﻿M. H. Wild on the Absorption of Light by the Air. 303 



Appendix, 



In close connexion with the preceding investigation is another, 

 of which I wish here to communicate a preliminary result. It 

 refers to the difference in the colour of the waters of lakes and 

 rivers in summer and winter, as well as of the warm and saline 

 water of the Gulf-stream as compared with the surrounding 

 water. I consider that the darker or deeper and more vivid 

 colour in summer, and in the Gulf-stream, is not to be ascribed 

 to a greater quantity of salt, but essentially to the higher tem- 

 perature. It is well known that in most bodies absorption in- 

 creases with increase of temperature. To decide whether this 

 is also the case with water, I first determined the coloration of 

 a paper surface illuminated by the sun, when looked at through a 

 layer of water 2*4 metres in thickness. To judge the colour 

 the better, the tube was only half filled, so that the uncoloured 

 white paper above was visible. Ordinary well-water cooled to 

 7° C. gave a bright greenish-white tone, and at 50° C. a de- 

 cidedly deeper bright green coloration ; distilled water, too, at 

 20° C. exhibited a bright bluish-green colour, which at 58° C. 

 passed into a deeper and greener colour. 



The transparency of water at two different temperatures was 

 directly measured. Between the photometer and the ground- 

 glass pane of a front window, at one aperture two glass tubes 

 (each closed by glass plates) of 50 millims. and 200 millims. 

 length were placed, at the other aperture of the photometer two 

 other such tubes of 100 millims. and 150 millims. length. 

 The relative intensities with the empty and the partially filled 

 tubes were ascertained; in one case the tubes of 50 millims. and 

 150 millims., and in the other those of 100 millims. and 200 

 millims. were filled with distilled water filtered through coarse 

 filtering-paper. The temperature of the water, of about 6° C, 

 was obtained by cooling the entire room to this temperature ; 

 with the higher temperature of 25° C, the room was at a tempe- 

 rature of 16° C. From the experiments made on the 4th and 

 5th of January, 1867, the following values w T ere obtained for the 

 coefficient of transparency of distilled water filtered through 

 coarse paper : — 



At24°-4C. a=0-91790, 



At 6°-2 C. a= 0-94769, 



referred to 1 decim. as unit of path. My former experiments in 

 Konigsberg had given for the same paper and a temperature of 

 about 17° C, and referred to the same unit of path, the value 



17° C. «=0-93968, 

 which agrees well with the above. Hence the transparency of 

 water at low temperatures is in fact greater than at higher ones. 



