408 & P. Lcmgley — Invisible Solar and Lunar Spectrum. 



rock-salt plate is covered by a sheet of lamp-black of such 

 thickness as to transmit less than one per cent of ordinary 

 white light, it transmits about 90 per cent of the radiations 

 belonging to these extreme wave-lengths. In other words, it 

 has become a transparent body to rays of this wave-length, 

 while it exercises an intermediate degree of absorption on in- 

 termediate rays, so that by the amonnt of their absorption by 

 lamp-black, we have a test by which these latter may be inde- 

 pendently identified. 



We have given the preceding study not only to map new 



lines and bands in the region between 3^ and 5^, but also to 

 verify the existence of an all but infinitesimal amount of heat at 

 much greater wave-lengths. We have applied great pains to this 

 latter, not so much on account of its own importance, as on 

 account of the important conclusions to be drawn from it later ; 

 for if it is true that in this extreme spectral region, correspond- 

 ing to temperatures much below the boiling point and even 

 below that of melting ice, the amount of the solar heat is 

 trivial, it is also true that the fact of its existing at all is of 

 very significant' interest to the meteorologist ; for it is obvi- 

 ously here, at temperatures below 100° C, that the rays which 

 make up the nocturnal as well as diurnal radiations from the 

 soil of our own planet are to be found. We observe that if 

 such rays can enter the air from the sun, they can go out even 

 from an icy soil (and still more from an ordinary one), to 

 whose radiations it hence appears the atmosphere is not im- 

 permeable. Heat, then, apparently escapes in some very mi- 

 nute degree even from the surface of the Arctic Zone, not 

 only by convection but by direct radiation through the at- 

 mosphere toward space. Meteorological questions of great 

 interest, to which we shall return in a later memoir, can best 

 be answered from a study of this spectral region, and even if 

 the problem of the lunar temperature did not interest us as 

 students of celestial physics, we should still find others depend- 

 ing on it of very practical import. 



As to the degree of accuracy obtainable in fixing its position 

 in the spectrum, we may remind the reader that the old formulae 

 of Cauchy and others being useless, as we have shown, we have 

 already undertaken a research* to enable us to determine such 

 wave-lengths in this region. We have determined all the wave- 

 lengths by the interpolating curve made from the observations 

 of 188T, which has been exhibited to the Academy at its April 

 meeting. In this memoir will be found all the explanations 

 we can offer relating to the degree of trustworthiness of the 

 values assigned to the present extreme wave-lengths. If they 

 are, as we believe, to be trusted within the limits there given, 



* On hitherto unrecognized wave-lengths. This .Journal. August, 1885. 



