1879.] on the Reversal of the Lines of Metallic Vapours. 355 



mixture of charred barium tartrate with aluminium was used, a 

 fourth dark line was seen with wave-length 5,535. This line was very 

 fine and sharply defined, whereas the other three lines were ill-defined 

 at the edges ; it is, moreover, the only one of the four which corre- 

 sponds to a bright line of metallic barium. 



Repeated experiments with charred tartrates of calcium and of 

 strontium mixed with aluminium gave no results, but on one oc- 

 casion, when some sodium carbonate was used along with the charred 

 tartrate of strontium and aluminium, the blue line of strontium was 

 .seen reversed, and on another occasion, when a mixture of charred 

 potassium, calcium, and strontium tartrates, and aluminium was 

 used, the calcium line, with wave-length 4,226, was seen reversed. 

 The fire in this case was fed with gas retort carbon, and the tem- 

 perature such that iron tubes, though well coated with fire-clay, gave 

 way in a few minutes. It appears, therefore, that the blue line of 

 strontium, and the above-mentioned violet line of calcium, are rever- 

 sible by this method, but not so easily or so certainly as the lines of 

 barium or its compounds above mentioned. 



In order to obtain higher temperatures than we could obtain in the 

 furnace used in our former experiments, we have made preliminary 

 experiments with lime crucibles heated (1) by a jet of coal-gas and 

 oxygen ; (2) by the electric arc. For this purpose a block of chalk 

 or lime had a vertical tubular hole bored into it about 6 or 7 millims. in 

 diameter, and for the gas jet a second lateral boring, meeting the 

 other boring at the bottom (fig. 1). For the electric arc two lateral 



Fig. 1. 



borings are made on opposite sides of the block, meeting the vertical 

 boring at its bottom (fig. 2). Above the crucible we place a mirror 

 inclined at 45°, so as to reflect the light from the vertical boring on 

 to the slit of a spectroscope, a plate of mica being interposed 

 between the mirror and crucible to deflect the stream of hot eas and 



