hy transmission through Fluids, 4-33 



reading, which requires much practice, will readily account 

 for an error of a few degrees, and the subjoined table (based 

 on the supposition that the original bulk was preserved as it 

 ought to have been by the addition of as much water as was 

 lost by evaporation) will show that in the last four experi- 

 ments O2 was probably mistaken for X. 



Yellow ray, depth 100 millimetres = 3*937 inches. 



The above table will give a general idea of the mode in 

 which the results are to be taken and tabulated. I find, as 

 might be expected, that solutions of sugar become decomposed 

 by keeping and by heat, but the effect is simply to diminish 

 the amount of rotation, and not lo alter the direction. I have 

 already adverted to the care required in observing the precise 

 point of obscuration, viz. the extreme boundary between the 

 blue and violet ray, and to the assistance to be derived from 

 observing the complementary tint. 



It is desirable to preserve one position during the observa- 

 tions with the face towards zero, and to look steadily in a 

 perpendicular direction, as the appreciation of the tint will 

 vary somewhat if the position of the eye be altered*. I ge- 

 nerally get an assistant to note the numbers, so that the ob- 

 servations may be made with as little looking off as possible. 

 The effect of gazing at all coloured objects must be remem- 

 bered, and it would be well to have the whole of the appara- 

 tus blackened, or, where convenient, to make the observations 

 in a. darkened chamber. The loss of light occurring from an 

 increased depth or strength of liquid, especially if such liquid 

 be coloured, must be also considered, and will frequently 

 prevent the experiments made at different depths from ac- 

 cording as they should do with each other. With constant 

 practice I can obtain a reading to half a degree, beyond 

 which I do not think it is possible in most cases to obtain 

 any certainty ; and of course the method does not furnish an 



* With some analysers the plane of polarization will be entirely changed 

 if the position of the eye be greatly altered. 



PhiL Mag. S. 3. Vol. 25. No. 168. Dec. 184.4. 2 F 



