Mechanics of Luminosity. 251 



If the resistances in 1, 2, 3, 4 are w^ iv 2) w s , «%, then 

 iv x : w 2 = w 3 : w 4 . But since w,, w 3 , w* 4 are known, we 

 have at once iv 2) the resistance of the platinum wire under 

 investigation. After the conclusion of each series of experi- 

 ments the constant of the galvanometer g x was determined by 

 means of a voltameter. 



The resistance of the platinum wire at the ordinary tempe- 

 rature of 15° was determined by means of an ordinary Wheat- 

 stone bridge, with one Leclanche cell to furnish the current, 

 so as to avoid heating. The resistances a, m } and g were 

 found in the same way. 



19. The first series of experiments had for its object to 

 determine how the intensity of light in the yellow increased 

 as the total radiation increased upon rise of temperature. 

 The region of the spectrum examined was close to the sodium 

 line. The platinum wire was at a distance of 45 millim. from 

 the slit. The amyl-acetate lamp was at a distance of 20 centim. 

 Then for various degrees of luminosity we obtain the resist- 



ances w t , their ratio to the resistance w n at 15° C. — , the 



Wl& 

 intensities of the current i and the quantities w t P proportional 



to the total radiation (assuming that the wire has the same 

 temperature at all points of its section), and, lastly, the bright- 

 ness J in the yellow. 



1010=10-2555 



tri =0-6314 6609 0-6923 0*7274 0*7736 

 i=l-997 2-195 2-478 2*843 3*383 



— =2*471 2*587 2*709 2*847 3-028 

 Wis 



iv t P= 2*518 3*184 4-251 5*879 8*854 



J = 0-025 0-0935 0*189 0*455 1*761 



According to the data of Sir William Siemens*, the ratio 

 w t /iv 15 = 2'471 would correspond to about 700° C, and that of 

 3028 to about 1000° C. ; with which the observation agrees 

 that in my experiments the brightness increased seventy times ; 

 whilst, according to M. Viollef, it increased for the portion 

 of the spectrum corresponding to the D-line from 0*05 to 3*6 

 between 715° C. and 1045° C, that is, about seventy-two 

 times. 



The above values show that the radiation in the visible 

 spectrum rises much more rapidly than the total radiation. 

 The former (J) increases 70 times, the latter (E, measured by 

 means of ici 2 ) 3^ times. 



* Proc. Roy. Soc. Lond. xxxv. p. 166 (1883). 

 t Comptes Rendus, xcii. pp. 866 & 1204 (188]). 



