Morrow — Spectra of Ilanganese, Lead, Copper, and lAthium. 273 



together with liydrogen lines and water-vapour bands, but none of nitrogen. 

 (Photograph II.) 



Tlie tubes were then opened at both ends by disconnecting the electrodes, 

 and each quartz tube was boiled with dilute hydrochloric acid, washed witli 

 distilled water, and dried by means of the Bunsen pump as described. A little 

 strong hydrochloric acid was then introduced into the lower limb of each 

 tube by means of a glass rod as before, the tubes were closed, exliausted, the 

 current passed and another set of photographs taken. Most lines had now 

 disappeared, but the ultimate lines can be seen on the original plate. 



The electrical evaporation of the electrodes was seen by the deposition of 

 tlie platinum as a mirror on the quartz tubes in the region of the electrodes ; 

 but it in no way affects the spectra obtained, as there are no lines showing 

 whicli are due to platinum. 



The panchromatic plates of Wratten and Wainwright were used for taking 

 the photographs, and were subsequently developed with amidol. 



The linear measurements of the lines were obtained by means of a micro- 

 meter reading to rsrrs of a^i inch, but being certainly accurate to ^woi: of an 

 inch. The cadmium line X4800*l was taken as the starting-point for the 

 measurements in each case; and the cadmium lines were measured at the same 

 time as the spectra of the chlorides. The cadmium lines as numbered by 

 Maseart were taken as the fiducial lines and their linear measurements 

 corrected from a curve previously obtained for the particular spectrograph 

 used. The corrections for the other linear measurements were obtained, 

 and the corresponding wave-lengths observed by interpolation between these 

 known cadmium lines. The wave-lengths obtained, and derived in this 

 manner, were compared with those of Eder and Yalenta , and the lines were 

 thus identified. In some cases there were no lines to correspond in their 

 wave-lengths, but these lines were recorded by Exner and Hasehek, and 

 in tliese cases I adopted their wave-lengths. 



The original photographic plate of Plate XXIV., No. II, was examined, 

 and the lines appearing on it were noted, and are marked ' U2 ' in the last 

 column, provided that they are not the ultimate lines of the metal which are 

 marked ' Ui' and appear on No. III. 



The comparative intensities of the lines of No. i are indicated 

 by ' S,' ' m,' ' f,' and ' ff,' signifying ' strong,' ' medium,' ' faint,' and ' very 

 faint.' In the last column tliese intensities are given more accurately by 

 numbers ranging from 10 to 1, the strongest line or lines being marked 10, 

 and those exceedingly faint 1. The intensities of the spark-spectra as observed 

 by the authors of the accepted wave-lengths are given in the next column 

 for comparison. 



