270 Plucker on Spectrum Analysis. 



first augments in intensity, gradually disappears and that of 

 metallic zinc comes into view. At last we see only the spectrum 

 of this metal, which is essentially composed of four lines emi- 

 nently brilliant and sharply defined ; one being red, more 

 refrangible than the red ray of hydrogen, and the three others 

 occurring in the regions of green and blue. If the tube is 

 permitted to cool, we notice the phenomena in inverse order, 

 the zinc spectrum disappearing first and being replaced by that 

 of chlorine. Excepting the non-coincidence of its bright rays, 

 chloride of cadmium comports itself like chloride of zinc. 



"Mr. Miller has lately presented to the Royal Society of 

 London very remarkable photographs of the brilliant bands of 

 the spectra of all the metals, but they do not seem to be as 

 sharply defined as ours. The difference may probably be 

 explained by the greater elevation of temperature of which his 

 were produced. I conclude, from the facts previously cited, 

 that these spectra were ' en marche ' towards the continuous 

 spectrum. 



" If we can employ a sufficiently powerful induction coil, we 

 may produce the spectral effects with gases having a pressure 

 of one atmosphere or more. We might even pass a continued 

 current of gas through the capillary tube instead of closing it 

 hermetically. A glass tube open at both ends and having an inch 

 or so rendered capillary in the middle, with platina wires thrust 

 up as far as the capillary portion, becomes a veritable chemical 

 analyser. We place one of its extremities in communication with 

 the apparatus in which the gas to be examined is developed, or 

 with the neck of a retort yielding any vapour, and the fluid gas or 

 vapour becomes incandescent as it passes the capillary portion 

 of the tube, where the platina wires are connected with an induc- 

 tion coil. If we wish to operate at ordinary pressures we let 

 the gas escape freely a,t the other end of the analyser, and if we 

 desire a lesser pressure it is easily obtained. The essential 

 character of the analysis thus briefly sketched is that it not 

 only enables us to recognize particular substances that may 

 enter into the composition of a given body, but to exhibit all 

 its elements. To do this in a sure and complete manner it is 

 necessary to ascertain for each body the changes which its 

 spectrum undergoes at each successive elevation of temperature. 

 We must also take account of the greater or less facility which 

 different substances offer for the transmission of the current, 

 and likewise not forget the transport of the substance of the 

 electrodes. 



" Up to the present M. Hittorf and myself have only touched 

 the borders of the chemical question, and of other questions 

 related to it, but the sphere of application of the new mode of 

 analysis appears to us great." 



