116 IOWA ACADEMY OF SCIENCE 



after a period of rest the wire tended automatically to reach the an- 

 nealed condition. However, at the time that this was noticed there was 

 no opportunity to give the wire sufficient rest, so the point was not 

 verified. 



When the writer had finished his observations with the wire in April, 

 1910. his attention was turned to other matters, and as a resiilt the wire 

 was not disturbed for a period of about eight months. When the last 

 experiments had been done on the wire, it had been vibrated a great 

 many times at large amplitude, and had not been annealed at the end 

 of the observations. So here was an excellent opportunity to observe if 

 the period of rest had had any effect in restoring the wire to its elastic 

 condition. The first experiment following this long rest was very care- 

 fully made, with the initial amplitude at only about 50°. This was 

 necessary, for a much larger amplitude would have destroyed the an- 

 nealing effect of the long rest, and so the effect sought for would have 

 (been completely masked. The results of Oils and of the subsequent 

 experiments, to be discussed later, will best be observed by reference to 

 the accompanying figure. Ill this figure there are platted two sets of 

 curves, one which we are now examining, and the other to be discussed 

 a little later. Curve (a) represents the relation between period and 

 amplitude in the lower ranges of the amplitude for the experiment 

 where the wire had been vibrated to an amplitude of 720°. The relation 

 between these same two quantities after the eight months' rest is shown 

 in curve (d). The curve showing the relation between these quantities 

 after the wire is annealed, is shown in curve (e). It is interesting to 

 see" that the rest has not only restored the wire to 'its annealed condi- 

 tion, but has actually carried the elastic condition beyong that point. I 

 believe that this is the first time that such a phenomenon has ben ob- 

 served. 



After this interesting observation, experiments were tried in order to 

 discover if possible the rate at which the wire was automatically re- 

 stored to its annealed condition. It necessarily follows that such ex- 

 periments must be very tedious, for it is necessary to leave the wire un- 

 disturbed for long intervals of time between readings. This will account 

 for the relatively few observations that I have recorded. In this 

 same figure will be found two other curves of this series: one, curve 

 (b), for an eight days' rest, and one, curve (c), for a, twenty-two 

 days' rest. It is hoped that further observations on this point may 

 be made. It will be observed that the recovery of the wire is really 

 very rapid at first, but -becomes gradually slower and slower. In fact 

 one is strongly reminded of the logarithmic form of the ordinary 



