The water proved to be not only a hard water in the sense of being very calcareous, but highly chlorinated too,—that is, one containing an alkaline chloride in quantity, and this fact viewed in conjunction with that of the imperfect operation of Clarke's process upon it led me to suspect that these chlorides can exercise a very decided solvent effect upon mono-carbonate of lime. That they have this effect was clearly shown by an investigation I carried out, and the result of which I now beg to lay before you as follows. First let me give an analysis of the hard water:—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

On behalf of any here who may have forgotten the mode in which Clarke's softening process effects what is desired of it, I will shortly describe it, as it is necessary to remember it in order that the bearing of the several statements to be made here may be fully apprehended.

This process then consists in applying lime-water to the water to be softened, by which all the lime existing therein in the form of bi-carbonate is brought into that of the mono-carbonate, and this compound being far less soluble in water than the original one, is in greater part precipitated.

The full meaning of the effects I proceed to state will now be readily gathered.

1.—A saturated aqueous solution of sodic chloride was mixed in equal volume with a strong solution of bi-carbonate of lime, and Clarke's softening fluid was added thereto in varying quantities until quite in excess. The liquid remained quite clear, and no precipitate formed in twenty-four hours.

2.—The same effects followed when either potassic or calcic chloride, potassic nitrate or sodic carbonate were substituted for the sodic chloride in experiment No. 1.

3.—The Oamaru hard water was mixed with sodic chloride at the rate of one pound to the gallon; the filtered solution when treated with lime-water in small or large quantity remained quite clear.

4.—The same water was charged with a slight excess of lime-water alone (sodic chloride being this time omitted) when an abundant precipitate formed. This precipitate, when analyzed and the results computed in grains per gallon upon the water taken, gave as follows:—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

5.—Solid carbonate of lime (as limestone) having been placed in contact for twenty-four hours with a cold aqueous solution of an alkaline chloride was found to be dissolved to a far greater extent than it would have been by the same volume of pure water, and further, the saline solution of lime thus made rapidly deposited a very considerable portion of this earth when heated to about 180° Fahr. Carbonate of soda was found to have the same effect upon limestone as sodic chloride.

The results of the first two experiments clearly show the solvent effects of strong solutions of several saline substances upon bi-carbonate of lime as newly made. The results of experiments Nos. 3 and 4 show that a naturally hard water when charged to only a comparatively small extent with sodic chloride refuses to part with any of its lime when treated with Clarke's process; they show further a certain extent to which this retention of lime may occur, a matter which will be fully considered shortly.

The results of the last experiment are confirmatory of much which has been stated before, by demonstrating the fact that cold aqueous solutions of an alkaline chloride or carbonate can dissolve the solid carbonate of lime, and that high temperatures are inimical to the retention of the dissolved earth by such solutions. Where necessary the solutions were boiled prior to use in order to expel any carbonic acid present.

I should state here that the common test for lime—oxalate of ammonia— could not be successfully used in examining for lime in experiment No. 5, as oxalate of lime is also very sensibly soluble in salts generally; this manifested itself to me very early in these experiments, by the fact that water containing a trace of sodic carbonate but otherwise pure, after being boiled with limestone and treated with this oxalate, afforded no reaction of lime, although this body was proved to be present in quantity sufficient to afford a good indication with the oxalate were the soda absent. This salt is, indeed, very similar to the bi-carbonate in respect to solubility in saline solutions.

Of all the salts tested for a solvent property of this kind the fixed alkaline carbonates appear to possess it to the largest extent.

The greater solvency of calcic carbonate in cold solutions of certain salts is an interesting fact and one about which the following particulars should be given. The finely powdered calcic carbonate was digested with a solution of the salt (first previously boiled for twenty-four hours) at a common temperature in a vessel closed from the air. The solution rapidly filtered off, affording a very perceptible precipitate when treated with oxalate of ammonia, and a still greater precipitate when slightly warmed. Another portion of the calcic carbonate was digested for the same time in hot solution of the alkaline carbonate, but the liquor gave no precipitate with oxalate of