154 S. P. Langley on the Selective Absorption of Solar Energy. 



sions as to the normal spectrum were unverifiable by direct 

 experiment by any means he possessed. He found it impos- 

 sible, with the most delicate thermopiles at his command, to 

 get sensible heat from the grating-spectrum without gathering 

 all that lay in the two halves of it together, and could conse- 

 quently only infer the result of complete measures. But it 

 followed that, if it ever became possible to measure amounts of 

 heat so minute, these conclusions could be verified on separate 

 rays of the diffraction spectrum. 



No one, so far as I know, has hitherto succeeded in mea- 

 suring the heat from a diffraction-grating except in the gross — 

 by thus concentrating, for instance, the whole upper half and 

 the whole lower half of its spectrum upon the pile, and so 

 reaching some results not without value, even as thus obtained, 

 but of quite other interest than those which may be expected 

 when we become able to measure with close approximation 

 / the separate energy of each wave-length; Having devoted 

 many years to the study of the solar radiant heat by means 

 of the thermopile, I was led to hope that, by my long appren- 

 ticeship to the precautions needed with this instrument and 

 the possession of the most delicate apparatus attainable, I 

 might succeed better than my predecessors. I found, how- 

 ever, that though I got results, they were too obscure to be 

 of any great value, and that science possessed no instrument 

 which could deal successfully with quantities of radiant heat 

 so minute ; for the average heat in the diffraction spectrum 

 does not, under the most favourable circumstances, reach one 

 tenth that in the prismatic one, and is usually much less even 

 than this. 



Impelled by the pressure of this actual necessity, I therefore 

 tried to invent something more sensitive than the thermopile, 

 which should be at the same time equally accurate — which 

 should, I mean, be essentially a "meter" and not a mere 

 indicator of the presence of feeble radiation, and was led by 

 nearly a year's continual experimenting to the construction 

 > of the Bolometer (/3o\?} /nerpov), an instrument the details of 

 y whose construction are described in the Proceedings of the 

 American Academy of Arts and Sciences, vol. xvi. (1881). 

 With this apparatus the experiments on the diffraction spec- 

 trum were resumed — the first entirely unquestionable evidence 

 of measurable heat, in a width so small as to be properly 

 described as linear, having been obtained on October 7, 1880. 

 Nearly the whole year 1880 passed in modifications of the 

 instrument, or in the making of those measures which gave 

 promise from the first of bringing results of value. 



It will be seen that they afford almost all the experimental 



