448 Prof. \V. X. Hartley. Pluuical Character* of the [Apr. 1 



III. "On the Physical Characters of the Lines in the Spt 

 Spectra of the Elements." By W. N. HARTLEY, F.U.," 

 Professor of Chemistry, Royal College of Science, Dubl 

 Receivea .M.-m-h 18, 1891. 



The properties of the atoms are a periodic function of the 

 masses, and the physical characteristics of the spectra of the eleiuei 

 appear to be an expression of the properties of the atoms ; for th 

 is undoubtedly an intimate connexion between the rays emitted 

 the self-luminous vapours of the elements and their chemical 

 physical properties. 



If we photograph the spark spectra of thirty or forty of 

 elements and arrange the spectra in groups following the perk 

 law, the arrangement will be seen to be a perfectly natural one. Tl 

 observation applies not only to the groupings of the lines, but also 

 the physical characteristics of the individual lines. In spark sj 

 the three most striking characteristics are (1) an extension of cei 

 lines above and below that part of the spectrum bounded by 

 points of the two electrodes ; (2) the nimbus which surrounds 

 extremities of the lines, even to some extent those portions whic 

 form an extension ; and (3) the continuous spectrum which forms 

 background to the lines. 



(1.) On the Extension of the Lines. The spark discharge, as shown 

 by Perrot, is composed of two parts, of which the fiery track, or 

 central portion, is a statical discharge, and the aureole, or flame, 

 is dynamical, and capable of electrolytic action. 



From careful observation of the sparks, and photographs of 

 spectra, I have come to regard all those spectra with lines extended 

 as spectra of different discharges taken simultaneously. The principal 

 lines lying between point and point of the electrodes are spectra 

 the fiery path of the spark ; the extension of the principal lines aboi 

 and below the points of the electrode appear to be spectra of 

 aureole. The principal observation which leads to this conclusion is 

 that the electrodes are seen to glow silently and continuously abo> 

 and below the points of the upper and lower electrodes, and frequentlj 

 slight roughnesses present the appearance of brightly but steadilj 

 shining dots ; particularly is this the case with those metals whic 

 exhibit the most extended lines, as for instance, cadmium, thalliut 

 and indium. The lines in many spectra are free from this extension 

 and no glow is observed on the electrodes. A study of about tl 

 different spectra of the metals and semi-metallic substances has 

 to the following observation. 



Elements which are difficult to volatilise, and those which are lad 

 ductort of elect. ~icity t do not exhibit tpectra with extended lines; and, con- 



