532 Prof. Clausius on the Conduction of Heat by Gases. 

 conduction of heat, wc obtain the following values for K ; namely, 



For atmospheric air . . . 44*06 . e 



For oxygen 41*90 . e 



For nitrogen 44' 71 . e 



For hydrogen 167*49 . e 



The complete numerical determination of these values requires 

 that the factor e should be known. A direct theoretical calcu- 

 lation of this quantity, according to the principles developed 

 above, is not possible, because for this it is necessary to know 

 the radius of the sphere of action p. We must therefore make 

 use of other data for the determination of e. Maxwell has calcu- 

 lated the mean length of excursion of the molecules from the 

 result of experiments on the friction of air in motion and on the 

 diffusion of gases, and in both cases has arrived at figures which 

 do not differ much from 



4607000 En s ,ish inch > or 16,000,000 metre - 



Without giving any opinion here as to the degree of confidence 

 which may be placed in this number, I am nevertheless of opinion 

 that we may employ it to give us an approximate idea of the 

 kind of magnitudes with which we have to do. Putting this 

 value into the above equation, we get for atmospheric air 



K- M - U ffiftt 



16,000,000 " 4,000,000' ' ' * { } 



This quantity denotes the quantity of heat, expressed in com- 

 mon heat-units, which would traverse a plane of one square 



metre during one second, if -j- were equal to — 1; that is, if 



the temperature decreased in the direction of the axis of abscissa? 

 near the point under consideration in such wise that, if a similar 

 decrease took place throughout a greater length, the tempera- 

 ture would diminish 1° C. in the length of 1 metre. 



§ 28. In order to compare this con ducting-power for heat 

 with that of the metals, we may make use of a result observed 

 by Peclet, who found, by measurement of the quantity of heat 

 which passed through a plate of lead, that, if a large mass of 

 lead were placed under such circumstances that a diminution of 

 temperature of 1° C. took place in a thickness of 1 metre, a 

 quantity of heat equal to 14 heat-units would then pass through 

 a surface of 1 metre square in one hour*. To compare this 

 number with that found for air, we must multiply the latter 

 by the number of seconds contained in an hour, it having been 



* Traite de la Chaleur, vol. i. p. 391. 



