52 



K. TSURUTA. 



§ 29. P. Bachmetjew (1) has made some experiments on the same 

 subject as that of the present paper. He describes two methods of 

 investigation, one of which is exactly the same as that used by Dr. 

 Cohn, and the other is to make a couple of two metals and to pull or 

 compress one of them, the thermo-electric current between the metals 

 themselves being compensated by another similar couple. Whenever 

 iron or nickel was coupled with other metals, such as copper, and was 

 alone subjected to tension, the curves showing the relation of E. M. F. 

 to load were found to be more or less curved, while in other cases 

 they were almost straight. 



The two methods are not essentially different, but the interest of 

 the second method lies in its enabling us to see, at least qualitatively, 

 the variation of the thermo-electric property of iron under longitudinal 

 tension with respect to other metals like copper. A few experiments 

 were tried on couples made of annealed iron wires and copper wires 

 annealed as well as hardened ; I found again the familiar form of the 

 thermo-electric curve. 



(1) Thermo-elektrische Untersuchungen, Repert. der Physik, Bd. XXVI. 1890. 



