402 Mr. R. Sabine on some Electrical Experiments 



In preparing the selenium plates, care was taken to obtain 

 as good a surface as possible. It was found that this might 

 be effected by melting the amorphous selenium between edge 

 strips of glass upon a piece of bright platinum foil, upon which 

 it was annealed ; the unequal contraction of the selenium and 

 platinum prevented any adhesion after removal from the an- 

 nealing-pot, whilst the selenium acquired a bright metallic 

 surface. 



Mercury Contacts. — It is necessary to discriminate between 

 the resistance of the selenium and the resistance of the junc- 

 tions. A plate of selenium was annealed at 200° C. with 

 platinum end wires. Each end was then inserted in the side 

 of a cup formed by a pill-box 7 where it was secured by means 

 of shellac in such a way that mercury could be introduced into 

 the cup to make connexion with the measuring-apparatus. 

 Before the mercury was poured in, the resistance between the 

 platinum wires was 124,300 ohms. After the cup at one end 

 was filled with mercury, the resistance was found to be reduced 

 to 82,300 ohms ; and when the second cup was supplied with 

 mercury, the resistance further diminished to 60,100 ohms. 



In a subsequent experiment, a plate of selenium with plati- 

 num-wire connexions, having a resistance of 20 megohms, 

 was found to have only 14 megohms resistance when the butt- 

 ends were in contact with mercury. The resistances of some 

 plates, however, were not greatly altered by this application 

 of mercury connexions at the ends. 



Resistance of Junctions, and want of homogeneity of the mate- 

 rial. — The resistance of the junctions between the selenium 

 and the platinum wires imbedded in it was found to be very 

 various, but not to depend upon the manner in which the wires 

 were imbedded in the selenium previous to its being crystal- 

 lized. It was also found that the conducting-power of the 

 selenium differed materially in different parts of the same plate. 



Fig. 1 represents a plate of selenium with 

 three platinum wires (1, 2, 3) imbedded 

 transversely in it. The resistance of one 

 of the wires was found as follows : — If x 

 is the resistance of junction 2 under exami- 

 nation, r and r' the resistances of the neigh- 

 bouring junctions plus whatever selenium 

 may be between them and the wire 2, the 

 measured resistances are : — 



l 



2 



3 



\ v 



1 2. 





J 



= r+x~\ 



= r+r' V,?; = 



-x + r f ) 



Between 1 and 2, II = r + x ") r> -r> . -d 



„ 1 „ 3, B I =r+^L= R ~? 1 + B ' - 

 „ 2 „ 3, R, 



