on the intensity of spectrum lines 429 



the paper normally, it is seen that the discharge current through 

 the tube which consists of the unrectified current from the secondary 

 of an Induction Coil is split into two streams and deflected to the 

 opposite sides of the walls of the tube. The brighter stream corre- 

 sponding to the break current is arranged to be on the side facing 

 the spectroscope. Since they are travelling in opposite directions 

 both the ions forming the discharge are deflected in the same 

 direction on to the glass and with a strong discharge in an intense 

 field the magnitude of the corrosion of the sides of the glass under 

 this sand blast action of the ions is considerable. Where this is 

 predominant the real effect gets mixed up with the effects of the 

 chemical disintegration and the phenomenon may get attributed 

 entirely to the latter cause. This conclusion is inevitable from the 

 intense fields and strong currents Messrs Kent and Frye used 

 without taking adequate precautions to exclude the complications 

 of the attendant chemical disintegration of the tube. 



To examine the extent of such disintegration heavy currents 

 and strong fields were employed for a few minutes and Fig. 2 

 shows an enlarged view of a side of the capillary of the glass dis- 

 charge tube in the region opposite the pole pieces. The deep 

 channels found to have been cut in the glass show that the dis-. 

 integration of the glass is considerable. The consequent compli- 

 cations are sure to lead one to erroneous conclusions in case proper 

 precautions are not taken to exclude them and make allowances 

 for such effects. To study the phenomenon free of such compli- 

 cations, this disruptive action of the ions should be reduced to a 

 minimum by the employment of a moderate field and current 

 and even then the tube must be of a material like quartz not 

 liable to such easy decomposition, arising from any local heating. 



In my experiments I used a quartz discharge tube with 

 aluminium electrodes the diameter of whose capillary was of the 

 order of about 2 mm. The current generally used was about 

 2 milliamperes and the field was of the order of about 5000 c.g.s. 

 units. Further the tube was in permanent communication with 

 a mercury pump fitted with McLeod's gauge and drying tube of 

 phosphorous pentoxide. Thus owing to the large volume in con- 

 nection with the discharge tube the pressure of gas in the tube 

 remained constant throughout the experiment and the disturbing 

 influence of the gradual absorption or emission of gas by the 

 electrodes or walls of the tube during the course of an experiment 

 was considerably compensated. Under such circumstances the 

 results obtained with the quartz tube are identical with those 

 obtained with the glass tube and hence the phenomenon cannot 

 be attributed to the chemical disintegration of the walls of the tube. 



Another possible view of the phenomenon put forward by 

 Messrs Kent and Frye and others is that the change is due to the 



