64 REPORT—1840. 
gant instrument*, and who has endeavoured to separate from its 
indications that part which is due to reflexion from the at- 
mosphere, finds by it, that at the summit of the Faulhorn, 
nearly 9000 feet above the sea, 30 per cent. of the vertical rays 
are already lost. With Herschel’s actinometer (calculating al- 
ways observations at different elevations by Lambert’s formula), 
he obtains only 26 per cent. of loss at the Faulhorn, or 32 per 
cent. at the level of the sea. This was in perfectly clear weather. 
73. M. Pouillet, employing likewise Lambert’s formule, and 
the modification of the actinometer already mentioned, finds, at 
different seasons of the year, an absorption varying from 21 to 
28 per cent., at a vertical incidence at Paris. 
74. Saussure seems first to have thought of comparing directly 
the intensity of solar heat at the top and bottom of a mountain, 
and he contrived a heliothermometer for that purpose; and by 
experiments on the Cramont, to the south of Mont Blanc, he 
actually proved the increased intensity of the solar rays as we 
-ascend, notwithstanding the diminution of temperature; un- 
doubtedly a very remarkable experiment for the period +. 
- 75. In 1832, Sir John Herschel kindly pointed out this pro- 
blem to my attention, and furnished me with two actinometers. I 
had the rare. good fortune to obtain the aid of Prof. Kamtz in 
making directly comparative experiments at the top and bottom 
of a column of air 6500 feet high, of known density, tempera- 
ture, and humidity, under the most unexceptionable circum- 
stances in point of weather. A provisional reduction of these 
experiments has given me 29 per cent. for the vertical loss at 
the level of the sea, a near agreement with the 32 per cent. in- 
dependently determined by the method of Bouguer and Lam- 
bert with the same instrument at the same time. 
76. Collecting these various results, we have for the absorp- 
tion of incident solar heat traversing the atmosphere vertically 
in clear weather, the following fractions (incident heat = 1) :— 
BORNE \ sh, ena) ata Goh csp he 
ata eri ee ete ae he ae HS EER 
ais OR GONE aE con nae 
MRC ee aes er ee rene 
LR LL oUt a oh ee aS ae” 
Kamtz and Forbes”, , . -: . s@9 
bE OR ie Me Mae ems 
Mean, omitting the two first . °277 
* Lehrbuch, iii. 10, &c. + Saussure, Voyages, 4to. iti, 310, and note. 
+ It appears from the remark of M., Pouillet, p. 8 of his Memoir, that he 
would make the fraction even lower for a perfectly pure sky. 
