244 OBSERVATIONS AT THE SUMMIT OF MONT BLA.NC. 



screen which woukl diminish in an enormous proportion the po\^ er ot 

 the ah'eady decliuing suu. 



At the surface of the earth one would (piickly enough perceive the 

 terrible phenomenon (supposing, at the low temperature the iiir would 

 have, there were any living beings to i)erceive it, which is im]»robabIe). 

 Temperature would everywhere fall, the climate of the iJoles would 

 advance toward the equator (already far more frigid than the poles 

 are now), and the conditions of vegetation and of animation would be 

 still more remote from fulhllment. 



It is true that by reason of the enormous mass of our central orb 

 and of the conditions which have i)resided over its constitution, we 

 still have at all events a long future before us before we shall witness 

 any such catastrophe. But let us not selfishly forget that the earth is 

 not alone over against the suu, that the happy family of the planets 

 counts sister planets whose mass is incomparably greater than ours, 

 the evolution of the phenomena of animation there being correspond- 

 ingly much slower, and that those planets need long futures to assure 

 their healthy and normal development. Sucli is the case with Jupiter 

 and Saturn, Avhose masses are so vast, and which appear to have passed 

 through as yet but a small fraction of their evokition. 



We have thus an interest, not for the moment, but in the future, for 

 our planetary world, in the atmosphere of the sun being so constituted 

 as to have a long future before it; that is to say, in its being free from 

 oxygen. 



But how shall we obtain intelligence as to the state of our interests 

 in this matter ? 



If we could transport ourselves without inconvenience to the limit 

 of our atmosphere, where it is on the border of the celestial void, the 

 solution of the question would be simple enough. We should receive 

 in a spectroscope a solar ray, and, by our knowledge of all the special 

 and characteristic modifications which oxygen gas produces in the 

 si^ectrum, should decide the question in an instant. 



But the problem is not so simple and easy. As things are, the 

 atmospheric air consists one-fifth of oxygen, to which we mainly owe 

 the presence of animal life on our globe, and when Ave wish to analyze 

 a ray, it has already traversed an almost infinite thickness of oxygen. 

 How, then, shall we distinguish between the spectroscopic effect on the 

 sun's atmosphere and that of the earth ? 



Here it is that the high stations become useful. 



Imagine that, with our eyes fixed upon the solar spectrum, we could 

 be buoyed up higher and higher in the air. We should see the oxygen 

 lines pale as we approached the limits of our atmosphere ; and since 

 these rays have not all the same intensity, but become weaker the less 

 their refraugibility, we should see the feebler lines disappear one after 

 another. If, then, it were made out that the decrement in the intensity 

 and number of lines corresponded strictly to the quantity of oxygen 



