IELTS Reading: sixty minutes, forty questions, no extra time

The oldest surviving water clock was found in the tomb of an Egyptian official and dates from around 1400 BCE. It is an alabaster vessel, wider at the top than the bottom, with a small hole near its base and a series of marks scratched on the inside wall. Water was poured in at dusk; as it drained away, the falling level passed the marks one by one, and the night could be divided into equal parts. The device answered a specific institutional need. Temple ritual required offerings at fixed hours, and priests could not read a sundial in the dark.

The principle is simple, but making it accurate is not. Water drains faster from a full vessel than from a nearly empty one, because the pressure at the outlet depends on the depth of water above it. A cylindrical container therefore runs fast at the start of the night and slow at the end. The Egyptian solution was elegantly empirical: they sloped the walls of the vessel so that it was narrower towards the base. As the pressure dropped, the falling level had less width to cover, and the two errors cancelled out. There is no evidence that anyone involved understood why this worked. It was arrived at by trial, adjustment and long observation.

The Greeks turned the water clock into a precision instrument. The critical advance is credited to Ctesibius, an engineer working in Alexandria in the third century BCE, who separated the problem into two parts. Rather than trying to make an irregular flow measurable, he made the flow regular. Water entered a small reservoir which was deliberately overfilled, with the excess running away through an overflow pipe. The level in that reservoir therefore never changed, the pressure at its outlet never changed, and the water leaving it dripped at a genuinely constant rate. This arrangement, the constant-head tank, is still used in laboratory equipment today.

With a steady flow available, Ctesibius could turn the measurement into a display. The escaping water was collected in a second vessel containing a float. As the float rose, it drove a rack and pinion, which turned a pointer against a dial. Later versions, described by Roman authors, added figures that moved, stones that dropped and trumpets that sounded on the hour. These were not decorative extras but a solution to a real problem: a dial must be watched, whereas a bell can be heard from the next room.

Water clocks reached their most sophisticated form far from the Mediterranean. In 1088 CE the Chinese official Su Song completed a clock tower in Kaifeng standing more than ten metres high, driven by a water wheel that was allowed to turn only when each bucket on its rim had filled to a set weight. That mechanism, halting a rotating wheel and releasing it at fixed intervals, performs exactly the function that the escapement would later perform in a mechanical clock, and it predates the European escapement by around two centuries. Su Song's tower also carried a rotating celestial globe and a set of figures that emerged to announce the hours.

Islamic engineers pursued the same line with equal ingenuity. The treatise completed by al-Jazari in 1206 describes water clocks of remarkable complexity, including one in the form of an elephant that combined Greek hydraulic principles with Indian and Chinese decorative traditions. Notably, several of his designs allowed the length of the hour to be adjusted through the year, because the systems of timekeeping then in use divided daylight into twelve parts whatever the season, so an hour in June was substantially longer than an hour in December.

For all this sophistication, water clocks carried a limitation that no amount of engineering could remove. The viscosity of water changes with temperature: it flows more freely when warm. A clock regulated in the afternoon will therefore run slow overnight, and in a cold climate it may stop entirely. This is one reason the technology persisted longest in warm regions and was gradually abandoned in northern Europe, where mechanical clocks driven by falling weights began to appear in monasteries and cathedrals during the fourteenth century.

The mechanical clock did not win because it was more accurate. Early weight-driven clocks were considerably worse than a well-made clepsydra, losing or gaining as much as fifteen minutes a day. They won because they were more convenient: they did not freeze, they did not need refilling, they did not have to be sited near a water supply, and they could be built into a tower where the whole town could hear them. Accuracy came later, with the pendulum. The water clock's three-thousand-year run ended not because the problem it solved had been solved better, but because the problem itself had changed.