105 



sun*. We think, that this circumstance has 

 but little influence on the accuracy of the cya- 

 nometrical measures made in the torrid zone; 

 for the greater the elevation the Sun has above 

 the horizon, the more uniformity there is in the 

 distribution of the aerial light -f\ It even ap- 

 pears, that a part of the sky may reflect a greater 

 or less quantity of light, without any indication 

 of a deeper or fainter tint by the cyanometer. 



I shall not enlarge any farther on the analogy, 

 that exists between the results obtained by the 

 cyanometer of Saussure and the lucimeter of 

 Bouguer. This subject is one of the most deli- 

 cate investigations in optics ; and the tint of the 

 sky deserves so much more the attention of na- 

 turalists, as the ingenious experiments of Mr. 

 Arago have recently proved, that the aerial light 

 is composed of rays not all of the same nature, 

 since it contains some that are insusceptible of 

 polarization. 



If the cyanometer indicate, I will not say the 

 quantity, but the accumulation and the nature 

 of the opake vapors contained in the air, the na- 

 vigator is in possession of a more simple manner 

 of judging of the state of the low regions of the 

 atmosphere. He attentively observes the color 

 and figure of the solar disk at it's rising and 



* Bouguer, Traite d'Optique, p. 71 and 367. 

 t Ibid, p. 74. 



