1888.] Development of Electricity by Atmospheric Oxidation. 185 



A, the siphon connecting this vessel with the other one in which 

 the oxidisable metal was immersed being bent so as to pass nnder the 

 rim of the jar a, a, a. It was found convenient to mount the jar on a 

 block of paraffin wax with a circular groove b, b, in which the bell- jar 

 stood, the groove being then filled with mercury so as to make a sort 

 of hydraulic lute; if required, the air inside the jar could be replaced 

 by oxygen, &c, by simply passing in a current of gas through one of a 

 pair of tubes c, c, introduced through a perforated cork in the neck of 

 the jar. Usually several aeration plates were separately arranged 

 in the same vessel, each one, B, being connected (by means of a 

 platinum wire d, d, d, imbedded in the paraffin wax, passing under the 

 rim of the jar), with a mercury cup e outside : in this way the plate 

 of oxidisable metal used,/, could be removed at pleasure for cleaning, 

 amalgamating, &c, and replaced without disturbing the aeration 

 plates, and could be opposed at will to any one of these by a simple 

 switch connecting the required mercury cup with the rest of the 

 circuit. 



On first setting up such an arrangement and taking readings 

 alternately with any one of the plates opposed to the oxidisable metal, 

 and a Clark's cell, the total resistance in circuit being the same 

 (usually several megohms to reduce the current density sufficiently) 

 values were obtained generally exhibiting progressive alteration 

 (sometimes increase, sometimes decrease) as time elapsed ; but after 

 periods varying in different cases from an hour or two to several days, 

 sensibly steady readings were obtained exhibiting little or no variation 

 for days and even weeks together; what variations were observed 

 were generally traceable either to temperature fluctuations or to slight 

 shaking or mechanical disturbance whilst renewing the opposed 

 oxidisable plate, or to slight differences in the latter. If, however, in 

 cells set up with dilute sulphuric acid air had free access, more or less 

 considerable alteration was often brought about after some time 

 through evaporation or attraction of moisture from the air, alteriug 

 the film of fluid in contact with the aeration plate; and this was still 

 more the case with cells set up with caustic soda solution through 

 absorption of carbonic acid, and with ammonia cells through vola- 

 tilisation of ammonia. 



The result of a large number of observations with cells of various 

 kinds was to show that the following general proposition holds : — 



If a cell set up with a given fluid, oxidisable metal, and aeration 

 plate generate an E.M.F. = e 1? then the effect of substituting 

 another aeration plate for the first is to alter the E.M.F. to 

 e 3 = e x + K T ; whilst that of substituting a different oxidisable metal 

 is to alter the E.M.F. to e 3 = e Y + K 2 ; the quantity K Y being inde- 

 pendent of the nature of the oxidisable metal used, but varying with 

 each kind of aeration plate employed, and also to some extent with 



