water, acidified with hydrochloric acid, is heated to boiling in a large beaker, and a solution of barium chloride run in until a precipitate ceases to form.; the liquid is allowed to simmer for a few minutes and then put on one side to deposit, which may take some little time; the clear liquid is decanted off, the deposit boiled up in a little distilled water, collected on a filter, washed, ignited in a platinum crucible and weighed; one part of barium sulphate is equal to 134 SO.

(122) Classification of Natural Waters and the Interpretation of a Chemical Analysis.

Great ingenuity has been displayed in the division and subdivision of waters. The simplest divisions are however the best, and it is quite sufficient to divide waters into: (1) surface waters, (2) spring waters, (3) river waters.

(1) Surface waters are those which are received almost direct from the heavens, that is they fall on what is termed a gathering ground and form streams, superficial wells, or lakes. Such waters are often more or less discoloured by peaty matters but contain only small amounts of dissolved salts. The waters of Dartmoor for instance often have a total residue of 5 or 6 grains per gallon, about 2 or 2.5 of which are due to lime carbonate; similarly the water of Loch Katrine supplying Glasgow only holds in solution 2.3 grains to the gallon. Such waters necessarily have very small amounts of chlorine, nitrates, and sulphates. There is another peculiarity, which is that they are largely influenced by season.

Occasionally it happens that these soft surface waters have an acid reaction, this acid reaction is said to be commonly caused by the rain washing through the sulphurous acid laden air of a town; at all events the best known instances of acid waters are derived from the north of England, and occur under conditions which are favourable to such an occurrence. An acid water will attack lead.

(2) Springs. These exhibit every variety of composition, according to the geological formation in which they arise. As a rule the waters from all springs are alkaline, and they contain in solution more or less carbonates of lime and magnesia; those from the chalk are particularly hard, but organically pure.

Deep springs are very constant in their composition, and do not exhibit strongly signs of seasonal variation.

(3) Rivers.-Most river waters partake of a mixed character, for they are derived both from deep springs and from the rainfall flowing off the watershed through which the river runs. The composition of river water exhibits distinct seasonal variations in composition. At times of greatest rainfall, a river is most impure and at times of least rainfall most pure.

(123) Interpretation of an Analysis.

There is no difficulty in condemning a water which is very bad or in pronouncing a water very good; the difficulty begins when the pollution is small in amount and doubtful in character.

A water that has an offensive smell, or exhibits many moving organisms or on cultivation shows a number of colonies of diverse species, is absolutely unfit for use.

A well water sunk near a house, showing a vast difference in composition from a neighbouring well sunk at the same depth in the same soil, must be looked on with suspicion.

If a number of wells in a village are examined all of which are of nearly the same depth and in the same soil, the purest may be taken as a kind of standard by which to judge the others.

Soft peaty waters may contain a considerable amount of organic matter and may be coloured rather highly, yet if the chlorides and sulphates be low, and agree well with a sample taken from a portion of the gathering ground which from its uninhabited character must be pronounced unpolluted by sewage, such waters cannot be pronounced injurious.

In forming judgment upon a doubtful water, reliance must not be placed upon any one factor of the analysis, but the report must be based upon a valuation of all the determinations, and a careful consideration of the general tendency of evidence. It was proposed by the late Mr. Wigner to assign a definite numerical value to each part of the component parts of an analysis. This idea is a good one, the chief difficulty in its practical adoption, being that the different classes of water cannot all be measured by the same scale.

large

""

022244

124

747

20

5 gr. per gallon =

2

from 4 to 8

100 gr. per gallon

= 1

⚫005 gr.

= 1

·001 gr.

·002 gr.

= 1

·010 gr.

5 grs. per gallon

= 1

= 6

= 12

Hairs and animal debris, 10 to 20 according to the quantity observed.

Mr. Wigner considered that a water which gave values not exceeding 15 an extremely pure water. The limit of a first-class water would be about 40; of a second-class water, 65, and anything beyond 65 would be a third-class water.

Dr. Frankland divides waters into two sections, according to the results of a combustion :

SECTION I.-UPLAND SURFACE WATER,

CLASS I.-Water of great organic purity, containing a portion of organic elements (organic carbon and organic nitrogen), not exceeding 2 part in 100,000 parts of

water.

CLASS II.-Water of medium purity, containing from 0.2 to 0-4 part of organic elements in 100,000.

CLASS III.-Water of doubtful purity, containing from 0.4 to 0.6 part of organic elements in 100,000.

CLASS IV.-Impure water, containing more than 0.6 part of organic elements in 100,000.

SECTION II.-WATER OTHER THAN UPLAND SURFACE.

CLASS I.-Water of great organic purity, containing a proportion of organic elements not exceeding 0°1 part in 100,000.

CLASS II.-Water of medium purity, containing from 0.1 to 0.2 part of organic elements in 100,000.

CLASS III.-Water of doubtful purity, containing from 0.2 to 0-4 part of organic elements in 100,000.

CLASS IV.-Impure water, containing upwards of 0.4 part of organic elements in 100,000.

The above mainly deals with interpretation of the organic elements of a water, but a water may be organically pure, and yet be unfit to drink because of objectionable mineral or saline constituents.

It may be at once laid down, that with regard to metals, if iron is in sufficient quantity to give a chalybeate taste, such water is not fit for a public supply. A water which contains no carbonates of the earths and has a slight acid reaction' will assuredly attack lead, and unless specially treated is not suitable for a public supply. A water which contains anything over 5 grains per gallon of magnesian salts, is not a good supply, and if there is choice another supply should be substituted.

(124) Purification and Softening of Water.

The most perfect method of purifying water is by distillation, a method which in the rainless districts of Peru and Chili is relied on to supply a considerable population, and one which is used on most of Her Majesty's ships at the present day. The other methods of purifying water are:-I. Subsidence; II. Filtration; III. Chemical treatment.

I. Subsidence.- Waters containing suspended matters are often very effectually cleared and rendered fit for use by being stored in reservoirs; the water being at rest, the suspended matters slowly or quickly according to their specific gravity sink to the bottom.

II. Filtration.-Filtration on a large scale has been already treated of. The best domestic filters are "the silicated carbon filter," "Spencer's filter," the filtering material of the "Sewage and Water Purification Company," and "Bischoff's spongy iron

1 Dr. Tidy has made experiments which tend to show that a water containing no silica attacks lead; if subsequent research should confirm this, it will be of great importance to determine the silica in water.

filter." All these not only act as efficient strainers, but also without doubt have some not quite understood chemical action; probably in most of them each little particle is a feeble galvanic couple, and by alternate reductions and oxidations induces chemical change. The test of a good filter is that it filters slowly, that when a solution of quinine sulphate, 8 grains to the gallon, is filtered the filtrate should be destitute of bitter taste and that it retains its properties for a year, or longer.

III. Chemical Treatment.-For the purpose of clearing turbid waters, which will not readily deposit suspended matters, the following chemical substances are in use:-The salts of iron, especially ferrous sulphate; the salts of aluminum, especially alum and aluminate of soda; and silicate of soda.

Most turbid waters are improved by a little ferrous sulphate, the salt breaks up, and the hydrated ferrous oxide falls slowly to the bottom entangling with it more or less organic matter; treatment with alum or sodic silicate acts in a similar manner, hydrated alumina oxide or silica falling down.

Softening water also purifies it. Clarke's process is the addition of a sufficient quantity of lime to the water to unite with the free carbonic acid, the result being that the lime which has been added falls down as lime carbonate, and also that which is already in the water and has been kept in solution by the free carbonic acid. The exact amount of lime to be added can be calculated if the amount of free carbon dioxide is known, or it may be found by adding lime water of known strength from a burette to a measured portion of the water until a sample tested with silver nitrate, gives instead of a white or yellowish precipitate a brown; as soon as this brown precipitate is produced, the addition of the lime water is stopped, and then the reverse process is followed, that is the hard water is added little by little to the same water already limed until the brown colouration ceases to be produced. Clarke's process of course only deals with "temporary hardness," any hardness due to sulphates or nitrates of lime and magnesia remains, but by the addition of a sufficient quantity of caustic soda to decompose the earthy sulphates, the permanent hardness is also removed.

In Clarke's process the carbonates thrown down are in a state of very fine division, and take some time to subside. There are, however, various forms of apparatus invented which quicken the process