104 Mr Swan on MetJiods of ascertaining 



to the incident ray ; and this is best done by turning the prism 

 round in a plane perpendicular to its refracting surfaces, until 

 the deviation of the transmitted rays is observed to be a mini- 

 mum. When this is the case, the incident and emergent rays 

 are equally inclined to the faces of the prism, and the index 

 of refraction can be deduced from the observed angles by a 

 simple formula. This arrangement was adopted by Newton, 

 and afterwards by Fraunhofer ; and the extreme exactness of 

 which it is capable, has been shewn by the singularly accurate 

 results obtained by the latter observer. If, therefore, it could 

 be simplified in its details, so as to lessen the trouble of obser- 

 vation attendant on it in its present form, it might be rendered 

 more generally available for increasing our knowledge of re- 

 fractive powers. From actual experience, in using an appara- 

 tus constructed after Fraunhofer's plan, I have found that 

 the manner in which the prism is mounted is productive of 

 great inconvenience. This arises from the adjustments of 

 the faces of the prism mutually affecting each other, so that 

 it is only after many approximate trials that both faces can 

 be made perpendicular to the limb of the theodolite, by which 

 the deviation of the refracted rays is measured. Another 

 inconvenience also attaches, as far as I am aware, to all the 

 existing modifications of this method, in as much as different 

 instruments are required for measuring the angle of the prism, 

 and the deviation of the refracted rays. 



WoUaston's goniometer has indeed been used to determine 

 both of these angles ; but for the latter purpose, it is a most 

 defective instrument, as it is only by the aid of marks made 

 on the walls of the apartment in which the experiment is con- 

 ducted, that the deviation of the refracted rays is ascer- 

 tained.* 



In endeavouring to simplify this method of ascertaining 

 refractive powers, I devised two different processes, and, as 

 one of these was the means of suggesting the other, I shall 

 describe them in the order in which they occurred to me. 



I. The first is a new application of the sextant, and will be 

 best understood by reference to figs. 1 and 2 (Plate XL), in which 



♦ Edinlburgh Encyclopaedia, " Optics," p. 506. 



