be carried over into the alkali. The common air ejected from the bottle containing the solution was collected and examined; but from the beginning to the end of the operation not a particle of carbonic acid was disengaged from the solution, although the chlorine was readily absorbed. Ultimately a liquid of a very pale yellow colour was obtained, being the same as M. Labarraque's soda liquor, and with which the investigations were made that will hereafter be described.
3. An experiment was then instituted, in which the effect of excess of chlorine, upon a solution of carbonate of soda of the same strength as the former, was rendered evident. The solution was put into two Woulfe's bottles, the chlorine well washed and passed through, until ultimately it bubbled through both portions without absorption of any appreciable quantity. As soon as the common air was expelled, the absorption of the chlorine was so complete in the first bottle, that no air or gas of any kind passed into the second, a proof that carbonic acid was not liberated in that stage of the experiment. Continuing the introduction of the chlorine, the solution in the first bottle gradually became yellow, the gas not being yet visible by its colour in the atmosphere above the solution, although chlorine could be detected there by litmus paper. Up to this time no carbonic acid gas had been evolved; but the Brat alkaline solution soon acquired a brighter colour, and now carbonic acid gas began to separate from all parts of it, and passing over into the second bottle, carried a little chlorine with it. The soda solution in the first bottle still continued to absorb chlorine, whilst the evolution of carbonic acid increased, and the colour became heightened. After some time the evolution of carbonic acid diminished, smaller quantities of the chlorine were absorbed by the solution, and the rest passing into the atmosphere in the bottle, went from thence into the second vessel, and there caused the same series of changes and actions that had occurred in the first. The solution in the first bottle was now of a bright chlorine yellow colour, and the gas bubbled up through it as it would through saturated water.
4. When the chlorine had saturated the soda solution in the second bottle, and an excess of gas sufficient to fill several large jars had been passed through the whole apparatus, the latter was dismounted, the solutions put into bottles and distinguished
