its Circular Components in the Faraday "Effect." 467 



obtained, the length of the prism (87 mm.) and consequent 

 distance between the pole-pieces prevented m) r obtaining a 

 sufficiently intense field to detect any resolution or widening 

 of the ray. A series of reflexions for increasing the effect 

 was not found feasible. 



The following arrangement admitted of repeated reflexions 

 and also less distance between the poles, so that the requisite 

 resolution was finally attained. 



> a 



In fig. 2, B and C are rectangular prisms of the heaviest 

 glass obtainable, ?/d = 1'903. These were placed with their 

 diagonal faces in contact with a plate of mica M. Two 

 auxiliary rectangular prisms of the same glass were placed, 

 one A near the lower end or base of the pair and the other 

 A' at the top, with their diagonal faces in contact with one 

 of the faces of B and C respectively. The whole system was 

 cemented together with a-monobromonaphthalin n I} = 1*6582 

 between A, B, M, C, and A'. A ray a from the slit passing 

 into A at the base of the combination, and then into B and (j 

 across M, was totally reflected by the outside face of at a 

 point below A! , thence again totally reflected by across M 

 and into B, where it was totally reflected to the other surface, 

 and so on around through the system, until it reached a 

 sufficient height to pass out through the second auxiliary 

 prism A' in the direction a' into the observing telescope. 

 The prismatic system was mounted with its refracting edges 

 vertically within an adjustable holder between the poles of 



