402 Transact io ns. — Ch emistry, 



this sparking across, as if every now and again the resistance to the passage 

 of the current from either side was in a measure increased, and rather than 

 the whole of it would pass though the vacuum tube a portion preferred passing 

 direct. 



My first experiment in this direction resulted in some slight damage to 

 one of the coils (the condenser being destroyed), and on my second experiment 

 of the same nature, through the accident resulting from the first, I was not 

 inclined to prolong the experiment beyond being able to satisfy myself that 

 the current still continued to pass in the manner described in the first 

 experiment, when the receiving poles were joined to their respective spheres 

 at G and H. 



It will thus be seen from the foregoing description of the experiments 

 above named, that there is some tangible ground for believing that the 

 currents in duplex telegraphy do pass one another. The subject yet, in my 

 opinion, requires further investigation before that point can be finally settled ; 

 possibly it may never be conclusively proved either one way or the other, or, 

 at any rate, not until such time as we are enabled to state what electricity 

 itself is. Looking at it from a common-sense point of view, however one 

 would naturally come to the conclusion that they must pass, or that the 

 currents, like two bodies meeting each other, exchange impulses, and thus 



signal 



to its destination. 



The foregoing method in working duplex is on the open-circuit principle; 



signal is made bv sendin 



wire with electricity from the battery for the time being), 



-w . 4fc _ _ 



It is possible, and 



accordance 



makin 



The plan is as follows :— The battery is placed at the back connection of the 



with the zin 



connections 



both cases, as shown. The 



battery, 

 1 be wh 



relay is held on the insulated stop by the current entering the relay, 



v^x. uiag 



magnetism 



magnetism on the side the 



on the side it enters, and strength 



insulated stop is on. As soon as the key is moved, 

 and contact made on the front anvil, through the current immediately leaving 



line 



the signal 



and thus records 



t> 



