INFLUENCE OF THE ROMAN EMPIBE 561 



It is to be regretted that Ptolemy, who had arrived at so 

 correct a knowledge of the complete insulation of the Caspian 

 (after it had long been considered to be open, in accoi dance 

 with the hypothesis of four gulfs, and even according to sup- 

 posed reflections of similar forms on the moon's disc),* should 

 not have relinquished the myth of the unknown southland con- 

 necting Cape Prasum with Cattigara and Thina? (Slnarum 

 metropolis], joining, therefore, Eastern Africa with the land 

 of Tsin (China). This myth which supposes the Indian Ocean 

 to be an inland sea, was based upon views which may be 

 traced from Marinus of Tyre to Hipparchus, Seleucus the 

 Babylonian, and even to Aristotle. f We must limit ourselves 

 in these cosmical descriptions of the progress made in the con- 

 templation of the universe, to a few examples illustrative of 

 the fluctuations of knowledge, by which imperfectly recognised 

 facts were so often rendered still more obscure. The more the 

 extension of navigation and of inland trade led to a hope that 

 the whole of the earth's surface might become known, the 

 more earnestly did the ever- wakeful imagination of the Greeks, 

 especially in the Alexandrian age under the Ptolemies, and 

 under the Roman empire, strive by ingenious combinations to 

 fuse ancient conjectures with newly acquired knowledge, and 

 thus speedily to complete the scarcely sketched map of the earth. 

 We have already briefly noticed that Claudius Ptolema3us, by 

 his optical enquiries which have been in part preserved to us by 

 the Arabians, became the founder of one branch of mathematical 

 physics, which, according to Theon of Alexandria, had already 

 been noticed, with reference to the refraction of rays of light, 



norum ad gentes, e rec. Bekkeri et Niebuhr., 1820, pp. 300, 619, 623, 

 and 628. 



* Plutarch, de Facie in orbe lunce, pp. 921, 19 (compare my Examen 

 crit., t. i. pp. 145-191). I have myself met, among highly-informed Per- 

 sians, with a repetition of the hypothesis of Agesianax, according to 

 which, the marks on the moon's disc, in which Plutarch (p. 935, 4) 

 thought he saw " a peculiar kind of shining mountains" (volcanoes 1), 

 were merely the reflected images of terrestrial lands, seas, and isthmuses. 

 These Persians would say, for instance, " What we see through tele- 

 scopes on the surface of the moon are the reflected images of our own 

 country." 



t Ptolem., lib. iv. cap. 9 ; lib. vii. cap. 3 and 5 Compare Letronn^ 

 in the Journal de Savans, 1831, pp. 476-480, and 545-555; Hun* 

 toldt, Examen cril., t i pp. 144, 161, and 329; t. ii. pp. 370-373. 



2 o 



