vom 29. October 1866. 653 



fröm a paper of Dr. Hartwig in the Astronomische Nachrichten 

 No. 1241. Ptolemy says, however, that it was exactly at Ba- 

 bylonian midnight, by which we must evidently understand 

 apparent midnight, a quarter of an hour after mean midnight. 

 Here then is an error of a füll hour in the time of Opposition, 

 answering to about 27' in Space; much the same error as in 

 the two other eclipses, and in the same direction. I know that 

 it may be objected that if such an error existed in Hansen's 

 tables, the eclipse of — 382 Dec. 22 could not have been visible 

 at Babylon at all. I accept the conclusion. I am satisfied that 

 the record of that eclipäe which Ptolemy has preserved is that 

 of an Observation at Athens, not at Babylon. The Athenian 

 astronomer who visited Babylon in the reign of Artaxerxes 

 Mnemon and broughtwith him the records ofBabylonian eclipses, 

 found only two eclipses recorded as having been observed there 

 in — 382 or thereabouts; and wishing for a third, in Order 

 that he might compare them with the three observed in the 

 years — 720 and — 719, he supplemented them with that which 

 he found recorded, and had perhaps himself observed at Athens 

 in the month Poseideon, in the archonship of Phanostratus. 

 The dates of the other two eclipses he reduced from Babylonian 

 to Athenian dates, in order to make them correspond with the 

 other; and it was in this form that Hipparchus obtained the 

 records of them all. The mistake of referring the first Obser- 

 vation to Babylon may have originated with Hipparchus; or 

 Ptolemy may have misunderstood what Hipparchus had said. 

 Now, if the eclipse be referred to Athens, the error in the tables 

 deduced from the Observation is in the same direction as is in- 

 dicated by the two eclipses of Merdocempedus and by that of 

 Sennacherib. The sun would rise at Babylon about 84 minutes 

 earlier than at Athens. The beginning of the eclipse was accord- 

 ing to Hartwig's computation 22 minutes later than it was by 

 the record as referred to Babylon, which would be about 62 

 minutes earlier than it was if the record referred to Athens, 

 answering to 35 minutes in Space. This is more than the error 

 indicated by the former tables; but it should be recollected that 

 in a calculation of the beginning or ending of an eclipse, the 

 latitude, which again partly depends on the place of the 



