262 Prof. R. W, Wood on the 



tried for tlie excitation, namely, the mercury arc, the active 

 line of: which is the bright green one at 5-ib'O, as can be 

 shown by interposing various ray-filters between the lamp 

 and the bulb which contains the iodine vapour. I now use 

 very large bulbs, 15 to 20 cm. in diameter, made from round- 

 bottomed flasks. The neck is drawn down in the blast-lamp, 

 a few small crystals of iodine are introduced, and the bulb 

 highly exhausted with a Gaede pump. The neck is then 

 sealed. If the light of the arc is focussed at the centre of 

 the bulb with a condensing lens, preferably the condenser of 

 a large projecting lantern, the fluorescence is so bright that 

 it can be seen from the back of the large lecture-hall. No 

 heating of the bulb is necessary. 



When illuminated with sunlight condensed by means of a 

 large Voigtlander portrait objective (F. 2*3), I have found 

 that the intrinsic intensity of the fluorescent light in the 

 green part of the spectrum was nearly ^ of that of the 

 Welsbach light. These experiments will be described in 

 detail in a subsequent paper, the object of the present 

 communication being merely to record the discovery of the 

 resonance spectra of the vapour. 



The quartz mercury arc is mounted at such a distance from 

 the bulb that the conjugate foci are equal. It is important 

 to use a very large condenser of very short focus, and it is 

 better to have the mercury arc tube perpendicular to the line 

 joining it with the bulb (i e. not end on). In this way we 

 illuminate a broad sheet of the vapour, and the fluorescence, 

 when viewed from the side from a point lying in the plane 

 of the illuminated sheet, has a considerable intensity. Its 

 colour is distinctly reddish, and if we examine it with a small 

 spectroscope designed for faint spectra, we shall find that, 

 instead of having an enormous number of lines forming the 

 characteristic bands of the iodine spectrum, we have a 

 beautiful series of isolated lines separated by distances 

 varying from 65 to 80 Angstrom units. The wave-lengths 

 have thus far been determined only with a Hilger "wave- 

 length " spectroscope, and can be considered correct only to 

 within perhaps 5 Angstrom units. It will be necessary to 

 photograph the spectrum with high dispersion, as has been 

 done in the case of sodium vapour. The general appearance 

 of the spectrum is shown in fig. 1. The lines indicated by 

 arrows are the lines of the mercury arc, which usually appear, 

 as a result of diffused light, unless great precautions are 

 taken. The green line always appears as it is the exciting 

 line, and the waves are in part re-emitted without change of 



