136 Prof. E. C. Pickering on the Polarization of 



One series, the first, is given in full in Table V., to show the 

 amount of variation of the different points observed. 



Table V. 



12. July 15. Sun's distance 90°, altitude 5°. 



M. D. Polar. 



75° 78-6 



60 74-3 



45 79-8 



30 770 



77-2 



30 74-8 



45 75-2 



60 76-8 



75 7^8 



Mean . . 767 



All these observations point to one very remarkable result — 

 namely, that the polarization is the same for a given solar dis- 

 tance for any meridian distance — in other words, that the polar- 

 ization is the same for all points equally distant from the sun. 

 The variations in the observations are to be ascribed partly to 

 errors of observation and partly to real irregularities in the at- 

 mosphere, as it is evident that they follow no regular law. The 

 means therefore give us the true polarization with much greater 

 accuracy. They are represented in fig. 5 by small crosses. The 

 next thing is to determine the law which connects the polariza- 

 tion with the solar distance in all these observations. A draw- 

 ing was made like fig. 5 enlarged, and a fine copper wire laid on 

 it and bent into such a shape that it should coincide as nearly 

 as possible with all the observations. The ordinates for every 

 10° were then read off, giving the results entered in column 7 

 headed « Observed" of Table I. 



A simple explanation of the polarization of the sky is to as- 

 sume that it consists of molecules of air or aqueous vapour 

 which reflect the light specularly, and whose index of refraction 

 differs only by a very minute amount from that of the medium 

 in which they float. The theoretical polarization would then be 

 at once given by Table II., making i equal to one half the solar 

 distance. The curve thus obtained is given in fig. 5 at A. The 

 polarization, according to this, should be complete at 90° from 

 the sun, while in reality it is only about 70 per cent. If, how- 

 ever, we multiply the ordinates of curve A by this fraction, we 

 obtain curve B, which agrees almost precisely with the curve 

 given in column 2 of Table XII. (/. c). Its ordinates are given in 

 column 3, and the differences in column 4. From the latter it 



