186 



Sir J. Conroy. On the Polarisation of [Feb. 4 r 



These results differ considerably from those obtained previously 

 with a natural face in water (Tables II, IV, and VI), and it therefore 

 did not appear worth while to make any further experiments with 

 artificial surfaces, as it seemed certain that the results would be 

 untrustworthy. 



The difference between the results obtained with this artificial 

 surface and with a natural surface of the crystal is too great to be 

 explained by supposing that the artificial surface was not cut absolutely 

 parallel to the direction of the cleavage, and must therefore be attri- 

 buted to some change produced by the polishing, possibly due to the 

 pressure employed (conf. Seebeck, " Pogg. Ann.," vol. xx, 1830, 27). 



The values of the azimuths and angles of polarisation given in 

 Tables I, II, III, IV, V, VI, VII, and VIII, were plotted on sectional 

 paper; the azimuths of the principal section of the crystal being taken 

 as abscissae, and the azimuths of the Nicol, and the angles of polarisa- 

 tion —40°, as the ordinates for the two sets of curves. 



In order to draw the smooth curves, a piece of plate-glass, rather 

 smaller than the drawing paper, was mounted in a soft-wood frame, 

 so that one surface of the glass was flush with the wood, the sectional 

 paper upon which the observations had been plotted was fixed to the 

 wood with drawing pins, and a sheet of ordinary drawing paper 

 placed over it, and fastened in the same manner. This glass drawing 

 board was then placed in front of a lamp, and smooth curves drawn 

 by eye in the ordinary manner. 



Professor Stokes pointed out to me that the experimental results 

 which had been obtained were well suited for reduction by means of 

 the harmonic analysis, and not only explained the method but himself 

 reduced the first set of observations made with a cleavage-face in 

 water. All the observations were accordingly reduced by this 

 method ; the determinations made at azimuths 1°, 11°, &c, and at 

 7° 20', 17° 20', &c, in the one series, and at 0°, 10°, &c, and at 

 5°, 15°, &c, in the second series, being kept separate. 



Owing to the fact that the principal section of the crystal is a 

 plane of symmetry, the periodic series for the development of the 

 azimuths of the planes of polarisation can contain sines only, and 

 that for the polarising angles cosines only, including the constant 

 term ; therefore the coefficients of the cosines in the former case, and 

 of the sines in the latter, were not calculated, except with the obser- 

 vations made with the artificial surface ; it seemed possible that the 

 process of polishing might occasion some want of symmetry, and 

 that therefore it was desirable to calculate the values of the co- 

 efficients for both sines and cosines. 



The observations in one set only having started at zero azimuth, in 

 the other three there was a small correction to be made in the 

 coefficients for the error thus produced; this was done by multiplying 



