Part I: Anatomy & General Mating Habits
Apis mellifera drones take 24 days to emerge from the cell. They spend the next 8-12 days maturing in preparation for mating with a virgin queen. Once mated, the drone dies and provides the queen with a modest amount of sperm that’ll eventually fertilize an egg and produce worker (or queen) bees. If the sperm is “good” (as defined by beekeepers), then the colony will produce “good” bees with desirable traits like high honey production, strong temperature tolerance, large brood numbers, low aggression and a-tendency-to-sting-your-neighbour’s-constantly-barking-cavoodle-named-“Hummus”.
Drones don’t get much attention beyond “they’re used for sex and nothing much else”. Even their minor role as contributors to colony thermal regulation is overlooked in beekeeping texts. Here at TheHoly Habibee Entomologically, I’m no different and will also overlook drones in favour of queen and worker dynamics. So why am I presenting a second article on the males? Short answer: it’s an expansion on the biology of a drone congregation area. Long answer:
There is a shirt that goes with article – designed & pressed by Yours Truly.
Drone Sperm Development & Mating Strategies
Drones are reared in larger brood cells than workers, resulting in bigger bodied males. Their production begins during in the warmer months, and 3-4 weeks prior to the emergence of new queens to maximize the numbers of sexually mature drones available for mating.
Readers interested in the honey bee lifecycle are directed to that article, or the excellent 2019 review by Rangel & Fisher II, which presents drone studies conducted on their non-reproductive functions… it won’t surprise you to know that what impacts drone health and behaviour will impact their spermatozoa production and mating fitness, so it looks like we’re locked in on the dogma of drones as only-being-good-for-sex. You’ll be pleased to know this is not the case for males in solitary bee species, and you can read about that here.
The most suitable drones for queen insemination are between the ages of 10-21 days (Czekonska et al., 2013). Younger males are not sexually mature and have not completed the prerequisites for successful mating (orientation flight, sperm transfer) whereas those drones older than 21 days have a viscous ejaculate with less seminal fluid, which translates to less successful little swimmers. A queen inseminated with older drone semen will have fewer spermatozoa to transfer into her spermatheca, and difficulty expelling the excess ejaculate out of her oviduct, leading to a premature plug and termination of the mating flight.
Sexual Maturation of Drones
Broadly, testes (“testicle”) formation begins at day three when the drone-to-be is in the larval stage (Lago et al., 2020). Various germ cells & spermatogonia are formed during the pupal stage (the building blocks of sperm), with meiosis (the process of sex cell production) being completed at the pre-pupal phase. When the cell is capped and the larva begins pupating, those building blocks undergo spermiogenesis, producing spermatozoa, the wiggly swimmers we all know (and love?).
The adult drone emerges with all the sperm cells he will ever produce in his (short) lifetime (reviewed in (Rangel & Fisher II, 2019)1. A drone is deemed sexually mature once the sperm migrates from the testes to the seminal vesicles. The processes described above are influenced by diet, hive environment, temperature and so on- the same case as any other caste. Sexual maturation has been reported as a wide range from 6-12 days with outliers going as far as 16, and can be broken down like this:
- Emergence to ~6 days: feeding and grooming by worker bees
- ~6-12 days: orientation flights
- Post-orientation flights: waiting for a mate within drone congregation areas
Testicular Development is Described as Four Stages
Anatomical descriptions of drone testes are hundreds of years old (one such source is from 1738 by the Dutch biologist Jan Swamerdam; Biblia Naturae). Broadly, the drone reproductive organs consist of paired testes, conducting tubules and ducts (vas efferens, vas deferens, and ejaculatory ducts), accessory sex glands (seminal vesicles and mucus glands), as well as the endophallus. The testicular development processes that occur in a drone post-emergence can be described as follows (adapted from Klein et al., 2021):
- Stage 1, post-emergence days 0-3: end of spermatogenesis
- Characterised by a halt in testicular growth and end of sperm formation
- Stage 2, post-emergence days 4-9: migration of spermatodesms from tubular lumens
- Testicle size begins declining, and the spermatodesms (“bundles of spermatozoa”) begin migrating to the into vas deferens and seminal vesicles
- Stage 3, post-emergence days 7-13: ongoing tubular and follicular cell atrophy
- Reduction of follicular cells and testicular tubules (used for producing the sperm) as the sperm are completely transferred into the vas deferens and seminal vesicles.
- Stage 4, post-emergence days 9-14: complete atrophy and collapse of testicular parenchyma
- Involution or “end stage” of the testicles, no further changes noted in the testes
Note that the length of Stages 2 & 3 shortens as the season progresses, and drones reared in the peak of spring & summer mature faster than those reared outside of this timeframe. This aligns with the reproductive season – if every other colony is producing new queens, your own will too,
Reproductive Competitiveness is Health-Based, not Territorial or Adversarial
Koeniger & colleagues reported that drones show no aggression to one another during the mating flights or at the congregation areas (Koeniger et al., 2005). They don’t fight or shove to get closer to the queen, instead adjusting their positions relative to their competitor and “defending” their fly zone. Mating flights have been described as “comet shaped” (and I hope that one day you’ll get to see them in person); when a drone leaves the comet, he is “disqualified” and another takes his place in an attempt to mate with the queen.
Her Majesty’s mating flight is suggested to have two phases: the first when she is unmated and travelling at her top speed. The second begins when the first drone latches onto her and mates, briefly slowing her down (Koeniger et al., 2005). This second phase may “motivate” the remaining drones to continue the the pursuit, as the queen is flying slower. From this, I infer:
- That the fittest drone is the one who mates first, suggesting that he holds the “best” genes for flight, pheromone detection and motor control. His daughter workers would likely have stronger flight muscles, and make for stronger foragers
- Alas, as sperm transfer into the spermatheca is inefficient (see below), not all the queen’s offspring will have those traits
- That the mating sign or total ejaculate load does not impact flight speed (for example, if it is 1 microliter per drone and she mates with ~12 drones, then this is the equivalent of a 12 milligrams weight increase pre-excess-load-expulsion). The average virgin queen’s weight varies from 110 mg to over 160 mg (Masry et al., 2015 working in Egypt), taking an average ~135mg (I know I’m being unscientific), we see that this this ~11% increase in bodyweight does not impact flight speed… Impressive muscles!
- For the record, when caged queens are introduced to a colony for requeening (i.e. those unrelated to her daughters) the heavier queens are more likely to be accepted, as weight is a measure of health and ovarioles size, muscle strength, abdomen size and so on. “Strength surpasses relatedness” – Al Kahtani & Bienefeld, 2021.
As such, drone reproductive competitiveness is in physical fitness- fighting over queens eliminates the two combatants from the mating comet and gives an opportunity for their less-hostile accomplices to inseminate the queen.
One Drone Provides a Lifetime’s Supply of Sperm to the Queen…
…So why does she still mate multiple times? The post-mating sperm transfer is not perfect. When the drone ejaculates, his sperm is stored in the lateral oviducts (rather than the spermatheca; described in Kraus et al., 2003). The queen’s “sperm pump” transfers semen from oviducts and into the spermatheca. This pumping system is inefficient; around 5-10% of the sperm reach the spermatheca, and the remainder is ejected through the sting chamber (Kraus et al., 2003 citing Ruttner 1956 & Woyke 1962).
Previously, the “sperm limitation hypothesis” described multiple matings as an adaptation for maintaining large colonies- after all, more sperm, more offspring (Cole, 1983). That hymenopteran queens less mating events was viewed positively, as it reduced the chances of predation and disease transmission (yes, insects have STIs too; Nosema in the case of honey bees). This hypothesis worked for ants, as some queens live for decades in colonies numbering in the tens of millions, but was not seen applicable to honey bees, as Woyke reported a single drone’s ejaculate being enough to give a lifetime’s supply of sperm to Her Majesty (Woyke 1962).
When Kraus and colleagues investigated the mating habits of A.m. carnica queens against those of Cape honey bee A.m. capensis, they found the latter to mate more frequently and suggested it was a bee-havioural adaptation (Kraus et al., 2003). A.m. carnica drones produce ~35% more semen per ejaculation than A.m. capensis, and one would expect that if the sperm limitation hypothesis was valid, the Cape queens would sense a greater semen load after mating with A.m. carnica drones, and reduce the number of mating flights. Their results found that it was not the case, as the A.m capensis queens continued to hold the same (higher) number of mating flights in the presence of Carniolan drones.
The authors argue that if sperm transfer wasn’t a limitation to queen’s mating frequency, then colonies would have selected for larger drones with larger ejaculates. Why this isn’t the case could be due to evolutionary ties of the bees, as the ancestor of A. mellifera is thought to be similar to bumble bees, species which produce smaller, annual colonies (as opposed to perennial in the Apis species; Kraus et al., 2003). As colonies grew larger, so did the demand for sperm to maintain fertilized egg laying. Coupling this with the inefficiency in spermatheca sperm transportation means suggests that the sperm-limitation hypothesis is also applicable to A. mellifera.
Comparing A. mellifera Drones to Other Species
By relative size, A. mellifera drones are three times heavier than A. cerana, 25% heavier than the giant honey bee A. dorsata’s drones, and 50% heavier than another cavity-nesting bee, A. koschevnikovi (see Koeniger & Koeniger, 2000). If you’re thinking that size has an impact on how fast drones, you would be mistaken- A. dorsata & A. cerana drones flew faster than A. mellifera & A. koschevnikovi (Koeniger et al., 2005). The authors proposed predation in their natural habitat as a factor for this flying speed, however they did not elaborate further. Another factor was suggested to be Her Majesty’s own speed- drones only need to go as fast the queen, and any faster would be deemed a waste of energy.
Variations also exist in the male genitalia, with Koeniger & Koeniger describing “the copulative organ is a membranous end phallus, which is differentiated into a broad vestibulum, with its cornua, the slender cervix and the thick bulb with its lobes” (Koeniger & Koeniger, 2004). If you find yourself having to distinguish drones on the basis of phallus size, look to where the hairs are positioned in each drone’s cornua & lobe. The variations are more apparent during mating, as the everted phallus bends one way or another and the hairy regions become more pronounced. Alternatively, you can walk away and never speak of this scenario ever again.
Stingless Bee Drones vs. Apis Drones
A brief mention on the stingless bees: polyandry is the exception rather than the norm within the Meliponini tribe (reviewed in Vollet-Neto et al., 2018). Exceptions were found in Melipona mondury queens from Brazil: the presence of the mating sign (i.e. drone genitalia) in the queen stimulates ovarian activation, as M. quadrifasciata queens mated under laboratory conditions showed ovarian activity only when the plug was removed after a few days of lodgement.
Some stingless males do not form drone congregation areas, instead gather outside colonies with a soon-to-emerge virgin queen and mate with her there. These strategies are suggested to reduce inbreeding and the production of diploid males, who are either sterile or father offspring that will be sterile (“triploid” [3n] offspring; Vollet-Neto et al., 2018). Vollet-Neto & colleagues report that when queens inbreed and produce diploid males, the queens die a few days after the diploids emerge. Her death could be the result of execution by the worker bees (“supersedure” would have been more polite), as the cuticular hydrocarbon profile of diploid males contains six compounds not found in males. Paternal regicide. Rough stuff.
Are Drones from a Laying Worker Just as Good as Drones From a Queen?
Short answer: No. Gencer & Kahya investigated this in A.m. ligustica and found that drones from queenright colonies are superior in size & reproductive capacity (Gencer & Kahya, 2011– see also Zaitoun et al., 2009). They report obvious visual differences in appearances, with queenright drones being 35% heavier than their laying-worker counterparts. Bigger body means better reproductive potential, and this was investigated through sperm analysis.
Laying-worker drones had reduced sperm counts, producing 10% less sperm per milligram of body mass compared to drones of the same size from queenright colonies2. Subsequent semen volume was lower, resulting in a higher ratio of seminal fluids to sperm. Sperm viability was the same, and the authors concluded that “drones from laying-worker colonies therefore produced less, but not poorer quality sperm than the drones from queen-right colonies” (Gencer & Kahya, 2011).
Is it the end of the colony if a queen mates with a drone from a laying worker colony? No, it means that she has a higher chance of being superseded sooner than queens mated with “normal” drones on the account of reduced sperm quantity.
I speculate that worker-laid drones are smaller due to the worker’s diet- she is not ingesting constant royal jelly and is instead subsiding on bee bread and nectar, a relatively poorer source of nutrition. Of course there could be hormonal changes regarding vitellogenin and juvenile hormones and egg development, or it could be a result of malformed worker ovaries (reviewed in Hartfelder et al., 2017 & also Yours Truly’s Factors Influencing Queen Bee Health).
Non-Sexual Roles of Drones Revisited: Worth Studying?
Another short answer: No. Drones lack the body structures for foraging behaviours (no pollen baskets or elongated proboscis for nectar extraction).
Drones do not actively help in regulating nest temperature- they are not fanning, collecting water, or heat shielding. That being said, Kovac and colleagues found that drones can generate heat and warm their surroundings, but since there are far fewer (older) drones reside in the brood area compared to workers, their overall contribution is small (Kovac et al., 2009). By chance, several drones might contract their muscles to warm themselves and incidentally heat nearby cells, however the majority of brood incubation is done by the workers.
Their primary (and perhaps only) function is reproduction, and this is reflected in the literature as most studies on drone biology are done in the context of reproduction.
How Queen Breeders Manage Queen Reproduction
I end Part One with a comment on the two ways that a queen breeder produces mated queens: artificial insemination using semen from drones with favourable traits, or by manipulating drone congregation areas. The former is self-explanatory; the latter works around identifying DCAs and then saturating them with drones from desirable colonies.
A DCA can be mapped by radar or triangulation methods (researchers know this well), or by using drone lures like Koeniger et al., 2005. Once mapped, a dozen or so “mother” colonies are placed within ~500m of the DCA, with drone production being staggered to maintain a constant supply of sexually-mature drones. Staggering can be achieved by stocking colonies with drone comb of various ages, for example colony B1 contains 14-day old drone comb, while colony B2 contains drone eggs.
If the goal is to control for certain traits, the breeder can set his colonies 10-15km away from other beekeepers’ apiaries. When researches perform studies involving colony reproduction, they house hives on small, remote islands where there are no other beekeepers. Seeley’s Honey Bee Democracy contains examples of such work, as does the work of Kraus and colleagues, covered above (Kraus et al., 2003).
Stay tuned for the genetics aspect of drone biology in Part 2 (which turned out to be general sex determination in honey bees).
Read the whole post and found puns, typos mistakes? Voice your displeasure with a strongly-worded message to admin@tgbee.com.au, or by filling this form.
Footnotes
- Curious about the ejaculate volume? The 2019 review of Rangel & Fisher II state it around 0.91-1.7 microliters per drone, with 3.6-21 million sperm cells per ejaculate. ↩︎
- For example, assume drones produce 1 microgram of sperm per milligram of bodyweight. If we compared two 100 gram drones from a queenright & laying worker colony, the queenright drone produces 100 µg of sperm, and the worker-laid drone produces 90 µg. ↩︎