12 SCIENCE IN GRECO-ROMAN ANTIQUITY 



periodically in the course of centuries. From 755 to 

 432 B.C. the eclipses succeeded one another in series 

 alternately of 5 and 6. In each series the eclipses took 

 place every six months, and the series were separated 

 by an interval of 17 lunations. 1 Thus it became 

 possible to make predictions at short notice, which 

 explains some of the inscriptions found on the cuneiform 

 tablets. 



As to the instruments of observation, we have little 

 information. The Chaldeans certainly made use of 

 the gnomon, which appears to be the most ancient 

 instrument used for studying the movements of the 

 stars, for it is everywhere mentioned before all others, 

 whether amongst the Chinese or the Chaldeans, the 

 Greeks or the Incas. It is, besides, a marvellously 

 simple instrument, composed of a vertical style stand- 

 ing on a horizontal plane. By reproducing the move- 

 ment of the sun, the extremity of the shadow pro- 

 jected by the style makes the division of the day 

 possible. 2 At first sight the precision of the observa- 

 tions made by means of the gnomon would seem to 

 increase with the length of the shadow, and therefore 

 with that of the style ; but in reality the shadow of 

 the style is not very sharp because of the penumbra. 8 

 Further, as the length and even the direction of the 

 shadow vary for the same hour on different days, it 

 was necessary to have recourse to some sort of table, 

 which gave for each month the length of the shadow at 

 different hours. 



Later on, the gnomon with the vertical style was 

 replaced by a gnomon with the style pointing towards 

 the pole. It was then only necessary to observe the 



1 2 Bigourdan, Astronomic p. 34. 



3 24 Sageret, Systdme, p. 95. 



8 In order to remedy this disadvantage Facundus Novus 

 had the idea of fixing a ball on the point of the gnomon. 

 Pliny, xxxvi, 72. 



