346 Dr. Lodge on the Rotation of the Plane of Polarization 



unfortunately the discharge is Hable to burst the covering. 

 In my helices p'=3p, which is rather excessive, because it 

 reduces the visible effect to one ninth of what it might be 

 if no covering were used. 



It is impossible to press this to its natural consequence of 

 using only one turn of thick sheet copper, because the resist- 

 ance of the rest of the circuit has been neglected. Taking it 

 into account, as r, the relation becomes 



lb 



(,')• 



4r 



r.iPPT 



K- .. , . ■ . . (6) 



which contains the product of the sizes of covered and un- 

 covered wire as well as the ratio ; and this product occurs in 

 the term containing r, the external resistance. Hence, to keep 

 this term small, it is desirable to use wire thin enough to 

 throw the major part of the resistance into the bobbin ; and 

 there is no limit to the thinness of the wire that may be ad- 

 vantageously employed, until the thickness of the covering 

 bears too high a ratio to the whole. And inasmuch as 

 insulation thickness may be more judiciously distributed 

 between layers than between consecutive turns, it is obvious 

 how extremely suitable a coil for the purpose is the secondary 

 of a Biihmkorff. 



I said that I did not know how to evaluate the complete 

 integral involved in (3) when 6 is not small ; but, as usual, I 

 sent the problem to my brother, and he speedily reduced it 

 to a form equivalent to this : — 



where 



«=-^fi=^-'^-. • • • • '^) 



A = SirknfiVo A / (y)' 



I have asked him to write a short appendix to this paper. 



Returning now to consider the meaning of these equations, 

 and attempting a numerical estimate of what to expect in 

 practice, we shall find that though the instantaneous rotation 

 expressed by (2) is enormous, being quite possibly 60 or even 

 180 degrees, yet the restoration of light expressed by (4) is 

 but feeble, and only some ten thousandth part or so of what 

 could be gained by rotating the Nicol. This explains why 

 it is fairly easy to analyse the restored light into a beaded 



