118 A SHORT HISTORY OF SCIENCE 



which the ratio of the radius of the excentric to the distance of its 

 centre from the earth could easily be determined for either planet. 

 Similarly, the three outer planets moved on excentric circles, the 

 centres of which lay somewhere on the line from the earth to the sun, 

 but these circles were so large as always to surround both the sun and 

 the earth. Dreyer. 



It seems probable that Aristarchus was led through this theory 

 to conceive of heliocentric orbits, and then to reflect that the earth, 

 too, might revolve about the sun as easily as the sun and planets 

 round the earth. 



EPICYCLES. Progress in observational astronomy increased 

 the number and magnitude of planetary irregularities beyond the 

 stationary points, retrograde motions, and variations, known to 

 Aristarchus, and apparently far beyond possible explanation by 

 the simple theory of excentric circles. The system was therefore 

 superseded by, or combined with, that of epicycles, not necessarily 

 as physically realized, but as at least a geometrical working hy- 

 pothesis, which should conform to and explain the observed phe- 

 nomena. 



The system of epicycles consists in superimposing one circular 

 motion upon another, and repeating the process to any needful 

 extent. The motion of the moon about the earth, for example, is 

 explained by assuming first a circle (later called the deferent) on 

 which moves the centre of a second smaller circle called the epi- 

 cycle, on which the moon itself travels. By varying the dimensions 

 of both circles and the velocities of the two motions, the observed 



changes, both of position and bright- 

 ness of the moon, may be more or less 

 satisfactorily accounted for and even 

 computed in advance. In particular, 

 the apparent retrograde motions of 

 the planets in certain parts of their 

 orbits may be explained. 



In the figure E denotes the earth, 

 the large circle is the deferent of a 

 planet, C the centre of the epicycle, PI, P 2 , Pz, P* different pos- 



