658 HEroRT— 1887. 



constitutes tlie film, the opposed plate being; an oxidisable metal (e.g. copper or 

 zinc), the former is the case ; instead of an oxidisable metal, we find that an 

 incorrodible plate {e.g. platinum) immersed in an oxidisable liuid (e.g. an acid 

 solution of ferrous sulphate, or an alkaline one of pyrogallol) may often be 

 employed. Gas-films of hydrogen coal-gas and similar oxidisable gases opposed to 

 incorrodible plates immersed in oxidising fluids {e.g. platinum in nitric acid or 

 alkaline permanganate solution) furnish cells of the second class. In all cases the 

 E.M.F of the combination falls rapidly with increasing current density, owing to 

 the using up of the gas-film more rapidly than the physical attracting power of the 

 supporting plate can renew it, even under the most favourable conditions, i.e. when 

 supported horizontally on the surface of the electrolytic fluid employed so as to be 

 simultaneously in contact therewith and with an atmosphere of the gas experi- 

 mented with. 



In most cases it is desirable that the gas-film plate should not be in direct contact 

 with the alterable fluid surrounding the plate opposed thereto : the cell then takes 

 the form of a two-fluid arrangement, the two liquids being on opposite sides of a 

 porous partition, or contained in separate vessels united by an inverted siphon or 

 asbestos wick, &c. For example, a plate or tray of spongy platinum opposed to a 

 piece of platinum foil, the former being arranged on the surface of a solution of 

 caustic soda in contact with the air, and the latter immersed in a solution of 

 pyrogallol in caustic soda conveniently protected from direct contact with air by 

 being enclosed in an inverted test-tube dipping into a somewhat more dense soda 

 solution connected by a siphon, &c., with the other solution. 



We find that with cells of the first class, especially those where air is the gas 

 employed, much more concordant valuations of the E.M.F set up when generating 

 only minute currents may be obtained, than might a priori be expected, provided 

 certain precautions are taken ; and that fairly accurate valuations are possible of 

 tlie effect produced by varying the nature of the ' aeration plate ' (plate supporting 

 the air-film), the strength and nature of the fluid surrounding it, and so on. 

 Thus the simplest cells of this kind (such as a plate of amalgamated zinc immersed 

 in caustic soda solution, on the surface of which the aeration plate is arranged) 

 appear in general to produce a higher E.M.F the stronger the solution. The efiiect 

 of varying the aeration plate, all else remaining the same, is independent of the 

 nature of the oxidisable metal, but is influenced to some extent by the strength of 

 the electrolytic fluid, and varies with its nature. For instance, the effect of sub- 

 stituting a tray of spongy platinum for an aeration plate of smooth platinum foil 

 is to cause an increment in the E.M.F of the cell, the numerical value of which is 

 sensibly the same whether zinc or lead be the oxidisable metal opposed, provided 

 the soda solution be the same ; but is not the same whether the soda solution be 

 strong or weak; and is widely different, if, instead of caustic soda, dilute sulphuric 

 acid be used as the electrolytic fluid ; and similarly in other cases. 



Somewhat unexpected results were obtained when certain of the more difficultly 

 oxidisable metals (mercury, silver, and gold) were opposed to aeration plates in 

 contact with appropriate electrolytic fluids : oxidation and solution took place 

 with celerity. Thus mercury and silver readily form mercurous and argentic sul- 

 phates if immersed in dilute sulphuric acid on which a platinum sponge aeration 

 plate is arranged, on completing the circuit through a moderately large external 

 resistance : the current produced is readily measurable by inclosing in the circuit 

 a small silver voltameter. Silver similarly dissolves in ammonia solution (prefer- 

 ably containing some ammonium chloride or sulphate) when opposed to an aeration 

 plate, as also does gold in potassium cyanide solution. Hitherto we have not 

 succeeded in dissolving platinum in any combination of the kind. 



II. Voltaic Circles producible by the victual Netjtralisation of Acid 

 AND Alkaline Fluids, supplem^nxed ax other Agencies. 



It has long been known that if an a'cid and an alkaline solution be united (by 

 means of a wick or siphon, &c.) a considerable current is producible for a short 

 time in the external circuit connecting two plates of platinum, &c., immersed in 



