﻿Spectrum of J\Jonatomir. Iodine Vapour. 



655 



The luminescence of the tube was excited either by con- 

 necting it directly with (he secondary of a Uuhmkorff 

 induction coil (15 cm. spark), or by using in connexion with 

 the induction coil a high-frequency transformer of the Oudin 

 type, and joining the tube to the circuit as shown schematically 

 in fig. 3 ; in the latter case a Wehnelt interruptor was used. 



Fig. 3. 



-• •- 



^tarth 



The temperature of the electric oven was measured with a 

 thermo-couple Pt— Pt/Rh. By means of two quartz lenses 

 (L in fig. 1) the light was concentrated on the slides of a 

 quartz spectrograph. 



At low temperatures the tube has shown a band spectrum. 

 H. Konen * finds that the band spectrum is almost continuous 

 in the ultra-violet part, having only few diffused maxima at 

 \ = 3300, 4270, 4520, 4760 A. On the contrary, I found 

 that this band spectrum has a very regular structure with 

 many well-defined maxima. There seems to be no discrepancy 

 between our results : this author has taken photographs 

 with long exposures, and in these circumstances, as I find, 

 all details disappear; on the contrary, my exposures never 

 lasted longer than a few minutes. For a more detailed 

 study of the band spectrum, I used a special quartz tube 

 with external electrodes, having a capillary part in the 

 middle. This tube being connected with the secondary of a 

 Uuhmkorff coil, I found distinct maxima at A, = 422, 404, 

 395, 387, 380, 373, 368, 363, 340, 325, 323, 320, 319, 312, 

 310, 306, 304, 302, 299 //-/*; for shorter wave-lengths the 

 maxima were too feeble to be measured. 



Raising the temperature of the oven, I found that the 

 transformation of the band spectrum into a line spectrum 

 occurred gradually. At 550° C. we have yet a band spectrum ; 



* Loc. cit. 



