562 COSMOS. 



in the Catoptrica of Archimedes.* We may esteem :t as an 

 important advance when physical phenomena, instead c f being 

 simply observed and compared toge ther (of which we have memo- 

 rable examples in Greek antiquity , in the comprehensive pseudo- 

 Aristotelian problems, and in lioman antiquity in the works of 

 Seneca), are intentionally evoked under altered conditions, and 

 are then measured.! This latter mode of proceeding charac- 

 terises the investigations of Ptolemy on the refraction of rays 

 in their passage through media of unequal density. >-t olemy 

 oaused the rays to pass from air into water and glass, and 

 from water into glass, under different angles of incidence, and 

 he finally arranged the results of these physical experiments 

 in tables. This measurement of a physical phenomenon 

 called forth at will, of a process of nature not dependent upon 

 a movement of the waves of light, (Aristotle, assuming a move- 

 ment of the medium between the eye and the object,) stands 

 wholly isolated in the period which we are now considering.;}; 

 This age presents, with respect to investigation into the ele- 

 ments of nature, only a few chemical experiments by Diosco- 

 rides, and, as I have already elsewhere noticed, the technical 

 art of collecting fluids by the process of distillation. Chemis- 

 try cannot be said to hnve begun until man learnt to obtain 

 mineral acids, and to employ them for the solution and libera- 

 tion of substances, and it is on this account that the distillation 

 of sea water, described by Alexander of Aphrodisias under 

 Caracalla, is so worthy of notice. It designates the path by 

 which man gradually arrived at a knowledge of the heteroge- 



* Delambre, Hist, de I' Astronomic ancienne, t. i. p. liv.; t. ii. p. 551. 

 Theon never makes any mention of Ptolemy's Optics, although he liver' 

 fully two centuries after him. 



t It is often difficult in reading ancient works on physics, to decide 

 whether a particular result has sprung from a phenomenon purposely 

 called forth, or accidentally observed. Where Aristotle (De Codo, iv. 4) 

 treats of the weight of the atmosphere, which, however, Ideler appears 

 to deny (M eteorologia veterum Grcecorum et Romanorum, p. 23), he 

 says distinctly, " an inflated bladder is heavier than an empty one." The 

 experiment must have been made with condensed air, if actually tried. 



Aristot., de Anima., ii. 7; Biese, Die Philosophic des Aristot., bd. 

 ii. s. 147. 



Joannis (Philoponi) <?ramraai/c7, in libr. de General., and Alexan- 

 dri Aphrodis., in Meteorol. Comment. (Tenet. 1527), p. 97, b. Compare 

 my Examen critique, t. ii. ?p 306- :12. 



