212 Mr. J. N. Lockyer. On the Classification^ [Jan. 10, 



took place in an electrically illuminated vacuum tube, containing 

 hydrocarbon and sodium. 



Before referring to this, however, I may mention an early experi- 

 ment of my own in connexion with this point. 



I described this experiment in the ' Manchester Science Lectures,' 

 1877 (p. 130), but it was made some years before. 



A mixture of meteorites taken at random was placed in a tube 

 attached to another tube with arrangements for passing electric 

 sparks, and this again was connected with a Sprengel pump. After 

 exhaustion, on passing the current under conditions which are 

 generally supposed to give a spark of low temperature, the spectrum 

 was seen to be that which Huggins, Donati, and others had observed 

 in the spectrum of the head of a comet. The gases occluded in 

 meteorites were thus shown to be exactly what we get in the head of 

 a comet. 



A Leyden jar was then included in the circuit, and the spectrum 

 of carbon was seen to have been replaced by that of hydrogen, from 

 the decomposition of hydrocarbons. Under low temperature condi- 

 tions, then, the spectrum was that of carbon, while under high tem- 

 perature conditions the spectrum was that of hydrogen. I also stated 

 that in my laboratory work I had come across other curious cases in 

 which compound vapours when dissociated only gave us one spectrum 

 at a time, meaning that in a vapour consisting of two well-known 

 substances, under one condition we only get the spectrum of one 

 substance, and under another condition we get the spectrum of the 

 other substance alone, so in others again of both combined. 



I had noticed this change very particularly during the researches 

 of Professor Frankland and myself, in 1869, on the spectrum of 

 hydrogen. In this case the two substances to be considered were 

 hydrogen and the mercury vapour from the mercurial air-pump which 

 was employed in the experiments. 



In the subliming experiments I also found that a carbonaceous 

 meteorite in vacuo gives off hydrocarbon vapour at the ordinary tem- 

 perature, as a weak electric discharge gives us the longest line in the 

 band spectrum of carbon without heating. On heating, the other 

 lines come in till the well-known bands are formed with more or less 

 completeness. If the discharge be a little less weak, the hydrogen F 

 line also appears, and sometimes C, and the F is brighter than the 

 carbon line. A non-carbonaceous meteorite, like the carbonaceous 

 one, also gives traces of continuous spectrum in the orange, yellow, 

 and green, with a weaker electric discharge. 



After describing the changes which took place in Comet Wells, 

 which I have already referred to, Hasselberg writes : 



" The above observations form an interesting addition to our know- 

 ledge of the physical peculiarities of the comet, arid give a new and 



