Drone Bees: Biology, Mating, and Expulsion
Let's start with the death, because it's the part that makes people put down their coffee.
A male honey bee - the drone - mates with a queen in midair, at speeds up to 22 miles per hour, between 33 and 131 feet above the ground, and the entire act takes approximately five seconds. During those five seconds, his endophallus everts - turns completely inside out - requiring most of the hemolymph (insect blood) in his body to force it outward. The ejaculation is so explosive that it's sometimes audible to the human ear as a small popping sound. Then his abdomen rips open. He's instantly paralyzed. He flips backward. And he falls to the earth, dead before he hits the ground.
This is the reproductive strategy that evolution designed. This is the plan.
A piece of his endophallus remains lodged in the queen as a "mating sign." The next drone to mate with her must remove it before inserting his own, which will also tear him apart. The queen will mate with 12 to 20 drones during one to several flights. Each one dies the same way. She stores roughly 87 million sperm in her oviducts, filters it down to 5-6 million in a specialized organ called the spermatheca, ejects the rest through her sting chamber, and never mates again for the rest of her life.
The drones got five seconds. She got enough genetic material to last five years.
The Most Expensive Eyes in the Colony
A drone is built for exactly one biological function, and his body makes this comically obvious.
His compound eyes contain up to 8,000 facets each - twice the size of a worker's or queen's. They're so large they meet at the top of his head, giving him a near-spherical field of vision optimized for spotting a queen against open sky. His body stretches 15-17 millimeters and weighs 200-300 milligrams - one and a half times the size of a worker. His flight muscles are proportionally massive, built to power aerial pursuits at 35 kilometers per hour.
And that's about all he's got. He has no stinger. He has no pollen baskets. He has no wax glands. For the first four to five days of his life, he cannot even feed himself - workers have to do it for him. A single drone consumes roughly three times as much food as a worker bee, which makes him an extraordinarily expensive mouth to maintain in a colony that runs on ruthless metabolic efficiency.
He takes 24 days to develop from egg to adult - three days longer than a worker, eight days longer than a queen. Then he spends another 10-12 days maturing before he can fly. During that time, he does almost nothing that resembles work. He doesn't clean cells. He doesn't nurse larvae. He doesn't guard the entrance. He sits on the comb, gets fed by workers, and waits for the weather to cooperate with his sex drive.
The traditional view of drones, held for roughly forever, was that they're lazy freeloaders who exist solely to mate. Recent research has complicated this picture somewhat - but only somewhat.
Where They All Go at 2 PM
Here's the part that genuinely puzzles scientists.
Every warm afternoon during mating season, drones from dozens of colonies converge on the same aerial spots - drone congregation areas, or DCAs - that occupy specific locations 10 to 40 meters above the ground. These spots are used year after year. The same areas. The same altitudes. The same invisible boundaries.
And no drone has ever been there before. Drones don't survive winter. Every drone using a DCA hatched that spring. Nobody taught him where to go. Nobody showed him the route. Yet he flies to the exact same spot his predecessors used, guided by mechanisms that researchers have spent decades trying to identify without reaching a satisfying conclusion.
The leading theories involve landscape features - drones appear to follow treelines and flyways - combined with possible magnetoreception (sensitivity to Earth's magnetic field) and solar orientation. South-facing locations seem preferred in the Northern Hemisphere. Wind-protected areas where flight is unimpeded get more drone traffic. But none of these explanations fully accounts for the precision with which drones find DCAs that exist in specific, bounded aerial volumes with no physical markers.
Scientists locate DCAs by flying helium balloons carrying either a caged virgin queen or synthetic queen pheromone (9-ODA) up to the right altitude and watching for the swarm. The synthetic pheromone attracts drones from as far as 800 meters downwind. A real queen draws them from about 420 meters. When a virgin queen enters a DCA, drones form what researchers call a "drone comet" - a streaming formation of approximately 31 males (the median count for Apis mellifera) that pursues her through the air, continually consolidating and breaking apart as drones join, exhaust themselves, and peel off.
The drone's mating flight averages 20 to 25 minutes before he has to return to the hive to refuel on honey. He makes multiple flights per day, and his entire adult life - roughly 60 days maximum - is organized around the afternoon window when conditions align for these flights.
Of the thousands of drones that patrol a DCA on any given afternoon, almost none of them will mate. Ever.
He Has No Father
The genetics of drone bees are, to use a technical term, deeply weird.
Drones are haploid - they carry only 16 chromosomes, a single set inherited entirely from their mother. Workers and queens are diploid, carrying 32 chromosomes, one set from each parent. Drones develop from unfertilized eggs through a process called parthenogenesis. The queen simply lays an egg without releasing sperm from her spermatheca. No fertilization. No paternal contribution. The drone has a mother and a grandfather, but no father.
This creates a genetic consequence that sounds like a riddle: because a drone has only one set of chromosomes, when he produces sperm, every sperm cell is genetically identical. There's no recombination. A drone passes 100% of his DNA to every single offspring. A worker bee, by contrast, shares only 50% with her own offspring - the same as any diploid organism.
This is the foundation of haplodiploidy, the genetic system that makes social insect colonies possible. Workers share 75% of their genes with full sisters (same father), making it genetically "cheaper" to raise sisters than to raise their own offspring. The math of kin selection explains why workers sacrifice their own reproduction to serve the colony - and why drones are the genetic linchpin that makes the whole system work, despite doing essentially no visible labor.
And then there's the diploid drone problem. Sex in honey bees is determined by a gene called the complementary sex determiner (csd). If a fertilized egg inherits two different csd alleles - the normal outcome when a queen mates with genetically distant drones - the result is a female. If the egg inherits two identical csd alleles, which happens when the queen and drone share genetic heritage, the result is a diploid drone: a fertilized egg that should have been a worker but developed male instead.
Worker bees detect diploid drone larvae and eat them alive. This is not a metaphor. Researcher J. Woyke observed in controlled experiments that "all the disappearing larvae were eaten alive by the workers." The colony recognizes these genetic dead ends by their abnormal cuticular chemical signatures and cannibalizes them before they consume significant resources.
Honey bees maintain roughly 87 csd alleles worldwide, according to research led by Hasselmann - far more than the 18 originally estimated. This diversity exists specifically to minimize the odds of diploid drones appearing. Inbreeding collapses that diversity. When it does, colonies weaken and die.
What Drones Actually Do (Besides the Obvious)
Research published in PMC documented something that challenges the freeloading narrative: drones contribute to colonial thermoregulation.
About 30% of drones aged eight days and older actively heat their thorax by more than 1 degree Celsius above their abdomen - endothermy, the insect equivalent of shivering to generate warmth. Younger drones do this less frequently, but they compensate through sheer numbers: their abundance on the brood nest is 3.5 times higher than that of the oldest drones, meaning their passive body heat contributes meaningfully to maintaining the 95 degrees Fahrenheit that developing brood requires.
Drones also generate substantial heat while warming up their flight muscles before departure - and that heat radiates into the brood nest. A colony maintaining 500-1,000 drones during peak season gains a non-trivial heat source during cool mornings and evenings.
A 2023 study published in Animal Behaviour found something even more interesting: "honey bee drones are synchronously hyperactive inside the nest" - suggesting coordinated in-hive behavior patterns that go beyond simple loafing. Workers stimulate drones to perform trophallaxis (food sharing) through vibration signals, indicating a more complex social role than "eat, sit, fly, maybe mate, die."
The resource cost is still significant. A colony with 1,000 drones consuming at three times the worker rate is effectively feeding 3,000 extra bees that produce no honey, collect no pollen, and build no comb. The colony's total drone investment runs approximately 200-400 grams of honey per season. That might seem modest until you remember that the colony is simultaneously losing 40-50% of its members to winter mortality and needs every gram of stored honey to survive.
The Mite Magnet
Varroa mites prefer drone brood at 8 to 12 times the rate they infest worker brood. Given a choice, a reproducing mite will overwhelmingly select a drone cell.
The reasons are mechanical and temporal. Drone brood remains capped for 15 days, compared to 11 days for workers - four extra days for the mite to reproduce. Mites produce 2.2-2.6 offspring per reproductive cycle in drone cells versus 1.3-1.4 in worker cells. The invasion window - the period during which a cell is attractive to a mite looking to enter - stretches 40-50 hours for drone brood versus 15-30 hours for workers. Larger cells physically accommodate more mites.
This preference has turned drone comb into a management tool. Drone brood removal - allowing bees to build drone comb, waiting for it to be capped, then removing and destroying it before mites complete their reproductive cycle - functions as a non-chemical varroa control strategy. The colony sacrifices some drones. The mite population drops. The math, while grim for the drones involved, works in the colony's favor.
The October Eviction
Every autumn, as nectar dwindles and the colony shifts into winter survival mode, workers physically drag drones out of the hive.
The drones resist. They're bigger than workers, so it often takes more than one worker to haul a drone to the entrance and shove him off the landing board. The evicted drones cluster near the entrance, trying to get back in. Workers block them. The drones can't feed themselves. They die of exposure and starvation within days.
This happens every year, in every colony, with the mechanical indifference of a system that runs on resource mathematics rather than sentiment. Drones consume food. Drones don't contribute to winter survival. The colony needs every gram of stored honey to maintain the cluster through January. The calculation is obvious, and evolution doesn't hesitate.
Drones don't reappear until late spring. When they do, the cycle starts again: 24 days of development, 10 days of maturation, a few weeks of afternoon flights to congregation areas that nobody alive has ever visited but everyone somehow finds, and - for the vanishingly small percentage who actually mate - five explosive seconds followed by immediate death.
The drones who don't mate live another month or two. Then October arrives. The workers come for them again.
It's tempting to feel sorry for drones. Their life looks like a raw deal from every angle - parasitized more than anyone else, evicted every winter, killed by the act of mating, eaten alive if their genetics are wrong. But the colony runs on their gamble. Every queen needs those 12-20 drones to mate with. Every generation of workers carries drone DNA. The 87 csd alleles that keep the species viable exist because drones carried them, flew to invisible spots in the sky that nobody taught them to find, and bet their bodies on five seconds of airborne contact.
The colony doesn't mourn them in October. But it raises a new generation in April, every single year, because the math only works if the drones show up.