This table shews the indications of the soap test already noticed in the first paper. The total hardness is the effect produced on soap by all the salts of lime and magnesia present, and all the carbonic acid and silica; the permanent hardness is that left after boiling, and is produced chiefly by the lime and magnesia not separated in the insoluble state, but still remaining in solution. Another column exhibits the proportion of Chlorine calculated as if it were all in the state of chloride of sodium or common salt. The chlorine is in small quantity except when tidal influence prevails.

So far as regards mineral constituents, the water of the Hooghly at Calcutta varies greatly according to the season. Compared with the waters supplying London, the solid contents during the rainy season are much smaller, and the total hardness much less; and even in January and February, these are somewhat under those of the London waters. As regards permanent hardness, the Hooghly water is very decidedly superior to the London waters probably all the year round, except possibly during the hot season at flood tide, though that latter point is at present somewhat uncertain. But the temporary hardness is easily removable; and for economical use, except during flood tide

It must be remembered that these results are for 100,000 grains water. For an Imp. gallon of 70,000 grains multiply by 7 and move the decimal point one place to the left.

of the hot season, as regards mineral constituents, the product of the Hooghly may be considered very good water. It will be compared with the Calcutta tank waters afterwards.

Organic constituents.

The attempt made by experiments with artificial mixtures to imitate the composition of the waters of the hot season, and ascertain the probable amount of change in the organic matter by keeping, as narrated in Part II. "Supplementary Observations," was not continued, partly because all the circumstances of the case could not be imitated, and partly because the plan did not seem to be considered satisfactory to those who objected to the correctness of my results in this particular. It appeared to be better to continue the observations, taking care to avoid delay in the process for estimating the organic matter more particularly. Besides, recently the objections to the correctness of my results have been in a great measure withdrawn,* and it is hardly necessary for me to do anything more in the way of directly answering objections, as it was never my object to criticise the labours of others, but simply to state my own, carefully obtained by methods of procedure the most correct and reliable known, up to the present time.

In the original paper I considered the various methods of ascertaining the nature and amount of organic matter in water, and discussed their several merits; and a few further remarks will now be made on the same subjects. The amount of organic matter by weight came first in order, but I shall at present postpone it, until the plan of oxidation by permanganate of potash has been noticed.

This plan has come greatly into favour, chiefly I suppose from its facility of application, a very valuable recommendation, provided its other merits be assured. In the original paper I gave it a qualified and guarded approval; the result of numerous experiments made since has not increased my estimate of its value, nor has that experience, and reflection thereon, led me to concur in the generally favourable estimate in which it is held. It is said "that it is not improbable that "the substances most readily oxidised, are just those most likely to be "injurious in their effects upon those who drink the water." This is Dr. Miller's remark. Others "believe" that the most pernicious are

* Indian Medical Gazette, Calcutta, 1st January, 1867, P. 14.

those that are most easily oxidized. These, it appears to me, are rather weak grounds on which to found the preference which is at present given to this mode of estimating the degree of organic impurity in water. Others speak of it as indicating the amount of putridity in the water, and this, in my opinion, comes nearer the truth. By this I understand that the amount of oxygen required is in proportion to the amount of certain products of the putrefactive fermentation of the organic matter in the water. This, however, as Dr. Frankland has stated,* furnishes no indication of the amount of organic matter actually present in the water. The offensive smell and other properties of these products make it more than probable that they are injurious to health; but even then it is not certain that there may not be other constituents, equally or even more injurious, but more difficult of oxidation. Nor is it even certain that these products of putrefaction are the only substances which are readily oxidized by the permanganate.

Moreover, a portion of these products are evidently of a very unstable character and quickly disappear, or at least lose their power of deoxidizing the permanganate. This was first brought particularly to my attention by the objections raised to my determinations of organic matter in the original paper, and has been noticed in the supplementary observations. Since then, I have made numerous observations on this point, and give a few selected ones by way of illustration. The details of the mode of observation are given in the original paper.

* Chemical News, March 23, 1866.

This table exhibits very plainly the rapid diminution of the amount of oxygen required, by keeping even for one day, and the more gradual diminution afterwards. I have not observed that any notice has been taken of this circumstance by the English chemists. Dr. Macnamara first directed my attention to it, and since then I have not only made many observations of the fact, but have also made experiments as to the cause. The analyses of the London waters published monthly are of the waters supplied by the water companies, therefore, all probably two or three days old. It is evident that in the recent water, there must be substances possessing active deoxidizing properties, which speedily undergo certain changes by which they lose these

properties. I have paid some attention to the subject, but am not at present prepared to discuss it. It will be matter for further examination.

At present, however, it has been brought forward to justify so far the comparatively unfavourable opinion I have expressed, of the value of the permanganate process as a guide to enable us to judge of the quality of a water as respects its salubrity. I could bring forward other reasons and adduce experiments, but as I do not intend to apply the method to the matters under investigation in this paper, only one other instance as an additional reason for rejecting it will be adduced.

I shall extract two or three numbers from the preceding table and place beside them a few others of waters from other sources, namely, from the Circular canal which connects the river at the northern extremity of the town with the Salt Water Lake. This Circular canal receives much the greater part of the sewerage of Calcutta. Reference will again be made to it and to the Salt Water Lake. Dalhousie Square Tank is filled from the river and the water is considered good; General's Tank is filled by the rains and is generally said to be the best drinking water in Calcutta.

General's Tank Water of 6th February, 3 hours old,

.2830

In his report on the London waters to the Chemical Society on March 13th, 1866,* Dr. Frankland is stated to have expressed surprise that the soft water supply from Loch Katr ne required more of the permanganate than any of the waters of th Metropolitan districts, but here is something more surprising still. The water of the Canal * Chemical News of 23rd March, 1866.