238 Royal Society: — 



is strikingly similar to that of iodine at a dull red heat. We have 

 first a bauk of continuous absorption in the blue with a sharp 

 boundary on the less refrangible side, and then a channelled-space 

 absorption throughout the entire green part of the spectrum reach- 

 ing to D. 



Lead. — This metal at first caused an absorption at both ends of 

 the spectrum ; shortly afterwards the whole spectrum was extin- 

 guished. As this is a readily oxidizable metal, special care was 

 taken to prove that a large excess of hydrogen was present. 



Thallium. — We are indebted to Mr. Crookes for a generous 

 supply of this metal. The characteristic green line of thallium was 

 observed bright, the light of the arc not being reversed ; and it may 

 be interesting to note that the vapour of this metal was incan- 

 descent five minutes after the withdrawal of the flame. 



Gold. — A distinct absorption in the blue and red was observed, 

 but there were certainly no traces of a channelled-space spectrum. 

 The spectral lines due to the oxyhydrogen flame were very con- 

 spicuous. It may be noted that the amount of gold volatilized was 

 only O01 oz. ; but this quantity of metal was sufficient to produce 

 an abundant supply of vapour. 



Palladium. — This metal caused a distinct absorption in the blue, 

 but no effect was noticed at the red end of the spectrum. There 

 was no channelled-space spectrum, and the lines caused by the 

 oxyhydrogen flame were barely visible. 



Selenium. — With the greatest thickness employed a channelled- 

 space spectrum was given by selenium. 



Iodine. — It will be remembered that, according to the results 

 already published by one of us, iodine vapour exhibits, at a low tem- 

 perature, a channelled-space spectrum, and a bank of absorption 

 in the violet. These later experiments showed that, at the more 

 elevated temperature, this bank was broken up and disappeared, 

 leaving a continuous channelled-space spectrum. 



These experiments, made at the Royal Mint, were often prolonged 

 for many hours consecutively. They involved much furnace-work 

 of a peculiarly trying nature ; and we have much pleasure in 

 acknowledging the assistance we received from Mr. Edward Rigg, 

 one of the assistant assayers, who conducted many of the tedious 

 manipulations with great skill and patience. We should also 

 mention that the care exercised by Joseph Groves, senior fireman, 

 in the preparation of the furnace and the lime-moulds, contributed 

 in no small measure to the success of the experiments. 



It appears to us that these experiments, conducted at the high 

 temperature of the oxyhydrogen flame, go far to support the con- 

 clusions which were drawn from the experiments at a lower tempe- 

 rature. First, in passing from the liquid to the most perfect gaseous 

 state, vapours are composed of molecules of different orders of 

 complexity; and second, this complexity is diminished by the 

 dissociating action of heat, each molecular simplification being 

 marked by a distinctive spectrum. There is also an intimate con- 

 nexion between the facility with which the final stage is reached, 



