in the Velocity of Propagation of Light in Water. 269 



II. The Methods of Observation. 



For the determination of the index of refraction many methods 

 may be adopted ; for almost every optical relation, in which the 

 velocity of the propagation of light is concerned, may be used as 

 the base of such a method. The most common of these methods 

 is that founded upon the measurement of the deviation which a 

 ray of light experiences in passing through a prism of the sub- 

 stance under investigation. This process is employed in my ex- 

 periments, and will be discussed subsequently. 



The second method, only employed by Arago* and Jamin 

 and lately also by Fizeau, is the method of interference. If we 

 allow two rays of common or homogeneous light to come together 

 under favourable circumstances in directions either parallel or 

 inclined at a small angle to one another, it is well known that 

 phenomena of interference occur which in general are parallel 

 bands when the common source of both rays is a line. If we 

 examine two rays of light which give rise to such phenomena, 

 we find, as is well known, a central bright band at the place 

 w r here both systems of waves have passed over exactly equal 

 lengths from their departure from the common source of light. 

 If, now, one of the rays is delayed, ; 4 it is later in reaching the point 

 where the central band was previously produced : this band can 

 no longer be found in the middle, but will be displaced a little 

 towards the side of the impeded ray. This fact enables us to 

 determine the velocity of propagation of light in a medium if we 

 know exactly the thickness of the impeding layer, and accurately 

 measure the displacement of the middle band by a micrometer 

 arrangement, or by angular measurement by means of a telescope. 



By this means we can detect extremely minute differences in 

 the refractive power ; and the method is very useful for differen- 

 tial observations. But the apparatus can no longer be employed 

 when the differences are great, unless we can effect a determinable 

 hindrance in the second ray as well as in the first. For this 

 purpose, in his apparatus Arago made use of a compensating 

 arrangement, the action of which depended upon the second ray 

 of light being made to pass at a measurable angle through a pa- 

 rallel-sided glass plate of known thickness, by which means it- 

 suffered an exactly determined retardation. We shall not describe 

 Arago's apparatus more minutely, because it has scarcely been 

 applied to quantitative experiments ; and we need only indicate 

 the principle of the modified interference-refractors which Jamin 

 employed for his observations f. This instrument depends upon 



* Arago's Sdmmtliche WerJce, deutsch, von Hankel, vol. x. p. 25/, ff. 

 t " Interferenzialrefractor," Cosmos, 1856, No. 10, p. 22/ ; and Pogg. 

 Ann. vol. xcviii. pp. 345-349. 



