532 



turpentine is, therefore, more dispersive of the planes of polarization 

 than the other. 



In reasoning on this experiment it occurred to the author that a 

 similar effect ought to be produced if the ray of polarized light, instead 

 of passing successively through two columns of French and American 

 oil of turpentine, were transmitted through a single column, composed 

 of a mixture of these two fluids. As they are identical in chemical com- 

 position, it did not seem probable that any chemical action could take 

 place ; and therefore, in accordance with the principle found to hold 

 good for mixtures in general, the effect of two columns of fluid, when 

 mixed together, ought to be the sum of the effects which they produce 

 separately. If, therefore, the lengths of the two columns be such that 

 they compensate each other for the mean ray, the fluid produced by 

 mixing them together ought to exercise a left-handed rotation on the 

 plane of polarization of the red ray, and an opposite effect upon that of 

 the blue ray. 



This anticipation was fully realized by the actual experiment. Some 

 difficulty was found in ascertaining experimentally the exact proportion 

 in which the fluids should be mixed together. The source of this dif- 

 ficulty is, as has been before stated, the impossibility of comparing with 

 exactness the intensities of lights whose colours are different, and 

 therefore of ascertaining the ratio of the lengths of two columns which 

 compensate each other for the mean ray \ and a small deviation from the 

 true proportion will render the rotation produced by the mixture either 

 wholly right-handed, or wholly left-handed. 



The best method of ascertaining this proportion is to measure suc- 

 cessively the actual rotations produced by each kind of oil of turpentine 

 in the planes of polarization of the red and blue rays. Let F r be the 

 rotation produced in the plane of polarization of the red ray by the 

 column of Bourdeaux turpentine ; F b that produced in the plane of po- 

 larization of the blue ray, and A r , A h , the corresponding rotations pro- 

 duced by American turpentine ; then, if m be the quantity of French 

 turpentine in the mixture, and n the quantity of American turpentine, 

 the mixture will be right-handed for the red ray (and therefore a for- 

 tiori for all the rest), if 



n F r 



- = or > — i 

 m A, 



on the other hand the mixture will be left-handed for the blue ray, 

 and therefore for all the less refrangible rays, if 



n F b 



— = or < — ; 



m A b 



and if the value of — be intermediate to these two values, the fluid is pos- 



sessed of double rotation. The phenomenon is rendered most striking by 

 giving to the two fluids in the mixture such a proportion that the left- 



