( 133 ) 



ILlll.— Experimental Inquiry into the Lams of the Conduction of Heat in Bars, and 

 into the Conducting Power of Wrought Iron. By James D. Foebes, LL.D., 

 D.C.L., F.R.S., V.P.R.S. Ed., Corresponding Member of the Institute of 

 France, Principal of the United College of St Salvator and St Leonard, St 

 Andrews. 



(Read 28tli April 1862.) 



1. The experiments to which I shall have to refer in this paper were all 

 made above ten years ago. A brief report of the general results, so far as they 

 were then obtained, was made to the British Association in 1852,* at whose 

 expense the instrumental apparatus had been provided. The causes of the delay 

 which has occurred in the publication of the numerical results will be presently 

 explained. 



2. I had previously, in 1831 and 1832, made some experiments on the con- 

 ducting power for heat of the metals, with Fourier's Thermometer of Contact, 

 though merely with a view to determine the exact order of the metals in this 

 respect. The results were not altogether decisive, particularly with regard to the 

 position of Iron in the series ; and the paper which was read to the Royal Society 

 of Edinburgh on the 7th January 1833, and which now lies before me, was never 

 published in full. The conclusions are, however, sufficiently stated (without 

 numerical results) in the " Proceedings" of the Society.f The most important 

 conclusion was this:—" That the arrangement of conductors of heat does not 

 differ more from that of conductors of electricity than either arrangement does 

 alone under the hands of different observers." The identity of the two series is 

 now generally admitted, although, at the time I speak of, it had not, I believe, 

 been suspected. 



3. In 1850 I commenced experimenting upon iron bars heated at one end, and 

 having thermometers inserted into holes at different points of their length. Expe- 

 riments in this form had been made by Lambert, Biot, Despretz, and probably 

 by others. But the peculiarity of the series now to be described consists in the 

 manner of reducing the observations, and in the nature of the results sought. In 

 the determinations of preceding authors, the law of communication of heat by 

 conduction within the bar, is assumed to be throughout proportional to the rate of 

 the variation of temperature (shown by the thermometers) at each point, :J: reckoned 



* Report for 1852 (Belfast), p. 260. f ^ol. i. p. 5 (1833). 



\ This is usually called the Newtonian Law, though it may be doubted how far Newton under- 

 stood it to apply to internal conduction. (See Phil. Trans. 1701, and Opuscula, torn. ii. p. 419.) 

 VOL. XXIII. PART I. 2 O 



