SPECTROSCOPIC DETERMINATION OE LITHIUM. 21 



alcohol the lithium is extracted from solution, and the other salts 

 are precipitated out, leaving the litliium behind in solution. If the 

 residue from evaporation of a water is treated as described, there is 

 almost no chance for loss of the smallest amount of lithium, and with 

 pure reagents none will be added to the solution. 



SPECTBOSCOPIC DETERMINATION OF LITHIUM. 



REVIEW OF METHODS DESCRIBED IN THE LITERATURE. 



Talbot ^ in 1826 suggested that if it should be shown that the 

 various colored lines of the spectrum were due to the different metals 

 and were characteristic of them, ''a glance at the prismatic spectrum 

 of a flame may show it to contain substances which it would other- 

 wise require a laborious chemical analysis to detect." Later ^ (1834) 

 he decided that the minutest portions of lithium and strontium 

 could be distinguished by optical analysis with as much certainty, 

 if not more, than by any other method. In 1836^ he said: ''It is 

 much to be desired that an extensive course of experiments should be 

 made on the spectra of chemical flames with accurate measurements 

 of the relative positions of bright and dark lines or maxima and 

 minima of light which are generally seen in them." He used chiefly 

 the spirit lamp for the production of spectral lines and mentioned * 

 the use of a stream of oxygen with the flame of the spirit lamp, by 

 which means he obtained much brighter flames with the salts of 

 sochum, strontium, or barium. 



KiRCHHOFF and Bunsen ^ published their work on analysis by 

 spectrum observations in 1860. Earlier observers had used ordinary 

 luminous gas flames or alcohol flames for the production of the 

 spectra wlule they used the burner described by Bunsen ® a few 

 years previously. Although Swan ^ had described the use of the 

 coUimator in 1856 and had suggested that the D line was due to 

 sodium, the thorough study and mapping of the spectra by Kirclihoff 

 and Bunsen is generally regarded as the beginning of the systematic 

 use of the spectroscope for analysis. They studied the spectra of 

 many salts of the alkalis and alkaline earths, using the flames of 

 sulphur, carbon bisulphid, alcohol, gas with the Bunsen burner, 

 carbon monoxid, hydrogen and oxy-hydrogen, and found that the 

 lines had the same position with all flames, but were brighter as 

 the temperature of the flames was higher. 



They made some tests also of spark spectra of the metals and 

 obtained the same lines. In order to obtain some idea of the sensi- 



1 Edin. J. Sci., 182G, 5: 77. 5 pogg. Ann., 18G0, 110: 101-189. 



2 Phil. Mag., 1834, (3) 4 : 114. « Pogg. Ann., 1856, 100: 85. 



3 Phil. Mag., 1836, (3) 5:3. ' Phil. Mag., 1860, (4) W: 173. 

 * Phil. Mag., 1833, (3) 3: 35. 



