392 Royal Society : — Prof. Norris on Molecular Changes 



tieular \Aire under use, even wlien maintained for some time at its 

 highest temperature. 



It is desirable that these facts should be borne in mind ; for no 

 sooner do we enter upon this research with more delicate and reli- 

 able apparatus, than we find that the kicks are but a small part of 

 the phenomena we are called upon to observe and ex])lain. We as- 

 certain at the very outset that iron and steel wire presents itself to 

 our examination under very different conditions, which demand a 

 rigid analysis. Thus, apart from its inherent quality, it may come 

 to us for experiment : — 



(«) In its usual commercial state. 



(6) After having been subjected to various degrees of annealing. 



(c) In various degrees of induced hardness. 



It is proposed to consider, in the first place, the phenomena pre- 

 sented by the ordinary steel wire of commerce, when first sub- 

 jected to the influence of heat. During the heating of such wires, 

 we notice that a first kick occurs at a point varying on the scale 

 from 29° to 33°, and that the wire subsequently proceeds to a high 

 expansion, which may be represented on the scale by the figures 

 Gi\°, 52^°, 55°, 50|°, and so forth, according to the quality and 

 state of the wire. In the act of cooling a kick commences at 

 points varying from 31^° to 22°^ and terminates at points varying 

 from 33° to 28°, when the finger proceeds on towards zero, but in- 

 var iahly falls short of reaching it hy amounts varying from |° to 5|° 

 {centimetres). [Vide fig. 1, p. 394.] 



In the next experiment with these \^-ires a much altered state of 

 things is found to obtain. Thus the first kick is lowered a little 

 in position, increased in range ; the expansion of the wire is ma- 

 terially diminished, being now represented by such figures as 58^°, 

 43^°, 45 1°, 50 J°, &c. respectively. In the cooling the second kick 

 is separated more from the first, being put back in the direction of 

 zero. It will also be seen that in cooling the finger passes beyond 

 zero, indicating now that the wire is permanently shortened, 

 whereas in the primary heating it \\2l^ permanently elongated. 



A comparison of these two series of experiments will show us 

 that, during the heating of these wires, there is going on at the 

 same time a whole series of phenomena, some of which, although 

 different in nature, are alike as to result, while others are different 

 both as to nature and result. We have, in the first place, the 

 ordinary expansion, which we will designate the " dynamical ex- 

 pansion." In the primary heating an amount of expansion will 

 be seen to occur which has no equiAalent in the dynamical contrac- 

 tion of cooling. The secondary heatings reveal to us the existence 

 of a contraction, which is also excited during the heating, and the 

 action of which is to limit the dynamical expansion previously 

 referred to. That it masks a portion of the dynamical expansion 

 is rendered obvious by the fact that, in cooling, the finger moves 

 beyond zero, showing that the wire has contmcted more in cooling 

 than it expanded in heating. 



In addition to this, we have also the phenomena of the kicks 

 themselves ; and these may be described as consisting of a temporary 

 contraction and expansion. 



