520 PROCEEDINGS OF THE AMERICAN ACADEMY. 



spectra, but appears to consider the light too feeble to permit of pho- 

 tography of the spectrum. 



The present paper contains the first part of a research on the spectra 

 produced by an arrangement similar to the Wehnelt interrupter, with 

 tables of the wave-length and approximate intensities of lines of the 

 following metals, together with comparison tables and plates of lines 

 produced in the arc and condensed spark : — 



Lithium. Carbon poiut in solution of lithium chloride. 



Sodium. Carbon point in solution of sodium chloride. 



Potassium. Carbon point in solution of potassium chloride. 



Magnesium. Wire in hydrochloric acid. 



Calcium. Platinum point in solution of calcium chloride. 



Strontium. Platinum point in solution of strontium chloride. 



Barium. Platinum point in solution of barium chloride. 



Aluminium. Wire in hydrochloric acid. 



Silver. Wire in hydrochloric acid. 



Zinc. Wire in hvdrochloric acid. 



Mercury. Platinum point in solution of mercuric nitrate. 



Tin. Wire in hydrochloric acid 



Lead. Wire in hydrochloric acid. 



It is the intention to present in this paper the more general study of 

 the phenomena, and those spectra have therefore been selected from the 

 plates at hand which are simplest and show marked points of interest. 



At first sight of the phenomenon one is reminded of the production of 

 spectra by allowing the spark to pass from a platinum point to a solution 

 of a metallic salt, and the spectra do show marked similarity to those 

 produced in this way. The researches on the mechanism of the in- 

 terrupter have shown, however, that the breaking of the circuit is caused 

 by the formation of a layer of vapor about the active point and that the 

 water of the solution is in large measure dissociated by the high tempera- 

 ture reached.* There seemed therefore the possibility that we might 

 possess in an arrangement of this kind another step in our scale of 

 spectra and possibly a temperature midway between that of flame and 

 arc, or arc and spark. 



The importance of a more thorough study of spectra produced under 

 water and in gases and vapors under pressure, and the fact that among 

 von Bolton's drawings of spectra of the Wehnelt interrupter there are 

 several which are banded, gave interest to a more exact study of the 

 phenomena. 



* Vid. Voller and Walter, 1. c. 



