Bee Buzz Box April 2021 The Honey Bee Queen Replacement Enigma – Part II Requeening Practice

Early European mini-mating nucleus colony setup*

Alan Wade, John Robinson, Jenny Robinson and Peter Robinson

In Part I of The Honey Bee Queen Replacement Enigma we reflected on how honey bee colonies replace their queens periodically and how colonies transitioning to a new queen can generate problems for beekeeper planned requeening. Having provided an overview of the issues this transition may present, we turn now to how problems with standard requeening practice can be best overcome.

Frontispiece: A novel, but less than successful, 1920’s queen mating nuclei setupi

Requeening practice

Many factors enhance requeening successii, not least feeding bees to make them more queen receptive and avoiding conditions where bees are highly defensive. In an insightful study, John Hogg discovered that the general conditions for acceptance of a single queen were identical to those of simultaneous acceptance of more than one queeniii. Once made queenless, he found that a colony would accept any number of queens, though maintaining any additional queen for any extended period of time is another matter.

It is especially worth noting that beekeeper initiated requeening results in the complete turnover of the colony stock where replacement bees are unrelated to that of the just replaced parent colony queen.

Dequeening and requeening Successful requeening practices have almost always relied on the emergency queen replacement impulse. The process of removing the colony queen results in bees immediately starting to raise new queens. With the loss of queen pheromone signalling the colony suddenly becomes highly queen receptive. However, instead of allowing the colony to raise its own new queen, you supply one.

Sometimes emergency, or perhaps supersedure, queen cells will appear after the introduction of a new queen (Figure 5). Their appearance is due to gradual diffusion of new queen pheromones and loss of signalling from the old queen: these cells are transitory and disappear by the time the new queen is released and has commenced laying.

Figure 5 Temporary supersedure (arguably emergency) queen cells developed shortly after queen introduction. Jerrabomberra Wetlands Apiary, 28 January 2016

The sleight of hand – supplying a caged queen in lieu of the colony raising an emergency queen – short circuits the ten or so days it takes for a colony to raise a virgin queen from already hatched eggs (Table 1) and the additional week it takes for her to be mated and to commence laying.

Table 1 Brood development schedule (days) for representative species of honey beeiv. Note the earliest emergency queen to emerge will arise from a 3-4 day old larva.

Requeening without dequeening Alternative approaches to requeening, employing the supersedure impulse, have been devised. They involve either creating conditions conducive to colonies raising queens themselves in the presence of the old hive queenv (Figure 6) or introducing select queen cells to the top of queenright coloniesvi. Here started cells are being transferred to strong builder colony also kept separate from the queen.

With this approach to requeening, there is always some risk that any old, and still well-functioning queen, or the colony worker bees themselves, will remove any new rival queen raised. When successful, however, there will be little or no interruption to the brood cycle and the techniques circumvent the requirement to locate and find the old queen.

Figure 6 Supersedure cells induced by isolating young brood above excluder. Canberra and Region Beekeepers workshop, Jerrabomberra Wetlands Apiary 30 September 2019

There are two well-reported instances of natural queen replacement by a queen from another colony. These arise when late swarms invade colonies of European races of the Western Honey Bee where the resident queen is usurpedvii and the more insidious parasitism of honey bee colonies displayed by some African honey bee races. Neither are likely to be ever encountered but demonstrate that queen supersedure can be used as a tool for requeening.

The African scenario is best illustrated by the Cape Honey Bee (Apis mellifera capensis). Its workers routinely drift to African Honey Bee (Apis mellifera scutellata) colonies and take over the role of the queen – as laying workers – steadily replacing the host African colony beesviii. In this extraordinary circumstance, where unfertilised workers produce workers parthenogenically, the colony is gradually repopulated by Cape beesix. However these laying Cape bee workers, even in aggregate, lay less efficiently than a fertile laying queen, so the colony dwindles and dies.

The slow return to the fully queenright condition

Where a new laying queen is successfully introduced, a delicate process, she will normally commence laying within days. The conventional wisdom is to delay inspection for uptake for at least a week to avoid the risk of her being rejected. Consequently some queen breeders advise leaving that check until around ten days.

Nevertheless a fairly prompt inspection is necessary since a colony sometimes rejects an introduced queen and you may not be able to determine that it has not raised one itself. In any case, the previous queen’s worker and some of her drone progeny will be entirely replaced within about six-to-eight weeks.

With self-requeening, that initiated by the bees, the situation could not be more different. It takes sixteen days to raise a queen from a newly laid egg and up to another week to ten days for that queen to mate and commence laying. In inclement weather the mating process may be delayed for up to a further week. With any further delay, however, requeening may fail entirely and a queen – if she survives – will become a drone layer. With all natural queen replacement, except supersedure, there will always be a period where the colony is functionally queenless.

Emergency queen colony recovery With emergency queen replacement it takes a minimum of ten days for the colony to produce a queen and another week or so for her to be mated and to commence laying (see again Figure 3 in Part I and Table 1 above). So with any raised emergency queen, such as that induced by splitting a strong hivex, only sealed brood will be present by the time the new queen is laying. Because emergency queens are often raised from older larvae, they often perform sub optimally and are often soon superseded.

Supersedure queen colony recovery The transition is much smoother where queen replacement is direct. While the old queen may disappear before a new daughter queen emerges and mates, there will be little or no interruption to the brood cycle. Often enough the old queen will survive until well after the new queen is established and both may continue to lay together for some time.

One of the enduring features of supersedure is that it usually goes unnoticed by the beekeeper. In the absence of regular hive surveillance, the only evidence that such queen replacement may have occurred will be renewed colony vigour and the finding of an additional queen or only a few cells from which queens have emerged. We mark all our queens and, having discounted instances where bees have swarmed, we identify supersedure by the presence of an unmarked queen.

Interestingly if a newly raised supersedure queen fails to return successfully to the hive after mating, a new batch of cells will be started. Thus with supersedure the risk of the colony ever becoming queenless is minimised.

Swarming queen colony recovery With swarming, however, a colony may be effectively queenless for many weeks, so it is quite possible for a surviving virgin queen to be present when all brood has long emerged.

At swarming preparation outset we have noted that the queen will have stopped laying. By this stage a fairly large number of queen cells, in varying stages of development, will have been started. Since the colony queen departs with the prime swarm, the swarm leaves virgin and emerging queens behind. The colony may then settle with a soon to be mated virgin queen or cast further swarms every three-to-ten days (Figure 7). Even where a colony swarms just once, there will be extended delay in the colony returning to its single queen condition.

Figure 7 Secondary swarm trajectory at Peter and Jenny Robinson’s garden apiary in Lyons on 20 October 2020:
(a) initial swarming flight preparation at 13.57.28 with swarm departing hive around 14.11.00;
(b) colony arriving at sedge thicket 20 m away around 14.20.00 and settling over several minutes from 14.20.35; and
(c) six-frame nucleus hive put in place around 14.25.00 with most swarmed bees having entered the hive by 14.30.20.

We found and marked two virgin queens (using different marking colours) in a single colony preparing to swarm at the Jerrabomberra Wetlands Apiary on 20 September 2020. As it turned out the colony, after supering, did not swarm but there were obviously more queens present as we subsequently found that the colony was headed by an unmarked queen.

On a related note we have been tracking queen succession in two-queen colonies.  Two-queen colonies revert to single-queen colonies of their own accord near the end of the season and likewise do so under drought conditions. Such reversions were common over the 2018-2019 and 2019-2020 summer drought seasons.

We have been investigating instances in two-queen colonies where the queens were clearly mismatched and where supersedure or swarm cells appeared in one half of the broodnest. To date, it appears that two-queen colony self requeening to retain the two-queen condition never occurs. Any new queen raised either does not survive or, if one does, it supersedes both the founding queens.

Despite the reputation of two-queen colonies rarely swarming, one of our two-queen colonies cast a very large swarm during recent 2021 spring build up period where conditions were particularly good and swarming was unusually prevalent. It reverted to this same single-queen condition despite the presence of numerous queen cells. Surprisingly however – while the swarm departed with one of the marked queens, the remaining parent colony queen – the one with the swarm cells was superseded. Farrar made the same findingxi noting that conditions conducive to initiation of supersedure or swarm cells were referred to the other brood nest:

Swarming is less of a problem in two-queen colonies than in strong single-queen colonies, but queen cells started because of a failing queen or crowding in either brood nest will stimulate production of queen cells in the other.

In another instance we attempted to unite two single queen colonies, both marked, to form a two-queen colony – a standard technique for their establishment. A perverse element was the fact that one of the single colonies already had supersedure cells, that is the queen was already in the process of being replaced. Instead of sensibly delaying the set up of the two-queen colony (until well after the queen had been replaced) we went ahead and united the two colonies anyway. Our clear expectation was that one queen would supersede the other and we would end up with a large single-queen colony. To our great surprise, we generated a two-queen colony containing both a marked queen and an unmarked supersedure queen. Here it seems likely that the failing queen was replaced.

These findings point to the more practical problem of attempting to requeen hives that are preparing to replace their queens naturally. It is essential to change the intent of the bees to either swarm or supersede their queen for requeening to be successful. Keep your requeening effort, and any attempt to sort out a colony attempting to self requeen itself, strictly apart.

Stock selection

Stock selection is an eternal quest and one that begs the question of desirable frequency for requeening and what to requeen with. While there are some advocates of employing wild stock and some even hold that one should never requeen – practises we do not support – we have found that using commercial stock (or raising a few queens ourselves and evaluating their performance) is the key to hassle free and productive beekeeping. Though never the silver bullet, use of commercial stock reasonably guarantees colony productivity, gentle temperament and a measure of disease, particularly chalkbrood, control. Home raised queens will always mate with drones of uncertain heritage and cannot be relied upon for the more valuable traits that queen breeders in the main have good control over.

Nevertheless we have found, where swarming is controlledxii, that within eighteen months 70% of our requeened colonies are headed by an unmarked supersedure queen. We conclude that colonies headed by young queens will regularly replace their queens irrespective of management practice. A key difference is that, while both supersedure and swarming result in a colony ending up with a younger and more fecund queen, with supersedure the colony workforce is retained.

Queen breeders expect the queens they raise to start laying at around the 10-12 day mark that is from the time of cell introduction. Noting that it takes one or, at the most, two days for ripe queen cells to emerge, this correlates with the roughly seven days it takes for a queen to be mated and the additional 3-4 days it takes for her to start laying.

According to Harry Laidlaw Jrxiii, a renowned authority on queen raising, catching queens is typically timed at around fourteen days. Our preferred queen breeder catches and cages his queens at around day seventeen the exact date pending the need to accommodate a new batch of cells and weather conditions.

This brings us to the question of the calibre of the queens available on the market. The vast majority of queens sold to beekeepers are untested, that is they are captured and caged as soon as they are laying. For the most part these queens will perform well and are relatively inexpensive. That said, all reputable queen breeders aim to produce queens that are productive and whose progeny are both gentle and disease resistant. As well they will ensure that raised queens (and supporting drone mother colonies) receive optimal nutrition.

You can, however, purchase tested queens, more expensive select tested queens or even artificially inseminated queens, those that are held for longer and that are evaluated for their performance and genetic traits. However their high cost is not warranted unless you have a large apiary or are planning to raise large numbers of queens for sale.

The traditional wisdom is that well-raised queens may live for several years but that they tend to be twice as swarm prone by the time they reach twelve months of age. Further their productivity – depending on how hard they have been worked – will suddenly decline once they reach between twelve and eighteen months of age.

Queensland bee researcher John Rhodesxiv demonstrated that queen acceptance can be improved if queens are held in their mating nucs for some weeks (Table 2). Rhodes also found that Australian queens tended to store less sperm than is recommended in Europe and North America noting that:

The 1999 experiment showed a positive correlation between increased queen survival and increased sperm counts with data suggesting that sperm counts of less than two million sperm per queen was associated with low survival of queens following introduction into honey production colonies…

Day kept in mating nucsSurvival rate (%) at 2 weeks#Survival rate (%) at 15 weeks#
715-28 [22 ± 9]10-25 [18 ± 11]
14-1748-70 [61 ± 12]15-70 [51 ± 31]
21-2480-93 [86 ± 7]63-88 [74 ± 13]
2885-95 [90 ± 7]60-87 [74 ± 19]
31-3588-95 [91 ± 4]73-88 [80 ± 8]

#Range [average and standard deviation]. Values are the average survival rates of several cohorts of 20 queens.
Table 2 Survival rate of mailing cage queens introduced to dequeened colonies illustrating dependence on time queens are held in mating nucs

The practical reality is that not all mailing cage queens will be accepted. It is also a mistake to believe that any well-raised queen will perform well beyond 12-18 months of age making it a good idea to requeen at least every second year.

Overall we find that at least nine out of ten queens, those we have bought from reputable queen breeders, are accepted and that they are laying well at first inspection. We attribute this seemingly high acceptance rate to the fact that we normally introduce caged queens to nucleus colonies or splits rather than to dequeened full strength colonies. Indeed the introduction of queens to nuclei rather than to full strength colonies – though we have no clear evidence for it – may enable queens to mature further in lieu of their being held longer in queen breeder mating nuclei.

From here we use these nucs, once well-established, in a flexible requeening program. This practice obviates the need to synchronise full strength hive requeening with arrival of caged queens in the post. Our two-step process gives us full control of requeening at an apiary level rather than at the level of the individual hive and provides insurance against the occasional queen introduction loss.


iRichter, M.C. (1922). Breeding and requeening: Are further importations necessary to improve our stock? Shall we requeen each year? Gleanings in Bee Culture 50(5):296-298.

iiWade, A. (October 2017). Bee Buzz Box October 2017. Establishing two queens instead of one queen in a honey bee colony. Part I Principles of introducing and running two-queen colonies.

iiiHogg, J.A. (1983a). Methods for double queening the consolidated broodnest hive: fundamentals of queen introduction. Part 1 The fundamentals of queen introduction. American Bee Journal 123:383-388.
Hogg, J.A. (1983b). Methods for double queening the consolidated broodnest hive: the fundamentals of queen introduction. Part 2 Conclusion. American BeeJournal 123:450-454.

ivOldroyd, B.P. and Wongsiri, S. (2006). Asian honey bees: Biology, conservation and human interactions. Harvard University Press. Australian National Library Dewey Number N 595.799 044

vVictor Croker and David Leehumis at Australian Honey Bee. (Pers. comm).

viForster, I.W. (1972). Requeening honey bee colonies without dequeening. New Zealand Journal of Agricultural Research 15(2):413-419. DOI: 10.1080/00288233.1972.10421270

viiMagnum, W. (2010). The usurpation (takeover) of established colonies by summer swarms in Virginia. American Bee Journal 150(12): 1139-1144. and Magnum, W. (2012). Colony takeover (ursurpation) by summer swarms: they choose poorly. American Bee Journal 153(1): 73-75.

viiiSchneider, S. Deeby, T., Gilley, D. and DeGrandi-Hoffman, G. (2004). Seasonal nest usurpation of European colonies by African swarms in Arizona, USA. Insectes Sociaux, 2004 51(4):359-364.
Danka, R.G., Hellmich, R.L. and Rinderer, T.E. (1992). Nest usurpation, supersedure and colony failure contribute to Africanization of commercially managed European honey bees in Venezuela. Journal of Apicultural Research 31(3-4):119-123.

ixHepburn, H.R. (2001). The enigmatic Cape honey bee, Apis mellifera capensis. Bee World 82(4):181-191. DOI: 10.1080/0005772X.2001.11099525

xWhere a colony with swarm cells is split, a swarm cell queen rather than an emergency raised queen will be the source of the new queen.

xiFarrar, C.L. (1953), p.l89. Two-queen colony management. American Bee Journal 93(3):108-110, 117 reprinted as Farrar, C.L. (1953). Two-queen colony management. Bee World 34(10):189-194.

xiiSwarming is largely preventable if colonies are regularly requeened, adequate space is provided for the queen to lay and for bees to cure nectar and store honey, and if colonies are divided (spit or Demareed) as spring bee populations explode.

xiiiLaidlaw, H.H (1979). Contemporary Queen Rearing, 1st Edn. p.95. Dadant and Sons, Hamilton Illinois.

xivRhodes, J. and Somerville, D. (May 2003). Introduction and early performance of queen bees: Some factors affecting success. Rural Industries Research and Development Corporation. RIRDC Publication No 03/049, 44pp.
Rhodes, J.W. (October 2008). Semen production in drone honeybees. Rural Industries Research and Development Corporation. RIRDC Pub. No. 08/130. 95pp.