THE LAWS OF CONDUCTION OF HEAT IN BARS. . 107 



in the present paper. I do not, however, regard this labour as wasted, for the 

 knowledge thus acquired of the nature of the remarkable curves of which it treats 

 will enable a future observer to attack the question in a far more direct manner, 

 and to obtain, with comparatively little trouble, numerical determinations of the 

 conductivity of the metals under ordinary circumstances, and adapted to most 

 purposes of theory or practice, 



123. Suggestions as to Experiments. — After mature consideration, I do not 

 think that the experimental methods require almost any modification. The 

 independence of the results of any moderate error in the thermometers seems 

 satisfactorily proved (Arts. 38 and 117) ; and if the object be merely to ascertain 

 the conductivity and " percentage decrement" for a number of metals, it may 

 easily be done without pushing the observations to the high temperatures used 

 in my experiments, which are always a fertile source of difficulty and error. If, 

 for instance, an extreme temperature of 120° or 140° Cent, only was aimed at, 

 shorter bars might be used ; the heat would be more manageable and more 

 quickly attained ; the thermometers would be more easily made, more easily 

 used, and subject to far smaller corrections ; and the dynamical experiments 

 especially, would be freed from an anxious and troublesome source of error, 

 arising from the irregularity of the primitive distribution of the heat in the cooling 

 bar (Arts. 25, 26, and 90). 



124. A more exact knowledge of the form of the statical curve of temperature 

 in any case may be obtained by using sources of heat of progressively lower tem- 

 perature, as explained in Arts. 27 and 58. 



125. It is probable that very good results might be obtained by simply using 

 boiling water as a source of heat at the hottest end of the bar, than which nothing 

 can be more manageable. The duration of the statical experiments could thus 

 be much reduced, and the temperature of the air of the apartment rendered more 

 stable. The difficulties referred to in Arts. 65, 109, as to the determination of 

 very small excesses of temperature next the cool end of the bar might thus be 

 in a great measure removed. Indeed, it would be a worthy object of study, in a 

 theoretical point of view, to determine the form of the Statical and Dynamical 

 curves for those low temperatures more accurately than I have done. I cannot 

 but suspect an anomaly in the conduction of heat when the temperature varies 

 with extreme slowness from point to point, which my observations rather indicate 

 than establish* 



126. Another experimental point of interest for the theory would be to estab- 



* I may be allowed to state here generally, that this anomaly would apparently assign a too 

 great conducting power to iron at low temperatures than we can readily admit. [The case of the 

 1-inch bar might rather lead to an opposite conclusion, but I have less confidence in the observa- 

 tions made on it for very small excesses of temperature ] Both the statical curve and the curve of 

 cooling deviate more and more from the logarithmic form as the temperature-excesses diminish. 



