This research was undertaken with a view to ascertaining if metallic chlorides have any tendency to combine with a further quantity of chlorine. The well-known fact that potassium iodide unites with iodine to form a tri-iodide suggested that potassium chloride might, under suitable conditions, form a similar compound with chlorine. The method of investigation which first presented itself was to expose crystals of various chlorides to the action of dry chlorine gas at low temperatures, when any chemical action taking place would be expected to show itself in changes in the appearance of the crystals. A few experiments were made with no apparent result, and the method finally adopted was to determine the quantity of chlorine absorbed by solutions of the salts. It was thought that in this way quantitative results could be obtained which would reveal any tendency towards the formation of perchlorides. If a body having the formula KCl3 could be shown to exist, it would strengthen the position of those chemists who contend that the generally accepted ideas regarding atomicity need modifying. The discovery of a compound having the formula KCl2 would be a still stronger argument. At an early stage, however, the research resolved itself into an investigation of the solubility of chlorine gas in solutions of metallic chlorides in water, the question of the existence of perchlorides becoming a side issue. The influence of salts in solution on the solubility of gases in water is a subject of considerable interest, as throwing light on the nature of solution in general. Of late, certain chemists are inclining to adopt the old view that solution should be included under the general head of Chemical Action.