Intelligence and Miscellaneous Articles. 143 



I have accordingly arbitrarily modified the exponent of X in such 

 manner as to make the results of observation agree with the 



formula — . 

 A« 



Let B be the intensity of the blue light diffused by the sky and 



S the intensity of the incident solar light ; then, putting the value 



of — for A =565 equal to 100, we shall have 



S 



B = *, 100= JL whence | = 100 », 

 8 A»' 565"' S U; 



n being a coefficient which is calculated by means of the values of 

 — corresponding to fixed values of A. 

 On differentiating we obtain 



tan a = — 100 n , 



and for the common ordinate = 100 which corresponds to A =565 



100 

 tan a.= — — — n : 

 565 



n being therefore proportional to the angular coefficient at the 

 point of intersection of all the curves which are obtained by making 

 n vary, that is to say, to the inclination of the curve at that point. 



I have compared the results calculated from the formulas r^ and 



1 



-^ with those of the observations of Lord Eayleigh in England * 

 A 



and of Yogelt at Potsdam, as well as with some of the results of 

 my observations at Montpellier. 



In the former case, I have constructed the curves of these obser- 

 vations and have determined the values of the ordinates corre- 

 sponding to 



A .... 635 600 565 530 510, 



and traced a new curve for which the ordinate corresponding to 

 A = 565 has been arbitrarily made =100. 



For the series of M. Vogel, who gives, not — , but p , that is to 



say the ratio of the intensity of skylight to that of petroleum, I 



have divided the ratios - by those of — or of the intensity of solar 



light to that of petroleum for the different A's found by M. Yogel, 



which gives the corresponding values of -~, and I have reduced the 



* Phil. Mag. ser. 4, xli. p. 107. 

 t Berl. Monatsber. pp. 801-811. 



