1903.] The Ultra-violet Spectrum of Radium. 299 



inside the bulb tube. A hole, CC, is made at one side of the bulb to 

 form a window, and a piece of platinum tube is put round the lower 

 platinum wire at D, so that its upper end is about 3 mm. above the 

 end of the wire it surrounds, and a millimetre below the centre of the 

 window, CC. The upper wire should be about 1 mm. from the top of 

 the platinum tube surrounding the lower wire. Any solution put in 

 the lower part of the bulb is sucked up by capillarity to the top of the 

 platinum tube, and an induction spark between the two wires gives a 

 spectrum of any metal in solution. When exposures of more than 

 ten minutes or so are required, it is necessary to introduce fresh 

 solution to keep the level in the bulb constant. This is easily done 

 during the exposure by means of a small pipette. The splash of the 

 spark throws drops of liquid for some distance, and the heat and 

 decomposition cause nitric acid to come off. These are objectionable 

 when falling on delicate apparatus, and they cause great waste of 

 solution, which with rare bodies is to be avoided to prevent this a 

 branch tube is sealed in near the top of the spark tube, dipping to the 

 bottom of a flask through a cork. Another tube passes just through the 

 cork and is connected with an aspirator. During the sparking opera- 

 tion a rapid current of air is drawn in through the window, and all 

 splashes and vapours are washed through the water in the wash-bottle. 

 Thus the acid fumes are kept from injuring the apparatus, and the 

 valuable salt is saved in the wash water. 



The presence of platinum lines is due to the platinum poles between 

 which the spark is taken. They are easily identified by photographing 

 a platinum spectrum and a platinum-radium spectrum on the same 

 plate, slightly overlapping one another. The platinum being common 

 to both, gives continuous lines across the two spectra, while the 

 radium lines only appear in one spectrum. Radium being purified 

 from barium by a tedious process of fractional crystallisation of the 

 bromides is almost certain to show barium lines in its spectrum. I 

 have only been able to use about a grain of perfectly pure radium 

 bromide, all other samples containing traces of barium. 



Owing to the length of the spectrum and the necessity of having 

 the lines near the position of minimum deviation to get the greatest 

 sharpness, each photograph is limited to a small extent of spectrum, 

 and eight exposures are needed to take in the whole ultra-violet 

 spectrum, and as far into the visible part as the plates are sensitive to. 



The following table gives the wave-lengths of all the lines ascribed 

 to radium by different observers. Many of the early observations are 

 necessarily imperfect, owing to the enormous difficulty of preparing 

 radium compounds of sufficient strength to show a photographed 

 spectrum. And- when sufficient concentration was obtainable, the 

 observations were necessarily limited, owing to the minute amount 

 available rendering verification difficult. 



