EXPERIMENTS ON LIGHT-SPECTRA. 491 



too large and not hot enough for other substances 

 besides common salt. 



A few grains of common salt are slightly moistened 

 with water on a watchglass or any small piece of clean 

 glass ; the platinum ring is then filled with the moist 

 salt and gradually brought into the flame ; a trans- 

 parent bead will be formed on the ring when the salt is 

 fused, which evaporates pretty rapidly, colouring the 

 flame strongly yellow, As long as there is any salt 

 left on the ring, the flame appears much larger than it 

 was before introducing the salt. The reason is, that 

 the common hydrogen flame is so little luminous that 

 only the hottest part in. the middle of the flame can be 

 seen ; but the vapour of the salt is so luminous that 

 the outer less hot portions of the flame are also ren- 

 dered visible. 



In order to view the flame through the disulphide of 

 carbon prism, the latter should be placed upon a support 

 formed of wooden blocks, books, &c., so as to have the 

 same height as the flame, with its refracting edge in a 

 vertical position, as shown in fig. 279. It is ira- 



f 



-o 



FIG. 279 (i real size}. 



material whether the refracting edge is on the right or 

 left of the observer ; the position of the spectrum will 

 in one case be simply the reverse from what it is in the 

 other case. In the experiments described, the refracting 

 edge is supposed to be to the left of the eye ; hence 

 the red will appear at the right end and the violet at 



