of the Atmosphere. 



171 



S A, and when we direct it to the 

 region of maximum polarization, 

 with the polarized bands parallel 

 to S A, S being towards the sun, 

 we shall see an interruption in 

 the bands somewhere between S 

 and A. If this interruption, or 

 point of compensation, is at the 

 point 2 in S A, I call S 2 the mea- 

 sure of the maximum degree of 

 polarization at the time of observa- 

 tion. After some practice I had 

 no difficulty in estimating by the 



eye when the neutral line was at 1 or 1^, or 2 or 2-|, without 

 placing marks at 1, 2, 3, &c. on a plate of glass at the end of 

 the tube. 



Having found that SO, SI, S2, &c. corresponded with degrees 

 of polarization, measured by the rotation of the plane of polari- 

 zation, I thus had a measure of the maximum polarization of the 

 atmosphere at the time of observation *. 



2. When it was necessary to measure very small degrees of pola- 

 rization, I preferred using a polarimeter with a single plate to one 

 with a pile of plates receiving the light at very small angles of inci- 

 dence. This instrument, shown in the annexed figure, consists 

 of two tubes of the same length, one of which, E F H G, moves 



Fig. 5. 



g a 2 1 * 



— ~^£ 



G 



l 



within the other, ABDC. One plate of glass, E L, longer than 

 AB, moves round a joint at E the end of the tube E F H G, 

 resting upon C (the end of the tube ABD C), so that when the 

 tube E F H G is pushed in, the plate E L forms a greater angle 

 with the axis of the tube, and when it is pulled out a smaller 

 angle. 



Now, if AB = EF, BF will be =AE, the tangent of the 

 angle ACE, AC being radius, or of the angle of incidence of 

 rays that fall upon E L parallel to the axis of the tube. The 



* See Phil. Trans. 1830, pp. 69, 133, 145, 287; and Trans. Royal Irish 

 Acad. vol. xix. part 2. 



