209 



Almost exactly similar numerical results were obtained for heat 

 from an argand lamp, from incandescent platinum, and from brass 



at 700°. 



The Third and longest Section of the paper, relates to the Index 

 of Refraction for Heat of different kinds compared to Light. The 

 method of observation adopted by the author, resembles in principle 

 Dr WoUaston's method of determining Indices of Refraction for 

 Light by the Critical Angle of Total Reflection. (Phil. Trans. 

 1802.) The practical part consists in determining the critical 

 angle of transition from partial to total reflection within a rock-salt 

 prism, having two angles of 40° and one of 100°. The sentient 

 surface of the thermal pile is so placed with regard to the prism, 

 that it continually receives rays coming from the source of heat 

 after undergoing two refractions and one reflection, whatever be 

 the angle of incidence. This is accomplished by mounting the 

 whole on a jointed rhomboidal frame, the surface at which total re 

 flection takes place being maintained by a mechanical contrivance 

 constantly perpendicular to the diagonal of the rhomboid, the inci- 

 dent and reflected rays being always parallel to its side, to which 

 the source of heat and thermal pile are fixed. In this way (a number 

 of precautions being taken, and modes of reduction adopted which 

 cannot here be explained) the following indices of refraction were 

 approximately determined, for the mean quality of the heat most 

 abundantly contained in the rays obtained from various sources : — 



Heat from 

 Locatelli Lamp, 

 Ditto, transmitted by Alum, 



Glass, . 



Opaque Glass, 



Opaque Mica, 



Incandescent Platinum, 

 Ditto, transmitted by Glass, 



Opaque Mica, 



Brass about 700°, 



Ditto, transmitted by Clear Mica, 



Mercury at 450°, 



Mean Lumikous Rats, 



Index of Refraction from Rock -Salt. 

 1.571 

 1.598 

 1.587 

 1.593 

 1.583 

 1.572 

 1.588 

 1.584 

 1.568 



1.577 

 1.672 



1.902 



The following general conclusions were deduced from the expe- 

 riments of this section. (1.) That the mean quality, or that of the 

 more abundant proportion of the heat from difi^erent sources, varies 

 within narrow limits of refrangibility. (2.) That these limits are 

 very narrow indeed where the direct heat of any source is employed. 



