TEMPERATURE OF THE SURFACE OF THE MOON. 23 



tlic liuvcr lialF of ;i slit, wlienee. tlic lif;lit passed tliron,<ili a eolliiiiatiiig lens, and fell upon a lar{;e, 

 Itutlierl'iii'd liiatinji' of 17,L'!)(i lines lo (lie ineli, wliose dltlVaeted ra,vs were viewed by an observing 

 telesoopo. Tlie iiielination of tl\e fjrating was deteiinined by a graduated circle and vernier, so 

 that by use of tlii' eiistoinary forauila tlie exact wave len.utli of tlie color or line in the center of the 

 tield could be eoni])ut(>(l. On the upper ]>art of the slit was a jirisni of total rellection which brought 

 in the rays from an Argand burner arranged to slide at right angles to the axis of the colliinating 

 teIes(!ope along a graduated scale. The amount of gas supplied to the burner was controlled by a 

 meter. Accordingly, a spectrum from a tlame of standard and constant brightness was formed by 

 the same grating in juxtaposition to the lunar s]>ectruui immediately under it in the api)arent field 

 and viewed by the same eye-piece. The lam|> was now withdrawn or apitroached until some i)ar 

 licular wave-length (c. (/., the yellow about 0''. (J) was Judged to be of like strength in either si)ectru in. 

 Under these conditions if the grating was rotated so as to bring in more of the blue end of both 

 si)ectra, the moonlight spectrum grew constantly brighter I'elative to that of the gas light, so that 

 it was necessary to strengthen the latter light to re establish equality. The field was limited by a 

 ■diaphragm to a narrow strip of both si>ectra, whost^ edges were brought as closely into juxtaposi- 

 tion as possible, and numerous seiies of comparisons were taken througluint the visible si)ectrum, 

 which after the requisite correctii)ns and reductions gave the relative intensity of the lunar spectrum 

 in each i)art to that of the gas. The same ajjparatus was used for the solar conipaiisou in the same 

 way, except that the stronger sunlight was allowed to enter through a smaller ai)erture and was 

 diflused, instead of concentrated, by being allowed to fall on a convex silvered mirror. It was 

 evident that the ])ropovtion of l)lue in the sunlight was greater than in the moonlight, as the fol- 

 lowing results show. 



Obsrrrafioii.s of June 'H)tli to 22d. 

 (Corrections for altitude have not been ai)plied.) 



For wave-length rj. .474. sunlight ii, 483,000 times moonlight. 

 For wave-length .581, sunlight U.'52,140 times moonlight. 

 For wave-length .625, sunlight 30,(i()0 times moonlight. 



These comparatively rough preliminary values are uot believed to have any great (juantita- 

 tive accuracy, but they at least show clearly that there is selective absorption of light (and hence 

 of heat) throughout the visible lunar si)ectruiu, of such a kind that the rays less transmissible by 

 glass will be found (so far as our investigation extends) in greater proportion in moon heat than 

 in sun heat, irrespective of any question as to scjisible radiation from the lunar soil. It was 

 evident that the photometric method was liable to error considerable enough to make, very con- 

 siderable discrei)ancies between the work of careful observers, and the general lesults only are 

 given above, because the work nf 1883 was supplemented by a tnore careful series of observations 

 in 1884, which we now proceed to give in detail. 



GENERAL CONSIDERATIONS. 



Ziillner has shown that, owing to the irregularities of its surface, the full moon does not 

 reflect as a smooth si)liere would do, but very nearly as a flat disk of like reflecting power, and 

 filling the same angle. Such a disk, if it presented, as seen from the earth, the mean serai- 

 diameter of li>' 3.5", and if it dillused all the solar energy which fell on it,* would send to us 

 lyY^Tjo of what the sun does, which is the portion of the solar energy which we should leceive from 

 such a moon, reflecting perfectly (hot specularly, but in all directions) all the solar energy which 

 fell on it. The moon, however, is far from being a perfect reflector. The color of its surface is 

 . comparable (as we recall) to that of sandstone rock, and hence it must reflect selectively, and, as 

 far as we can see, in such a manner that the longer wave-lengths are in larger proi)ortiou in the 

 reflected than in the original solar beam, in which, roughly speaking, the Inininous energy is 

 about one half of the uou-luminous or dark beat. Since the moon then only imperfectly re- 



' Consitleralde difft-reuce may exist even in values ol)taiued from such geometrical considerations. Thus Lam- 

 bert's formula f^ives the numljcr tijJuu, which is nearly that used Ijy Lord Kosse ; and George P. Coiid (Memoirs ot 

 American Academy, vol. viii) uses the value tsoiTO) where we have taken -grkuTi- 



