1 NOTES. 



from an old manuscript of the Almagest, omitting all the latitudes. As 

 Ptolemy was but imperfectly acquainted with the amount of the retrogression 

 of the equinoctial and solstitial poiuts, (Almag. vii. c. 2, p. 13, Halma), and 

 had assumed it about 0'28 too slow, his table (Ideler, in the work above cited, 

 p. 34), which he intended to correspond to the beginning of the reign of 

 Antoninus, gives the places of the stars for a much earlier epoch (i. e. for the 

 year A.D. 63). Compare also considerations and tables for facilitating the 

 reduction of modern star places to the time of Hipparchus, given by Eucke in 

 Schumacher's Astr. Nachr., No. 608, S. 113 to 126. The earlier epoch for 

 which, unknown to its author, Ptolemy's table represents the firmament, 

 coincides very probably with the epoch to which we may refer the Catasterisms 

 of the Pseudo-Eratosthenes, which, as I have already remarked elsewhere, are 

 later than the Augustean Hyginus, appear to be taken from him, and are 

 unconnected with the poem of Hermes by the true Eratosthenes. (Eratos- 

 thenica, composuit God. Bernhardy, 1822, p. 114, 116, and 129.) These 

 Catasterisms of the Pseudo-Eratosthenes contain barely 700 separate stars 

 distributed among the mythical constellations. 



( 178 ) p. 93. Kosmos, Bd. ii. S. 260 and 433; English edition, p. 224, 

 and Ldx. note 354. The Paris Library possesses a manuscript of the Ilkha- 

 nian Tables, written by the hand of the son of Nassir-Eddin. They take their 

 name from the title llkhan, assumed by the Tartar princes who reigned in 

 Persia. Keinaud Introd. de la Geogr. d'Aboulfeda, 1848, p. 139. 



( 179 ) p. 93. Sedillot fils, Prolegomenes des Tables Astr. d'Olong-Beg, 

 1847, p. 134 ; Note 2 ; Delambre, Hist, de 1'Astr. du moyen age, p. 8. 



( 18 ) p. 93. In my examinations into the relative value of astronomical 

 determinations of geographical positions in the interior of Asia, (Asie Centrale, 

 T. iii. p. 581596), I have given the latitudes of Samarcand and Bokhara 

 according to the different Arabian and Persian manuscripts in the Paris Li- 

 brary. T have shewn that the latitude of Samarcand is probably above 39 52', 

 while the greater number and best MSS. of Ulugh Beig have 39 37', and the 

 Kitab al-athual of Alfares, and the Kanun of Albyruni, 40. I think it right 

 again to call attention to the importance to geography and to the history of 

 astronomy, of a new and trustworthy determination of the latitude and longi- 

 tude of Samarcand. We know the latitude of Bokhara by culminations of 

 stars from Burnes' Travels ; they make it 39 43' 41*. This would give the 

 errors of the two fine Persian and Arabian MSS. (Nos. 164 and 2460) in the 

 Paris Library at only 7 to 8 minutes ; but Major Rennell, generally so happy 

 in his combinations, would have been in error 19' for the latitude of Bokhara. 



