470 Mr. S. Bidwell. On the Effect of Tension [Apr. 24, 



under tension. Wires of several different sizes and qualities were 

 suspended inside a magnetising coil, and were loaded with various 

 weights ; and in each case observations were made of (1) the smallest 

 magnetising current which caused sensible change of length ; 

 (2) the current producing maximum elongation (if any) and the 

 value of such elongation ; (3) the critical current which was without 

 effect upon the length of the wire; and (4) the contraction pro- 

 duced by a certain strong current. 



The results indicated that the maximum elongation became smaller 

 as the load was increased, disappearing altogether when the tension 

 exceeded a certain limit ; and that contraction began to take place at 

 a correspondingly earlier stage in the magnetisation. 



These results were chiefly of interest as disproving Joule's conjec- 

 ture, which has often been quoted as if it were an experimental fact, 

 that, under certain critical tension (differing for different specimens 

 of iron, but independent of the magnetising force*), magnetisation 

 would produce no change whatever in the length of the wire. 



The subject, however, seemed worthy of more complete investiga- 

 tion, and I have lately undertaken a series of experiments in which 

 the changes of length undergone by a stretched iron wire were traced 

 continuously as the magnetising force was gradually increased from 

 a small value up to about 375 C.Gr.S. units. Similar experiments 

 were also made with a nickel wire and with a thin strip of cobalt, 

 the behaviour of these metals under tension never having been 

 previously studied. 



Apparatus. 



The apparatus employed was the one described and figured in my 

 former paper. The diagram there given, together with a short 

 description, is, for convenience, here reproduced (see fig. 1). 



Experiments. 



For reasons which need not be repeated here, it was found neces- 

 sary to support the magnetising coil in the manner shown in the 

 figure, its whole weight being borne by the experimental wire. The 

 minimum load on the wire was therefore represented by the weight 

 of the coil, together with the pull of the lever, the two amounting to 

 1'36 kilo. Greater tension was produced by attaching weights to the 

 hook H. 



The iron used was a piece of soft annealed wire, 0'7 mm. in 

 diameter and 10 cm. in length, between the clamps. The weights 



* At least within tne limits of the forces employed by Joule, estimated to range 

 from 7 to 114 C.G.S. units. See ' Roy. Soc. Proc.,' No. 242, 1886, p. 112. 



