SECT. XXV POLARIZATION OF CALORIC. 215 



crown and flint glass prisms transfer that point respec- 

 tively to the red and to its limit. M. Melloni, observing 

 that the maximum point of heat is transferred farther 

 and farther toward the red end of the spectrum, ac- 

 cording as the substance of the prism is more and more 

 permeable to heat, inferred that a prism of rock-salt, 

 which possesses a greater power of transmitting the 

 calorific rays than any known body, ought to throw the 

 point of greatest heat to a considerable distance beyond 

 the visible part of the spectrum, an anticipation which 

 experiment fully confirmed, by placing it as much be- 

 yond the dark limits of the red rays as the red part is 

 distant from the bluish green band of the spectrum. 



In all these experiments, M. Melloni employed a 

 thermo-multiplier, an instrument that measures the 

 intensity of the transmitted heat with an accuracy far 

 beyond what any thermometer ever attained. It is a 

 very elegant application of M. Seebeck's discovery oi 

 thermo-electricity; but the description of this instrument 

 is reserved for a future occasion, because the principle 

 on which it is constructed has not yet been explained. 



In the beginning of the present century, not long after 

 M. Malus had discovered the polarization of light, he 

 and M. Berard proved that the heat which accompanies 

 the sun's light is capable of being polarized ; but their 

 attempts totally failed with heat derived from terrestrial, 

 and especially from non-luminous sources. M. Berard, 

 indeed, imagined that he had succeeded ; but when his 

 experiments were repeated by Mr. Lloyd and Professor 

 Powell, no satisfactory result could be obtained. M- 

 Melloni lately resumed the subject, and endeavored to 

 effect the polarization of heat by tourmaline, as in the 

 case of light. It was already shown that two slices of 

 tourmaline cut parallel to the axis of the crystal, trans- 

 mit a great portion of the incident light when looked 

 through with their axes parallel, and almost entirely ex- 

 clude it when they are perpendicular to one another. 

 Should radiant heat be capable of polarization, the quan- 

 tity transmitted by the slices of tourmaline in their for- 

 mer position ought greatly to exceed that which passes 

 through them in the latter, yet M. Melloni found that 

 the quantity of heat was the same in both cases : whence 



