372 



NA TURE 



[Feb. 1 8, 1886 



vacuum-tubes. Now it is quite possible that in one of before. But this is precisely what we cannot do in the 



these vacuum tubes as we pass the spark through it, we conditions under which we are placed, nevertheless it can 



may have various atomic structures, some of which if we be done in the atmospheres of the sun and stars. Prof. 



could carry them away to a separate place might on 1 Pierce has shown that in such atmospheres where gravity 



cooling present us with something we had never seen ] is very powerful, the heavier molecular structures will 



naturally separate themselves from the lighter, and seek 

 a lower level. Lockyer therefore imagines that in such 

 atmospheres there is the separation of molecular struc- 

 tures always going on, the heavier falling downwards 



until they reach a region of higher temperature where 

 they become dissociated or broken up, and the lighter 

 mounting upwards until they reach a region of lower tem- 

 perature, where they combine together, and hence become 



heavier. This kind of sifting process must not be con- 

 founded with the rapid motions of ascent and descent 

 of the various solar currents, sometimes carrying a body 

 of particles downwards, andthus heating it in the process 



and sometimes carrying it upwards and cooling it in the 

 process. Both of these causes raust be regarded as at 

 work together in the solar atmosphere, and they give us no 

 doubt the best explanation of a very pecuUar circumstance 



