[cale] destruction OF FLUORESCENCE 259 



As the rays causing fluorescence were in the invisible region a 

 nickel glass filter H was used which transmitted from X = 4078 A 

 to X = 3342 A and a little in the extreme red. The visible light 

 scattered by distilled water could be detected with a rested eye and 

 was calculated to be only 1/1000 of the fluorescent intensity of the 

 standard aesculin solution. 



Any slight variation in the intensity of the arc afïected both 

 solutions in the same way as the intensity of fluorescence varies 

 directly with the intensity of the exciting light. This has been verified 

 recently for extremely great intensities.' 



III. Experiments 



Preliminary experiments were made to test the sensitivity of the 

 instrument. The standard solution was taken as 4/100,000- and the 

 intensity of fluorescence of solutions as dilute as 1/10,000,000 could 

 be detected and measured. 



The spectrum of the blue fluorescent light emitted by the aesculin 

 solution is shown in Plate lA. It is seen to be a broad band extending 

 from about X = 5461 A to X = 4078 A. As quartz was used for all the 

 optical apparatus ultraviolet fluorescence, if present, would have 

 appeared. The mercury light used to excite fluorescence was reflected 

 from the walls of the vessel giving the line spectrum shown. 



On long expousre to the ultraviolet light the solution was trans- 

 formed from a colourless liquid exhibiting a bright blue fluorescence 

 to a pale amber-coloured non-fluorescent liquid. 



The fluorescing solution had an absorption band in the extreme 

 ultraviolet and another with a maximum about X = 3400 A (Plate IB). 

 The latter absorption indicates the removal of the light causing 

 fluorescence from the incident beam and is not present in the ab- 

 sorption spectrum of the decayed solution. The wave-length of the 

 exciting light is seen to be shorter than the wave-length of the fluor- 

 escent light, thus verifying Stokes' Law in the case of aesculin. No 

 intermediate stage with an absorption band differing slightly from 

 that of the fluorescing solution as found by Wood for some fluorescent 

 substances was detected in this case. 



A solution was exposed to ultraviolet light for about 2 hours and 

 measurements of the fluorescent intensity made every 15 minutes. 

 After removing to a dark room for several hours the excitation was 

 resumed. A characteristic set of readings is given in Table 2 (I and 

 II) and Fig. 4. From the graph it is seen that the fluorescent inten- 

 sity was less after the period of rest than at the cessation of exposure. 



8R. W. Wood, Phil. Mag., pp. 757-765, April, 1922. 



