POLARIZED SPECTRA OF DOUBLE CHLORIDES. 



115 



and becomes clearly double at 2,128.1 and 2,199.3. The d y and e a series 

 of the same salt are doubled and the ba series of +20, which was 

 assumed from the relations of the spectrum to be an unresolved doub- 

 let, is separated into a bw and an aw series at 185. Other examples 

 of doubling may be noted in the case of C g , c g , a g , and B w of uranyl 

 ammonium chloride, a a , a w , and bw of the rubidium salt, and bw, D g , and 

 d a of the caesium salt. 



laloo 



isloo 



20|00 



21 00 



URANYL POTASSIUM CHLORIDE -195 



C C C 



GREEN 



E A B 



-U-L 



E A B 



I I I 



dea c d e a c 



WHITE 



C c 



B i _ . B j e ab c e ab c 



EA. E * I !!!! !!!! 



I I I I I 



URANYL AMMONIUM CHLORIDE 



C C C 



D J D 



GREEN Cl I EA Cl , E A c , 

 1 1 I I I 1 1 III i 



d e a c dea 

 C III ! ! I I 





ill i: 'ii n 



* e' b c e b 



III ! 



URANYL RUBIDIUM CHLORIDE 



c c c 



GREEN B 



A B 



I I 



A B 



I . 



d. ab c d ab 



WHITE 



URANYL CAESIUM CHLORIDE 

 C C 



e b 



e b 



GREEN 



DO EA 



il I I 



dO E A 

 Tl I I 



c d e ab c d e ab 



: ! 1 I j I 



J ' _\ } ,- 1 i. 



WHITE 



B A 



6 A 



J l_ 



e ba, c e ba. 



IflOO 



_L 



19100 



_L 



2000 



2IJOO 



FIG. 74. Polarized bands of fluorescence and absorption; four contiguous groups showing the 

 relation between the green and white components at 185. Dotted lines show computed 

 positions of absorption bands; solid lines above the base indicate fluorescence, below the 

 base absorption. 



ON THE FREQUENCY INTERVALS OF FLUORESCENCE AND ABSORPTION. 



During the preliminary study of the fluorescence and absorption of 

 uranyl ammonium chloride described in Chapter V, the symmetry of 

 the spectrum was such as to lead to the suspicion that the various 

 homologous series would be found to have the same constant-frequency 

 interval. The final tabulation of results, however, after many redeter- 

 minations of what seemed to be discordant values, showed that while 

 the departures from uniformity were in general scarcely larger than the 

 errors of observation, they were to some extent systematic and indi- 



