34 



Dr. E. N. da C. Andrade on the Electrical 



§ 10. Relation hetween Conductivity and Pressure. 



The experiments on conductivity were made with a streak 

 of metallic vapour (strontium) as described in §3: the 



electrodes, of platinum foil, were L*2 cm. apart, and tlie 

 streak stood between them, without touching cither. As 

 only the alteration of the current is here required, the action 

 of the glowing negative electrode produces no disturbance, 

 when we consider that it was shown that at atmospheric 

 pressure for this position of the streak the current is almost 

 the same for a hot as for a cold negative electrode. The 

 galvanometer was read first with the streak present, and then 

 for the pure flame, and the difference of the readings taken 

 as proportional to the conductivity of the sail vapour. 



It has already been shown (§ 3) that, if the whole streak 

 lies between the electrodes, without touching them, the 

 current is proportional to the applied potential in the experi- 

 mental range of to 100 volts. Measurements were there- 

 fore made mostly for one fixed potential difference. 100 

 volts. The results are exhibited in the following table : — 



Table IIT. 



Pressure 

 (Atmospheres). 



Conductivity 

 (arbitrary units). 



Pressure X 

 conductivity. 



1 



10-4 



MM 



16 



62 



93 



1-95 



36 



72 



24 



23 





265 



24 



61 



31 



13 



40 



40 



080 



3-2 



405 



080 





60 



0-65 





62 



0-50 



2-6 



Tim relation between conductivity and pressure is shown 

 in fig. n (a). From curve (/'). where the conductivity is 

 shown against the reciprocal of the pressure, it is I 

 thai the conductivity diminishes faster than it would if it 

 were proportional to the reciprocal of pressure. The cause 

 of this may possibly be the Fad that the diffusion is very much 

 greater at atmospheric pressure. This makes it possible that 

 at atmospheric pressure Borne of the metal atoms diffuse as 



