476 



Profs. G-. D. Liveing and J. Dewar. [Apr. 3, 



A lining of lead gave the lines \4058, 3683*3 and 3639*3 strongly, 

 and these lines were frequently developed, though less strongly, when 

 there was no lead lining ; the metal being without doubt derived from 

 the leaden washers used to make the ends of the tube air-tight. 



A strip of silver gave the lines X, 3381*5 and 3278, and these lines 

 were sometimes reversed. No trace of the channelled spectrum of 

 silver was developed even when silver oxalate was put into the tube, 

 and furnished plenty of silver dust after the first explosion. 



A magnesium wire about 2 millims. thick and two-thirds the length 

 of the tube gave the b lines very well ; that is to say \ and & 3 were 

 well developed, and & 4 was also seen, but as the iron and magnesium 

 components of & 4 are very close together, and the iron line had been 

 observed before the introduction of the magnesium, it was not possible 

 to say with certainty whether or not the magnesium line were present 

 too. ~No other magnesium line could be detected. The blue flame line 

 was carefully looked for, but could not be seen. The photographs 

 showed none of the magnesium triplets in the ultra-violet, nor any 

 trace of the strong line \ 2852, which appears in the flame of burning 

 magnesium, and is yet more conspicuous in the arc when that metal 

 is present. 



Metallic manganese, introduced into the tube in coarse powder, gave 

 the group at wave-length about 4029 with much intensity, but no other 

 manganese line with certainty. In the visible part of the spectrum 

 the channellings in the green due to the oxide were visible. 



A lining of zinc produced no zinc line, and zinc-dust gave only a 

 very doubtful photographic impression of the line \3342. A strip of 

 cadmium gave no line of that metal either in the visible, or in the 

 ultra-violet part of the spectrum. 



Tin, aluminium, bismuth, and antimony, also failed to produce a 

 line of any of those substances, and so did mercury which was spread 

 over copper foil made to line the tube. 



Thallium spread as amalgam over the copper lining gave the lines 

 X 3775-6, 3528-3 and 3517*8. 



Chromium was introduced as ammonium bichromate, which of 

 course left the oxide after the first explosion. This gave the 

 chromium lines with wave-lengths about 5208, 5205, 5204, 4289, 

 4274*5, 4253*5, very well and persistently, also the lines with wave- 

 lengths about 3605, 3592*5, 3578*5. 



Sodium salts (carbonate, chloride) developed the ultra-violet line 

 A, 3301 ; and potassium salts give the pair of lines about wave-length 

 3445 ; but no more refrangible line of either metal was depicted on 

 the photographs. Lithium carbonate gave, besides the lines in the 

 red, orange, green, and blue, the violet line, \ 4135*5 ; but no more 

 refrangible line. 



Photographs of a flame of mixed coal-gas and oxygen, in which an 



