WOOD. — SPECTRA OF SODIUM VAPOR. 239 



series are absent, and there are six or more series present. The fluores- 

 cence spectrum with white-light excitation is shown on Plate 1, e, which 

 is from a negative made with the twelve-foot grating. As will be seen, 

 the bright lines coincide with the bright lines of the magnetic spectrum, 

 though much broader. It will be easier to explain how the series were 

 picked out after we have commenced the study of the fluorescence. I 

 have added to Plate l spectrum i, the magnetic spectrum with the series 



Gkeen Rotation Spectrum. 



indicated. There appear to be five distinct series and a number of lines 

 which thus far have not been brought into any definite relation with 

 one another. These series we will number 1, 2, 3, 4, and 5. All the 

 lines belonging to the first series have one dot under them, those be- 

 longing to the second have two dots, etc. These series are shown 

 separated on the chart (Plate 5) at the top. Absent lines are indicated 

 thus : 0- 



The fifth series is at the top, the fourth next, and so on down, the 

 extra lines being indicated in the lower row. This arrangement is 



