in following these dances would be as much a ‘passenger’ in the swarm as would be the drones. The queen would at all times be surrounded by, yet following the workers, as a safety precaution against predators. By releasing 9HDA pheromone, the queen would contribute to the clustering behaviour of the bees in the swarm (Winston et al. 1982). Thus both queen and worker pheromones would be integrated into facilitating the flight of the swarm, and possibly, together with sight and sound, enable the drones to attach themselves to the swarm.
A similar perplexing observation involving small hive beetles (Aethina tumida) by Dr Lundy in Pretoria where he witnesses a flailing ball travel through the air as if in a swarm directly to one of his garden hives where the beetles ‘melted’ inside (Tribe, 2000). Hours later the bees absconded due to the overwhelming numbers of beetles. How this ‘swarm’ of beetles emerged from one hive possibly a considerable distance away and were able to move en masse in a co-ordinated manner and to invade another hive is unknown.
References
Allen, M.D. 1965. The effect of a plentiful supply of drone comb on colonies of honeybees. Journal of Apicultural Research 4(2): 109-119.
Avitabile, A. and Kasinskas, J.R. 1977. The drone population of natural honeybee swarms. Journal of Apicultural Research 16(3): 145-149.
Burgett, D.M. 1974. Drone honey bee flight from clustered swarms. Annals of the Entomological Society of America 67: 683-684.
Esch, H. 1967. The sounds produced by swarming honey bees. Z. vergleich. Physiol. 56: 408-411.
Hogg, J.A. 2006. The anatomy of reproductive swarming. American Bee Journal 142(2): 131-135.
Lindauer, M. 1955. Schwarmbienen auf Wohnungssuche. Z. vergleich. Physiol. 37: 263-324.
Schmidt, J.O. 1994. Attraction of reproductive honey bee swarms to artificial nests by Nasonov pheromone. Journal of Chemical Ecology 20(5): 1053-1056.
Schmidt, J.O., Slessor, K.N. and Winston, M.L. 1993. Roles of Nasonov and queen pheromones in attraction of honeybee swarms. Naturwissenschaften 80: 573-575.
Seeley, T.D. and Buhrman, S.C. 1999. Group decision making in swarms of honey bees. Behavioral Ecology and Sociobiology 45: 19-31.
Tribe, G.D. 2000. A migrating swarm of small hive beetles (Aethina tumida Murray). South African Bee Journal 72(3): 121-122.
Wenner, A.M. 1992. Swarm movement: a mystery explained. American Bee Journal 132(1): 27-31.
Winston, M.L. and Otis, G.W. 1978. Ages of bees in swarms and afterswarms of Africanized honeybee. Journal of Apicultural Research 17(3): 123-129.
Winston, M.L., Otis, G.W. and Taylor, O.R. 1979. Absconding behaviour of the Africanized honeybee in South America. Journal of Apicultural Research 18 (2): 85-94.
Winston, M.L., Slessor, K.N., Smirle, M.J., and Kandil, A.A. 1982. The influence of a queen-produced substance, 9HDA, on swarm clustering behaviour in the honeybee Apis mellifera L. Journal of Chemical Ecology 8(10): 1283-1288.