POLARIZED LIGHT AT THE SURFACE OP A UNIAXAL CRYSTAL. 
619 
isotropic refracting media agree with Fresnel’s formulae for the relative positions of 
the planes of polarization of the incident and reflected rays, though not for the 
refracted. 
The experiments having all been conducted with the same piece of spar do not 
afford data for deciding whether the differences observed are functions only of the 
angle of incidence, or whether, as is more likely, they depend partly on it and partly 
on the position of the face of incidence with reference to the axis of the crystal. 
Though in no case large, they are certainly considerably greater than the possible 
errors of experiment. 
The experiments have been conducted by Lord Bayleigh’s kind permission in one 
of the rooms of the Cavendish Laboratory, Cambridge, and the apparatus used is 
chiefly the property of the Laboratory. 
In conclusion, I would refer to a paper on the same subject by F. E. Neumann 
(“ Beobactungen iiber den Einfluss der Krystaleflachen auf das reflectirite Licht.” 
Pogg. Ann., xlii.). 
He polarized light by passing it through a tourmaline, and then allowed it to fall on 
a prism of Iceland spar at a known angle of incidence, observing the position of the 
plane of polarization when only one ray passed through the crystal. The light of a 
lamp was used and the crystal prism achromatised approximately for the ordinary ray. 
The prism was capable of rotation about a normal to the face on which the light fell so 
as to alter relatively to the axis of the crystal the position of the plane of incidence. 
The theoretical values of the position of the plane of polarization were calculated 
from formulae given by Neumann (Abhandlungen der Akad. der Berlin, 1835), and 
MacCellagh (Irish Transactions, 1837), which are identical with those deduced from 
the electro-magnetic theory. In the eight observations recorded, the differences between 
the calculated and observed values of the azimuth of the plane of polarization range 
from —8' to + 4'. 
Each observed value is the mean of 40 observations, but it is not stated how far the 
observations differ among themselves ; observations were only made for a small number 
of values of <f>, the angle of incidence, two at most for each position of the plane of 
incidence relatively to the axis of the crystal, and the values of <f> chosen were 40°, 
45°, 50° and 55°. 
Now, it will be seen on referring to my tables, that for these values of <j> the 
differences there recorded are small. Nothing is said in Neumann’s paper as to the 
method adopted to determine the values of <f/ and </F, the angles of refraction of the 
*■ 
ordinary and extraordinary rays. In fact, they are both indeterminate, seeing that the 
experiment was made with white light. These are the only other experiments made 
to test the theory so far as it concerns the refraction of light. 
The same paper of Neumann’s contains a series of experiments on the position of 
the plane of polarization of light reflected from the surface of the crystal under various 
circumstances, the results of which are compared with theory. 
