1881.] 



into Electric Distribution, Sfc. 



321 



tributions are the same when the experiment is begun with a plate 

 covered with either ion, or when the action is continued for a longer 

 time. But a change in the strength of solution, or in temperature, or 

 in strength of current alters the distributions. Speaking generally, 

 there is an increase in the range of the positive ion with increase of 

 temperature, increase of current strength, and decrease in strength of 

 solution ; while the range of the negative ion decreases with increase 

 of temperature, increases to a certain extent with increase of current, 

 and varies irregularly in the case of change in the strength of solu- 

 tion. 



The mutual influence of plates, end to end and side by side, the 

 distribution on open tubes, and on plates encircled by them or copper 

 gauze, show general analogies with what is known of the action of 

 static electricity. 



As long as the plates are perpendicular to the electrodes, the distri- 

 butions of one ion on the two sides is the same, but when the plate 

 forms an angle with a line joining the electrodes, this no longer holds 

 true, and the boundary lines of the same ion become markedly concave 

 on one side and convex on the other. 



With plates parallel to the electrodes, hardly any action took place 

 until the shape of the plate was altered, and it is shown that the 

 relation between the length and breadth of a plate of given area has 

 a very material effect in determining the accumulation and distribution. 

 And it was also found that the surface of a plate in this position for a 

 measurable distance from its edges, and still more from its angular 

 projections, shows no signs of electrification even when the body of 

 the plate does so to a sensible degree. These results afford a new 

 basis for investigating the field of electrolytic action both as regards 

 physical differences and the relative direction in which the action is 

 transmitted. 



With electrodes of the same breadth and depth as the cell, the 

 same distributions were obtained in different parts of the field of 

 action, but when the electrodes were not of the same breadth and 

 depth as the cell, differences were obtained in absolutely corresponding 

 parts of the field. These differences increased as the electrodes 

 became narrower. It is evident that the smaller electrodes give a 

 non-homogeneous field. 



The distributions are altered by the interposition of an ebonite 

 screen between an electrode and an analysing plate even in a field 

 otherwise homogeneous, and some instances of this are given. Analogies 

 are found between the effect of these screens and that produced by 

 diminishing the size of the electrodes. 



The distributions are divisible into two classes — parallel and non- 

 parallel. The first class includes all cases which show a similarity 

 in the distribution on the two sides of the analysing plate ; the second 



