One phenomenon that ancient astronomers had difficulty explaining was the retrograde motion of the planets. Over the course of a single night, a planet will move from East to West across the sky, like any other celestial object near the ecliptic. Most objects in our sky appear to rise somewhere on the Eastern horizon and set somewhere on the Western horizon. The only exceptions are stars near the North celestial pole, that stay above the horizon all the time and appear to make counterclockwise circles around the celestial pole. As one travels further North, the region of the sky that remains above the horizon at all times becomes larger, until the entire sky appears -to an observer at the North Pole- to be simply circling the North star. As one ravels South, the region that remains above the horizon becomes smaller, diminishing to zero size for an observer on the Equator. If one continues South of the Equator, one would observe a progressively larger region surrounding the South celestial pole that remains above the horizon at all times. Stars in that region would appear to circle the South celestial pole in clockwise circles.)
If observed from one night to the next, however, a planet appears to move from West to East against the background stars most of the time. Occasionally, however, the planet’s motion will appear to reverse direction, and the planet will, for a short time, move from East to West against the background constellations. This reversal is known as retrograde motion.
The model of the solar system developed by Ptolemy (87 – 150 A.D.) was a refinement of Aristotle’s (384 – 322 B.C.) universe. This model consisted of a series of concentric spheres, with the Earth at the center (geocentric). The motions of the Sun, Moon, and stars was based on perfect circles. To account for the observed retrograde motion of the planets, it was necessary to resort to a system of epicycles, whereby the planets moved around small circular paths that in turn moved around larger circular orbits around the Earth. This accounts for retrograde motion according to the ancients.
In its final form, the model was extremely complicated, requiring many nested levels of epicycles, and with even the major orbits offset so that they were no longer truly centered on the Earth. Despite all of this fine tuning, there remained significant discrepancies between the actual positions of the planets and those predicted by the model. Nevertheless, it was the most accurate model available, and it remained the accepted theory for over 13 centuries, before it was finally replaced by the model of Copernicus.
Copernicus replaced the geocentric universe of Ptolemy with one that was centered on the Sun (heliocentric), with only the Moon orbiting the Earth. His model was still based on circular orbits and therefore still required further refinement), but it was able to achieve superior precision than the Ptolemaic model without the need for epicycles or other complications. The explanation for retrograde motion in this system arises from the fact that the planets further from the sun are moving more slowly in their orbits than those closer to the sun. The retrograde motion of Mars occurs when the Earth passes by the slower moving Mars.
When combined with the refinements of Kepler (elliptical orbits with the sun at one focus, relationships between distance from sun and orbital speed – both within a single orbit and between orbits) this does, in fact, provide the correct explanation for the observed retrograde motion along with precise predictions of the positions of the planets.