1886.] Magnetisation in the Length of Iron Wires. 263 



well cause variations in the readiness with which the arrangement 

 would respond to a movement equivalent to less than one-tenth of a 

 scale division. Nevertheless it is clear in spite of one or two dis- 

 crepancies that a greater magnetising force is necessary to cause a 

 sensible elongation when the load is great than when it is small. In 

 one case, that of the thin wire under a load of 14 lbs., there was no 

 evidence of any elongation. It is probable, judging by analogy, that 

 a maximum elongation, too small, however, for the instrument to 

 detect, would occur with a current of about 0'12 ampere. Whether 

 any load, however great, would render the preliminary elongation of 

 the wire too small to be measured by an ideally perfect instrument is 

 uncertain. 



The second determination could be made with far greater accuracy. 

 But the load had the effect of flattening the apex of the elongation 

 curve in such a manner that the actual maximum was not so sharply 

 denned as in the case of free rods. 



The third determination, that of the magnetising current under the 

 influence of which the original length of the wire was unaltered, was 

 susceptible of great accuracy, and was the most important for the 

 purpose of the present investigation. 



The measurement of the amount of retraction caused by a given 

 strong current was also perfectly easy and certain. 



The figures recorded in the table disclose the following facts : — 



1. The effects produced by magnetisation upon the length of an 

 iron wire stretched by a weight, are in general of the same character 

 as those which have been shown in my former paper to occur in the 

 case of a free iron rod. Under the influence of a gradually increasing 

 magnetising force such a wire is at first elongated (unless the load 

 be very great), then it returns to its original length, and finally it 

 contracts. 



2. The maximum elongation diminishes as the load increases 

 according to a law which seems to vary with different qualities of 

 iron. If the ratio of the weight to the sectional area of the wire 

 exceeds a certain limit, the maximum elongation (if any) is so small 

 that the instrument fails to detect it. 



3. The retraction due to a given magnetising force is greater with 

 heavy than with light loads. 



4. Both maximum elongation and neutrality (i.e., absence of both 

 elongation and retraction) occur with smaller magnetising currents 

 when the load is heavy than when it is light ; retraction, therefore, 

 begins at an earlier stage. Thus the anticipation expressed in my 

 former paper is justified. 



5. The effects both of elongation and of retraction are, as might be 

 expected, greater for thin than for thick wires, and for soft than for 

 hard iron. 



T 2 



