ing if it boils. This is most simply and easily ascertained. The remainder of the process is the same whether lead bath or any other be employed. The instrument, into which a little well-ignited asbestos has been introduced, is gradually and carefully introduced into the bath and fixed in position, the end of its fine tube dipping sufficiently deep below the water in the little trough. - The graduated tube stands close by, filled with water and inverted in the trough. The cork, d, is now inserted, and it is necessary to wait till the little column of water, extending a m.m. or two up the fine evolution tube, remain quite stationary, no more bubbles and no further motion occurring. This is a sign that a uniform temperature obtains within. The cork is now taken out, and the little tube with substance dropped in, and the cork quickly replaced again, whereby a bubble or two of air escape by displacement. These are not heeded, but immediately afterwards the inverted tube is placed quickly over the point of the evolution tube, a. Very soon the substance below is vapourised, and displaces the air of the apparatus above, which issues in a rapid stream of bubbles into the graduated tube. This evolution at an end (generally occupying 10 to 15 seconds), the little water column will first assume a tranquil condition; others gradually show signs of receding. It will recede after some little time a certain distance up the narrow tube, a, the effect of diffusion, set up below, between vapour and air. As soon as the gas evolution is over the cork is removed, the thumb placed over the orifice of the graduated tube, and the latter removed and completely immersed in a large deep cylinder filled with water and containing a thermometer. The barometer and thermometer in the water are now, after waiting a minute or two, read off, and then all observations are at an end. The following simple formula gives the vapour density

D=S760(1+0.0036651),
(B-w) V0001293

or gathering together the constants

where

D=S(1+0.0036651)*587780
(B-w) V,

S=weight of substance. t-temperature of water in cylinder, i.e., of the room. B=Barometric column reduced to o°. w= tension of aqueous vapour at to. V measured volume of air, in grad. tube. In the case of substances slightly decomposed by atmospheric air at their boiling and vapourising points, dry nitrogen gas is introduced into the apparatus, and the vapour density is thus determined in nitrogen. V. Meyer prepares the nitrogen according to the Gibbs-Böttger method by boiling a solution of I part potassium bichromate, I part ammonium nitrate, I part commercial sodium nitrate, and in 3 parts of water. The gas was also passed over a layer of ignited metallic copper before conducting into the gasholder. From the gasholder it passes first through drying tubes, and then down a long narrow glass tube reaching to the bottom of the vapour density apparatus, itself perfectly clean and dry. The instrument is thus filled by displacement and most simply and expeditiously.

The wide range of this apparatus is shown in the numerous examples of determinations made by Prof. V. Meyer's assistant, Herr Carl Meyer, by its means. Chloroform boiling at 61° is first determined (in steam), and, lastly, in the lead bath, the perchlor-diphenyl of Merz, boiling considerably higher than 440°. This determination was made in an atmosphere of nitrogen.

Results.

To demonstrate the great value of the method for ascertaining the molecular weights of certain high boiling hydrocarbons, whose compositions may lie so closely together that the centesimal numbers obtained by an organic combustion will fail to distinguish them, it may just be mentioned that the body chrysene, C18H12, as regards the numbers furnished by organic analysis, cannot be distinguished from dinaphthyl, C20H14. The calculated vapour densities of the two bodies are respectively 7.89 and 8.77. The numbers found by Victor Meyer's method were--for chrysene, 812; d'tto for dinaphthyl, 8.73. (I am told the substance was not quite pure.)

I am told the apparatus for the above method is to be had on application to Herrn H. Fischli, Zürich, Ramistrasse 45.

Zürich, Feb., 1879.

PROCEEDINGS OF SOCIETIES.

CHEMICAL SOCIETY.
Thursday, February 6, 1879.

Dr. J. H. GLADSTONE, F.R.S., President, in the Chair.

AFTER the announcement of visitors, the minutes of the previous meeting were read and confirmed.

The Treasurer, Dr. RUSSELL, then announced that he had received a bequest of £1000 from the executors of the late Mr. Sidney Ellis, of Nottingham.

The following certificates were read for the first time:A. W. Stokes, T. L. Teed, C. H. Hutchinson, G. Rait, W. Stone, T. Griffiths, W. Palmer.

During the meeting the following gentlemen were balloted for and declared duly elected Fellows of the Society:-F. R. Japp, C. F. Cross, H. Wilson, C. E. Cassel, A. E. Menke, J. J. Broadbent. E. A. Letts, E. H. Rennie, W. Stevenson, S. Spencer, C. W. Smith, A. J. G. Lowe, T. Gough, R. Gracey, P. P. Bedson.

The PRESIDENT said that it was scarcely necessary to remind the Fellows that the discussion on Dr. Tidy's paper had been adjourned to this evening, and that it was his duty to preside over the champions of the three processes for determining the organic matter in potable waters, and to see-if he might be allowed the expression

that the lists were clear and the fight was fair. The question to be discussed this evening was that of the relative merit of the processes, and there was the question behind, viz., how far these processes tell us whether a water is fit to drink or not. It was one thing to determine the carbon and nitrogen, and quite another to decide whether such quantities so detected were hurtful or not. He should like to mention that two curves which formed part of Professor Tidy's paper had not been published with the letterpress on account of the delay which had The curves would be arisen during some corrections. publ shed in the forthcoming number. He would call on Mr. Riley, as he moved the adjournment, to open the discussion. Mr. Riley, however, not being present, Dr. Frankland opened the discussion.

Dr. FRANKLAND said that in responding to the President's invitation to open the discussion he wished to express the gratification which the perusal of Dr. Tidy's paper had afforded him. The author had performed his task in a truly philosophical spirit, with fairness and candour. He had frankly told us of the alteration of some of his former opinions, and, above all, he had contributed a large amount of important and laborious work to the subject of water Calc. Found. Boiling pts. analysis. He thought that chemists generally were much 4'13 4'13 indebted to Dr. Tidy for the clear and impartial style in which he had handled an intricate and exciting theme. over 440 As he would have to criticise some parts of the paper

17:24

17'43

61°

68

Water Analysis.

adversely he wished to record at the outset his very | favourable opinion of it as a whole. To proceed with the details of the paper, he quite agreed with the author that if it were possible to estimate the organic matter in the water before evaporation, such a mode of estimation would be preferable to any operations of the kind upon a water residue; but unfortunately he knew of no method by which this could be accomplished. He also agreed with the author's opinion that potable waters contained a great variety of organic matters, although he would not go so far as to say that the variety was infinite." He entirely dissented, however, from the author's suggestion that non-volatile matter may be removed mechanically during evaporation. He had proved by actual experiments that mechanical removal only took place during violent agitation or the breaking of gas bubbles at the surface of the liquid. The opinion that solid or liquid particles were mechanically removed during quiet evaporation was emphatically negatived by almost every quantitative determination in analytical chemistry. He was still more astonished, however, to find Dr. Tidy, who is a physician as well as a chemist, suggesting that the poisonous constituents of sewage may be volatile :-" Remembering how virulent that poison is, I should myself be far from astonished if it were of a highly volatile nature, just as we know that arseniuretted hydrogen far exceeds arsenious acid in intensity of action on the human body." All the knowledge we have hitherto acquired about the infectious matter of epidemic disease points to the inevitable conclusion that the propagating material is not merely organic, but organised; and that its virulence, unlike that of arsenic or other similar poisons, resides in its vitality and its power of multiplying itself almost indefinitely in the human body. Now it is impossible to conceive of organisation in substances whose molecules are free to move as they list :-An organised liquid, gas, or vapour, is a physical impossibility, and hence there is no foundation whatever for the opinion that the poison of sewage is volatile, or that it is not contained in the residue left on evaporation. Indeed, the author, somewhat inconsistently, admits this generally-accepted view of the nature of morbific poison when he says, soon after making the foregoing suggestion,-" But it must be remembered that the evils arising from drinking contaminated water are probably the work of invisible particles-germs or whatever else you like to call them." We cannot, however, estimate separately the disease producing organic matter of polluted water. It is probably, even in the worst waters, so small in quantity as to be almost imponderable. Dr. Tidy appeared to be equally inconsistent in his great fear of loss of organic matter by oxidation in a reducing atmosphere of sulphurous acid, when he afterwards went on to show the extreme difficulty of oxidising such organic matters by an acidified solution of potassic permanganate! Moreover, the oxidation of organic matter by no means necessarily increases its volatility. The difficulty which the author has occasionally expressed in the reduction of nitrates when these salts occur in large quantity appears to have arisen from his fear of oxidation by sulphurous acid, and his consequent neglect of the precautions given. He says, "I cannot conceive that the organic matter in the residue can survive direct actual treatment with a sulphurous acid solution. I never do this part of the experiment without a feeling of despair." This conception and this feeling were, however, utterly opposed to facts. Hundreds of combustions of the residues of shallow wellwaters containing large quantities of nitrates have demonstrated that the organic matter does survive this treatment. Moreover, this difficulty is in practice more imaginary than real; for when a water contains these excessive quantities of nitrates, it is to be condemned on this ground alone. With regard to the remaining obstacles mentioned in connection with the combustion process, it is easy to procure suitable cupric oxide, either by oxidising sheet-copper in a muffle, or by procuring roasted copper scraps from the makers of sulphate of

CHEMICAL NEWS,

{February 14, 1879.

copper, The speaker always procured these scraps from Messrs. Tennants, London Road, Manchester, and it was only necessary to heat them to redness in an iron tube for an hour or so in a current of air to fit them for use. There cannot be any occluded nitrogen in the copper, as suggested, because the copper is always heated to redness in vacuo before the combustion begins. The author is now convinced of the accuracy of the determination of organic carbon, and the speaker confidently anticipated that further acquaintance with the combustion process would prove to him that the corresponding determination of organic nitrogen was quite accurate enough for all practical purposes. With regard to the albuminoid ammonia process, his experience coincided with that of Dr. Tidy. It had been put forward as capable of answering the question, "Is the water wholesome or is it not?" It does not, however, answer this question: It condemns water containing only peaty matter, and it acquits water containing urea and uric acid! If we look at the results of its application to the waters supplied to London and recorded in this paper, could anybody believe that in 1876, for instance, the Chelsea Company's water, drawn from the Thames, was "safe" in January, "dirty" in March and August, "safe" again in November, and "of extraordinary purity" in December, when it contained no less than 0.423 part of organic elements in 100,000 parts of water; and when three other samples drawn from the same source were denounced as "dirty" by the same process. Again in 1871, whilst the West Middlesex Company's water

But

· was of extraordinary purity" in January, April, and May, the Chelsea Company's water was "dirty" in those months, although quite "safe" in December. Floods undoubtedly make Thames water dirty, but the albuminoid ammonia curve pursues the even tenor of its way quite regardless of them. The very instructive comparative curves, given at the close of the paper, prove conclusively that there is no sort of connection between the pollution of water and the proportion of so-called albuminoid ammonia. He now came to the most important part of the paper, viz., that treating of the oxygen or permanganate process. Experiments had recently been made in his laboratory by Mr. Woodland Toms which showed that there was not, as suggested by the author, any substantial difference between Dr. Tidy's method of performing this process and his own, which was that originally described by the late Dr. W. A. Miller. The author speaks of this process as if by it "everything capable of oxidation" was completely oxidised. there must be some mistake here, because of the numerous organic matters hitherto submitted to it cxalic acid was the only one completely oxidised. The author had made numerous experiments on the influence of various mineral salts upon the test, and he had shown that the presence of nitrates made no difference with its indications. It was far otherwise, however, with sulphuretted hydrogen, ferrous salts, and nitrites. The first two were easily detected and removed, but the nitrites presented a formidable difficulty, which, in the speaker's opinion, the author had not overcome. He (Dr. Frankland) admitted that they were not often present in potable water; but unfortunately they were especially liable to be present in well-waters, and the oxygen process might thus condemn deep well-waters of most excellent quality. The author's own figures, obtained with the Kent Company's deep well-water, showed this, especially when the very erratic curves of this very uniform water were studied. Thus, in 1870, for instance, the oxygen consumed by this water varied from o to 50, whilst the organic elements varied only from 1 to 3'3. In 1871, the oxygen consumed varied from 1 to 7, but the organic carbon and nitrogen only from I to 2.6. In 1874, the oxygen consumed was remarkably and uniformly low, whilst the combustion regults were, for this water, remarkably and uniformly high. Again, in 1875, the oxygen consumed varied from 1 to 7, but the organic carbon and nitrogen only from 1 to 2; and, lastly, in 1876, the oxygen consumed varied from 1 to 5, but the

organic matter of water is probably a very small fraction of the whole. A water may contain much organic matter, and be nearly, if not quite, colourless, or it may contain but little, and be deeply tinted. Thus, as mentioned by the author, a sample of water which had been in contact with precipitated alumina was, when examined by the speaker, of the very light blue-green tint of pure water when seen in a two-foot tube, although it yielded by combustion o 160 per 100,000 of organic carbon, and another sample which had been in contact with mine mud was, when clarified by subsidence, colourless in a quart decanter, although 100,000 parts of it yielded on combustion no less than 0544 part of organic carbon. The author, however, said "The peculiar tint of the water is an indication of the kind of organic matter, and the tintdepth is an indication of the quantity of organic matter in the water." The speaker was not aware of the grounds upon which this statement was made. It was not im possible that spectroscopic examination might help us. The absorption-bands of brown urine and of peat might be different, and it was surprising that the author, who evidently attached so much value to the colour test, had not pushed his investigation in this direction. For his own part he had been unable, except in waters from the same district, to guess the approximate portion of peat from the tint-depth of the water. The tinctorial power of the peat in one district seemed to differ from that in another. But it was scarcely necessary to pursue the subject further, as the author himself concludes his advocacy of the 2-ft. tube by virtually condemning it as a quantitative instrument. The quart decanter, or rather flask, was, in the speaker's opinion, more useful than the 2-ft. tube, because if water appears coloured or turbid in such a vessel the consumer would have just cause of complaint, but not otherwise. He cordially agreed with the author's strictures on superficial and rapid analyses of waters, in which the determinations were "chosen with far greater regard to the analyst's time than to the needs of the inquiry." The so-called partial analyses were often very partial, and often brought disgrace upon the profession of chemistry. It was satisfactory to know that the author not only preached but practised this doctrine; the fulness and minuteness of his analyses were well worthy the attention and imitation of chemists. He also agreed with Dr. Tidy in deprecating a hard and fast division of waters into classes, although a suggestive classification was often useful. He would prefer to modify somewhat the division into four classes adopted by the author, making a separate and more liberal scale for upland surface water, which rarely contained anything hurtful, and a somewhat stricter one for spring, well, and river water, the organic matter in which was very liable to come from dangerous sources. He would therefore propose the following classification :

organic elements only from 1 to 3. There is a strong, matters are coloured, and the coloured part of the probability that the organic matter in the Kent Company's water is acted upon but very slightly by permanganic acid, and that the effects recorded by Dr. Tidy were chiefly produced by nitrites. The speaker had observed on one occasion that a sample of this water, on its first arrival at his laboratory, consumed three times as much oxygen as it did after standing a couple of days, although he had proved that the organic matter in this water underwent no diminution during eleven days' agitation with atmospheric air. Nevertheless there was undoubtedly a remarkable approximation between the curves representing for many years the organic carbon in the river waters supplied to London and those recording the oxygen taken up from potassic permanganate; and the speaker was not without hope that as a short and easy cut to the valuation of organic carbon the process might be made available for potable waters generally, but he thought the author had, as yet, failed to establish it in this position. There were also some other remarks about the capabilities of the oxygen pro- | cess, which must not pass unnoticed. Dr. Tidy admitted that it certainly does not affect all organic substances; but he says it affords the most positive evidence (evidence to my mind beyond dispute) of the relative quantity of matters in the water likely to be injurious (for it is the offensive and deleterious we wish to hurt out, and not the inert and the harmless), and enables us to speak with confidence in advising the use of, or the rejection of, a water for drinking purposes." Where is the evidence of this beyond dispute? Dr. Tidy has not given us the slightest fragment of it. If it exists it ought to be produced, because it would put the oxygen process upon the firm and trustworthy basis which it, at present, certainly lacks. No process of water analysis can distinguish between "offensive and deleterious" and "inert and harmless" organic matter; but there is a method which can take account of starch, sugar, gum, gelatine, uric acid, and urea. Dr. Tidy said, and he quite agreed with him, that these substances are in themselves quite innocuous; but how do they get into potable water? People are not in the habit of throwing sugar, starch, gum, or gelatine into rivers. There is a sufficient observance of the proverb "Waste not, want not 99 to prevent such a use of these valuable materials. Care is taken that these substances first pass through a canal | before they reach a river, stream, or well, and that canal is the alimentary canal! Will Dr. Tidy assert that these substances embalmed in the secretions of that canal are ever fit for human consumption, or are not, at times, charged with the potentiality of disease? Hence he (Dr. Frankland) said that no process of water analysis is trustworthy which does not take full account of these substances, inert and harmless though they may be, when they leave the grocer's shop. Another statement in the paper in reference to the oxygen process ought not to pass unchallenged, although the author had ceased to act upon it. He said, "In the case of the London waters, however, I think I have abundant experiments to prove that the oxygen used, multiplied by eight, denotes very nearly the actual quantity of organic matter present; nor would it be difficult to obtain further evidence (if it were required) on this point from the work of others." Will Dr. Tidy mention the details of some of these experiments? To chemists generally there is no method known by which "the actual quantity of organic matter present" in London water can be even approximately determined. How, then, can that quantity be proved? If the author knows of a process why does he not disclose it? Such a discovery would go far to sweep the oxygen, albuminoid ammonia, and combustion processes into oblivion, for it could scarcely fail to be equally applicable to other waters besides those consumed in London. With regard to the adjuncts to the oxygen process, the speaker remarked that he considered the colour test to be useful only from a sentimental point of view. Comparatively few organic

:

SECTION I.-UPLAND SURFACE WATER. Class I.-Water of Great Organic Purity.-Containing a proportion of organic carbon not exceeding o‘2 part in

100,000.

Class II.-Water of Medium Purity.-Containing from 0'2 to 0'4 part of organic carbon in 100,000. Class III.-Water of Doubtful Purity.-Containing from 0'4 to 0'6 part of organic carbon in 100,000. Class IV.—Impure Water.-Containing upwards of o‘6 part of organic carbon in 100,000. SECTION II.-WATER OTHER THAN UPLAND SURFACE. Class I.-Water of Great Organic Purity.-Containing a proportion of organic carbon not exceeding o't part in 100,000.

Class II.-Water of Medium Purity. Containing from o'1 to 0.2 part of organic carbon in 100,000. Class III.-Water of Doubtful Purity.-Containing from 0'2 to 0'4 part of organic carbon in 100,000. Class IV.-Impure Water.-Containing upwards of 0'4 part of organic carbon per 100,000.