380 Dr. J. H. Gladstone on some Points connected 



of a physical nature. (1) Of course the cooling of such a 

 nebulous mass must take place from its outer portion^ giving 

 rise to a distribution of temperature greatest at the centre and 

 gradually decreasing towards the circumference. One of the 

 results of this would be that the least volatile constituents near 

 the circumference would condense and sink towards the centre 

 of gravity, forming eventually the liquid or solid nucleus, 

 while the most volatile would still extend to the outermost 

 portions of the nebula ; and the rest would arrange themselves 

 in the order of their volatility, condensing into cloud at various 

 distances from the intensely heated centre. (2) But there is 

 another law regulating the arrangement of the gases, which has 

 been pointed out and mathematically proved in a paper com- 

 municated to the Royal Society by Mr. G. Johnstone Stoney 

 as far back as May 1867*. He concludes it ^^ to be a neces- 

 sary consequence of the molecular constitution of gases that 

 in such an atmosphere, decreasing in temperature from within 

 outwards, the various constituent gases are not everywhere 

 equally mixed, but that in the upper regions those that have 

 the lightest molecules rise the furthest, so that the gases over- 

 lap one another in the order of the masses of their molecules." 

 We may therefore expect to find the chemical constituents 

 arranging themselves according to their molecular masses so 

 long as they retain their gaseity, and also according to their 

 volatility when condensation has ensued. In either case, how- 

 ever, the separation between the constituents would be far from 

 perfect. The lower stratum of the gaseous envelope will, of 

 course, consist of all the gases diffused together ; and while 

 the lighter gases outreach the heavier ones, still the molecules 

 of a heavier gas, shooting about in all directions, will often 

 approach the outer limits of a lighter one. Every one also 

 who has performed a fractional distillation knows how difficult, 

 or rather impossible, it is to separate entirely one body from 

 another by means of the difference in their boiling-points. 

 ISTow this exactly represents what we actually find in the sun 

 itself. Stretching far beyond its luminous sphere there is an 

 enormous atmosphere of hydrogen, w^hich is by far the light- 

 est of all the gases, and at the same time the least condensable 

 of those which have been recognized in the sun. Far into 

 this atmosphere of hydrogen rise small quantities of sodium 

 and magnesium, both volatile metals, and both giving vapours 

 of exceptional lightness. Lower down we find in large quan- 

 tities the vapour of iron, certainly a less volatile metal, and 



* Proceeding's of the Royal Society of London, vol. xvi. p. 25, and 

 vol. xvii. p. 1. 



