600 PROCEEDINGS OF THE AMERICAN ACADEMY. 



In all of the photographs taken magnesium was used as comparison 

 metal, and lines in the luminescence spectrum which fall near sharp 

 magnesium lines can therefore he measured with considerable accuracy. 

 It is on these lines that the values in the column headed " Maximum 

 Error " are based. For example, the line at A 4347, a sharp strong line 

 in the thermoduininescence spectrum of each of these fluorites, lies near the 

 A 4352 line of the magnesium spark spectrum, and the error of measure- 

 ment on several plates is not greater than the value given. The line at 

 A 5893 coincides exactly with the unresolved sodium lines in the com- 

 parison spectrum. 



Plate 3 is a reproduction of a photograph of the thermo-luminescence 

 spectrum of the Westmoreland fluorite, with the comparison spectrum of 

 the magnesium spark beside it. The magnesium spectrum is much over- 

 exposed in this photograph, and the exposure for the luminescence spec- 

 trum has been perhaps somewhat too short. A longer exposure brings 

 out the diffuse background rapidly, and the lines are not made easier to 

 measure. The chlorophane spectrum has so much of the continuous 

 background as to make it difficult of reproduction, and doubtless many 

 weaker sharp lines escaped measurement from this same cause. 



Conclusions. 



It would appear proven that the sharp-line portion of the spectrum 

 from the two fluorites is the same. All of the strong lines agree in 

 wave-length and in relative intensity within the limit of accuracy of the 

 measurements. It should be remarked that " intensity " measurements 

 of lines nearly hidden in the diffuse bands are of little significance, but 

 they have been graded on a scale of 1 to 10 for the purpose of a 

 qualitative comparison. 



The diffuse bands of the two spectra are, however, entirely different, 

 and the color of the thermo-luminescence light is determined by the pre- 

 ponderance of one or the other series of diffuse bands, the sharp-line 

 portion being in either case comparatively weak. 



The change of color of the luminescence light during the heating can 

 now be explained. It is due to the persistence of one of the sources 

 of luminescence — the one causing the appearance of the sharp lines 

 or the one causing the appearance of the diffuse bands — after the 

 other source has ceased to be active. In the Westmoreland fluorite 

 the sharp lines appear to weaken and disappear before the broad bands, 

 and the same seems to be true of the chlorophane, although the question 



