consider this question to be settled, at any rate not authoritatively so; thus the degree of solubility of calcic carbonate in pure water, as assigned by Fresenius, is one part in 10,601 parts of water, and by Bloxham, one part in 35,000 parts, while one in about 23,000 parts is generally taken as the datum wherewith to compute the quantity of this compound present in seawater, or about three grains per gallon.

Now these statements are, as you perceive, of a very conflicting character, and upon coupling the knowledge of this fact with that of the very decided solvent effects of many alkaline salts upon calcic carbonate (as shown in my last paper to this Society,^* I became impressed with an idea that in the case of the earthy carbonates generally those results upon which the higher ratios of solubility were based had been vitiated by the action of one or more of such salts in the test liquid used, a certain quantity of the carbonate tested being dissolved thereby, which, added to that capable of being dissolved by pure water, raised the quantity dissolved to one proportionately greater than that proper to assign to such carbonates.

Under this impression I investigated the matter for myself, and whether my hypothesis be right or not, it was soon ascertained beyond doubt that these higher ratios are wholly erroneous, and not only this, but that even in most cases the lower ones are greatly over-stated.

The actual results of this investigation, together with the method used therein, I will now proceed to state as succinctly as possible.

First, then, as to the method:—

I make up a solution of the earth, the carbonate of which is to be tested, to a certain strength, as a chloride, also a very weak solution of carbonate of soda or of lithia. Both solutions are boiled for a long time to expel any free carbonic acid present, and afterwards allowed to cool to a common temperature. The solution of earthy chloride I then dilute with freshly distilled water applied successively in small and measured quantities, until the last portion sampled therefrom when mixed with the carbonate manifests but the faintest cloudy appearance (indicating a precipitate of carbonate) when viewed in a large bulk and after the lapse of a considerable time.

Now it is obvious that so long as the slightest cloud is formed under these circumstances, the whole of the earthy carbonate which is produced by the interaction of the salts used is not dissolved, a minute quantity being thrown out of solution, and so by computing the quantity of earthy carbonate present in a certain volume of the liquid we have an approximate estimate of the solubility in water of the earthy carbonate with which we