16S On the Diffusion of Utnt 



bodies where the temperature is higiier to those where it is lower, 

 it is not, therefore, a necessary result, but is only contingent, 

 requiring the presence of another condition, which may be want- 

 ing, and actually is wanting, in many instances ; — this is, that 

 the quantity of heat in the system should be given, and should 

 not admit of continual increase from one quarter, nor diminu- 

 tion from another. When such increase and dimiiuition take 

 place, no such equilibrium can be attained. In proof of this, he 

 mentions the fact, that a bar of iron thrust into the fire, though 

 red-hot at one extremity, will not i)ecome so at the other in 

 any length of lime, but eac'i part of It will have a fixed tempera- 

 ture, lower as it is further from th& fire, but rf mainiiig invaria- 

 ble while the condition of the firf», and of the surrounding me- 

 dium, continues the same. He illustrates it also more fully by 

 the following example : Let A, B, C, D, &:c. be a series of con- 

 tiguous bodies, or let them be parts of the same bodyf and let 

 us suppose that A receives fro'.n some cause, into the nature of 

 which we are not here to inquire, a constant and uniform supply 

 of heat. It is plain that heat Avill fiow continually from A to B, 

 from B to C, &c.; and in order that tliis :r.?.y take place, A must 

 be hotter than B, B than C, and so on ; so that no unitorm dis- 

 tribution of heat can ever take place. The state to which the 

 system will tend, and at which, after a certain time, it must ar- 

 rive, is one in which the momentary increase of the heat of 

 each body is just equal to its momentary decrease, so that the 

 temperature of each individual body becomes fixed, all these 

 temperatures together forming a series decreasing from A down- 

 wards. This is then apphed to the argument in the following 

 manner: " If heat be conniumicated to a solid mass, like the 

 earth, from some source or reservoir in its interior, it must go 

 off from the centre on all sides tov.ards the circumference. Ou 

 arriving at the circumference, if it were hiiidcred from pro- 

 ceeding further, and if space or vacuity presented to heat an 

 impenetrable barrier, then an accumuhitiou of it at- the surface, 

 and at last a unifiirm distribution of it through the whole mass, 

 would inevitably be the consequence. But if heat may be lost 

 and dissipated in the boundless fields of vacuity, or of ether, 

 which surround the earth, no such equilibrium can be established. 



much of the beauty and unity of the system !■; lost. These seem to me to 

 require the assunii^tion jf a ctntrat heat, or tjeiieral reservnii- tU'htat ca- 

 pable of extendiiic; its action to every part of the circumterence, always 

 exiitini;, thougli not equally active in its apparent effects. It i'5 to this 

 Tiew ot tiie subject, consistent, I believe, with the original stiitcniiiit oftiie 

 theory by Dr. llutlon, that the armniieiit applies. It is this which Mr. 

 Playfair admit*., and on tlie admission of which, indeed, his reasoning is 

 founded : and, strictly spe.ikint:, it is to his reasoning only that the obser- 

 vations in this paotr are directed. 



The 



