MORSE. — SPECTRA OF WEHNELT. 531 



BANDS IN THE WEHNELT SPECTRUM OF BARIUM. 



Wave-lengths. Wave-lengths. 



4570 5272.0 (red edge of same fluting). 



to 5325.0 head of fluting (red edge). 



4575. 5446.0 

 4598 to 



to 5462.0 



4605. 5477.0 centre of band. 



5134.0 centre of band. 5407.0 

 5167.0 centre of band. to 



5242.0 head of fluting (violet edge). 5519.0 (line. Intensity 6). 



The spectrum produced about a platinum point in a solution of barium 

 chloride is in many points similar to the spark spectrum. Many arc 

 lines, absent from, or extremely weak in the spark spectrum, make 

 their appearance. X 3910.0, A 4283.0, X 4351.0, A. 4580.0, X 5425.0, 

 X 5778.0, are examples. At the same time many strong arc lines are 

 absent in the Wehnelt spectrum, which has the appearance of a rather 

 incomplete composite of arc and spark spectrum so far as the li7ie spectra 

 are concerned. The brightness of the " oxide " bands in the green is 

 remarkable. Altogether the spectrum of the Wehnelt is much like 

 that produced by sparking from a platinum point to a solution of a 

 barium salt. (Vid. Lecoq, 1. c.) 



Aluminium. 



The spectrum of an aluminium point as active electrode in an acid or 

 an alkali or a salt solution, and that from a platinum point in a solution 

 of an aluminium salt, seem to be exactly the same as that obtained by 

 allowing a spark discharge without condenser to pass in air between 

 aluminium terminals. The author has obtained the same banded sjjectrum 

 in tubes containing oxygen at a pressure of a few millimeters of mercury, 

 and with the same distinctness in tubes containing hydrogen at low 

 pressures, by allowing the spark, condensed or uncondensed, to pass 

 between aluminium terminals. In the case of tubes filled with hydrogen 

 the bands persist for only a comparatively short time, and the spectrum 

 of hydrogen remains. The same spectrum, with the exception of some 

 lines not belonging to the bands, is obtained from aluminium in the arc. 



Hasselberg * has measured a great number of lines in these flutings, 



* Kon. Svensk. Akad., 24. 



