472 



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



elusion. Our next experiments were made on the flash of the com- 

 bining gases inclosed in an iron tube, half an inch in diameter and 

 about 3 feet long, closed at one end with a plate of quartz, held in its 

 place by a screw-cap and made tight by leaden washers. Two narrow 

 brass tubes were brazed into the iron tube at right angles to the axis^ 

 one near each end, and one of these was connected with an air-pump; 

 the other with the reservoir of gas. Into one of these brass tubes was 

 cemented a piece of glass tube with a platinum wire fused into h% 

 whereby the electric spark was passed to fire the gas. 



The tube was placed so that its axis might be in line with the axis 

 of the collimator of a spectroscope, and the flash observed as it 

 travelled along the tube. 



It was seen at once that more lines made their appearance in the 

 iron tube than in the glass vessel, and one conspicuous line in the 

 green was identified in position with the E line of the solar spectrum. 

 Several other lines were identified with lines of iron by comparison 

 with an electric spark between iron electrodes. There could be no 

 doubt that the flash in an iron tube gave several of the spectral lines 

 of iron. We supposed that this must be due to particles of oxide 

 shaken off the iron by the explosion, and proceeded to try the effect 

 of introducing various substances in fine powder, and compounds, 

 such as oxalates, which would give fine powders by their decompo- 

 sition in the heat of the flame. Several interesting observations were 

 made in this way. When some lithium carbonate was introduced, not 

 only were the red, orange, and blue lines of lithium very brilliant, but 

 the green line hardly less so. After the lithium had once been intro- 

 duced into the tube, the lithium lines continued to make their appear- 

 ance even after the tube had been repeatedly washed. When the 

 lithium had been freshly put in, the red line was observed to be 

 much expanded, very much broader than the line given by lithium 

 in a Bunsen burner reflected into the slit for comparison. The light 

 was dazzling unless the slit was very narrow ; and it was noticed that 

 if the spark by which the gas was fired was at the distant end of the 

 tube, so that the flame travelled along the tube towards the slit, there 

 was a reversal of the red line ; a fine dark line was plainly visible in 

 the middle of the band. When the spark was at the end of the tube next 

 the slit, no reversal was, in general, seen. Later observations showed 

 that some other metallic lines might be reversed in this way, and 

 photographs taken of the reversals. These observations with the eye 

 on the reversal of the red lithium line were made with a diffraction 

 grating, and were repeated many times. They show that there are 

 gradations of temperature in the flame, and that the front of the 

 advancing wave of explosion is somewhat cooler than the following 

 part. The combination of the gases is not so instantaneous that the 

 maximum temperature is reached at once. When some magnesia was 



