of dark Heat-rays by Gases and Vapours. 25 



would only have to be slightly altered, as may be seen from 

 the research of Jacques.) The discrepancy will be still 

 further increased by the improbable assumption that absorp- 

 tion takes place only in the part of the spectrum investigated, 

 and that therefore the percentages of absorption found are 

 distributed over the whole higher portion of the spectrum 

 without undergoing increase there. 



When, therefore, Tyndall finds for any medium an absorp- 

 tion of x per cent, by use of a source of heat at 270° C, the 



CO 



same length of the same medium must absorb at leat -p? per cent, 

 of the solar radiation. 



With the question thus simplified, the discussion of the 

 meteorological results becomes very simple. The most accu- 

 rate measurements are those of Violle *, who, as is well known, 

 made experiments on the intensity of solar radiation on the 

 summit of Mont Blanc and at its foot, on the Grlacier des 

 Bossons. He obtained an absorption of 16 per cent, in the 

 intervening air. 



We can easily calculate from this what absorption one metre 

 of air such as that on the top of Mont Blanc would exert, if 

 we employ the law of absorption, as we may certainly do for 

 such an approximation as this. 



The height of the barometer on the summit was 430 mil- 

 lims., on the Glacier des Bossons, 661 ; hence the intervening 

 column of air, reduced to atmospheric pressure, would have 

 a thickness of 2428 metres. From the formula 



where A denotes the emergent heat, E the incident heat, and 

 d the thickness of the layer, in metres, x is found to be 0*00007 ; 

 and hence for one metre A = 99*9930, or there is an absorp- 

 tion of 0*0070 per cent. But now, according to Tyndall, a 

 column of air of one metre thickness should absorb 6*086 per 

 cent.f, and then Violle should have found for this length at 

 least 0*086 x £, or 0*0147 per cent. Hence it follows that the 

 absorption by air alone is sufficient to account for the absorp- 

 tion of the sun's rays in our atmosphere. 



After what has been said, Tyndall 's result, that aque- 

 ous vapour in a length of one metre absorbs 4 to 6 per 



* Compt. Bend. 1876, i. pp. 662 & 729. The heat received from the 

 sun amounted to 2*392 cal. per minute and square centimetre on the 

 summit of Mont Blanc, and 2022 at the foot of the mountain. 



t This number is calculated from Table I. p. 80. There, exceptionally, 

 it is possible to calculate the percentage, since ammonia is given as com- 

 pletely opaque, the corresponding number 1195 representing therefore the 

 total radiation. 



