970 Mr. Beer and Prof. Tyndall on Mano metric 



Discussion of Results. 



The above experiments show most clearly that the body of 

 the arc possesses no net charge, and that this statement 

 holds to within a very short distance from the anode and 

 probably — though on this point the evidence is not so 

 definite — from the cathode as well. 



Compared with any slope of potential method of in- 

 vestigating space-charge distribution in the- arc, the wind- 

 pressure method is extraordinarily sensitive. In the body 

 of the arc, therefore, the balance between the two ionic 

 streams is perfectly maintained. 



If the negative ions are electrons, the ratio of their 

 mobility to that of the positive ions is presumably at least 

 500 ; so that, except close to the electrodes, the current 

 carried by positive ions cannot be more than one-fifth per 

 cent, of the total. If, however, any of the negative ions are 

 of atomic magnitude, this percentage is too low. 



The view that the arc is in a highly dissociated state is 

 therefore confirmed. Production and recombination of ions 

 is continually taking place, and the ions in drifting to the 

 electrodes under the influence of the field give rise to space- 

 charges near the electrodes. It has been shown above that 

 the pressure observed at the anode is in quantitative agree- 

 ment with the view that it is produced by the motion of a 

 space-charge of negative ions in a thin layer close to the 

 electrode ; but the data are not sufficient to give the 

 magnitude of the charge or the thickness of the layer in 

 which it is appreciable. We may, however, make a rough 

 guess if we take a current of 10 amperes and assume 

 the anode pressure at an infinitesimal hole to be, let us say, 

 50 dynes/sq. cm. (or five times that observed at the smallest 

 hole used). If we further assume the current to be carried 

 entirely by negative ions of mobility 10,000 cm./sec./volt/cm., 

 it may be shown from the wind-pressure formula that the 

 thickness is 10~ 3 cm. The thickness would be far less if 

 the negative ions in the arc are of molecular and not elec- 

 tronic magnitude. On the other hand, the presence of 

 positive ions taking part in the flow of the current would 

 increase the estimated thickness. 



That some positive ions are present at the anode surface 

 is almost certain on thermionic grounds. Also they afford 

 a simple explanation of the small reversed readings, referred 

 to above and obtained in some cases with a displaced arc. 

 The region outside the luminous area where the main 

 current arises is red-hot, and is therefore presumably emitting 

 positive ions. If these form an unbalanced stream of positive 



