368 Prof. Wood and Mr. Moore on the Fluorescence 



words, the near end o£ the horizontal image of the slit was 

 brought onto this projection. The line of fluorescent light 

 was thus tipped at one end with a small point of light similar 

 to the light which produced the fluorescence. The spectrum 

 of this small illuminated spot, which was of course a narrow 

 band (practically an illuminated point), fell alongside of the 

 fluorescent spectrum, enabling a comparison to be made at 

 once. Any exception to Stokes's law would make itself 

 e^-ident as an extension of the fluorescent spectrum on the 

 more refrangible side of the position occupied by the small 

 spot of illuminating light. 



Some very remarkable and significant results were obtained 

 with this dis|K)sition of the apparattLs. Starting with A-iolet 

 illmuination there appeared through the grating only the 

 small comparison-spot of light which moved along as the 

 wave-length was increased. As soon as wave-length 4600 

 was reached the fluorescent sjiectrum appeared. Its appear- 

 ance is indicated in fig. 3. Plate XIII. (a). A strip of blue 

 fluorescent light continuous with the reference-spot showed 

 that light of the same wave-length as the absorl^ed light was 

 being emitted by the vapour. A region of darkness occurred 

 on the less refrangible side (usual sense of the term) and 

 then came a broad green band with a maximum of intensity 

 on the end towards the yellow. Then came another gap 

 extending considerably above the D lines, followed by a very 

 faint trace of the red fluorescent band (not shown). On 

 c^raduaUv increasinor the wave-length of the licrht the follow- 

 ing changes were noted : — The spot of reference light, accom- 

 panied by its fluorescent prolongation, moved up the spectrum, 

 pushing the dark region before it. while the point of maximum 

 fluorescence in the wide green band moved down the spectrtmi 

 to meet the advancing spot. These changes are represented 

 in fig. 3 {b) (c) (d), Plate XIII.. the point of maximum fluores- 

 cence nearly coinciding in position with the reference-spot in 

 (d). On still further increasing the wave-length the fluores- 

 cence became very faint, and appeared to extend a Kttle 

 farther down the spectrum than the position occupied bv the 

 small spot of Hght. This means an exception to Stokes's 

 law when the wave-length of the illuminating light (green) 

 has the maximum value which still yields fluorescence. 



In fig. 4, Plate XIII.. an attempt has been made to represent 

 these changes in the conventional way (Miiiler Pouillet, 

 Lehrhuch der Physih, ii. 1, p. 368). Evervthing shown in 

 this diagram is, however, due to fluorescence, the deviated 

 continuous spectrtmi of the light exciting fluorescence having 

 l^een omitted. It will be seen that there is an emission of 

 fluorescent light of the same wave-length as that of the excit- 

 ing light, which gives the diagram an appearance not unlike 



