In the geocentric system, which prevailed from the Babylonians to Copernicus, we distinguish two cycles of rotation:
- The diurnal cycle: all the stars revolve around the Earth in 24 hours. They rise in the East and set in the West. This movement is obvious, but it does not allow us to distinguish the planets from other stars.
- The cycle specific to each planet: what differentiates the planets is that after 24 hours, we do not find them in the same place on the fixed sphere. They have moved, they are wandering stars. We then measure the time it takes for each planet to go around the celestial vault and return to the starting point. For example, according to this view, the Sun moves about one degree per day, and it takes a year to go around. For the ancients, it took one year to go around the Earth, even if we say today that it is the Earth that goes around the Sun in one year. The Moon, for its part, takes 29 days.
By classifying the stars according to their period of revolution on the fixed sphere, we have the following order:
| Star | Period of revolution |
| Saturn | 29 years |
| Jupiter | 12 years |
| March | 687 days |
| Sun | 365 days |
| Venus | 225 days |
| Mercury | 88 days |
| Moon | 29 days (around the Earth) |
This table gives the modern vision (period of revolution around the Sun). Indeed, the period of revolution of a planet against the background of the sky seen from the Earth is not constant, since the Earth is moving. It was to explain these irregularities that the Ancients introduced epicycles into their geocentric system.
The Ancients had made the (correct) hypothesis that the slower a star rotates around the Earth, the further away it is. The irregularities of the geocentric vision do not call into question the general order.