NEWS under such circumstances, that the two analyses have differed in regard to fat to the extent of five or ten hundredths of a per cent. This difference I have traced to the method adopted by the inspectors in dividing the samples. The milk, when purchased, is put into a jug, and from this three bottles are filled. Many minutes often elapse between the making of the purchase and the subsequent dividing and sealing of the samples, during which time a partial separation of cream takes place, to an extent depending on the condition of the milk at the time, and, as a consequence, the bottle first filled will contain a milk somewhat richer in butter than the others. This difference would not have occurred in the divided samples obtained as they used to be under the old Act, for each sample was then taken in a bottle to the analyst, and by him divided after having been thoroughly mixed by shaking, which can be done in a bottle but not in a jug. I have also found a want of accordance in the results of milk analyses arising from another and very different cause. I have been anxious to satisfy myself as to the usual amount of reduction in the proportion of fat which takes place in large vessels from which milk is being dipped out for sale both by retail and wholesale; and having had several experiments made in which milk has been dipped in successive quantities out of the tin churns containing about 16 gallons, and also out of the counterpans or tureens containing about 10 gallons each, I have found that when such quantities of milk have, by successive dippings, been divided into two parts, one consisting of what has been dipped out, and the other of what remained, on taking a sample from each as a milkman would take it, and estimating the fat, the mean of the two extremes has not coiucided with the proportion of fat contained in the original milk. In fact, it is impossible, in dealing with large quantities of milk, as the milkmen do, and using the same vessels, to get concordant results of that description. I have always found the mean of the extremes to be below the quantity present in the original sample, if in taking the sample the measure or dipper be plunged in vertically to the depth of an inch or two below the surface, for in that case the richer stratum of milk and cream on the surface is broken and dispersed and the measure filled from the poorer portion beneath. A milkman can easily supply a sample of milk from a counter-pan that shall be either richer or poorer in fat than a fair average sample of the whole would be by varying the mode of taking it, but it would be impossible, in the usual mode of serving milk, to ensure that the sample shall represent the mean of the whole. My object in communicating this note is to show that samples of milk taken in the usual way from the large vessels in which it is kept and conveyed for sale, hardly can, and generally does not, fairly represent an average of the bulk from which they are taken in regard to the proportion of fat present.-I am, &c., 17, Bloomsbury-square, December 28, 1875. T. REDWOOD. ESTIMATION OF MINUTE TRACES OF COPPER. To the Editor of the Chemical News. SIR,-Will you allow me to draw attention to the fact that a process for the estimation of minute traces of copper, after precipitation of iron with ammonia, exactly similar to that given by Mr. Carnelley in your issue of this week (vol. xxxii., p. 308), was proposed by me in the CHEMICAL NEWS, vol. xxxii., p. 3, in an article on the analysis of minium, of which an abstract appears in the Journal of the Chemical Society for December last. It is fair to own that the principle was suggested by Mr. Carnelley's process for the estimation of iron, and I frankly confess that, with the splendid opportunities of experiment and investigation offered by the laboratory of Owens College, he has treated the subject with a thoroughness and precision which were unattainable by me, owing partly to want of leisure, and partly to the fact that the method was improvised under pressure of the sudden necessity for giving a complete analysis of a series of remarkably pure samples of red lead. He will find, however, on referring to my paper, that I have drawn attention to the disturbing influence of ammoniacal salts, which I have proposed to neutralise by the preparation of a comparison liquid from one-half of the solution to be examined. It would appear that the usefulness of colorimetry, and also of judgment by turbidity, which may be provisionally termed "nephelometry," might be widely extended by the elaboration of a thoroughly reliable process for determining the exact equivalence of columns of liquid of different lengths by the correspondence of their colour or appearance as viewed from the ends; such devices as the preparation of comparison liquids, or the intentional addition of foreign salts, might then be entirely dispensed with as follows: Let two equal columns of the solution to be tested be taken, and let their exact measure be known; then let a different proportion of a standard solution of the ingredient to be determined be added to each-it will probably be convenient to add twice as much to one as to the other-and let the columns be made equivalent by shortening one of them. It will be evident that if the original solution was quite free from the substance sought, the comparative lengths of the columns will correspond with the proportions of standard solution added to each, or in the case supposed will be represented by the fraction n =; but, if the original solution did contain the substance to be determined, the value of the fraction will be altered, and the quantity present in the longer column, the bulk of which is known, may be determined by a simple equation. 2 " where n, of course, represents the known quantity o ingredient added to the longer column by means of the standard solution. The application of this principle may be rendered clearer by an example. The difficulty of obtaining distilled water absolutely free from ammonia for use in Wanklyn's process of water analysis is well known, and it would be obviously very convenient to determine once for all the proportion of ammonia in a given sample; we might proceed thus: two columns measuring four ounces each are taken; to one of them is added standard ammonia solution equal to 0'0005 grain, to the other equal to o'oor grain ammonia, the column of stronger tint is shortened, after Nesslerising, until the colour, viewed from above or below, by means of an appropriate apparatus, exactly corresponds; if the distilled water was pure one column will be exactly onehalf the length of the other, if ammonia was present the fraction will be greater. Suppose it is, then9 (x + 0.001) = x + 0·0005 14 which will give x = 0.0004 grain. The" appropriate apparatus" mentioned above is, so far as I am aware, still a thing of the future.—I am, &c., THOS. P. BLUNT, M.A. Oxon., F.C.S. Shrewsbury, January 1, 1876. COPPER IN BREAD. To the Editor of the Chemical News. SIR,-Dr. Edmunds in his contribution to the CHEMICAL NEws of December 31 (vol. xxxii., p. 311), throws doubt 8 Liverpool Soda Tests. CHEMICAL NEWS, Jan. 7, 1876. on the fact that copper has ever been used as an adulterant | samples. The analysts with whom I had to deal profor bread, and suggests that the blue colour developed on bringing bread ashes to a red heat may have given rise to the idea. I have myself found copper some years since in Belgian bread by the wet process (though never in English), but believe that that adulteration no longer is continued at the present day. I cannot imagine how Dr. Edmunds can suppose that a chemist could mistake a blue colouration at a bright red heat to be due to the presence of copper. I am, &c., To the Editor of the Chemical News. SIR, Amongst the points in connection with the commercial analysis of soda ash, which have not been touched upon by your correspondents, there is one of some importance, namely, that of filtering the solutions before testing. It is evident that any matter, insoluble in boiling water, which neutralises test acid, causes the ash to appear by so much of more value than it really is to the consumer. The following cases, not selected, but being simply the last three parcels of ash we have used, show what this may probably amount to in good ashes. The alkali in the well washed residue is reckoned at its equivalent of Na2O (Na = 23). The slight difference between Messrs. Teschemacher and Smith's analyses and my own is no doubt very easy of explanation. Now as the ash is charged at so much the unit according to the percentage of soda, and in III. we have o'93 useless, out of 5111 there is a loss to the consumer of over 1 per cent. of the money value; or, if we compare the make of I. and II. with III., as there is a difference of o5 out of about 50, there is a loss of 1 per cent. on buying III. as compared with the others.-I am, &c., SAMUEL HALL. East London Soap Works, Bow. December 29, 1875. SODA TESTS. To the Editor of the Chemical News. SIR,It must strike any disinterested person who has perused your correspondence on the subject of commercial alkali testing, that the reputation of the analysts referred to is somewhat under a cloud, and that it is incumbent upon them as they value their commercial integrity to offer some reason for the discrepancies which have been laid to their charge. It is no secret that for many years the "Liverpool test" has been regarded by all but those more immediately benefitted by it, as thoroughly anomalous; and now that the subject has been ventilated in your columns, if no explanation be forthcoming, there would seem to be no alternative but to suppose that it does not pay to expose its fallacies. The tests must either be right or wrong. If the upholders of it believe the former, why do they not furnish grounds for their belief? If the latter, why do they not explain that it is simply a case of "trade custom," with which they do not consider it their business to interfere? Some time ago I acted as manager of a Lancashire works, and had occasion to make many tests of soda-ash and caustic soda. And I need not add my results seldom or never corresponded with the Liverpool test of the same fessed, I believe, to standarise their test acid on the 24 equivalent system; but, even adopting the same equivalent, I had to make it a rule to add 1 per cent to my results. The consequence was that I quickly found it to be quite unnecessary to send out 70 per cent caustic soda higher than 69 per cent, or 55 per cent soda-ash over 54 per cent of my test, feeling sure that it would turn up all right in Liverpool. It is unpleasant to seem to impugn the commercial honesty of members of a profession which, of all others, appeals directly to the conscience, but neither is it agreeable to know that the possibility exists of purchasing ash by London, Glasgow, or Newcastle test and selling it at a profit by the Liverpool one. Not merely, however, is the testing objectionable, the sampling is extremely lax. It used to be no uncommon thing to sample parcels of fifty to a hundred casks or drums by merely drawing small portions from two or perhaps three at most. I have in fact known cases where, from the sampler being in a hurry to catch a train, he has been content with sampling a single one. So unsatisfactory indeed are both the sampling and analyses, that it is usual for manufacturers, when the first sampling and test does not accord with their expectations, to order a second or even a third sampling, till the test comes out of the required strength. I have an instance in my mind of a batch of some fifty casks of ash being sampled three successive times by the Liverpool chemist, and certified successively as 53 per cent, 52 per cent, and 54 per cent alkali; the latter being the strength at which the manufacturer had invoiced the parcel previously to its being despatched from the works. It is my impression that many of the brokers do not discourage this state of affairs, and that in numerous instances, so long as a certificate of some sort is forthcoming to send it represents the true strength or not. out to America with the invoice, they do not care whether shipped from Liverpool, and sold entirely on the LiverBut as many thousands of tons of alkali are annually pool test, how important the matter is will at once appear. And it is therefore devoutly to be hoped that the present state of affairs may speedily have an end. I am, &c., JOHN MORRISON. Pottery Lane, Forth Banks, Newcastle-on-Tyne. December 28, 1875. LIVERPOOL SODA TESTS. To the Editor of the Chemical News. SIR, I have read in the CHEMICAL NEWS the address of Mr. John Pattinson to the Tyne Chemical Society, in which he makes some remarks respecting Liverpool soda tests. I have also read the correspondence on the same subject published in your paper. As one Liverpool chemist, engaged to a considerable extent in the examination of alkali products, I must emphatically deny that my test differs materially from what Mr. Pattinson terms the "Tyne test," or that it ranges higher than the ordinary commercial test, which is, as Mr. Pattinson states, "based on the assumption of the old chemists that pure carbonate of soda contains 59 per cent of soda" (the old equivalent of soda being taken as 32 and that of carbonate of soda 54). Some seven years ago I had a long correspondence with Mr. Pattinson respecting soda tests, when the subject was fully and fairly discussed between us. As a result of this correspondence, and of numerous discussions with other persons familiar with the origin and working of the commercial soda assay, I then arranged a method of working, which I have ever since employed, and which translates the percentage of soda ascertained on the basis of the equivalent more generally accepted by modern chemists (carbonate of soda 53 containing soda 31) into the commercial standard, in such a way as to obtain NEWS figures that are essentially the same as those given by Whilst working by the method I have alluded to, and testing for some of the oldest and largest firms in the alkali trade, buyers as well as sellers, and my reports sent not only to numerous places in the United Kingdom, but to very many places abroad, I can safely say that for fully six years past I have not received one single intimation that my reports have been too high, but I have had occasional complaints that my results were too low, and in one or two cases business has been withdrawn from me on the distinct statement that such was the case. I should be sorry, sir, to make such an apparently egotistical statement did I not feel that as a "Liverpool chemist" I am included in the remarks of Mr. Pattinson, and therefore called upon to defend myself from the imputation, which apparently includes all Liverpool chemists who test alkali products of reporting higher figures than are warranted by following the "ordinary commercial standard." With regard to the letter of Mr. George Warner, F.C.S., I can say that I believe he will not object to my stating that I have several times exchanged samples with him, and that our results have closely agreed. One word in conclusion. It is undoubtedly inconvenient that the commercial standard differs from the chemical standard now generally accepted by chemists, but manufacâurers and buyers based their trade custom on the statements of the older chemists, and it is inconvenient to be continually changing a trade standard. Still, the difference between the chemical and the trade standard being known, it is possible that sales based on the chemical standard may gradually be made until that standard only is adopted. But in this case buyers must expect to pay a corresponding increase in price. I very often make tests based only on the 31 equivalent of soda for sellers and buyers who distinctly understand on what basis they are working, and I believe that such tests are satisfactory to both parties as well as to myself. Apologising for the length of my communication, I am, &c., A. NORMAN TATE. 7 and 9, Hackins Hey, Liverpool. SOURCES. Mr. Pattinson mentions a case in which a sample of CHEMICAL NOTICES FROM FOREIGN caustic soda, tested in Liverpool, was certified by a Liverpool chemist to contain 60 per cent of alkali. Í he "identical sample," he says, tested in his laboratory was found to contain only 57 per cent. He adds that " no notice was taken by the Liverpool chemist of the request to the buyer that the sample should be re-examined." Inasmuch as I make it an invariable practice, although using every precaution to ensure accuracy in my work, to fully consider any objection that may be raised to my analyses and assays, and fully sift all possible sources of error, I am able to say that the case referred to by Mr. Pattinson has not been brought to my notice. I therefore, without hesitation, disclaim any connection with the transaction. But just a word or two in reference to it. It would be interesting to know whether or not it was a sample of "cream" caustic soda, and, if so, whether it was the very same identical bottle sample, or a sample drawn from the identical bulk. I can well understand a considerable difference in samples of cream caustic soda drawn from the same bulk, although I will not go so far as to say the difference would extend to three per cent. Different strengths of cream caustic soda may be drawn from the same drum. It is an alkali product that should be sampled with extreme care, in order to arrive at the average strength of the bulk. And this leads me to remark that one of the most constant sources of disagreement in the reports of different chemists lies in the sampling operation. The drawing of samples is a most important matter, requiring the most scrupulous care, whereas it is very frequently conducted in a most careless and incomplete manner, and then differences in tests must result. To compare the results of different chemists, tests of portion of the very same sample must be insisted on. A further source of difference lies in difference of manipulation, but on this point I would say that having tested alkali products, either in alkali manufactories, or as a professional analyst, for nearly twenty years, I cannot state any reason for the wide differences referred to by Mr. ! NOTE. All degrees of temperature are Centigrade, unless otherwise expressed. Comptes Rendus Hebdomadaires des Seances de l'Academie des Sciences. No. 23, December 6, 1875. Constitution of the Phosphates.-MM. Berthelot and Louguinine.-- In this memoir the authors examine the forundertake an alkalimetrical study of phosphoric acid; mation of an insoluble phosphate, that of baryta; they and, finally, they seek to define the displacements and reciprocal distribution of an alkaline base among phosphoric acid and the nitric, hydrochloric, and acetic acids. They conclude that phosphoric acid is not a tribasic acid of the same kind as citric acid, as the third equivalent of a soluble base is separated from phosphoric acid by the feeblest actions, and even by dilution. With ammonia it happens that this third basic equivalent does not combine with phosphorie acid, or if it combines at first it does not remain definitely united to the acid, but is gradually separated spontaneously and completely. Neither is phosphoric acid a bibasic acid in the same sense as are the sulphuric, oxalic, or tartaric acids. The second base, as alkalimetrical operations show, is not neutralised by phosphoric acid, and is entirely separated by the hydrochloric and nitric acids, and gives indications of division even with acetic acid. In short, the three equivalents of base united in the phosphates considered as normal are combined in different and unequal manners. Phosphoric acid must be regarded as a monobasic acid of a mixed function. Colouring Matter of the Fruits of Mahonia, and the Characters of the Wine which may be obtained from them by Fermentation.-I. Pierre.-Upon cotton the juice of this fruit gives fugitive red stains, which with ammonia give a variety of unstable greens. The wine 10 Chemical Notices from Foreign Sources. CHEMICAL NEWS, obtained from the berries is admitted to have a disagree-, rather to the absorption of atmospheric ammonia than able flavour, but the author suggests that it might be used the direct assimilation of free nitrogen. for giving a factitious colour to inferior wines. The Jaborandi (Polycarpus pinnatus).-M. E. Hardy. Note on a Process of Magnetisation called Double-The essential oil of the Polycarpus, as regards its Touch.-M. J. M. Gaugain.-Not adapted for abstraction. boiling-point and certain of its characters, resembles the Temperature of Elevated Regions of the Atmo- iso-terebenthenes, and is exceedingly analogous to the oil sphere.-M. D. Mendeleef.-It appears that in the upper of citron. strata of the atmosphere there exists a source of heat, for the observed heat is always higher than the calculated temperature. This source of heat is doubtless found in the aqueous vapours of the atmosphere. Distribution of Magnetism in Circular or Elliptical Discs of Steel.-M. E. Duter.-The author's results are that the sum total of magnetism diffused over circles or ellipses is proportional to the surface. This magnetism may be considered as distributed on threads which affect a hyperbolic form. The non-transverse axes of these hyperboles are the axes of symmetry of magnets perpendicular to the neutral line. Certain Properties of Gallium.-M. Lecoq de Boisbaudran. Already inserted in full. Hydrated Derivative of Cellulose.-M. Aimé Girard. 6'4 515 The equivalent of water fixed in this substance resists desiccation. The substance, for which the author proposes the name hydro-cellulose, is characterised by its extreme friability. Constitution of the Albumenoids.-M. P. Schützenberger. The author finds that the differences of constitution of the proteic matters are of the second order, and that they all contain a common nucleus, around which are grouped variable proportions of less important groups, such as urea, oxamid, tyrosin, and the glutamic acid of gluten. Bulletin de la Societe Chimique de Paris, Reply to the Objections of M. A. Gautier, relative to the Part played by Carbonic Acid in the Spontaneous Coagulation of Blood.-M. E. Mathieu and V. Urbain. The authors maintain that carbonic acid is the cause of the spontaneous coagulation of blood, and during life the fibrin dissolved in the plasma does not coagulate, because the acid gas, in the same manner as oxygen, is combined with the red globules. Constitution of Benzin.-MM. A. Rilliet and E. Ador. -A hypothetical paper, not adapted for abstraction. Extraction of Sulphur.-C. F. Sestini.-Having observed, in the Romagna, a workman carefully picking out crystals of gypsum mixed with sulphur ore, before its introduction into the doppione (a furnace receiving a double row of retorts), he was told that in the furnace gypsum destroyed the sulphur. On investigating the reactions ensuing when gypsum and sulphur are heated together, he found that at 130° the gypsum lost all its water and became anhydrous. At higher temperatures, up to 444°, the sulphur reduced the sulphate of lime to a sulphide, and escaped as sulphurous acid : CaSO4+2S=2SO2+CaS. Does the Mould developed on Organic Bodies assimilate the Nitrogen of the Atmosphere ?-MM. F. Sestini and J. del Torre.-The authors having reviewed the conclusions of Boussingault, Boischode, Selmi, and others, and having repeated their experiments under varied conditions, express the opinion that the slight increase of nitrogen found after the growth of Penicillium was due Solution of Platinum by Sulphuric Acid in the Process of Concentration.-M. A. Scheurer-Kestner.Already noticed. Nitronaphthyl-Sulphurous Acids and their Derivatives.-M. P. T. Cleve.-The author's object was to examine the acid which Laurent obtained by the action of fuming sulphuric acid upon nitro-naphthalin, to compare it with those obtained by the at ion of nitric acid upon a and ẞ naphthyl-monosulphurous acids, to compare these acids, and to transform them into the corresponding dioxynaphthalines and dicarbonated acids. Bisulphate of Lithia.-M. H. Lescœur.-This salt is obtained by the use of monohydrated sulphuric acid. It forms deliquescent crystals, which when dried consist of LIOSO3 51'87 per cent, free sulphuric acid 46.85 per cent. Theory requires sulphate of lithia 52.88 per cent, to 47'12 per cent of sulphuric acid. Its melting-point is about 120°. Biacetate of Thallium.-M. H. Lescœur.-This s is obtained by exposing a solution of acetate of thallium in monohydrated acetic acid to spontaneous evaporation. Its formula is TIO,C4H303,C4H404. It contains 64.82 per cent of thallic oxide, and 18.84 per cent of free acid. The salt effloresces slightly in dry air, and melts at about 64°. Biacetate of Lithia.-M. H. Lescœur.-The solution of neutral acetate of lithia in glacial acetic acid deposits, on spontaneous evaporation, crystals in the form of a hopper. Its formula is LiO,C4H3O3,C4H4O4, and it contains 1154 per cent of lithia, and 49'05 of acetic acid. It is deliquescent, and melts at 99°. If heated suddenly in a small platinum capsule it gives off vapours of monohydrated acetic acid, which take fire and burn. Part played by Carbonic Acid in the Phenomenon of Coagulation.-Dr. F. Glenard.-The author infers, from the experiment described, that carbonic acid plays no part in the coagulation of blood. Transformation of Starch by the Action of Diastase, and the Production of a New Saccharine Matter.-M. A. Petit.-If at a constant temperature of 50° a gramme of diastase is allowed to act for some hours upon a kilo. of starch paste at 1-10, the liquid, after filtration and prolonged ebullition,-in order to stop the action of the diastase,-contains, besides maltose capable of fermentation and of reducing Fehling's liquid, another fermentable sugar, without action upon Fehling's test, even after ebullition for five minutes in contact with sulphuric acid at 1-100. New Alcoholometric Formula.-M. C. Delavaud.— Not adapted for abstraction. Correspondence from St. Petersburg Nov. roth, Session of the Chemical Society Sept. 11 to 13.-The minutes of the Session contain a very interesting pape by M. N. Zalomanoff, on the absorptive power of soils M. Beilstein announces that in the action of a mixture of hydrochloric acid and chlorate of potassa upon benzoic acid there is formed, besides, B-dichlorobenzoic acid and the ordinary a-dichlorobenzoic acid. M. Menschoutkine, on behalf of M. Lissenko, handed in a note on the memoir of M. Jatzoukowith on the action of oxygen on coals and paraffins. M. Menschoutkine also announced, on behalf of MM. Tscheque and Steiner, that the action of monochloracetic ether upon xanthogenate of potash gives rise to xanthogenacetic ether. THE CHEMICAL VOL. XXXII. No. 842. ON THE more delicate results to be obtained than the use of a NEWS. stronger standard does. ESTIMATION OF VERY SMALL QUANTITIES By M. M. PATTISON MUIR, F.R.S.E., As I have lately been occupied with experiments upon the action of saline solutions upon lead and copper, which involved the measurement of very small quantities of these metals, I thought it might be well to test the accuracy and delicacy of the method employed. The method itself is in no way new, being that described by Wanklyn in his book on "Water Analysis." The depth of colour produced by the addition of sulphuretted hydrogen water to a known volume of the liquid under examination is compared with the colour produced, by the same means, in an equal volume of water, to which a known amount of lead or copper, in solution, has been added. In comparing the colour of the liquid under examination with the standard liquid, I find it preferable to employ stout glass tubes, holding about 100 c.c., and having a diameter of about 15 c.m., rather than white porcelain dishes as recommended by Wanklyn. contents of the tubes are thoroughly mixed by means of glass tubes on the ends of which bulbs have been blown. (See Thorpe, "On a Method of Estimating Nitric Acid, &c." Fourn. Chem. Soc. [2], xi., 547.) The Wanklyn recommends the use of standard solutions, I c.c. of which is equal to 1 m.grm. of copper or of lead: he employs 70 c.c. of the water to be tested. If, therefore, the colour produced on adding sulphuretted hydrogen water to 70 c.c. of the liquid under examination is found to be equal to that produced by the addition of the same reagent to 70 c.c. of distilled water to which I c.c. of the standard has been added, we shall have I grain per gallon of lead in the water. But 1-10th of a grain of lead per gallon is generally considered hurtful when present in a drinking water; to estimate this we should require to use only o'I c.c. of the standard: a very small error in reading the burette measurements would introduce a comparatively large error in the result. Thus, in the case of a water containing 1-10th grain of lead per gallon, an error in reading of 0.05 c.c. would introduce an error in the quantity of lead equal to one-half of the total quantity to be estimated. The first point, therefore, to investigate appeared to be the strength of the standard solutions. I shall describe the experiments made with copper. Standard used, I c.c. = 1 m grm. copper. The second point to be determined was the limits of accuracy of the method, and first as to the lower limit. From the experiments with copper already detailed it will be seen that o'5 m.grm. of copper per litre could be estimated by using 50 c.c. of the liquid. Standard used, 1 c.c. = o'I m.grm. copper. Solution contained o‘25 m.grm. of copper per litre. Details as before. In this experiment it was very difficult to determine the exact point at which the colours were the same, as the intensity of colouration produced was very slight. A further addition of o'05 c.c. of the standard could hardly be said to produce a noticeable change in the depth of colour. I think, therefore, that o'5 m.grm. of copper per litre 0035 grain per gallon, is the smallest quantity which can be accurately estimated by this process when working with 50 c.c. of the liquid under examination. The amount of lead which can be estimated with accuracy is less minute than the amount of copper. Standard used, I c.c. = o'I m.grm. lead. The use of a standard, 1 c.c. of which is equal to 1 m.grm. of copper or of lead enables more accurate and = • A Paper read before the Manchester Literary and Philosophical Society. = I m.grm. of lead per litre 0'07 grain per gallon, is, therefore, the smallest quantity which can be accurately estimated by this process when working with 50 c.c. of the liquid under examination. By the evaporation of 1 litre of water to 50 c.c., a quantity of copper so small as o'025 m.grm. per litre, or of lead equal to o'05 m.grm. per litre, can be estimated by this process. In other words, the process will estimate I part of copper in in 2,000,000 parts of water, or I part of lead in 1,000,000 parts of water. Secondly, as to the upper limit. 20 m.grms. of copper per litre 14 grains per gallon is the largest quantity which can be estimated by this method when working with 50 c.c. of the liquid under examination. With lead the following results were obtained : Expt. No. 15 16 .. 10 m.grms. of lead 07 grain per gallon is therefore the largest quantity which can be estimated by this method when working with 50 c.c. of liquid. In making these determinations I found that the colours of the liquids might be compared immediately after the addition of sulphuretted hydrogen. The colours did not become intensified on standing. |