On the Investigation of Young's Modulus for Wires. 539 



Taking as a basis the quantity transmitted in nine weeks, 

 the coefficient of diffusion equals 28' 7 x 10 -7 . This result, and 

 others not recorded, indicate that the errors due to the 

 various causes studied are, at any rate, not enormous. 



To diminish the variations of temperature, in one experiment 

 the whole apparatus was put inside a copper cylinder; but, 

 unfortunately, some air-bubbles were accidentally allowed to 

 form within the compartments, and the quantities of copper 

 sulphate transmitted were irregular and of large value. At 

 present the author is experimenting with the tubes full of 

 a weak gelatine jelly, with the intention, ultimately, of 

 stopping each tube with a short plug of jelly, and thus pre- 

 venting circulations of all sorts. The plugs will probably be 

 made insoluble with formaldehyde. Mr. Hibbert has helped 

 me with the analyses, and the results are hopeful ; but, as 

 already indicated, progress is inevitably slow. 



University College. Sheffield. 



LI. The Application of an Interference- Method to the In- 

 vestigation of Young's Modulus for Wires, and its Relation 

 to Clianges of Temperature and Magnetization; and a 

 further Application of the same Method to the Study of the 

 Change in Dimensions of Iron and Steel Wires by Mag- 

 netization. By G. A. Shakespear, 1851 Science Scholar , 

 Trinity College, Cambridge*. 



MUCH work has been done with the object of elucidating 

 the relation between temperature and elasticity, and 

 more remains to be done. 



The problem presents many difficulties, not a few of which 

 arise from the fact that it is not easy to maintain at a constant 

 temperature, more or less high, a sufficient length of the 

 material under investigation. Again, very slight differences 

 of temperature between the wire undergoing traction and the 

 scale on which the alteration of length is measured may easily- 

 lead to erroneous results. 



These difficulties may be lessened by having the length of 

 wire small ; but this involves the necessity of some means of 

 measuring accurately very small changes of length ; inter- 

 ference supplies such a means. In the present work it has 

 not been found necessary to go to a higher order of accuracy 

 than -^ of the distance between successive interference- bands 

 of sodium light, which represents a length of about '000003 cm., 



