1S76.] L. Schwendler — On the General Theory of Duplex Telegraphy. 9 



If we now substitute in #, for K the value B, and in S 2 for B the 

 value K we get 



\v \/b 

 ^x = eq w 8 K 



while 



S 2 =e q ^§B 



3 ~ e 1— IT ° x 



remains the same. 



S. has an absolute maximum for b = — - — : S» for a = — - — and 

 1 3 3 



^3 for b = a + d as stated before. 



Hence we know what relations between the different variable, should 

 not exist. 



This is all we can get from the function S. For further relations we 

 must look to the function D. 



For Station I we have* 



_ e' _ZP_ A' 



which again, with respect to the variations of K', B', and A.' may be written 

 in three different forms : 



„ , e' K" X' v's/b' 

 e B K fx ^ a > 



and 



e B Jo. /x. 



D'=—. — — - — 8 X' 

 3 -«" •# ^'v^ 



as low a resistance as practice allows. But this is clearly wrong, for if It is made very 

 small as compared with K, the balance becomes unstable. This fact explains, to a certain 

 degree, the failure which has attended the application of the compensation method for 

 Duplex working, because the method was tried under the most unfavorable quantitative 

 arrangements. 



* When investigating the minimum absolute magnitude of S, the terms could be 

 taken without an accent, because S contains only terms belonging to the same station. 

 When investigating D this cannot be done as D contains also terms belonging to the 

 other station. 

 2 



