EFFECT OF TENSION ON CERTAIN ABNORMAL METALS. 43 



elsewhere.* The wire under tension merely replaces the wire under 

 pressure of the previous experiments. 



The wires used were small, and a load of a few kilograms was suffi- 

 cient in all cases. The wire was mounted vertically, and attached at 

 the upper end to one arm of an equal arm balance, to the other arm of 

 which known weights could be applied. For most metals it was most 

 convenient to use the ordinary solid weights, but if there were hystere- 

 sis effects, as in the case of nickel, it was necessary to apply and remove 

 the weight continuously; to accomplish this a water weight was used. 

 In the case of antimony, because of its excessive fragility, a special 

 arrangement was necessary which will be described in detail later. 



The lower end of the wire was attached to a bracket, supported from 

 above. This bracket, with the wire, dipped into an oil reservoir about 

 12 inches high and two inches diameter. The top and bottom of the 

 reservoir were connected through side tubes with a turbine stirrer, 

 by means of which a continuous stream of oil was maintained past the 

 specimen. The wire itself was about 6 inches between potential 

 terminals. In most cases the measurements were made at room 

 temperature only, and the stirrer adequately maintained approximate 

 equality of temperature. A correction could be easily determined 

 and applied for the change in resistance due to drift of temperature of 

 the oil bath. In a few cases, however, when more careful regulation 

 of temperature was necessary, a large bath of water, maintained at 

 constant temperature thermostatically, was raised around the oil 

 bath. The walls of the latter were of thin brass, and exchange of heat 

 between the oil and the water was sufficiently rapid to maintain con- 

 stancy of temperature in the oil. 



The magnitude of the tension applied was usually considerably less 

 than the elastic limit. The behavior of the resistance gives a sensitive 

 test of the perfect elasticity under the applied tension. The wire 

 was usually seasoned by a number of applications of a tension higher 

 than that of the final measurements. The behavior of the resistance 

 beyond the elastic limit is complicated, and would make an interest- 

 ing study on its own account. I felt this to be beyond the scope of 

 the present work. I have, however, in nearly all cases determined at 

 least the sign of the permanent change of resistance produced by 

 exceeding the elastic limit, and in some cases have examined the 

 phenomena a little more in detail. It appears that in all cases the 

 permanent change of resistance beyond the elastic limit is an increase. 



4 P. W. Bridgman, Proc. Amer. Acad. 52, 571-646, 1917. 

 « 



