THERMAL CONDUCTIVITY. 17 



years afterwards, and it was only by consulting the original 

 MS. in the possession of the Institute that Poisson and 

 others were able to study the results which had been arrived 

 at by Fourier. Fourier defines conductivity in a manner 

 which is easiest explained by means of a simple illustration. 

 Suppose that I have a solid wall separating two chambers 

 which are of different temperatures. Suppose that on one 

 side of this wall the temperature is 2 higher than it is on 

 the other, then we know that heat will pass from the hot 

 side to the cold side. The question is, how much heat will 

 pass ? Fourier says, let us continually keep the two sides 

 at a constant temperature, so that when heat passes out 

 from the outside an additional amount of heat is always 

 being added to keep the temperature constant, while at 

 the cold side let heat be continually drawn away, let cold 

 matter be continually added, so as to keep the temperature 

 there constant also. Then, if we can measure the amount 

 of heat which passes through this wall, we are able to 

 measure exactly its conductivity. You see that, the longer 

 the time that elapses the greater is the amount of heat 

 which will pass through. Therefore, if we call the quantity 

 of heat which passes through, or the flux of heat, by the 

 letter F, the flux of heat is proportional in the first 

 place to time, and we will call the time T. Then, 

 again, suppose we take another wall of exactly the same 

 substance, the temperature of the two sides of which are 

 also separated by two degrees, the highest temperature 

 in this case being the same as the lowest in the case we 

 considered before. The circumstances are almost identical, 

 the difference of temperature is the same, and the thickness 

 of the wall is the same, consequently the same quantity 

 of heat will pass through this wall in the course of a given 

 interval of time. Now let us place these two walls against 

 each other, the one overlapping the other, and let us have 

 the temperature of the two vessels now differing by 4, 

 then half way through these two walls we shall have the 

 mean temperature, and consequently each of these walls 

 will be in identically the same condition as it was when 

 there was only a single wall employed ; the difference in 

 temperature of the two faces of each wall will be exactly 

 the same as it was before, and the same quantity of heat will 

 pass through. Consequently, if we double the difference 



VOL. II. C 



