342 



region, where there is less air than in the 

 vticuum of our air pumps ; and where (at 25,000 

 toises high) the mercury in the barometer would 

 not rise to 0*012 of a line. We have ascer- 

 tained the uniform mixture of atmospheric air 

 to 0 003 nearly, only to an elevation of 3000 

 toises ; consequently not beyond the last stra- 

 tum of fleecy clouds. It might be admitted, 

 that, in the first revolutions of the globe, gaseous 

 substances, which yet remain unknown to us, 

 may have risen toward that region, through 

 which the falling stars pass : but accurate expe- 

 riments, made upon mixtures of gasses, which 

 have not the same specific gravity, prove, that 

 we cannot admit a superior stratum of the at- 

 mosphere entirely different from the inferior 

 strata. Gaseous substances mix and penetrate 

 each other with the least motion ; and a uni- 

 formity of their mixture would have taken place 

 in the lapse of ages*, unless we suppose in 

 them the effects of a repulsive action unexam- 

 pled in those substances which we can subject 

 to our observations. Farther, if we admit the 

 existence of particular aerial fluids in the inac- 

 cessible regions of luminous meteors, falling 

 stars, bolides, and the Aurora Borealis; how 



* See my experiments on a mixture of hydrogen and oxy- 

 gen, or on an atmospheric air with base of hydrogen, in a 

 paper on refractions inserted in my Astron. Obs. Vol. i, 

 p. 117—120, 



