878 Wenrich—A St ml;/ of the Time Interval in a 



There was no relay at hand sensitive enough to put this 

 apparatus into practical effect, ami since a working theory was 

 the main point aimed at, a galvanometer, as a detector, was 

 considered sufficient. A curve, showing the relation between 

 the current from B, and the time it took to move the galva- 

 nometer needle 7 divisions, was plotted. See fig. 7 for this 

 curve, and Tahle II for the data. This shows a good inverse 

 time element. With B, shunted off the line and a relay in the 

 place of the galvanometer, we would have a working scheme. 

 It was predicted that the time required to effect a certain 



Fig. 6. 



R/GHT £7V£ VIEW 



throw in the galvanometer could he shortened by placing a 

 reflector over the tube 6H. For this purpose the tube was 

 exhausted so as to get rid of convection currents. A nickel- 

 plated sheet of copper in cylindrical form was used as a reflector. 

 The reflector was mounted on a sliding frame OF, fig. 4. For 

 an end view see fig. 5. With the same currents from B, and 

 B 2 as were used for obtaining the curve, no difference in time 

 required to effect a certain throw could be detected as the 

 reflector was moved along the tube. This arrangement gave 

 no "time element'' adjustment and hence the reflector was 

 abandoned. 



The bridge current used was O065 amperes and the heating 

 current ranged from this to about twice the amount. The 

 heating current was furnished by a storage battery with a 

 variable resistance in series. 



In the frame OF, figure 4, in the place of the reflector there 

 was placed a wire heater of cylindrical form. See figure 5 for 

 an end view. The dots on the inner circumference represent 



