Intelligence and Miscellaneous Articles. 383 



is thus a compound bar, and presents dissimilar metals at its acute 

 angle. 



Now by fixing a little cocoa-nut tallow on the free extremity of 

 each rod, and applying heat in a uniform manner to the acute junc- 

 tion, with the two compound bars in a horizontal position, the least 

 difference in the rapidity of conduction in the two directions of the 

 similar compound bars could be ascertained with satisfactory accu- 

 racy by the difference of time at which the tallow upon the two ends 

 melted. 



The source of heat was a very small and steady flame of a spirit 

 lamp carefully protected from all currents of air, and the acute junc- 

 tion of the double pair was just in contact with the flame in such a 

 manner as to apply the heat equally and uniformly to the pointed 

 extremity of each pair. 



In nearly every case the experiments were repeated to ensure 

 correct results, and in those cases where the results were less certain, 

 rods of different diameters and lengths were tried in order to render 

 the effects more conspicuous : the rods varied from T Vth of an inch 

 in diameter and 3 inches long, to ^th of an inch in diameter and 6 

 inches long ; and the following were the results obtained : — 



Experiment 1 . In this experiment each arm of the pair was com- 

 posed of brass and nickel-silver wires, each metal being 3 inches 

 long and -^-th of an inch in diameter ; on applying heat to the junc- 

 tion for a short time, the tallow upon the nickel-silver extremity 

 melted about 10 or 12 seconds before that upon the brass end, 

 indicating that the heat conduction was more rapid from brass to 

 nickel-silver than vice versd: this experiment was repeated with 

 similar results. 



Exp. 2. Wires of copper and nickel-silver, each being 3 inches 

 long and -jig-th of an inch in diameter, the conduction of heat was more 

 rapid from copper to nickel-silver than vice versd by 10 or 15 

 seconds : repeated with like results. 



Exp. 3. Iron and nickel-silver wires 3 inches long and -^-th of an 

 inch in diameter, heat travelled faster from iron to nickel-silver by 

 2 or 3 seconds : repeated with wires 2 inches long and -J^th of an 

 inch in diameter with similar results. 



Exp. 4. Wires of nickel-silver and lead, each 2 inches long and 

 -|-th of an inch in diameter, heat conduction faster from'nickel-silver 

 to lead than vice versd by 6 or 8 seconds : repeated several times 

 with similar results. 



Exp. 5. Zinc and nickel-silver wires, each 2^ inches long and 

 Y^th of an inch in diameter, conduction of heat faster from zinc to 

 nickel-silver than vice versd by about 60 seconds : repeated with 

 wires 2 inches long with similar results. 



Exp. 6. Wires of nickel-silver alone but of different diameters, 3 

 inches long and -J-th and -J-th of an inch in diameters, heat travelled 

 faster from the large to the smaller wire than vice versd by 10 or 12 

 seconds : repeated several times with similar results ; also with wires 

 of xoth and ^th of an inch in diameters and 2-J- inches long, several 

 times with similar results. 



