amf Al>so)'ptio7i Spectra of SoiUuin Vapour. 'Mu 



increased this spot increastHl in brilliancy, still reniainino- orccii, 

 })asscd through a maximum, ami then orradually faded away, 

 disapjuniring entirely when the illuminatino- licrht became 

 yellow. The vapour remained dark until the wave-length ot" 

 the light exceeded that of the D lines, when the reil 

 fluorescence gradually ileveloped. passing through a maximum 

 in the same manner and then fading away. 



The fluorescence of the vapour produced in this way was 

 much less brilliant than in the former experiment, since 

 the total radiant energy thrown into it was very much less 

 than when direct sunlight was employed, still the light 

 sent down the lateral tube was intense enough to give a 

 spectrum when the Steinheil spectroscope was directed down 

 the tube. This spectrum was very feeble however, and it 

 was only with great difficulty that the changes which took 

 place when the wave-length of the illuminating beam was 

 changed could be followed. T\'hen blue light was emploved 

 the complete green fluorescent spectrum seemed to be present, 

 with a maximum intensity at the yellow end. As the wave- 

 length of the light was increased the point of maximum 

 intensity in the fluorescent spectrum moved towards the 

 blue. The important bearing which the phenomenon has on 

 the theory of fluorescence made a more careful study of the 

 relation between the wave-lengths of the absorbed and emitted 

 light very desirable. To accomplish this the arrangement of 

 the apparatus was changed in the following manner. The 

 monochromatic illuminator was rotated throiioh a rio-ht anole 

 so that its emitting slit was horizontal instead of vertical. 

 The dense sodium vapour above the mouth of the crucible 

 was now illuminated with a thin horizontal sheet of mono- 

 chromatic light (the image of the horizontal slit of the illu- 

 minator projected in space by a lens) . This arrano-ement of 

 course limits the fluorescence to a thin horizontal layer of 

 vapour, which when viewed through the lateral tube appears 

 as a narrow line of bright green light. By viewing the sheet 

 of vapour edgewise we get a much more intense light, and 

 since its width is small we can dispense with a spectroscope 

 entirely, simply viewing it through a prism or grating. 



A small transmission o-ratino- of 14480 lines to the inch. 

 which gives a first-order spectrnm nearly as bright as a 60° 

 flint prism, was mounted in front of the lateral tube, and the 

 line of fluorescent light viewed through it, the head being- 

 covered with a black cloth to cut out all extraneous light. 

 It was found that by slightly inclining the large tube to the 

 illuminating beam it was possible to illuminate a minute 

 projection on the inside edge of the lateral tube with the 

 monochromatic light which caused the fluorescence: in other 



