Power of Lamp- and Platinum-Black. 171 



This radiant energy was estimated in two ways : by measur- 

 ing the energy received (1) when the radiant energy was 

 absorbed, and (2) when the reflected energy was absorbed. 

 A cover-glass^ 0*1 mm. thick, placed in front of the thermo- 

 pile slit, absorbs the energy radiated from the heated surface, 

 but transmits the radiation which passes through 1 cm. of 

 water. On the other hand, a plate of rock-salt 0*5 mm. 

 thick, transmits the radiant energy, but absorbs a large and 

 measurable portion of the fluorspar Reststrahlen. 



The following values were found for the heating effect : — 



Lampblack (0'205 mm. thick) 0*95 % of the incident energy. 



Platinum-black (deposited 15 mins.) . . 00 „ „ ,, 



1i r> \ v " « ) • • v'O „ „ „ 



A part of the reflected radiation escapes through the 

 opening in the hemisphere by which the incident radiation 

 enters. The slit image being near the centre of the hemi- 

 sphere in order to obtain good definition, a portion of the 

 light regularly reflected escaped in this way. This loss was 

 considerable in the case of thin deposits, as was seen by 

 mounting the black surface so inclined that the regularly 

 reflected radiation could not pass through the aperture. 

 With the thickest layers, however, inclining the surface thus 

 did not cause any appreciable increase in the reflected 

 energy * ; further than this it was not found possible to 

 estimate the correction for the aperture in the hemisphere. 



After all other corrections have been applied, the results 

 are to be multiplied by the reciprocal of the reflective power 

 of german-silver, for before the radiation reflected from the 

 black surface falls on the thermopile it undergoes a reflexion 

 at the german-silver hemisphere f . The reflective power of 

 german-silver has been determined by Paschen t, and the 

 corrections are made using the following values : — 



X 0-8 ^ . 



71 per cent 



&7 fi . , 



. . 94-5 „ 



25-5 p . . 



. . 100 



51 fi . . 



. 100 



* Consequently, the values given in the preliminary communication 

 (loc. cit.) remain unaltered for the thickest layers only. For the thinner 

 layer of lampblack, the apparent reflective powers there given are but 

 56'0 per cent, of their true values. The correction could not be deter- 

 mined for the thin lampblack layers, for they were destroyed in measur- 

 ing their thickness before this correction was found to be so considerable : 

 the correction is probably smaller than that found above for the medium 

 thickness and larger for the thinnest layer of lampblack. 



t The polish of the hemisphere employed was not all that could be 

 desired; this fact is of little account for the longer wave-lengths, but 

 the correction to be applied for A = - 8/u. is uncertain for this reason. 



% Paschen, Ann. d. Phys. iv. p. 301 (1901). 



