of Iron when under Stress or Strain. 49 



been previously permanently lengthened by the 1 kilo, weight 

 at the bright red temperature, was rendered suddenly more 

 positive relatively to the unstrained wire. When the wires 

 were allowed to cool there was, after a short pause, an equally 

 sudden deflection in the opposite direction. This experiment 

 was repeated with three different pairs of wires and with the 

 same results. 



It has been mentioned in Experiment I. that as soon as the 

 temperature reached a bright red the deflection of the galva- 

 nometer frequently altered by fits and starts, and was even at 

 times strongly reversed in direction. This circumstance con- 

 stituted a difficulty in all the experiments, and was presently 

 found to be due to slight shiftings of the flame of the burner 

 from slight draughts in the room. Hot iron at any tempera- 

 ture is thermoelectrically negative to cold iron ; but when the 

 temperature reaches a bright red there is such a sudden 

 accession of negativeness that if we lay a piece of iron at 

 this temperature upon another piece of cool iron there is 

 developed an electromotive force which, thermoelectrically 

 considered, is very large. 



Experiment III. 



In fig. 2 A is a piece of the same iron wire connected with 

 one terminal of a condenser C having a capacity of one third 



Fig. 2. 



of a microfarad ; B is a second piece of the same iron wire 

 connected by insulated copper wire with one terminal of a 

 reflecting-galvanometer Gr having a resistance of about 6,000 

 B.A. units ; the other terminals of the condenser and galvano- 

 meter are connected together with insulated copper wire. 

 The free end of B was heated in a burner nearly to a white 

 heat, and then, having first been removed from the flame, was 

 quickly placed on the free end of A; immediately there was a 

 deflection of the spot of light on the galvanometer-scale 

 Phil. Mag. S. 5. Vol. 25. No. 152. Jan. 1888. E 



