WOOD. — SPECTRA OF SODIUM VAPOR. 243 



made with the prism spectrograph. As much of the detail is lost 

 during the process of reproduction, I have prepared a very careful 

 drawing of one of the groups of bands at wave-length 5200 (Plate 4, 

 Figure 5). The drawing was made from a print with the aid of a 

 hand magnifier, and the peculiarities shown are found throughout the 

 entire spectrum. 



The bright lines are sharp, and quite as narrow as the iron lines of 

 the comparison spectrum. We must remember, however, that the 

 slit was not very narrow, 0.2 mm. perhaps, and it is quite possible 

 that a further contraction would not decrease the width of the 

 fluorescent lines. Each bright line is in general accompanied by 

 two lateral wings, which terminate quite sharply, the narrow spaces 

 between every two adjacent wings appearing as narrow dark lines. 

 These wings do not in general appear with strictly monochromatic 

 stimulation. In the work of last year, when studying the remarkable 

 changes which occur in the spectrum stimulated with the fairly 

 homogeneous light furnished by the monochromatic illuminator with 

 very narrow slits (see previous paper), I observed that as the wave- 

 length of the light was very gradually altered, the fluorescent lines 

 appeared with wings first on one side and then on the other, the 

 change in the appearance of the line reminding one of a flag flying 

 first on one side of the mast and then on the other. With the strictly 

 monochromatic illumination obtained with the isolated metallic arc 

 lines, the fluorescent lines are usually devoid of wings, though in 

 some instances the wings are found, and sometimes the wings appear 

 without the lines. These circumstances appear to indicate that the 

 wings are due to the stimulation of the electron by frequencies slightly 

 greater and slightly less than its own natural frequency. 



I have not yet had time to repeat last year's experiments with 

 the monochromatic illuminator, and plan to make a further study 

 of the changes which accompany very gradual changes in the wave- 

 length of the exciting light. The observations are very difficult and 

 uncertain, as the light furnished by the monochromatic illuminator is 

 not very bright when its slits are made as narrow as possible and the 

 fluorescence spectrum can only be observed by carefully rested eyes. 

 Probably by using a large prism spectroscope as an illuminator better 

 results can be obtained. What is most desired is a light siren ! 



Analysis of the Complex Spectrum. Stimulation with 

 Monochromatic Light. 



It was found impracticable to use the monochromatic illuminator 

 for the study of the simple spectra which made up the complex 



