408 Mr. R. Sabine on some Electrical Experiments 



ance of which was smaller. The following are the observa- 

 tions:— 



Current in 

 selenium. 



Measured resistance. 









Current direct 



Current inverted. 



microvvebers. 



ohms. 



ohms. 



0-6 



40090 



4U130 



30 



40020 



40300 



60 



39800 



40400 



90 



39600 



40440 



14-4 



39240 



40450 



21-6 



38800 



40370 



50-4 



37570 



39790 



720 



36900 



39350 



1080 



36130 



38720 



In each case it was found that the resistance of the selenium 

 had somewhat increased during the measurement, due probably 

 to heat generated by the current. 



The behaviour of the selenium in these experiments is open 

 to one of two interpretations. If due to resistance, the altera- 

 tion is possibly a consequence of the operation of the Peltier 

 effect ; if due to electromotive force, it is probably a simple 

 consequence of electrolytic polarization. 



When the current passes through the selenium, it encounters 

 a compound conductor analogous to that in which Peltier 

 observed the cooling and heating effects of the junctions be- 

 tween antimony and bismuth. The crystals round the point 

 of contact which cools contract, and have a tendency therefore 

 to recede and to make the contact with their neighbours still 

 worse ; the crystals about the heated contact expand and 

 improve the conduction. The effects are opposite, but not 

 necessarily equal, because one junction may be more susceptible 

 than the other. It is of course questionable how far this 

 heating and cooling of an uncertain number of points of bad 

 contact between the crystals of a selenium plate near the 

 electrode would be sufficient to produce a sensible effect. 

 But it is necessary to bear in mind the nature of the doubtful 

 contacts which we are probably dealing with, and which the 

 smallest conceivable approach or recession may improve im- 

 mensely or break entirely. 



When the direction of the current in a compound conductor 

 is reversed, the heat or cold produced at one of the junctions 

 has to disappear, and the opposite effect to be developed, before 

 the resulting change can be completely observed. It was 

 found that when the bridge resistance was adjusted, in anti- 



