XX.] THE MOVEMENTS OF THE EARTH. 341 



what is termed a second. Consequently, the shadow would 

 return to the same place night after night, in just 86,164 

 seconds. If an accurate clock, beating seconds, had a dial, 

 thg circle of which was. divided into 86,164 parts, and a 

 single hand, which should cover one of these divisions, at 

 every beat, the motion of the hand would exactly keep pace 

 with that of the star-shadow. And, if any point of the 

 dial were marked twelve, when the star-shadow was at 

 any point of its course ; whenever the star-shadow returned 

 to that place, the hand of the clock would again mark 

 twelve. 



Such a clock would keep what is called sidereal, or star- 

 time, and the 86,164 seconds (or 23 hours 56 minutes and 

 4 seconds) would be a day by the " star-clock." As the 

 apparent motion of the stars is due to the rotation of the 

 earth on its axis, the hand of the star-clock would travel 

 round the dial, in exactly the same time as the earth turns 

 on its axis; which period of time (86,164 seconds) is 

 termed a sidereal day. 



For practical purposes, however, this clock would be of 

 very little use. Unless we happen to be astronomers, when 

 we ask the time, it is not from any wish to learn how far the 

 earth has turned on its axis, in reference to a particular star ; 

 we want to know the time of day, or the time of night, 

 whether it is forenoon or afternoon, or the like. To 

 apply to our sidereal clock for an answer to these questions, 

 would be worse than useless. For, supposing that on any 

 particular day, twelve by the star-clock exactly corresponded 

 with noon by the sun ; the day after, the star- clock would 

 mark twelve nearly four minutes earlier ; and, the day after 

 that, earlier still by a similar amount ; so that, in a quarter of 

 a year, twelve by the star-clock would be six hours before 

 noon and so on. In short, twelve by the star-clock 



