482 M. H.Weber on the Heat-conducting Power 



Hence, in the determination of heat-conducting power, atten- 

 tion must be directed especially to the simplicity of the pheno- 

 mena from the observation of which it is to be deduced. Now 

 this simplicity appears to have been attained in a far higher 

 degree by a method which F. Neumann has given in his Lec- 

 tures than by any other hitherto employed, even in a higher 



degree than by the one made use of by Angstrom (described in 

 Pogg. Ann. vol. cxiv. p. 513), to which Neumann's stands in 

 the closest relation. 



The following observations, made according to this method 

 with the greatest care possible, for an exact determination both of 

 the internal and the external conducting-power of iron and German 

 silver, have lully established the superiority here ascribed to it. 



Determination of the Heat-conducting Power after Neumann's 

 Method. 



o 



Angstrom used in his experiments a bar of everywhere equal 

 cross section, and of such a length that the temperature of one 

 end did not perceptibly vary when the other end was exposed to 

 different high temperatures. The one end was now, at equal 

 intervals of time, alternately brought to two diflPerent high tem- 

 peratures, and thereby a periodical state of temperatures produced 

 in the bar, from the observation of which the value of the in- 

 ternal conducting-power was finally obtained. In order to de- 

 duce the mathematical expression for that state of the bar, 



o 



Angstrom represented the temperature-state of the alternately 

 heated and cooled end, by means of a series of sinuses, as a func- 

 tion of the time. If we denote the same by (f>{i), by Uq and Ui 

 the temperatures which the end-surface alternately assumes, by 

 T the interval of time after which each change occurs, then 



Neumann has now shown that the same problem can be 

 treated in another way, and one more favourable for observation, 

 — and that it is more suitable for the determination of the two 

 conducting- powers to subject not merely one end of a bar re- 

 garded as unlimited, but both ends of a bar of infinite length to 

 the same periodic change of temperature — in such manner that 

 when the two ends in the 0th period have the temperatures Uq 

 and Wj they in the next following take the temperatures u^ and 

 Uq, in the 2nd period (on the contrary) they are in the same 

 state as in the 0th, and in the 3rd as in the 1st, &c. 



In order to determine the temperatures which are hereby 

 produced in the different parts of the bar after the lapse of a 

 series of periods, we have first to determine the temperatures 



