THE GREEK ASTRONOMY. 509 



record is in the first Book of Herodotus's History (chap. Ixxiv.) He 

 Bays that there was a war between the Lydians and the Medes ; and 

 after various turns of fortune, " in the sixth year a conflict took place ; 

 and on the battle being joined, it happened that the day suddenly be- 

 came night. And this change, Thales of Miletus had predicted to 

 them, definitely naming this year, in which the event really took place. 

 The Lydians and the Medes, when they saw day turned into night, 

 ceased from fighting ; and both sides were desirous of peace." Prob- 

 ably this prediction was founded upon the Chaldean period of eighteen 

 years, of which I have spoken in Section 11. It is probable, as I have 

 already said, that this period was discovered by noticing the recurrence 

 of eclipses. It is to be observed that Thales predicted only the year 

 of the eclipse, not the day or the month. In fact, the exact predic- 

 tion of the circumstances of an eclipse of the sun is a very difficult 

 problem ; much more difficult, it may be remarked, than the predic- 

 tion of the circumstance of an eclipse of the moon. 



Now that the Theory of the Moon is brought so far towards com- 

 pleteness, astronomers are able to calculate backwards the eclipses of 

 the sun which have taken place in former times ; and the question has 

 been much discussed in what year this Eclipse of Thales really occur- 

 red. The Memoir of Mr. Airy, the Astronomer Royal, on this subject, 

 in the Phil. Trans, for 1853, gives an account of the modern exami- 

 nations of this subject. Mr. Airy starts from the assumption that the 

 eclipse must have been one decidedly total ; the difference between 

 such a one and an eclipse only nearly total being very marked. A 

 total eclipse alone was likely to produce so strong an effect on the 

 minds of the combatants. Mr. Airy concludes from his calculations 

 that the eclipse predicted by Thales took place B. c. 585. 



Ancient eclipses of the Moon and Sun, if they can be identified, are 

 of great value for modern astronomy ; for in the long interval of be- 

 tween two and three thousand years which separates them from our 

 time, those of the inequalities, that is, accelerations or retardations of 

 the Moon's motion, which go on increasing constantly, 1 accumulate to 

 a large amount ; so that the actual time and circumstances of the 

 eclipse give astronomers the means of determining what the rate of these 

 accelerations or retardations has been. Accordingly Mr. Airy has dis- 

 cussed, as even more important than the eclipse of Thales, an eclipse 

 which Diodorus relates to have happened during an expedition of 



Or at least for very long periods. 



