334 Adeney — Photographs of Spark Spectra from the Rowland Spectrometer. 



ships of the component lines of spectra, but also to the modes of vibration of the 

 matter, which is the cause of such radiations, the evidence of such molecular 

 movements being the phenomena exhibited when the lines are under powerful 

 magnetic stress, it is clearly of importance that published reproductions of photo- 

 graphed spectra should show any characteristics of the component rays as well 

 as the constitution of the spectra generally. In confirmation of this statement it 

 may be remarked that the late Dr. Preston, f.r.u.i., f.r.s., found the photographed 

 spectra, which form the subject of this communication, and others belonging to 

 the same collection, of great assistance in his important work upon the influence 

 of a strong magnetic field upon radiating matter. It is desirable, therefore, that 

 prism spectra should be compared with grating spectra, obtained in the same 

 manner from the same elements, and attention may be directed to the repro- 

 ductions of prism spectra of iron, nickel, cobalt, and gold.* Platinum -was first 

 published in the Journal of the Chemical Society in 1882.t 



The production of powerful sparks by means of dynamos, producing alternating 

 currents, has been one of the causes of the variation in the character of spark 

 spectra, as for instance in the work of Liveing and Dewar, Trowbridge and 

 Sabine, Kayser and Runge, Eder and Valenta, and by the use of special forms of 

 induction coils by Eugene Demar^ay de Gramont, also Exner and Haschek. The 

 last named workers used a current transformer combined witli a coil, such as is 

 suitable for Tesla's experiments. 



The Occurrence of Air-lines in Spark Spectra. 



The metliod of photographing the spectra of metals without the usually 

 accompanying spectrum of air, which has been particularly described by 

 Hemsalach,J also gets rid of the short lines and dots observed in many spectra, 

 for instance in that of zinc. The process of self-induction, which is very simple, 

 apparently lowers the temperature of the s^Dark, and the spectra then more nearly 

 approach the character of those obtained by the arc. 



As these photographs were taken before the method of Hemsalach was published, 

 it was impossible to employ it ; it is advisable, however, to state that the air lines 

 serve to definitely fix the positions of neighbouring lines of other elements, and in 

 certain parts of the spectrum they prove to be a great convenience. 



Numerous observations have also shown that, no matter how dense the spectrum 



*Sci. Trans. Eoy. Dublin Soc, vol. i., scr. 2, p. 231, 1882. 



f Journ. Chem. Soc, vol. xli., p. 84. 



X Comptcs Rendus, vol. cxxix., p. 285, 1899. 



