﻿the Poles of the Electric Arc. 769 



independent of the earth's magnetic field, we shall find 

 also a numeral concordance. For instance, by extrapolating 

 for 11 amp. : — 



From Duffield's Table VII. (Burnham), 



L = 3*5 mm., anode P = 1'55, cathode F — 1*75, 

 i(P + + P_) = 1-65 dyne; 

 From D.'s Table VII. (Davis), 

 mean value L = 2*5 mm., anode P — 2*02, cath. P = 1*78, 



KP++.P_)= 1-90 dyne; 

 From Table I. (above), 



mean value L = 2'5 mm., anode P = 3*49, cath. P = 0'29 



i(P+ + P_)= 1*88 dyne. 



There is, on the contrary, a remarkable difference between 

 the separated values P + and P_, which may be due to the 

 difficulty in eliminating V *. The readings from Table VII., 

 which are almost unaffected by V and E, show, according to 

 my results, P + >P_, i. e. a greater effect upon the anode. 



5. In order to find out how other circumstances may 

 influence the pressure P, I have tried some experiments 

 with cored carbons, finding an increased effect upon the 

 negative pole and a reduced one upon the positive. This 

 behaviour is to be attributed to the metallic salts of the core 

 (8., p. 77). 



I have also noticed that the readings for ascending and for 

 descending current are often a little different, as happens, 

 for instance, in P. D. measurements, for both the shape and 

 the matter of the carbon (occluding gases, metallic salts, 

 grain, &c.) are altered by burning. 



The diameter of the electrodes has no great influence in 

 the present research, of course only while it remains large 

 relatively to the crater size. For, putting a carbon rod 

 12 mm. in diameter against a similar one of 3 mm., when 

 the latter is acting as anode, the arc hums and the repulsion 

 becomes greater (S., fig. 8). 



Further remarks, made also with thick carbon rods either 



* In Duffield's experiments, setting- approximately H v — 47 (England) 

 and taking 0\[=OB = ll cm. from his tigs. 1 and 17, formula (1) would 

 give V = 0"26I. Instead of 0'26, W2 get from fig. 8, 014, and from, 

 p. 124, 0-48. 



Phil Mag. S. 6. Vol. 44. No. 262. Oct. 1922. 3 D 



