13 min read

Bigger queens, better queens - part 0

Why are bigger queens, better queens? It may be more to do with queen mating success and colony fitness than simply the number of eggs she lays.
Cell bar frame with developing queen cells
Cell bar frame with developing queen cells

This is the prequel to Bigger queens, better queens - part 1 which was all about the maternal effect in honey bees.

The maternal effect, although well-known in other species, has only recently been demonstrated in honey bees. Essentially, it involves the queen preferentially laying larger eggs in queen cells than she lays in worker cells. These larger eggs develop into larger and heavier queens.

Perhaps unsurprisingly - though also reassuringly - these larger queens have more ovarioles and lay more eggs than smaller queens. Finally, these larger queens appear to pass on desirable traits to their progeny; second generation queens are also larger, as are the workers.

Part 1 in this series of 3 (or perhaps now 4) posts described the demonstration of the maternal effect. In this post I'm going to explore - inevitably briefly due to the availability of some of the historical literature - why larger queens might be better.

It's important to emphasise that none of the sources used for this post were aware of the maternal effect in honey bees. Although there may have been hints such a thing occurred in studies dating back to the 1970's, the evidence was circumstantial or - at best - correlative, and it wasn't demonstrated until 2019.

Factors affecting the value of a queen

This heading is lifted verbatim from Laidlaw and Eckert's Rearing queens (1962). Despite now being over 60 years old, the majority of this book - at least in terms of the practical beekeeping - remains relevant. Unsurprisingly, there's no mention of Varroa which wasn't identified in the USA until 1987 {{1}}.

The value of a queen depends, largely, upon her ability to lay a sufficient number of eggs to maintain a strong colony appropriate to the various seasons of the year (Laidlaw & Eckert, 1962).

The egg-laying ability of the queen is the primary measure of her value. It should be obvious why this is; a queen that lays too few eggs, or too many unfertilised eggs, or who does not start laying early enough in the season, or who stops laying too soon in the autumn, will result in an undersized colony that is too small to thrive.

Slow spring expansion means that the colony is unlikely to be big enough to swarm (reproduce).

As a beekeeper you might think this is "a good thing" ... it isn't.

An undersized colony will collect less nectar, will be less able to defend itself against robbing, and may not collect sufficient stores to survive the winter. Winter survival is also compromised by too few winter bees ... numerous studies have shown that large colonies (in late autumn, not necessarily in midsummer) have better overwinter survival.

Quality and quantity

Laidlaw and Eckert qualified their definition of the value of a queen by also commenting on how she is responsible for the characteristics of the colony.

A queen that laid 'sufficient number of eggs to maintain a strong colony' is of little value if the colony is aggressive, susceptible to disease, or that swarms as soon as your back is turned {{2}}.

However, these qualities are largely inherited and/or determined by environmental conditions.

If you allow an aggressive colony to requeen itself, do not be surprised if the colony remains aggressive once the worker population is replaced with the new queen's brood. In this case, the aggressive trait is inherited.

Similarly, if it's raining, or nearly midnight (or, for spectacular effect, both), and you lift the roof and crownboard, even a normally placid colony will react defensively. This is the influence of the environment. Exactly the same things applies in a nectar dearth, or if the colony is queenless.

Remove the roof at your peril

Although it might be preferable to be faced with a smaller number of psychotic bees at midnight on a rainy night, the reality is that the qualitative traits - in terms of colony survival and productivity - are generally secondary to the production of sufficient eggs to maintain a strong colony.

Reproductive potential

The reproductive potential of the queen includes, but is not restricted to, her egg laying. Reproductive potential is defined in terms of:

  • fecundity - the number or rate at which eggs are laid
  • fertility - (related to above) but perhaps better defined as the number or proportion of fertilised eggs produced
  • longevity - the period over which the queen is productive, defined both in terms of years, but (in my view) more importantly, in terms of the months of productivity per season {{3}}

Not all queens are the same size. My native dark bees have relatively small queens, whereas my bait hives sometimes attract swarms headed by enormous yellow queens. This is the natural variation between different strains - and hybrids of those strains - of honey bees.

More relevant to reproductive potential is the demonstration that queens of different sizes can be produced from a single strain of bees.

This has been exploited to study the reproductive potential of honey bees. These scientific studies largely post-date the general acceptance that 'bigger queens are better queens', but provide tangible evidence supporting this view.

All that potential

The fate of a fertilised egg depends entirely upon its environment. This is termed polyphenism - a combination of poly- (more than one) and phenotype (the observable characteristics of an individual) - which means:

The capacity of a genotype to produce two or more distinct phenotypes in response to an environmental factor such as temperature, photoperiod, or nutrition.

In honey bees, it is nutrition that determines the fate of the egg.

If the egg is laid in a horizontal cell - a worker cell - it is fed worker jelly and, after pupation, emerges as a female worker (the blue line in the cartoon below).

In contrast, if the cell orientation is vertical, the nurse bees feed the developing larvae on a rich diet of royal jelly (the red lines). This induces an extensive range of DNA modifications (methylation) which result in changes in the expression of a wide range of genes. After pupation, a queen emerges.

It is (predominantly, but not exclusively) the additional sugar in the royal jelly that leads to queen development

Fortunately for clumsy amateur beekeepers and those of us who rear queens {{4}} (and for the honey bee colony) there is a bit of flexibility during the first 0 to 3 days of larval development that allows older larvae to also be reared into queens, under certain circumstances.

Queen (red) and worker (blue) development - numbers are days after the egg is laid

If a young larva is transferred from a horizontal worker cell to a vertically-oriented queen cell, it will be fed royal jelly and reared as a queen. Similarly, if the queen is lost, the bees will re-engineer a worker cell to be vertically oriented, feed the larva it contains royal jelly, and thereby produce the queen.

It's this 'magic' that allows us to rear queens and that enables the colony to usually recover from the - otherwise catastrophic - loss of a queen due to cackhanded beekeeping or disease.

After 3 days of larval development, this ability to reprogramme a worker into a queen is lost.

Function not form

However, there is a direct relationship between the amount of royal jelly fed to a developing larva, and how queen-like the resulting adult bee is.

The longer a larva is fed royal jelly, and therefore the more she consumes, the more queen-like the adult will be (indicated by the thickness of the red lines).

Developing queen larvae are in a perpetual 'all you can eat' restaurant, whereas developing workers 'enjoy' less of a less rich diet.

In reality, it's a bit more complicated than that, but at a first approximation that'll do.

Seminal studies conducted by Jerzy Woyke (1971) demonstrated that queens raised from worker larvae of increasing age had diminished reproductive potential; defined in terms of their weight, ovariole number, spermatheca diameter, and stored sperm counts.

This is why it is important to use very young larvae when queen rearing. It's also why colonies preferentially choose 3 day old eggs when rearing queens under the emergency response.

Unfortunately, Woyke's studies are hidden behind the Journal of Apiculture Research's paywall and even my 'access all areas' university account can't get to a copy of the paper.

Never mind ... Woyke was studying the phenotype - the physical characteristics - of the queen. As far as practical beekeeping is concerned, it's what the queen does with that extra weight, those extra ovarioles and the increased stored sperm that matters.

And, it turns out, there isn't necessarily a direct relationship between the physical attributes of a queen and her reproductive potential. Some subsequent studies have produced contradictory results; larger or more ovarioles don't always equate to more eggs, though Nelson & Gary (1983) demonstrated a direct relationship between heavy queens, greater brood area and increased honey production.

Colony fitness and queen size

Typically, these sorts of studies are done by rearing queens from young (day 0) or old (day 2) larvae and comparing the performance of the resulting queens, or the colonies that they head.

What other characteristics of a colony could be measured as a means of determining whether larger queens are desirable?

I've previously discussed polyandry and colony fitness, and the interesting concept of hyperpolyandry and larval selection:

Who’s the daddy?
I’ve recently discussed the importance and influence of polyandry for honey bee colonies. Briefly, polyandry – the mating of the queen with multiple (~12-18) drones – is critical for colony fitness e.g. ability to resist disease, forage efficiently or overwinter successfully. Hyperpolyandry, for example resulting from instrumental insemination of the

These posts are worth (re)reading ... my writing might be as poor as ever, but the underlying science is excellent, and the results are fascinating.

Queens go on several mating flights and mate with multiple drones. It is important that they do as it increases the genetic diversity within the colony. This genetic diversity is critical to ensure colony fitness; this is a catch-all term that describes how well the colony survives and reproduces.

A fitter colony will exhibit better disease resistance, will overwinter well and will build up strongly to swarm (reproduce) the following season.

So, how do 'high' and 'low' quality queens - reared from young or old larvae respectively - compare when their mating success is analysed?

It takes two to tango

One study, by Tarpy et al., (2011), aimed to investigate the mating behaviour and success of high and low quality queens, an important functional quality related to colony fitness.

Tarpy measured the physical characteristics of the queens (reared by grafting 0 day or 2 day old larvae, but otherwise treated in an identical manner).

It's worth mentioning here that the 'low quality' queens were so small that many could get through the queen excluder on the hive entrances, so preventing accurate recording of the total number of mating flights.

The low quality queens were significantly lighter (~87% that of high quality queens), had a narrower thorax (~94%) and a reduced spermatheca diameter (~92%) and hence volume. As a consequence, the 'low quality' queens stored significantly less sperm (~65%), though the overall level of sperm viability was the same.

Why was there less sperm? Probably because the queens had mated with fewer drones. Both the observed mating number and the effective paternity frequency - determined by genetic analysis of resulting brood to quantify the distinct patrilines in the colony - were significantly lower in the 'low quality' queens (at ~82% and ~78% respectively).

Do these differences matter?


Although not measured in this study, colony fitness - as defined above - is related to the genetic diversity of the colony. If the queen mates with fewer drones the diversity is inevitably reduced, which may well be detrimental. Furthermore, if the volume of sperm stored is insufficient it may limit the reproductive longevity of the queen.

Consequences for the colony

In an extension of these studies, Tarpy and colleagues (Rangel et al., 2013) measured certain characteristics of colonies headed by 'low quality' or 'high quality' queens.

The queens were reared using standard techniques and then used to requeen a package of bees (US readers will be familiar with these ... it's a box containing bees but no comb or brood). The resulting packages were hived on frames with or without foundation, so that the bees could draw worker or drone comb based upon the needs of the colony. Colonies were monitored from May to September, and their subsequent overwintering survival recorded. A further 40 colonies (50/50 low or high quality queens) were set up the following season to repeat the analysis of overwintering survival.

Colony performance (from Rangel et al., 2013)

Colonies headed by 'high quality' queens (solid lines in the graphs) built ~34% more worker comb (a) and ~144% more drone comb (b) and stored more pollen and nectar (c), all of which were statistically significant. Although there were numerical differences in both worker and drone brood production (e and f, respectively) these were not statistically significant. Finally, colonies headed by 'high quality' queens were estimated to contain ~41% more workers (f) than those headed by 'low quality' queens.

Bigger queens, bigger colonies 😄.

Survival of the fittest?

However, the overwinter colony survival is, frankly, a bit of a mess and is confounded by the way it is presented in the paper. Essentially, in the first year most of the colonies perished, and there was no significant difference between colonies headed by 'high' or 'low' quality queens.

Not good 😔.

In the repeat of this part of the study, 75% of the colonies headed by 'high quality' queens survived overwinter, in contrast to only 30% of the 'low quality' queen colonies.

Confusing 😕. Such diametrically opposing results cannot be meaningfully interpreted.

I think this part of the study needs to be repeated.

One interesting observation was the number of times that queens were superseded.

My expectation would have been that the colonies headed by 'low quality' queens would have been superseded much more frequently. This would fit with previous observations on the supersedure of poorly mated queens, or those failing due to disease. However, although supersedure was observed, there was no significant difference in the numbers of 'high' or 'low' quality queens that were replaced.

Relevance to practical beekeeping

All beekeepers rear queens. Not all know they rear queens, but the fact that queens have a finite lifespan and the beekeeper continue to keep bees means that queens must be being reared {{5}}.

If you rear queens passively, by which I mean simply letting colonies requeen during swarm control, or by splitting colonies, then it's worth remembering that the bees will preferentially choose 3 day old eggs and that the resulting queens need to be well-fed during their early development.

If there is a dearth of nectar or a shortage of pollen, feed the colony. Trust the choices the bees make in terms of the larvae they rear as queens. They know what they are doing.

However, if the colony has little or no eggs or very young larvae then provide some ... they can only work with what they have in the hive and if, for whatever reason, the only larvae are 2-3 (or more) days old they will only be able to rear small, low quality, queens.

Active queen rearing

If you rear queens actively i.e. by preparing a cell raising colony, adding larvae and then moving the resulting queens (or cells) to mating nucs, then it's not worth going to all that palaver unless you can ensure that; a) the larvae are as young as possible, and b) the cell raising colony is packed with young bees and has ample nectar and pollen.

Grafted larvae flanked by open brood, lots of pollen and a frame feeder

I use the Ben Harden queenright queen rearing method. This involves establishing a four-frame 'chimney' above a busy brood box. One of the frames contains the larvae - usually grafted, though they don't have to be - one is full of open brood, to draw up nurse bees from the brood box, and the remaining two are pollen-filled. And next to that is a frame-feeder to which I add syrup on a daily basis until the cells are sealed.

All sorts of other tricks have been used to try and guarantee good larval nutrition. Laidlaw and Eckert (1962) and David Woodward (in his book Queen Bee: Biology, Rearing and Breeding (2010), Northern Bee Books) both describe 'double grafting'. This method is designed to ensure the larvae are well fed from the moment they are transferred to the cell raising colony, perhaps even before they get the attention of the nurse bees.

Larvae are grafted into cell cups and added to the cell raising colony for two days. During this period they are fed copious amounts of royal jelly. After two days, these initial larvae are removed (but the royal jelly is not) and replaced with another very young larva.

Whilst I understand the logic, this isn't a method I use. Laidlaw and Eckert imply that larval feeding before grafting is probably more important than in the immediate period after grafting before the nurse bees get busy.

Picking winners, part 2
Some larvae are nutritionally deprived and may produce suboptimal queens. Grafting may miss the best larvae the colony would select for rearing as emergency queens.

Recent studies on larval choice by workers suggests that the bees also consider well-fed larvae are the best starting material for new queens ... the colony from which larvae are sourced should also be well fed if needed.


Part 2 in this poorly sorted series will follow next month once a few final experiments have been conducted. I think it's almost certain that I will need a final post (part 3?, I've lost count) to discuss some of the gaps in our understanding of the maternal effect.

Informative? Useful? Entertaining? ... choose any three.
Please support further articles by becoming a sponsor or funding the caffeine that fuels my late night writing ...

Thank you


Nelson, D.L., and Gary, N.E. (1983) Honey Productivity of Honeybee Colonies in Relation to Body Weight, Attractiveness and Fecundity of the Queen. Journal of Apicultural Research 22: 209–213 https://doi.org/10.1080/00218839.1983.11100589.

Rangel, J., Keller, J.J., and Tarpy, D.R. (2013) The effects of honey bee (Apis mellifera L.) queen reproductive potential on colony growth. Insect Soc 60: 65–73 https://doi.org/10.1007/s00040-012-0267-1.

Tarpy, D.R., Keller, J.J., Caren, J.R., and Delaney, D.A. (2011) Experimentally induced variation in the physical reproductive potential and mating success in honey bee queens. Insect Soc 58: 569–574 https://doi.org/10.1007/s00040-011-0180-z.

Woyke, J. (1971) Correlations Between the Age at Which Honeybee Brood was Grafted, Characteristics of the Resultant Queens, and Results of Insemination. Journal of Apicultural Research https://www.tandfonline.com/doi/abs/10.1080/00218839.1971.11099669.

{{1}}: Though it was undoubtedly there before then as it was found in at least 12 states that autumn.

{{2}}: That's not a scientific definition, but one that applies all too frequently 😉 .

{{3}}: A 7 year old queen that only laid eggs in May and June is of little use to anyone.

{{4}}: I belong to both groups.

{{5}}: OK, yes they could be buying them in from Greece every season, but they'd be nuts to do that, wouldn't they?

To read or contribute to the discussion you need to be a subscriber to The Apiarist. Free and paid tiers are available. All subscribers receive the weekly post from The Apiarist as an email newsletter. You can opt out at any time and your email is not used for anything else.