Technology & Medicine › Ancient world › Greek and Roman engineering
Heron's aeolipile (steam toy)
c. AD 60 (Alexandria)· Invention or discovery· Tools, energy & materials
Mostly agreedGood evidence; some details, dates or numbers uncertain.
A spinning steam-powered sphere described by Heron of Alexandria around AD 60, the first recorded steam engine, built as a curiosity.
See it on the Technology & Medicine map →How it works
A hollow bronze ball sat on pivots above a boiler. Steam rose into it and escaped through two bent nozzles on opposite sides, pushing the ball round — the same push-back principle as a rocket. Heron described it in his Pneumatica alongside other gadgets: a coin-operated holy-water dispenser, temple doors that opened when an altar fire was lit, and a wind-driven organ. None did useful economic work.
Why it mattered
It proved steam could make motion 1,600 years before practical steam engines, a classic case of an invention arriving before society had a use for it.
How we know
Heron's own Greek texts, Pneumatica and Mechanica, preserved through Byzantine and Arabic copies. Modern replicas work but spin fast with little torque (turning force).
Still unknown or debated
Why no one developed it further is debated; Heron's own dates rest on a lunar eclipse he mentions, identified as AD 62.
Connections (3)
▸Greek and Roman engineering included Heron's aeolipile (steam toy)Well documented
The connection
Heron worked in Alexandria's tradition of pneumatics — devices driven by air, water and steam — following Ctesibius and Philo. The aeolipile was one entry in his Pneumatica, a catalog of clever mechanisms for temples and entertainment, showing both the skill and the limits of ancient engineering: brilliant ideas rarely put to economic use.
When and how
Alexandria, Roman Egypt, around AD 60, in Heron's Pneumatica.
How we know
The Pneumatica survives and describes the device step by step; Vitruvius (about 25 BC) mentions a simpler steam-blowing aeolipile.
Debates and open questions
Whether any aeolipile was ever built for practical work, such as turning a spit or opening doors, is unrecorded.
▸Heron's aeolipile (steam toy) demonstrated steam power about 1,650 years before Newcomen's atmospheric engineMostly agreed
The connection
Heron's aeolipile was a sphere spun by jets of steam, a toy rather than a working engine. Newcomen's engine used steam differently, condensing it to create a vacuum so air pressure could push a piston. The direct link is weak, but Heron's writings, printed in Latin from 1575, encouraged early modern inventors to experiment with steam.
When and how
Heron described the device in Alexandria around AD 60; Newcomen's engine ran near Dudley, England, in 1712, after steam experiments by Denis Papin and Thomas Savery.
How sure are we?
Heron's Pneumatica survives and describes the device, but no source shows Newcomen read Heron, so the link is one of precedent, not influence.
Debates and open questions
Why the ancient world never developed steam power is debated: cheap slave labor, costly fuel, weak metalworking or simply no practical need are all proposed.
▸Heron's aeolipile (steam toy) shared its reaction principle with the Jet engineMostly agreed
The connection
The aeolipile spun because steam rushing out of its nozzles pushed the sphere the opposite way: action and reaction, later stated as Newton's third law (1687). A jet engine works the same way, compressing air, burning fuel in it and blasting hot gas out the back to push the aircraft forward. The link is shared physics, not direct descent.
When and how
Alexandria about AD 60; England (Frank Whittle) and Germany (Hans von Ohain) in 1937; first jet flight, Heinkel He 178, 1939.
How sure are we?
Engineering histories routinely cite the aeolipile as the first reaction turbine; the physics matches, though Whittle built on gas-turbine work, not Heron.
Debates and open questions
Some argue the aeolipile is closer to a rocket or a reaction steam turbine than a jet; calling it an 'ancestor' is mainly a teaching device.