376 v Scientific Intelligence. 



the bulb and then continued to one terminal of the secondary of 

 a spark coil. The primary coil was fed with an alternating 

 current. Thus the discharge tube was simply a long vacuum 

 tube bent through 180° at the middle of its length. The outlet 

 tube was joined to a second discharge tube which was spherical 

 in form and which was provided with two lateral tubes each con- 

 taining one electrode. These side tubes were in the same vertical 

 line and hence at right angles to the several sections of connect- 

 ing tubing. The inlet and outlet tubes of the spherical discharge 

 bulb lay along a horizontal diameter, and each was furnished with 

 a glass stop-cock. The last outlet tube led to a Gaede mercury 

 pump. Usually the pump was regulated so as to maintain a 

 pressure of about l mm of mercury in the region of the inlet of 

 the U-tube, and this corresponded to a rate of flow of the gas of 

 about 8 meters per sec, at the same place. 



When all three cocks were open and the U-tube alone was suit- 

 ably excited, the following striking phenomena could be observed. 

 Just below the inlet tube the discharge was' yellow with a tinge 

 of orange, and this was succeeded by a bluish violet region lower 

 down in the limb. Still further down in this branch and through- 

 out the entire curved portion of the U a greenish yellow light 

 was emitted. In the second leg, and immediately below the out- 

 let tube, the gas radiated bright red. If the electric current was 

 decreased, while the flow of gas was maintained constant, the 

 regions of different hues grew appreciably longer. Increasing 

 the current shortened the colored segments. On the other hand, 

 when the electrical conditions were kept invariable, the aforesaid 

 regions increased or decreased in length according as the rate of 

 flow of gas was made larger or smaller. 



That these phenomena are due to successive stages in the dis- 

 sociation of nitrogen dioxide may be shown by the aid of the 

 spherical discharge tube. Without exciting either tube, the gas 

 is drawn slowly through the entire system for some time, and 

 then the cocks on both sides of the spherical bulb are closed. On 

 sending the electric current through this bulb the discharge first 

 assumes a reddish yellow color, which gradually passes over into 

 bluish violet. Suddenly the color changes to greenish yellow and 

 this, in turn, slowly gives place to bright red. Thus the various 

 stages of dissociation which are seen simultaneously, but spread 

 out linearly, in the U-tube are presented in succession, but in the 

 same general region, in the closed vacuum bulb. 



The question of the physical significance of the various color 

 changes has been investigated spectroscopically by J. Zenneck 

 and B. Strasser.* The first spectrum, of orange yellow color, 

 belongs either to the tetroxide or the dioxide of nitrogen, since 

 both of these vapors seem to give practically the same radiation. 

 The second spectrum is due to some " labile" oxide intermediate 

 between the dioxide and nitric oxide, perhaps nitrogen trioxide. 

 Nitric oxide gives rise to the third spectrum, while the fourth is 



*Phys. Ztschr., No. 26, p. 1201, Dec, 1911. 



