261 Resonance Spectra of Inline. 



reminded of the resonance spectra of sodium previously 

 studied : — 



a Series. 

 A Differences. 



65 

 60 

 65 



67 



70 



65 

 69 







6 Series. 



A. 





A Differences. 



5325 



l 





a 5330 



i 





a 5395 



4 





a 5460 



10 Ex. 



Life. 



a 5525 



10 





5545 



2 





l 5567 



2 





a 5593 



5 



, ; . 68 



b 5635 



3 



, ' 70 



a 5660 



10 " 



b 5705 



3 



8 



'. . 69 



a 5730 





6 5774 



5 



....... 63 



a 5796 



10 ' ' 





6 5837 



1 





a 5865 



7 





5962 



2 





6017 



3 





In the earlier work it appeared as if the green 546 line 

 was the only one operative in stimulating fluorescence. I 

 have since found that if the light of the mercury arc is 

 filtered through an eosine screen, which removes the green 

 light completely, we still have a fairly bright fluorescence of 

 a red-orange colour. The spectrum has not yet been photo- 

 graphed, but visual observations with a Hilger wave-length 

 spectroscope show that it is apparently made up of the a and 

 b series. Two yellow lines, coinciding with the exciting 

 yellow lines, are very conspicuous (resonance radh.tion) ; 

 they are of equal intensity, while in the resonance spectrum 

 excited by the total radiation of the lamp the longer line is 

 very much brighter. Bordering these two yellow lines, and 

 symmetrically placed to the right and left, are several very 

 faint pairs of lines, the components of each pair apparently 

 separated by a distance equal to the distance between 

 the yellow exciting lines. It is probable that one of the 

 lines of the ik a " series excited by the green line coin- 

 cides with the longer of the two yellow Hg lines, at 

 least within the limits of resolution of the spectroscope 

 used. The "a" series can thus be regarded as stimulated 

 at two points, 516 and 5796, when we employ the total 

 radiation. The u b" series we can provisionally ascribe 

 to excitation by the shorter of the two yellow lines. By 

 employing a screen of neodymium chloride, we can remove 



