590 PROCEEDINGS OF THE AMERICAN ACADEMY. 



since succeeding results are to be published in the Proceedings, and the 

 connection between the facts of fluorescence and those obtained from 

 other methods of excitation is a very close one. 



In Plate 1, the general appearance and something of the detail of the 

 fluorescence spectra from two crystals of fluorite may be seen. These 

 are crystals which have been numbered 4 and 5, and the cuts show the 

 spectra produced when they are excited by the light from the spark 

 between electrodes of various metals. For the tables of measured wave- 

 lengths reference should be made to the paper in the Astrophysical 

 Journal. 7 The facts exhibited by the tables and the plates may be 

 briefly summarized as follows : — 



1. The light of the fluorescence excited in a crystal of fluorite by the 

 light of the condensed spark between electrodes of certain metals shows 

 in its spectrum many sharp lines and narrow bands. 



2. A single fluorite crystal is able to emit a very great number of 

 these lines and bands, a part only of which appear under excitation by 

 the spark from one metal, another part under excitation by another 

 metal, and so on. The entire fluorescence spectrum of a fluorite crystal 

 must then be considered the sum of all the lines excited by all the 

 different sparks. 



3. Certain lines appear to be common to the fluorescence spectra 

 excited by different sparks. This statement holds true to the limit of 

 accuracy of the measurements on the photographs obtained. 



4. Certain of these lines and bands are evidently peculiar to the light 

 excited by a single metal. No error of measurement could account for 

 the differences of distribution of the strong lines in the various spectra 

 emitted by the same crystal under excitation by different sparks. 



5. The fluorescence spectrum varies from crystal to crystal, not only 

 in minor points, but even in the strongest lines. 



6. Metals having strong lines in the ultraviolet part of the spark 

 spectrum excite extended fluorescence spectra. 



7. The exciting source for the sharp lines of the photographs is light 

 between wave-lengths A 3000 and A 2000, since it is absorbed completely 

 by a thin layer of glass, but not by a considerable thickness of quartz. 



8. The strong sharp lines of the fluorescence spectra lie between 

 A 5700 and A 6400. No sharp lines whatever have been observed of 

 shorter wave-length than A 4700. 



9. The lines in these fluorescence spectra do not appear to belong to 



» Astrophys. Jour., 21, 83 (1905). 



