944 



that which we find in some of the later Assyrian texts. Of a know- 

 ledge of the saros there is not the least indication in these texts; 

 everything indicates on the contrary that the astrologers of that 

 period were not acquainted with the saros so as to be able to 

 forecast a lunar eclipse with certainty some time beforehand" ^) and 

 he concludes from this: "Before the 8"^'' century the eclipses of the 

 moon (and sun) were not observed with that care which is neces- 

 sary for establishing their period, which was still unknown in the 

 7^'' century." In how far the first part of this sentence is correct 

 will be shown presently. 



KuGLF.R points out, moreover, very rightly, that the discovery of 

 the saros is not so easy as is often thought ^). In the first place 

 because this period does not embrace a round number of days, but 

 8 hours more. If we take a list of the lunar eclipses') visible at a 

 particular spot, say Babylon, and then look up those which occur 

 one saros period later, they will all be seen to take place 8 hours 

 later, chiefly therefore by day : most of them will not be visible 

 in Babylon. On the other hand there now appear a large number 

 of eclipses the predecessors of which were not visible 18 years 

 earlier, because they occurred before the beginning of the night. 



Experience, therefore, by no means points in the direction of the 

 saros period. To get eclipses at about the same daytime the period 

 must be trebled; after 54 years the visible eclipses return to a great 

 extent in the same order. If we arrange tlie visible eclipses in series 

 of 54 or of 18 years, it is then not difficult to establish (he existence 

 of the saros period. But it is quite another matter to find or to 

 discover this period. If someone who knew nothing of this period 

 was given the task of finding from a complete list of lunar 

 eclipses, e.g. from .Oppolzkr's Canon, a period after which they would 

 return in the same manner, he would certainly find it a very 

 difficult one. And how much more difficult the discovery must have 

 been in Babylon may be seen when we consider the conditions that 

 had to be fulfilled. 



The first essential would be to have a complete list of all the 

 visible eclipses. Now it is doubtful whether in the Assyrian period 

 (8^'^ and 7''i centuries B. C.) a continuous list could be made of 

 the eclipses observed, as, at least in all oin' extant reports of the 

 astrologers, the year is not noted. Let us, however, suppose that for 



1) F. X. KuGLEK, ibid Erganzungen, p. 132. 



2) Ibid. Vol. II. p. (i5-67. 



3) We do not mention sun eclipses here, because for them owing to the 

 influence of the parallax, the regularity is still much more diHicult to find. 



