ASTRONOMY ON MONT BLANC. 183 



Kiev, found 3.5, and Knut Angstnhn, who observed on Teneriffe, 

 proposed to adopt 4 calories.'^ 



These discordances result in part from a diversity of modes of 

 observation, but partly also by the variability of the condition of 

 the atmosphere, owing to the presence of water vapor, of dust, and of 

 snoAv, carried by the wind. Such impurities, often invisible, are 

 chiefly found in the lower layers of the atmosphere, and as they 

 cause a powerful absorption, which diminishes the energy of the 

 rays, it is sought to avoid them b}' observing at high altitudes and in 

 calm, cold weather. 



In 1896 Messrs. Crova and Houdaille made experiments ac 

 Chamonix (altitude 1,050 meters) and at Grands Mulets (3,020 

 meters). These observations Avere repeated the following year by 

 M. Hansky, a young Russian attached to the staff of the observa- 

 tory at Meudon. He observed successively at Brevent, Grand 

 Mulets, and the summit of Mont Blanc, employing the apparatus of 

 M. Crova, and imder the hitter's advice. The discussion of the 

 observations led him to the value of 3.4 calories. He repeated the 

 work in 1898, 1900, and 1904, and has reached the result 3.3 as the 

 most probable value of the solar constant. He believes that it is 

 certainly between 3.0 and 3.5 and in any case above the value 2.5 

 calories," which Langley has lately obtained with his bolometer. It 

 is clear that the physicists are not in accord as to the true value of 



" It seems fair to Angstrom and to Langley to state the following additional 

 facts. Angstrom proposed the value 4 calories not as a result of his observa- 

 tions of 1895 and 1896 on Teneriffe. but as a result of inferences which he made 

 in 1890 regarding the influence of the carbonic acid gas of the atmosphere in 

 dinihiishing the solar radiation. The progress of investigation convinced him 

 (if the error of these inferences, and he publicly and unreservedly withdrew the 

 proposed value 4 calories in a footnote to an article published in 1900 (Annalen 

 der Pliysik, vol. 3, p. 721, 1900). 



Langley's value, 3 calories, depends on an inference which he made in 1881 

 of a failure of Bouguer's formula for allowing for atmospheric absorption, even 

 when applied to homogeneous rays. Nevertheless he made no correction in his 

 later papers for this, like that by which, in 1881. he raised his value of the solar 

 constant from 2.2 to 3 calories, nor is any justified. The value of 2.5 calories is 

 taken from an article published by him in 1903, and this single value was dis- 

 tinctly stated to be only provisional in character. In 1904 he published (Astro- 

 ])hysical Journal, Vol. XIX, p. 305. 1904) a list of twenty-five values, all lying 

 between 1.93 and 2.28 calories, which constitutes his last word on the subject, 

 and this list includes the result which had been stated at 2.5 calories, but which 

 on a more careful reduction came at 2.2 calories. The substantial correctness 

 of these values is confirmed by simultaneous observations in 1905 and 1906 at 

 Washington and Mount Wilson. From the apparent accuracy of the work, and 

 the comparison with temperatures upon the earth's surface he was led to con- 

 sider that there might have occurred a real variation of the sun of nearly 15 

 per cent, as indicated by the divergence of the values. — (Translator.) 



