38 -/. 0. Thompson — Law of Elastic Lengthening. 



to deal with the superposition of different after-effects, and the 

 problem becomes much more complex. The after-effect in the 

 case of a strip of india-rubber is approximately proportional 

 to the magnitude of the stretching.* Probably this applies to 

 the metals as well, and if this is the case, the after-effect be- 

 comes of no account in the question we are discussing, pro- 

 vided the measurements are made always a definite time after 

 the stretching of the wire. Moreover, since the after-effect in 

 my experiments was invariably small I shall introduce in my 

 results no correction for that factor. 



Whenever the heaviest weights were used I set the cathe- 

 tometer 13 sec. after placing the weight on the pan. This 

 could be done by setting the cathetometer approximately 

 beforehand. Careful measurements showed that this precau- 

 tion was necessary only when the heaviest weights were used. 

 Ln other cases it made no apparent difference whether the read- 

 ing was made at the end of 13 sec. or a few- seconds later. The 

 measurements were made with rare exceptions at intervals of 

 two minutes. In this time the after-effect following the 

 release of the wire had almost entirely disappeared, as the 

 results of two series of experiments given on pages 40 and 42 

 will show. 



Still I wished to know exactly how soon this after-effect dis- 

 appeared, and accordingly used the following plan : A fine 

 flexible linen thread was tied to the bottom of the pan and 

 then passed through a pulley beneath which was fastened to 

 the floor. The thread was then made fast to the axle of a 

 wheel, and thus by turning the wheel I could impart to the 

 wire any desired tension. 



In order to compare the influence of the after-effect with 

 that of the thermal changes I proceeded as follows : By means 

 of the thread a tension corresponding to a certain weight was 

 imparted to the wire. The wire was kept stretched for about 

 13 sec. and then, by turning the wheel, released. During this 

 process it was not necessary to remove my eye from the tele- 

 scope. The wire came to rest without vibration in about one 

 second, and a second later I had the telescope sighted on the 

 mark. Then by means of the micrometer screw on cathe- 

 tometer II, I measured the change in position of the mark 

 in the following 11 sec, then the change from the 13th to the 

 30th second. By this time the mark had come back to its 

 previous position. The results of the measurements with the 

 brass wire are given in the following table : 



*F. Kohlransch, Pogg. Ann., elviii, p. 360, 1876. 



