VARIOUS KINDS OF HEAT. 347 



Rumford, becomes, through the discovery of its mechanical 

 equivalent, a certainty. 



The form of force denoted by the name &quot;heat&quot; is plainly 

 not single, but includes several distinct, though mutually 

 equivalent, objects, three principal forms of which are distin 

 guished in common language : namely, I. Radiant Heat ; II. 

 Free (sensible) Heat, Specific Heat ; and III. Latent Heat. 



There can be no doubt that radiant heat must be regarded 

 as a phenomenon of motion, especially since the recent detec 

 tion of phenomena of interference in the radiation of heat. 

 But whether there really exists, as is commonly assumed, a 

 peculiar aether, of which the vibratory motion is perceived by 

 us as radiant heat, or whether the seat of this motion is the 

 particles of material bodies, is a question that is not yet made 

 out. 



Still greater obscurity hangs about the essential nature of 

 specific heat, or what goes on in the interior of a heated body. 

 Not only does the unanswered question of the aether enter 

 again here, but, before we can be in a position to form any 

 clear ideas on this subject, we require to have an exact knowl 

 edge of the internal constitution of matter. We are, how 

 ever, still far from having reached this point ; for, in particu 

 lar, we do not know whether such things as atoms exist that 

 is, whether matter consists of such constituents as undergo, no 

 further change of form in chemical processes. 



But a span of that time which stretches both backwards 

 and forwards into eternity is meted out to man here on earth, 

 and the space which his foot can tread is narrowly bounded 

 above and below : so also his scientific knowledge finds nat 

 ural limits in the direction of the infinitely small as well as of 

 the infinitely great. The question of atoms seems to me to 

 lead beyond these limits, and hence I consider it unpractical. 

 An atom in itself can no more become an object of our inves 

 tigation than a differential, notwithstanding that the ratio 

 \vhich such immensely small auxiliary magnitudes bear to 



