146 The Ohio Journal of Science [Vol. XVIII, No. 5, 



hence the effect of these holes could be entirely neglected, par- 

 ticularly since the coupling was always very loose, viz., 11 cm. 

 On account of the tension of the strings and the thinness of the 

 plates, the plates were slightly bowed, that is convex toward 

 each other. This effect was never greater than S parts in 1100, 

 hence could be neglected. 



c 



tH (t 



-c j; „■ J 



TiecBitftr LjiJ Lacher Systtm |_j i] Oscillitor 

 Figure 1. 



11. Variable Factors. 



The following variable factors required study. First, 

 position of the bridge for each tone; second, influence of the 

 spark-gap length; third, length of the oscillator; fourth, edge-on 

 distance, y, between the plates; fifth, change of coupling, that is, 

 change of the face-on distance, .r, between plates; sixth, oscillator 

 deterioration, that is, irregularity due to deposit of soot upon 

 the metal rods of the spark-gap in oil; seventh, possible change 

 in the sensitivity of the thermocouple even though soldered. 

 While any one variable was being investigated, if another 

 variable suffered a change it constituted a source of error for 

 the time being. 



Since the tension on the three circuits, oscillator. Lecher 

 and receiver circuits, was kept always uniform, small variations 

 of the lengths of the circuits could either be neglected or taken 

 into account as needed. Again as heretofore the receiver 

 circuit was always just half of the Lecher circuit. The fifth 

 variable above, .r, has been studied, but its importance demands 

 a special paper. 



Our method of handling the first two variables was as follows : 

 With a given x and 7/, and a given oscillator length, having 

 set the spark-gap at about the position for maximum intensity 

 for a given tone, a bridge curve was taken across a given peak 

 forward and backward. Then the bridge being placed at the 

 peak of a curve, a spark-gap curve was taken. Without 

 stopping to plot the observations at the time, the correct 

 position of the bridge could be readily told to well within 0.5 mm. 

 by mere inspection of the observations. 



