POLARISATION OF LIGHT. 



49 



ing this fact was read at the Royal 

 Society on the 19th of May of the same 

 year. This letter was followed by a 

 series of papers containing an elaborate 

 inquiry into the laws which regulate 

 this new class of optical phenomena. 

 A brief and general view of these facts 

 will constitute the subject of this chap- 

 ter. 



1. On Cylinders of Glass with one 

 axis of double refraction. If we take 

 a cylinder of glass about half an inch 

 thick and an inch in diameter, and trans- 

 mit heat from its circumference to its 

 centre, it will exhibit \vhen exposed to 

 polarised light in the apparatus, Jig. 39, a 

 system of rings traversed by a black rect- 

 angular cross exactly like those shown in 

 Jig. 25 : and by turning round the analys- 

 ing plate, C, 90, we shall see the comple- 

 mentary set shewn in Jig. 26. As the 

 cylinder must be held at a distance of at 

 least six inches from the eye, the rings 

 and cross will appear as it were in the 

 interior of the glass. If we cross the 

 rings with a plate of sulphate of lime, we 

 shall find that they exhibit the same phe- 

 nomena as the rings formed by calca- 

 reous spar, and therefore the double re- 

 fraction is negative, or repulsive. When 

 the heat is uniformly distributed over the 

 glass, the system of rings entirely disap- 

 pears, and the doubly refracting struc- 

 ture no longer exists. 



If the same cylinder of glass, when 

 cold, is immersed in boiling oil, or 

 equally and strongly heated in any other 

 way, and is made to cool rapidly by ap- 

 plying a cold and good conductor to its 

 circumference, it will exhibit a similai 

 system of rings which will also disappear 

 when the glass returns to an uniform 

 temperature. This system of rings, 

 however, is positive, like those produced 

 by zircon and ice. 



By comparing the value of the tints 

 with their distances from the axis of the 

 cylinder, Dr. Brewster found that if T 

 is the tint corresponding to any distance 

 D, t the tint corresponding to any other 

 distance, d, it will be thus expressed : 



Td* 

 t = -, that is, the values of the tints 



vary as the squares of their distances 

 from the axis of the cylinder. 



If the polarised light is transmitted 

 through the circumference of the cylinder 

 when the system of rings is complete, 

 a system of fringes will be seen parallel 

 to the base of the cylinder, and similar 

 to those which are next to be described. 



2. On Rectangular Plates of Glass 

 with two series of axes of double re- 

 fraction. Take a well annealed rect- 

 angular plate of glass, E F C D, fig. 53, 

 which exhibits no tints whatever when 

 examined in the apparatus, Jig. 39, and 

 place its edge C D on a bar of iron A B 



Fig. 53. 



nearly red hot. Let the glass plate resting 

 on the iron be placed at D E, in fig. 39, so 

 that the polarised light, A C, may pass 

 through everj' part of it. The moment 

 that the heat of the iron enters the sur- 

 face CD, fringes of brilliant colours 

 will be seen parallel to C D, and nearly 

 at the same instant, and before a single 

 particle of heat has reached the upper 

 edge E F, or even a b, similar fringes 

 will appear at the edge E F. Light at 

 first white and then yellow, orange, &c., 

 will appear at a b, and these colours will 

 be separated from the other two sets of 

 fringes by two dark lines M N, O P, in 



which there is neither double refraction 

 nor polarisation. Between MN and 

 O P the double refraction and polarisa- 

 tion is negative, like that of calcareous 

 spar, while, on the other side of MN and 

 O P, the double refraction is positive. 

 Similar fringes are seen through the 

 thickness of the plates and also in the. 

 direction of their lengths. When the 

 plate of glass is of a square form, as in 

 fig. 54, and the thickness through which 

 the polarised light passes about one-third 

 the length of one of the sides, the tints 

 have the form shown in fig. 54. 



When the thickness is increased so as 

 E 



