bining with a second dose of carbonic acid, making up bicarbonate of lime.

"Now, if a solution of the nine ounces of burnt lime, forming lime-water, and another solution of the one pound of chalk and the seven ounces of carbonic acid, forming bicarbonate of lime, be mixed together, they will so act upon each other as to restore the two pounds of chalk, which will, after the mixture, subside, leaving a bright water above. This water will be free from bicarbonate of lime, free from burnt lime, and free from chalk, except a very little, which we keep out of account at present for the sake of simplicity in this explanation. The following table will show what occurs when this mutual action takes place:

"A small residuum of the chalk always remains not separated by the process. Of 174 grains, for instance, contained in a gallon of water, only 16 grains would be deposited, and grains would remain. In other words, water with 17 degrees of hardness arising from chalk, can be reduced to 14 degrees, but not lower.

"These explanations will make it easy to comprehend the successive parts of the softening process.

"Supposing it was a moderate quantity of well-water from the chalk strata around the metropolis that we had to soften, say 400 gallons. This quantity, as has already been explained, would contain one pound of chalk, and would fill a vessel 4 feet square by 4 feet deep.

"We could take 9 ounces of burnt lime made from soft upper chalk; we first slack it into a hydrate by adding a little water. When this is done, we would put the slacked lime into the vessel where we intend to soften, then gradually add some of the water in order to form lime-water. For this purpose at least 40 gallons are necessary, but we may add water gradually till we have added thrice as much

as this; afterwards we may add the water more freely taking care to mix intimately the water and the lime-water or lime. Or we might previously form saturated limewater, which is very easy to form, and then make use of this lime-water instead of lime, putting in the lime-water first and adding the water to be softened. The proportion in this case would be one bulk of lime-water to ten bulks of the hard water.

"It is of importance that the lime, or lime-water, that is the softening ingredient, be put into the vessel first, and the hard water gradually added, because there is thus an excess of lime present up to the very close of the process, and this circumstance is found to render the precipitation of the carbonate of lime produced in the process more easy.

"But what you will wish to know now is, by what mark is the conductor of the process to find out when there is enough of water to take up the last of the excess of lime, so as to be enough, but no more.

"This is done by what has been called the silver test, the only test necessary to the operator after the process is fairly set a-going. This test is a solution of nitrate of silver, in twice-distilled water, in the proportion of an ounce per pint. In making use of the silver test with ordinary waters, we get a white precipitate; but if the water have in it a notable excess of lime-water, there is a light reddish-brown precipitate produced; but if the excess be very slight, we only get a feeble yellow precipitate. The way we make use of the test is to let two or three drops of it fall on the bottom of a white tea-cup; then add the water somewhat slowly; then if there be the slightest excess of lime, a yellow colour will show itself."

It may be here mentioned that a more delicate test than the silver test for ascertaining if there is an excess of lime in the water, consists in using a solution of cochineal, the natural colour of which is yellowish-red, which is turned violet in the presence of alkalies; and other agents are now used to show, by distinct colour, or its absence, if there is an excess or not of lime in the softened water.

According to Dr. Clark's scale, one degree of hardness means that there is one grain of chalk in a gallon of water. According to the scale introduced by Dr. Frankland, in the sixth report of the Rivers Pollution Commissioners, parts per 100,000 are used, or one grain of chalk in 100,000 grains of water, so that it is necessary, in considering the reports of the Rivers Pollution Commissioners, to bear in mind this difference of degrees of hardness. To reduce the hardness to parts per gallon, or to the Clark scale, it is necessary to multiply by 7 and divide by 10.

The amount of soap

Hard water decomposes soap. ascertained by Dr. Clark to be wasted before softening the water, is equivalent to 2 ounces for each degree of hardness for every 100 gallons. Dr. Clark introduced a soap test, or a means by which a solution of soap is made to at once indicate the degree of hardness of a water. When pure chalk is burnt into lime, one pound is converted into 9 ounces of lime, and this quantity is soluble in 40 gallons of water. Beyond this, lime is not soluble in water, so that clear lime-water always possesses a known composition. This amount of lime is equivalent to 98 43 grains per gallon. As one particle of lime will remove 1'777 of chalk, it follows that 98 43 × 1777 174'9, or the number of gallons of water one degree of hardness which one gallon of lime water will soften. In practice, however, while theoretically 175 gallons of water of one degree of hardness may be softened by one gallon of lime-water, owing to the impurities in the lime, probably not more than 130 to 150 gallons would be softened, so that, to arrive at the amount of lime-water necessary to soften hard water, if we divide 130 by the degrees of hardness according to Clark's scale, it will, generally, roundly represent the number of gallons of water which can be softened by one gallon of lime-water.

It is found in practice that neither by boiling nor by the time process can all the hardness which is termed “temporary hardness" be removed; in fact, a small quantity of chalk, to the extent of one part in 50,000 parts, is soluble in water, and still remains in solution after the process. In

practice, however, 10-11ths of the whole of the temporary hardness may be removed by the lime process. In carrying out the Clark process, the lime-water is usually applied owing to the fact that it is a standard liquid containing a known quantity of lime, although cream of lime is sometimes. used with advantage. Large tank space is required to carry out the Clark process, three tanks ordinarily being required, one for filling up, another for drawing down, and a third in reserve while cleansing is going on, and each tank should hold a day's supply.

It is a point of importance to know that the salts of iron may be readily removed by the application of lime. This the author found in the case of a water works at Horsham, where the water was so highly charged with iron, that everything it touched was discoloured; but by the application of lime the whole of the iron was removed. It should also be noted that the process has a marked effect in removing organic matter from water. This was shown by the Rivers Pollution Commissioners and by the analysis published by Professor Wanklyn, especially in the marked diminution in the amount of albuminoid ammonia. The great objection raised against soft water has been its liability to produce lead poisoning; but the consensus of opinion is that water softened by the Clark process is not liable to attack lead, which is a point of very considerable importance in favour of the process.

Dr. Clark's process was first carried out by the Woolwich and Plumstead Waterworks Company, where it was shown that the water was successfully softened from 23 degrees to 7 degrees. These works were subsequently bought by the Kent Waterworks Company, and the process of softening the water was discontinued. An experiment was also made at the Chelsea Waterworks Company's works, the result of which is set forth in the following Table :

In the course of experiments made in softening various waters, it was observed that, if the water was at all tinted, the softening process did not clear it; but there was a tendency for the matters separated to remain in suspension in such water, so that it was considered expedient, in softening river-water, that the water should be filtered until quite bright before it undergoes the softening process.

It may be here mentioned that, in 1852, a patent was applied for (W. B. Bowditch) to treat water with clay and alkali, and subsequently filtering.

Mr. Phillip H. Holland, M.R.C.S., late Inspector to the Burial Act Office, suggested, as an addition to the Clark process, the use of oxalate of ammonia or soda, to further reduce the hardness of the water after treatment by this process. The use of carbonate of soda for softening water has been known throughout the whole country by teamakers from an early period. This salt added to water acts on the bicarbonate of lime and magnesia, and precipitates chalk and carbonate of magnesia. It also decomposes the sulphates of lime and magnesia, precipitating the lime and magnesia, while the soda remains in solution, so that the permanent hardness of the water is reduced by the use of this salt. The Rivers Pollution Commissioners of 1874 state, in their sixth report, with