S. P. Langley — Invisible /Solar and Lunar Spectrum. 407 



a supplementary research to be given elsewhere.) The princi- 

 pal lunar heat then is found here, at a point of the spectrum 

 corresponding to the maximum radiations from melting ice, 

 but its maximum amount is probably less than 1 per cent of 

 the corresponding solar heat, which we have just found to be 

 itself so small. That we can detect the lunar heat at all under 

 these circumstances is due to the fact that we are here able to 

 employ for it very short-focused mirrors and lenses, which con- 

 dense it into a very short and relatively hot spectrum (there 

 being no fear of their diffusing extraneous heat, since none 

 worth mention exists). In the case of the sun we must employ 

 a wholly diiferent optical train, forming a far longer spectrum. 

 It will be easily understood that these means, which enable us 

 to determine the position of the solar and lunar heat maxima 

 here are not favorable to a determination of the relative 

 amounts of heat received from the sun and moon under such 

 different conditions. We can only say that these ratios are 

 themselves utterly changed from what they are in the visible 

 spectrum, where we all know that the solar light is something 

 like five hundred thousand times moonlight. It is probable 

 that the solar heat received in this part of the spectrum is less 

 than five hundred times the lunar ; but the actual ratio is only 

 very roughly determinable by our present means. 



By comparison with the " heat " spectra given in a previous 

 memoir, we may also note the fact that some of the wave- 

 lengths given from ice, are identifiable in the solar spectrum, 

 nor (in view of the now established fact that the ratios of the 

 heat, at like parts of the spectra of two unequally hot bodies, 

 are functions of the wave-length) need it surprise us, that 

 we have also found that this part of the spectrum of the sun, 

 is not incomparably hotter than the corresponding part of the 

 ice-spectrum. 



It may be asked if we can, after all, feel sure of the charac- 

 ter of such minute amounts of heat in the presence of the 

 already described reflected and diffused heat from the upper 

 spectrum, considering the possibility that something may go 

 wrong in the elaborate arrangement of the sifting train, and 

 leave us (as everything we study now is invisible) without 

 ocular warning of the fact. I may reply that we have lately 

 found an admirable check upon the efficiency of our optical 

 devices in the behavior of that familiar substance lamp-black, 

 which all physicists use either on thermometers, thermopiles, 

 or bolometers. All of us know probably that it is not abso- 

 lutely non-selective, as it used to be thought, and that it has a 

 tendency to transmit the infra-red with greater freedom than 

 the visible spectrum, but the statement I am about to make 

 may excite surprise. It is, that when a very perfectly polished 



