1887.] Transmission of Sunlight through the Atmosphere. 171 



standard illuminating value of the spectrum was taken from observa- 

 tions made in Switzerland at 8000 feet altitude on September 15 at 

 noon. The other observations were made at South Kensington. It 

 was found that with the wind in the proper quarter the sky at the 

 latter place was as pure in colour as in the country, and that measures 

 made on the days on which there was apparently no haze gave results 

 which when combined together gave a minimum value for k of 0*0013. 

 A mean of the results showed that Jc — 0*0017. 



The author then discusses the value of the area of the curves so 

 obtained, and shows that astronomers who have used the ordinary 

 logarithmic formula of V = Ia~ sec ZD have not erred in so doing, but 

 that these results are perfectly concordant with the results obtained 

 by taking the different values of absorption over the whole visible 

 spectrum. He further shows that with the coefficients of trans- 

 mission for different wave-lengths which Langley has published, the 

 above formula is applicable, which is contrary to the theory which 

 Langley propounded. The author further shows that when using the 

 part of the spectrum to which various photographic salts are sensitive, 

 the areas of the curves of intensities also fall in with the logarithmic 

 formula. 



He also points out that if the value of jx in the formula V= Ier-M* 

 (from which the formula I' = Ia secZD is deduced) be divided by 104 

 in the case of the optical value, or by 255 in the case of the photo- 

 graphic value, when bromoiodide of silver is employed on the 

 sensitised plate, the value of h is obtained. 104 and 255 represent 

 l/\ 4 of 5570 and 4450 respectively. From this he deduces the fact that 

 the illuminating value of any part of the visible spectrum on any day 

 at any hour can be ascertained by taking the optical and photographic 

 values of total sunlight. This plan not only enables the light which 

 undergoes general absorption to be calculated, but also the values of 

 loss of intensity for each ray. He further indicates that a double 

 photographic observation, in one of which the less refrangible end of 

 the spectrum, and in the other of which the more refrangible is used, 

 will give similar results. 



Experiments on the photographic methods are described, and the 

 results agree with those which theory indicated would arise. 



The author states that the high altitudes in the Alps may not be 

 suitable for observations of wave-lengths in the infra-red ; but that 

 they are especially suitable for observations in the visible part of the 

 spectrum, since the atmosphere is very often free from dust, though 

 it may not be free from aqueous vapour, and the latter affects the 

 dark rays vastly more than it does the visible rays. He then points 

 out the probable cause of the presence of particles which scatter light,, 

 and briefly discusses the differences which he obtains in his value for 

 the transmission of light, as compared with Bouguer, Seidel, 



