It is unnecessary at present to say more respecting the quantity of organic matter in the Hooghly water near Calcutta. It does not appear to exceed two grains per gallon even in the most unfavourable circumstances of season and tide, and during ebb tide is only about half that or less. Dr. Parkes says that it should not exceed one and a half grain per gallon in drinking waters, and the London authorities seem to be endeavouring to reduce it almost to nothing. In view of the difficulty of judging of its nature, it is desirable to have as little of it as possible.

In the original paper I have discussed the nature of the organic matter, and have not now much to add. The season of the year since that communication is the least favourable to the examination of the impurities in the water, as during these months-October to February -all the waters are at their purest. I formerly expressed an opinion on the amount of ammonia being probably a good indication of the impurity of a water, and gave reasons for it. The following are the results of some examinations of the river water, during these months, for ammonia.

TABLE VII.

Showing proportion of Ammonia in 100,000 fl. grs. of water.

.0250

18th February, 1867, Salt Water Lake Marsh,

The quantities are small, yet consistent enough with the previous observations. Reference may be made to them in remarks that follow.

But the Ammonia is not the only thing to look to in connection with. organic matter. As formerly observed, there are the products of oxidation of nitrogenised matter, including that of ammonia itself,

namely, nitric and nitrous acids, and there are nitrogenised substances in progress of decomposition, as well as non-nitrogenous substances, usually called vegetable, carbohydrates, and hydrocarbons more or less oxidised. Reasons have been given in the first paper for not attaching very much importance to the estimation of nitric acid, yet the enquiry is interesting, to account for the destruction of the nitrogenised matter. If the nitrogen is oxidised, it ought to be found as nitric or nitrous acid, unless it be supposed that it escapes in some gaseous form. Time has not permitted me to investigate this subject, but the few trials I have made for the detection of nitric acid have not been at all successful,-possibly it may be for want of sufficient care. On the other hand, the test recommended for nitrous acid, Price's test by starch, acid, and Iodide of Potassium, at once gives indications of that acid. But unfortunately it equally has given me free indications in the distilled water used, though means have been taken carefully to free the water and all the re-agents employed from it. At present I can give no explanation of these difficulties: the point is reserved for enquiry.

The most difficult part of the subject is the estimation of the nature of that more fixed portion of the organic matter which has undergone comparatively little change. The determination of the amount of nitrogen in this is one mode of examining it: I have not at present attempted this, as the waters have during the last four months been in the least favourable condition for such examination, and my time has been occupied with other parts of the enquiry. But I can only confirm what was before stated, that the organic matter evidently varies considerably in its nature: that of the rainy season resembles more that contained in tank waters; that of the dry season is more like that contained in sea water. The former evidently contains more matter of vegetable origin, but so far as I have been able to form a judgment, this only partially accounts for the difference.

But before proceeding to discuss these points, it may be better to have the whole of the data before us, by including the composition of the tank waters as well as of the river water. In the first paper one table exhibited some of the principal characteristics: now, additional information can be given. It will be exhibited by the following tables,

The first three are well known tanks referred to in my first paper; Ramdhone Ghose's Tank, Jaun Bazar, is that also noticed there,* near Mr. Dall's school, the water of which has since become much cleaner, and the Baranagar Tank is also that referred to there, the well being an old one on my own premises.

As to the remainder, some explanation is necessary. I was requested by the Municipal authorities in December last to examine some waters in the northern part of the town, with reference to a proposed public tank to be excavated there, and the results are included in the table. Of these Nyan Chund Dutt's Lane Tank and Well and Blaquiere's Tank are old sources of supply; the three latter "wells" were simply holes dug in the ground 8 or 9 feet deep to collect water for examination: the water from the aqueduct was supplied at my request for comparison.

Another table will exhibit a few more points for comparison.

* Called Dhurrumtollah Tank by mistake.

And to get a full insight into the nature of such waters, a complete analysis was made of two of them, of which this is the result for 100,000 fluid grains of water.

Phosphoric acid: very distinct indications of in both.

This was also found in smaller quantity in the other four of these waters, and in still smaller quantity in the water of the aqueduct.

The relative proportions of these constituents, so different from those of ordinary spring or even river waters, point clearly to their origin. The large quantity of alkaline salts, compared with those of Lime and Magnesia, and the large proportion of Potash and of organic matter, indicate that they are derived from the decomposition of vegetable and animal substances: the phosphoric acid and perhaps the chlorine being more particularly characteristic of their animal origin. In fact, it is the composition of sewage water, and differs from some other analyses of English town sewage* most remarkably in the large proportion of potash present, no doubt the product of the vegetable food of the mass of the population. It is indeed simply sewage water, deprived in great part of its bad smell by filtering through the earth. And the partial analyses of the other tanks and wells (except the three first tanks of the table) indicate that the waters all partake more or less of the same character, and contrast strikingly with the superior purity and different characteristics of the river water during the cold season.

The large quantity of Ammonia as exhibited by Table VIII. in the two waters fully analysed is also very striking.

Now to return to the consideration of the organic matter which has undergone comparatively little change, and for the proper examination of which we have no very ready or suitable chemical processes: the best means of judging of its nature and properties have already been discussed in my first paper. These are, chiefly, the general appearance and smell of the residue obtained by evaporating the water, the smell on burning, and the estimation of the amount of nitrogen by the Soda lime process. This last, for reasons already stated, I have not applied. But I may observe again, that these bad tank waters more resemble the river water of the rains than of the hot season, in the appearance and properties of their organic contents. No doubt this proceeds in part from the larger proportion of matter of vegetable

Lawes and Gilbert, Journal of Chem. Soc. Ser. 2, Vol. IV. p. 118 for 1866. Way's Report on Sewage of Towns quoted in Parkes's, Manual of Practical Hygiene, 2nd edition, 1866, p. 325.