KNOBLAUCH ON RADIANT HEAT. 417 



tion of 24°*25, when reflected by black paper produced a deflec- 

 tion of 34°-5. Hence the proportion was really inverse. 



When the rays of heat reflected by two surfaces could not be 

 distinguished from each other, as e. g. those reflected by white 

 and black satin, which passed through the diathermanous media 

 used in the same manner (see p. 391 to 392, 410 to 413), if this 

 depended upon both absorbing the rays of heat sent to them 

 in the same proportion, the relation of the intensities with which 

 they reflect the heat should remain unchanged under the influence 

 of radiation from any source. 



This was also shown by experiment to occur most distinctly. 

 Thus the heat of the Argand lamp, when reflected by white satin, 

 produced a deflection of 31°; after reflexion from black satin, 

 of 27°*5 ; and that of the cylinder at a dark red heat, in the 

 former case a deviation of 27°'25, in the latter of 23°'5, Hence 

 white satin reflected each kind of radiant heat better in the same 

 degree than black satin. 



Yellow leather and brown Spanish leather, which also reflected 

 heat in such a manner that it was transmitted by diathermanous 

 bodies in the same proportion (see p. 395, 411 to 413), reflected 

 both the heat of the Argand lamp and that of the heated cylinder 

 with the same intensity. Under the influence of the rays of the 

 former, a deflection of 28°'37 to 28°'5 was obtained in the case 

 of each of these surfaces, and a deviation of 18°'75 with the rays 

 of the metallic cylinder. 



To ensure still greater certainty in these experiments, in ad- 

 dition to the heat of the Argand lamp and of the metallic cylin- 

 der, I caused also that of red-hot platinum and the flame of 

 alcohol to be reflected from all those of the surfaces which have 

 been previously mentioned, which 1 had of the same size (8 

 centim. square). 



The experiment consisted simply in exposing these surfaces 

 with their normal at an inclination of 32° to that of the longitu- 

 dinal axis of the thermoscope, and their centre at a distance of 

 7 inches from the latter and 4*5 from the source of heat, seriatim 

 to the above four sources of heat, and observing the deflections 

 which the heat reflected by them produced in the thermo-mul- 

 . tiplier. The numbers obtained in this manner (each the arith- 

 metical mean of two observations)* are contained in the following 

 tables : — 



• Accurate to within 1°'5. 



VOL. v. PART XIX. 2 F 



