How Fast a Bee Can Fly at Each Stage of the Trip

A honey bee heading out to a feeding station covers ground at about 7.5 metres per second, roughly 17 mph or 27 km/h. Coming home with a full crop she drops to about 6.5 m/s, near 14.5 mph. Both figures come from Adrian Wenner's 1963 paper in the Journal of Apicultural Research, which timed individually marked foragers over courses of 210 and 420 metres with wind speed recorded alongside each flight. They describe ground speed, for Apis mellifera, under stated conditions. Measure a bumblebee, or measure airspeed instead of ground speed, or send the same bee into a headwind, and the number moves. The question has a solid answer at each stage of the trip and no honest single answer for the trip as a whole.

Which bee is being measured, and by what method

Trivia lists publish one top speed for "a bee" because one number fits the format. Honey bees, bumblebees, solitary bees and Africanized honey bees are different animals flying different missions, and the researchers who measured them used incompatible instruments.

The table below sets five published figures next to the quantity each one actually describes. Read down the third column before comparing the second.

| Species | Published figure | What was measured | Method and source | |---|---|---|---| | Honey bee (Apis mellifera) | 7.5 m/s unladen, 6.5 m/s laden (17 and 14.5 mph) | Ground speed over a fixed course | Timed flights to feeders at 210 m and 420 m, wind recorded; Wenner, 1963 | | Buff-tailed bumblebee (Bombus terrestris) | Sustained flight tested to 4.14 m/s (9 mph) | Airspeed imposed by the tunnel | Closed-circuit wind tunnel, Pitot tube; Biology Letters, 2025 | | Orchid bees (10 euglossine species) | 5.32 ± 0.57 m/s (12 mph) | Average maximum airspeed against turbulent flow | High-speed video at an outdoor air jet in Panama; Combes and Dudley, 2009 | | Yellow-legged hornet (Vespa velutina) | 1.56 ± 0.29 m/s (3.5 mph) | Mean speed on a tethered arm | Flight mill; Sauvard, Imbault and Darrouzet, 2018 | | European hornet (Vespa crabro) | Up to 6 m/s (13 mph) | Free flight speed | Cited by Sauvard's team from Spiewok and Schmolz, 2006 |

The honey bee tops that table partly because Wenner measured ground speed in open air while Combes and Dudley measured how fast a bee could be pushed before it lost roll stability and crashed. The orchid bee number is a ceiling under duress. The bumblebee number is the fastest setting the apparatus was run at, not the animal's limit. None of the five was produced to be compared with the other four.

Airspeed and ground speed are two different quantities

Airspeed is speed through the air. Ground speed is speed over the field below. A bee with a tailwind can post a high ground speed while working no harder than usual, and a bee fighting a headwind can be flat out and appear slow from the hedge.

Wenner classified each flight by wind angle, counting a bee as flying with the wind at 135–180 degrees and against it at 0–45 degrees, and he found that wind altered flight speed far less than it would if the wind were simply added to a constant airspeed. The mechanism turned up four decades later. Andrew Barron and Mandyam Srinivasan, then at the Centre for Visual Science at the Australian National University, reported in the Journal of Experimental Biology in 2006 that bees hold ground speed steady by keeping the rate at which the ground streams across the eye constant, raising airspeed to compensate for headwinds as strong as half their maximum recorded forward velocity. Compensation stayed almost complete even when the visual texture they had to work from was sparse.

The consequence is blunt. A bee that looks unchanged on a windy afternoon is burning more fuel to look unchanged. Field tracking inherits the same problem: the harmonic radar work Riley's group at Rothamsted Research published in Nature in 2005 records tracks over the ground, so those trajectories are ground speeds too.

What a full load actually costs

Wenner's laden bees lost about one metre per second, 7.5 down to 6.5. Under neutral winds specifically, the pair was 7.8 and 7.0 m/s. That is the most quoted load penalty in the literature, and it is not a constant subtractable from any bee on any trip.

The load itself varies enormously. Feuerbacher and colleagues, publishing in the Journal of Experimental Biology in 2003, recorded nectar foragers carrying a mean of 40 percent of body mass and pollen foragers 27 percent. An unloaded worker leaving the nest runs roughly 65 to 75 mg, and a 2019 Apidologie review compiling direct crop measurements found returning foragers holding anywhere from 6 to 60 microlitres, with Calderone and Page reporting an average crop content of 25.5 ± 15 mg. A pair of pollen pellets weighs 8 to 15 mg.

Type matters as much as mass. The same 2003 study found pollen foragers running hovering metabolic rates about 10 percent higher than nectar foragers whatever the load, and calculated that carrying pollen raised mechanical power output by 16 to 18 percent. Nectar rides inside the body; pollen rides on the legs, where it adds drag. A bee carrying 25 mg of nectar and one carrying 15 mg of pollen are not solving the same aerodynamic problem, which is why a fixed mph deduction for "a laden bee" fails on contact.

Wingbeats are not miles per hour

The most repeated bee number is a frequency. Douglas Altshuler, then at the California Institute of Technology, put it plainly in University of California outreach: "the much larger honeybee flaps its wings 230 times every second." His 2005 paper in PNAS recorded that rate alongside an unusually small stroke amplitude near 90 degrees, and a 2017 laser vibrometry study in the Journal of Experimental Biology measured 234 ± 13.9 Hz independently.

Frequency does not convert into speed, and the clearest evidence is what bees do when they need more force. In Altshuler's low-density heliox trials the bees held wingbeat frequency nearly constant and increased stroke amplitude by close to 50 percent. Feuerbacher's loaded foragers likewise did not raise frequency or amplitude at all; they generated more lift per stroke. Bumblebees beat slower than honey bees, in the region of 130 to 160 Hz, on larger wings, and fly at broadly comparable speeds.

Frequency is also unstable in ways speed is not, and the literature disagrees about how. Sotavalta reported in 1963 that wingbeat frequency in Apis mellifera and Bombus pascuorum was independent of temperature. Unwin and Corbet found a negative correlation across bees and bumblebees. Woods, Heinrich and Stevenson, writing in the Journal of Experimental Biology in 2005, found a slight but statistically significant increase between 19 and 37 °C, holding below 31 °C. Parmezan and colleagues, using optical sensors on Brazilian bees and wasps for a 2021 Apidologie paper, found a positive correlation in every species they tracked except A. mellifera. Nobody has resolved this, and a number that unstable cannot be the basis for a speed claim.

Bees against wasps and flies, and why the comparison keeps flipping

The related searches want a ranking. The measurements refuse to produce a stable one, and the sharpest example sits inside a single paper. Daniel Sauvard's team at INRA Orléans, publishing in PLOS ONE in 2018, clocked yellow-legged hornet workers on flight mills at a mean of 1.56 ± 0.29 m/s across a 0.90 to 2.10 m/s range. In the same paper they cite free-flying European hornets at up to 6 m/s, from Spiewok and Schmolz's 2006 work in Physiological and Biochemical Zoology. That is a near fourfold gap between two hornets, and Sauvard's team volunteered the reason: flight mills disrupt optic flow and add mechanical friction, so their own figures likely understate field performance.

Houseflies are worse served. The widely circulated 7.2 km/h, about 2 m/s, traces to compiled animal-speed databases rather than to a primary measurement; the free-flight reference researchers actually cite is Wagner's 1986 work in the Philosophical Transactions of the Royal Society B.

Spiewok and Schmolz added a variable nobody expects. Hornet workers slowed down in low light while drones did not, and worker speed proved independent of temperature while drone speed correlated negatively with it. A single species produced different answers by sex and by illumination.

Can a person outrun a bee?

Over three seconds, a healthy adult can. Over three minutes, nobody can.

An untrained adult sprints somewhere near 12 to 15 mph and holds it for one to three seconds. Usain Bolt reached 27.8 mph between the 60 and 80 metre marks of his 9.58-second world record, averaging about 23.4 mph across the whole run. Wenner's unladen honey bee sits at roughly 17 mph and does not decelerate. Jogging pace, five to eight mph, loses to a bee immediately.

Distance is the part that decides outcomes. The Smithsonian Institution and the University of Arkansas Cooperative Extension training manual MP451 both put Africanized honey bee pursuit at up to a quarter of a mile, around 400 metres, against a couple of hundred feet for European colonies. No peer-reviewed measurement of Africanized airspeed appears in the literature at all; the 12 to 15 mph figure that circulates is popular-press material. What is documented is response threshold and recruitment. Greg Hunt of Purdue University's Department of Entomology reported in the Journal of Insect Physiology in 2007 that in colonies made up of 50 percent of each type, Africanized bees supplied 81 percent of the first few bees to sting a leather flag waved overhead. Extension guidance is consistent about the response: run in a straight line to enclosed shelter, cover the face, and stay out of water, because the bees will wait.

How to read a bee speed claim

Railway rule books handle this awkwardly on purpose. A permissible line speed is never published on its own; it arrives attached to gradient, curvature and braking distance, because the bare number is the thing that causes the accident. Bee flight speeds have the same shape and are almost never published that way. Six questions recover the missing conditions:

1. Name the species. Apis mellifera, Bombus terrestris and a euglossine are not interchangeable, and a honey bee result cannot be transferred to a bumblebee.
2. Establish airspeed or ground speed. Field studies and radar give ground speed. Wind tunnels give airspeed. The two diverge by the wind vector.
3. Check the load. Ask whether the bee was outbound or returning, and whether the payload was nectar or pollen, since the two cost differently.
4. Check wind and temperature. Wenner recorded wind angle for every flight. Air temperature matters because thorax temperature must reach roughly 30 °C before takeoff, and foraging bees hold the thorax near 38.5 °C across a wide ambient range.
5. Identify the instrument. Timed feeder runs, harmonic radar, wind tunnels, flight mills and high-speed video each impose their own ceiling.
6. Ask whether the value is a mean or a maximum. Combes and Dudley's 5.32 m/s is an average of maxima under duress. Wenner's 7.5 m/s is a regression through ordinary commuting.

One trap remains after all six. Foraging range is not speed. Eric Mussen, the University of California Extension apiculturist at Davis, described colonies "flying as many as four miles from the hive in their quest for water, nectars, pollens and propolis, a fifty-square mile potential area of coverage," while typically working within two to two and a half miles. That describes how far, not how fast, and the two get merged constantly in the same paragraph.

Questions people actually ask

How many miles does a bee fly in its lifetime?

The 500-miles-per-bee and 55,000-miles-per-pound-of-honey figures trace to the National Honey Board's trivia sheet; no extension or peer-reviewed source carries them. Neukirch's 1982 paper in the Journal of Comparative Physiology found something narrower and better grounded: each worker has a roughly fixed total flight allowance, which heavy daily flying spends sooner.

Do bees ever get tired of flying?

Flight capacity changes with age rather than fatigue in the human sense. Gilgenreiner and Kurze's 2024 flight-mill study of bumblebees in Proceedings B found distance rising roughly sixfold between days seven and fourteen, then declining by day twenty-one. Neukirch reported that old honey bee foragers lose the ability to rebuild flight-muscle glycogen.

How fast do bees fly when chasing you?

No published study has clocked a pursuing bee. The nearest measured figure is Wenner's 7.5 metres per second, about 17 mph, for an unladen honey bee flying a straight foraging course. Pursuit is not straight, so treat 17 mph as an upper bound on the speed a person is being followed at.

Are bees faster than humans?

Over three seconds, no. An untrained adult sprints at roughly 12 to 15 mph, and Usain Bolt peaked at 27.8 mph between the 60 and 80 metre marks of his 9.58-second record. Over three minutes, yes: a honey bee holds about 17 mph, and a person cannot.

How fast can killer bees fly?

No peer-reviewed measurement of Africanized honey bee airspeed appears in the literature; the 12 to 15 mph figure circulates on popular sites without a primary source. What is documented is behaviour. Greg Hunt of Purdue University reported that Africanized bees delivered 81 percent of the first stings in colonies split evenly between the two types.

What is the fastest bee?

By the best-documented maximum, orchid bees. Stacey Combes and Robert Dudley recorded an average maximum of 5.32 metres per second, about 12 mph, in wild euglossines flying against a turbulent air jet in Panama. Honey bees post higher ground speeds in field studies, which measure a different quantity under easier conditions.

How fast can bumblebees fly?

A 2025 Biology Letters wind-tunnel study held Bombus terrestris workers in sustained free flight at airspeeds up to 4.14 metres per second, about 9 mph, measured with a Pitot tube. That was the fastest setting tested rather than the bee's limit, so treat it as a documented minimum.