POLARIZED LIGHT AT THE SHRFACE OE A DNTAXAL CRYSTAL. 
615 
Behind the Nicol was placed a large lens, K, of somewhat short focal length, to act 
as a condenser, and at its focus the source of light, C, used in the experiment. 
The lens M, the screen and lamp L, were of course removed, and when the narrow 
parallel beam of light from the lamp, C, after passing through the Nicol fell on the 
slit of the collimator, I knew that it traversed the Nicol in a direction parallel to the 
axis of rotation. 
By means of a long handle and a Hook’s joint the tangent screw of the circle of the 
polarizer could be turned by the observer when his eye was at the telescope. 
The experiments were conducted in a somewhat different order to that, adopted in 
the first set. The analyzing prism was placed so that the light fell on it at a known 
angle of incidence and the Nicol turned by hand until the two black bands appeared 
fairly close together in the ordinary spectrum. The polarizing circle was then clamped 
and turned by means of the handle until the bands were brought exactly into coincidence. 
The reading N 0 of the Nicol circle was taken and also the deviation D 0 of the dark 
band in the ordinary spectrum. The Nicol circle was then turned, keeping the angle 
of incidence on the analyzing prism the same, until the dark bands were seen in the 
extraordinary spectrum and coincidence was established as before; the reading N E of 
the Nicol circle was taken, and also the deviation D E of the band in the extraordinary 
spectrum. The analyzing prism was then moved so as to alter the angle of incidence, 
and another set of observations taken, only in the reverse order. A series of values 
of the angles of incidence and the quantities N 0 , N E , D 0 , D E were thus observed. Each 
single observation was repeated five times and the mean taken and used in the 
calculations. 
For a certain known angle of incidence, D 0 is the deviation of the light that is 
quenched in the ordinary spectrum. It is also however the deviation in the ordinary 
spectrum of the light that is missing from the extraordinary, when the coincidence 
between the black bands is established there. 
D 0 is therefore the angle of deviation we must use in calculating the position of the 
plane of polarization of the incident light, when the ordinary ray only is transmitted. 
The experimental value of the angle determining this position, measured from some 
unknown zero, is N E , for this gives us the position of the Nicol when the extraordinary 
wave is quenched. 
Of course the theoretical values of the positions of the plane of polarization might 
have been determined as before from the values of the angle of incidence and 
ordinary refractive index, by the use of the interpolation Table I. 
I wished, however, to make this series of observations entirely independent of the 
first, and so recalculated the interpolation table. 
Since the position of the zero of the polarizer circle with reference to the analyzing 
prism is unknown, an unknown constant will, as before, come into our tables. 
Let us determine it so that the theoretical and experimental values of the position 
of the plane of polarization in the first experiment on the ordinary wave agree. 
MDCCCLXXXII. 4 K 
