436 Rormell — Dielectric Constant of Paraffins. 



extending over a meter rod, thus indicating the position of the 

 bridge. 



P is a Pliicker argon tube mounted in such a way that it 

 may be moved at will along the three mutually perpendicular 

 directions. Tubes filled with other gases were tried but none 

 of these glowed with such brilliancy. Professors John Trow- 

 bridge and T. W. Richards* have called attention to the 

 unusual sensitiveness of an argon tube to oscillating discharges. 

 In order to observe the glow of the tube to the best advantage 

 it is necessary to work in a darkened room. 



Some Preliminary Experiments. 

 (a) Determination of the Line Constants. 



In this adjustment of the oscillator the primary was made 

 of copper wire 2 mm in diameter and was furnished with dis- 

 charging balls 4 mm in diameter. The diameter of the coil was 

 4 cm . The total length of the secondary, including the leading 

 out wires to the first bridge, was about 25 cm . In every case the 

 secondary is made of number 18 copper wire. The wave 

 length given out by such a combination when the primary was 

 •3 mm from the secondary was 6L-33 cm . In order to determine 

 the exact wave length one may proceed as follows : Place the 

 argon tube under the wires, in contact with them, and about 

 14 cm from the first bridge. Slide the movable bridge along 

 the wires until the first position of maximum glow is reached ; 

 this will be 27*80 cm from the first bridge. When the bridge is 

 in the proper position, the maximum glow of the tube may be 

 easily noted from the extension of the glow into the sealing 

 off tube. 



Coolidgef found it necessary to connect the first bridge 

 with the earth in order to lead off the irregular vibrations of 

 the oscillator, which otherwise masked the position of the node 

 of the fundamental wave. 



I have not found this necessary, however, and attribute it 

 largely to the accurately symmetrical arrangement of the 

 primary and secondary. 



Table I contains a set of readings which were taken for the 

 determination of the line constants. In this as in all other 

 cases ten readings were taken in locating the position of a 

 node. The average of two or three such groups for the 

 same node never varied beyond a half of one per cent. 



The readings for the nodes most distant from the first 

 bridge are not so concordant, but the percentage of error 



*This Journal, vol. iii, p. 19, 1897. 



f Anoalen der Physik, lxvii, p. 582, 1899. 



