434 Mr. F. J. Smith on Neiv Methods of Investigating 



the wire caused a large movement of the end of the lever, 

 which traced a line on the smoked surface of paper rotating 

 on an ordinary physiological chronograph cylinder. A thermo- 

 couple made up of two wires, one of platinum, the other of pla- 

 tinum alloyed with 10 per cent, of rhodium, was twisted round 

 the wire where it was heated ; the thermo-couple was in circuit 

 with a Depretz D'Arsonval galvanometer. By means of this 

 combination the temperatures at which the changes of length 

 of the wire took place were read. The movements of the beam 

 of light reflected from the galvanometer were recorded on a 

 moving photographic film. 



The outcome of the experiments made with this arrange- 

 ment was that the changes of form of the metal under 

 examination took place at the times of change of temperature, 

 so that a curve so traced on the smoked paper can be used as 

 an index of the changes of form and the changes of tempera- 

 ture. While an experiment was going on, electrically con- 

 trolled styli marked the time on the smoked paper and on the 

 photographic record second by second. 



Two copies of curves traced by this method are given, 

 numbered I and II. They show that on heating the wire it 

 expands up to a point B, and then stops, although the heat 

 is supplied to the wire at a uniform rate, it next expands up 

 to G ; heat then being taken off, it contracts to D, where for a 

 short time it ceases to contract, it then contracts to E. The 

 time is marked on the zigzag line TT every two seconds. 

 So far two critical points are made evident. The wires used in 

 the curves I and II were differently loaded; this 



tracing 



