262 THE MONT BLANC OBSERVATORY. 



plains, tliirteeii or fourteen doubles can be seeu, not reckoning that 

 which is known as the head of B. 



But even at Chamonix— tliat is, at an altitude of 1,050 meters — the 

 thirteenth double is very difficult to niakeout,andat the Grands-Mulets 

 (3,050 meters) it is only possible to see from the tenth to the twelfth, 

 while at the summit of Mount Blanc I could hardly go beyond the 

 eighth. 



It is not to be supposed that we establish a proportionality between 

 the numerical diminution of the doubles and that of the atmospheric 

 action. The law is evidently of a much more complex character. But 

 this diminution, especially when considered in connection with the 

 experiments made with tubes full of oxygen, and able to re-produce the 

 series of atmospheric phenomena to which we have referred, is sufficient 

 for us to conclude that the B group would totally disappear at the limits 

 of our atmosphere. It is remarkable, however, that if we take the 

 coefficient 0-5(!G that represents the dimijiution of atmospheric action 

 at the summit of Mount Blanc according to barometric pressures 



/ —^^^ = 0*506 ) and multiply it by thirteen — the number that repre- 

 sents the doubles clearly visible on the plain — we obtain 7-4 as the 

 result — that is to say, very nearly the number (8) doubles that can be 

 seen be me on the summit of Mount Blanc. 



This result is certainly remarkable, but I repeat that, in my opinion, 

 it is only by the comparison with tubes reproducing the same optical 

 conditions as nearly as possible that any definite conclusions will be 

 obtained. These comijarative experiments have already been com- 

 menced in the laboratory of Meudon Observatory, and they lead to the 

 same result, viz, the disappearance of the groups A, B, and a at the 

 limits of the atmosphere. On account of the importance of the ques- 

 tion, however, the experiments will be repeated and completed. 



The question arises as to whether the high temperatures to which 

 solar gases and vapors are subjected are not capable of modifying the 

 power of selective absorption, and particularly whether the absorption 

 of oxygen w^hich takes place in the sun's atmosphere would not be alto- 

 gether different from that indicated by the experiments which have 

 been made at ordinary temperatures. 



I have already instituted experiments with the idea of replying to 

 this objection. I shall give an account of them to the academy in due 

 course, but I may say that the absorption spectrum of oxygen, either 

 the line spectrum or the unresolvable bands, do not api)ear to be modi- 

 fied in an appreciable manner when the oxygen is raised to tempera- 

 tures of about 400 or 500 degrees. 



On the whole, I think that observations made on the summit of 

 Mount Blanc give a new ami much sounder foundation to the study of 

 the question of the purely telluric origin of the oxygen groups in the 

 solar spectrum, and lead to the conclusions previously stated. 



