between Optical and Electrical Qualities of Metals. 169 
A similar proof for the agreement of experiment and 
theory results from Table III. Here the values of (100—R) 
determined experimentally are placed beside those computed 
from formula (4), and solely dependent on the electric con- 
ductivity and the wave-length A=12y. 
Excepting the values given for bismuth, the agreement 1s 
good, particularly when we consider that the numbers 
calculated from formula (4) are absolute values and do not 
contain any arbitrary coefficient. 
1G, 
The Emissive Power of the Metals for \ = 25'5pu. 
The formula (4), found by our experiments relating to 
reflexion and deduced from Maxwell’s theory, was now to be 
examined for rays of much greater wave-length ; for instance, 
for the “ residual rays”’ * of fluorite. For the accomplishment 
of this purpose, the investigation of (L00—R) by aid of the 
reflecting-power is not an advantageous method, as was 
mentioned above. or all metals the reflecting-power R 
approaches 100 per cent. asymptotically with increasing 
wave-lengths, and the difficulty of experimentally deter- 
mining (100—R) increases accordingly f. 
But if—instead of the reflecting-power—the emission- 
power is made the object of research, the course of investi- 
gation becomes much easier. In that case the metal surfaces 
require merely to possess the same temperature for the com- 
parison of their radiating-power with that of an absolutely 
* By “residual rays” we mean the still surviving part of the whole 
radiation of a source of heat after having undergone several reflexions on 
surfaces of a certain substance; for instance, of rock-salt, sylvine, or 
fluorite. In the infra-red these substances possess rather sharply-limited 
bands of metallic absorption, at which the reflecting-power attains very 
high values (sometimes more than 90 per cent.), whilst for other wave- 
lengths it is small. The residual rays of sylvine attain their maximum 
energy at a wave-length of about 61», those of rock-salt at 514. The 
residual rays of fluorite are less homogeneous than those of rock-salt and 
sylvine. ‘They begin at 22, attain a sharply-defined maximum at 
A=24 p, and decrease first rapidly, then slower towards the longer waves. 
On account of the unsymmetrical form of the energy-curve the mean 
wave-length of the assemblage of rays—after three reflexions from 
fluorite—amounts to 25°5 w (H. Rubens, Wied. Ann. lxix. p. 576, 1899). 
For further particulars on residual rays, vde H. Rubens & EH. F. Nichols, 
Wied. Ann. lx. p. 418 (1897); and H. Rubens & E. Aschiinass, Wied. 
Ann. |xv. p. 241 (1898). 
+ Cf. H. Rubens & E. F. Nichols, Wied. Ann. 1x. p. 418 (1897). 
