Detection of Small Quantities of Arsenic in Fabrics, Yarn, and Wall Papers. By R. FRESENIUS and E. HINTZ (Zeit. anal. Chem., 27, 179-182).-25 grams of the material is placed in a litre stoppered retort of Bohemian glass, and 250 c.c. of hydrochloric acid of 119 sp. gr. is added. The neck of the retort is bent so that the part near the bulb is inclined upward, whilst the other part slopes downward. It is connected with a condenser, which is also fitted airtight to a tubulated receiver, and this again to a Peligot's tube. The receiver and tube contain water, and are kept cold. After digestion for an hour, 5 c.c. of a cold saturated solution of ferrous chloride is added, and the liquid is slowly raised to boiling, which is continued until frothing stops the distillation. A further quantity (100 c.c.) of hydrochloric acid is then added and distilled over. The united distillates are diluted to 800 c.c., and saturated with hydrogen sulphide, first warm and then cold. The arsenious sulphide, which contains organic matter, is filtered off on an asbestos filter formed in a stopcock funnel. After partial washing, it is treated with a solution of bromine in hydrochloric acid (1·19), and the solution, washed through with the same acid, is again distilled with ferrous chloride in an apparatus similar to the former, but smaller. The distillate now gives arsenious sulphide free from organic matter, and requiring only to be purified from sulphur. Test analyses gave satisfactory results. The residue in the retort was found to be free from arsenic.

M. J. S.

Some Methods of Separating and Determining Arsenic, Antimony, and Tin. By E. LESSER (Zeit. anal. Chem., 27, 218221). The oxalic acid method of F. W. Clarke (Chem. News, 21, 124) gives good results. To the solution, which should be neutralised as far as possible, 35 to 40 parts of oxalic acid are added for each part of tin present. The liquid is then heated whilst being saturated with hydrogen sulphide. The precipitate is dissolved in ammonium sulphide, and the solution acidified with oxalic acid is again saturated whilst hot with hydrogen sulphide. The two filtrates which contain the tin are concentrated, mixed with ammonia, ammonium sulphide, and acetic acid, and the precipitate converted into stannic oxide for weighing. The arsenious and antimonious sulphides are dissolved from the filter by ammonium sulphide, oxidised with hydrochloric acid and chlorate, and the arsenic precipitated as magnesium arsenate, after addition of tartaric acid. The precipitate requires resolution and reprecipitation to free it from basic magnesium tartrate.

Vohl boils the strongly acid solution of the three metals, and adds sodium thiosulphate until the precipitate becomes white. The precipitate contains the arsenic and antimony. Lesser finds the separation of tin and arsenic by this method satisfactory, but that of tin and antimony inaccurate.

De Clermond and Frommel's method of separating arsenic from the other two metals by boiling the sulphides with water until they are completely converted into oxides does not give good results, as the oxides of tin and antimony are not absolutely insoluble. Neither is the separation of arsenic from antimony by strong hydrochloric acid satisfactory.

Lesser weighed his antimony as tetroxide, but recommends weighing as sulphide, for which purpose the precipitate, freed from sulphur by carbon bisulphide, is washed into a weighed porcelain crucible by ammonium sulphide, and dried in the covered crucible in the air-bath at 200-230°, for at least nine hours. M. J. S.

Separation and Estimation of Boric Acid. By H. N. MORSE and W. M. BURTON (Amer. Chem. J., 10, 154-158).—The substance, if insoluble in water, is heated with pure potash in a nickel crucible for two hours, so as just to maintain the whole in a state of fusion. The mass is extracted with hot water, which if iron is present, must amount to 100 parts for every part of potash employed, and the whole is heated on a water-bath for a considerable time; this precaution is necessary, as otherwise the iron is not completely removed, and ferric sulphate being soluble in alcohol, its presence is not admissible. The filtered solution is evaporated to 10 or 12 c.c., and after addition of a few drops of tropaolin, 00, a not too dilute sulphuric is added, until exactly neutral. The solution now containing only neutral salts of the alkalis, boric, silicic, and carbonic acids, and not exceeding 20 c.c. in volume, is dried by slowly stirring into it pure copper sulphate (free from iron and from chlorides) that has been dehydrated at 150°. The powdered mass is transferred to a filter-tube, and extracted with absolute alcohol (dried over copper sulphate) in six lots of 15 c.c. each. The alcoholic filtrate is run into a quantity of standard baryta solution, the excess of baryta is converted into the carbonate by a current of carbonic acid, and the whole is then evaporated and ignited in a platinum dish over a good burner. The residue is a mixture of barium metaborate and carbonate; the amount of boric anhydride is found thus :-as the molecular weight of boric anhydride minus the molecular weight of carbonic anhydride is to the molecular weight of boric anhydride, so is the weight of residue minus the theoretical weight of the barium carbonate obtainable from the baryta used, to the weight of boric anhydride present. The analyses. quoted are exceedingly concordant and satisfactory. H. B.

Estimation of Silver in Alloys of Silver and Copper. By H. RÖSSLER (Dingl. polyt. J., 267, 570–572).—Gay-Lussac's wet assay is recommended in cases where great accuracy is required, providing that the alloy to be assayed is melted, and a sample taken from the fused mass. This is necessary, as alloys of silver and copper in solidifying do not remain homogeneous throughout, the inner and outer parts of the alloy containing different proportions of silver.

D. B.

Estimation and Separation of Metals by means of Sodium Pyrophosphate. By G. VORTMANN (Ber., 21, 1103-1106).-The behaviour of metallic salts towards sodium pyrophosphate and acetic acid can be employed as a means of separating the metals, and the pyrophosphates thus obtained, being insoluble in water, dilute acetic acid, and solutions of ammonium salts, can be made use of for quautitative determinations.

Copper salts give with sodium pyrophosphate a bright blue precipitate, soluble in excess of the reagent; on adding acetic acid, a bright blue crystalline precipitate is obtained, the precipitation is, however, incomplete, and can be entirely prevented by the addition of sodium tartrate or sodium thiosulphate.

Cadmium salts give a precipitate soluble in excess. Acetic acid reprecipitates the salt almost completely even in the cold; by evaporating to dryness and digesting the residue with water, reprecipitation is complete. The addition of sodium tartrate or thiosulphate does not hinder the precipitation.

Manganese salts yield a precipitate, soluble in excess, but completely reprecipitated by acetic acid; sodium tartrate does not prevent the precipitation.

Zinc salts behave similarly, but reprecipitation is complete only when the solution is evaporated to dryness and the residue taken up with water; sodium tartrate retards reprecipitation.

Cobalt salts give a gelatinous precipitate, soluble in excess; on shaking or heating gently, the solution becomes gelatinous but not if sodium tartrate is added. Acetic acid reprecipitates the cobalt salt. Presence of sodium tartrate does not prevent reprecipitation.

Nickel salts behave similarly, but the addition of sodium tartrate prevents the reprecipitation with acetic acid.

Ferrous salts yield a precipitate soluble in excess, but completely reprecipitated on addition of acetic acid.

Ferric salts give a precipitate soluble in excess and not precipitated by acetic acid; on addition of acetic acid and sodium sulphite, reprecipitation is complete.

Aluminium salts give a precipitate soluble in excess, but completely reprecipitated by adding acetic acid and boiling; the addition of sodium tartrate prevents reprecipitation.

Uranic salts give a precipitate soluble in excess and not reprecipitated by acetic acid.

Chromic salts give a bright green precipitate which is scarcely soluble in excess even on boiling. Acetic acid prevents the precipitation of chromic salts, but the solution becomes turbid; the addition of sodium tartrate prevents the turbidity.

From the above results it will be seen that by means of sodium pyrophosphate copper can be separated from cadmium, cobalt from nickel, manganese and zinc from ferric salts, manganese from aluminium and uraninm, and ferrous salts from aluminium and uranium, possibly also from chromium and ferric salts. Cadmium, zinc, manganese, cobalt, nickel, possibly also iron and aluminium, can be estimated as pyrophosphates. F. S. K.

Reduction with Lead. By F. STOLBA (Chem. Centr., 1887, 1240, from Listy. Chem., 11, 225-226).-Under certain conditions, lead reduces iron, chromium, and tin chlorides almost as quickly and completely as zinc. Lead, however, only reduces stannic chloride to stannous chloride. Small quantities of nitric acid do not influence the reduction.

J. P. L.

Colorimetric Estimation of Minimal Quantities of Iron. By SABANEEFF and KISLAKOWSKI (Chem. Centr., 1888, 84, from Pharm. Zeit. Russ., 26, 776-777).-For the estimation of minimal quantities of iron in mineral waters, &c., the colorimetric method with ammonium sulphide leaves nothing to be desired, either from the point of rapidity or precision. It is also free from the errors which organic substances may cause in titrating iron with potassium permanganate or alkaline chromate. J. P. L.

Determination of Iron in Iron Ores by the Tartaric Acid Method. By L. BLUM (Zeit. anal. Chem., 27, 146-151).—A source of error in the precipitation of iron by ammonium sulphide from solutions of ores containing magnesium and phosphoric acid, exists in the simultaneous precipitation of magnesium phosphate, the phosphoric acid of which subsequently comes down with the ferric oxide when the solution of the sulphide is precipitated by ammonia. With ores containing only a small amount of phosphoric acid and for technical determinations, the error can be practically got rid of by largely diluting the solution, heating it before adding ammonium sulphide, and before filtering allowing it to remain only just long enough for the ferrous sulphide to subside (about half an hour).

Fresenius points out in a footnote that a more certain way would be to precipitate the iron with most of the alumina as basic salts, before separating by ammonium sulphide in a tartrate solution, or else to allow the solution made alkaline by ammonia to remain for some time and to filter before the addition of ammonium sulphide.

M. J. S.

Estimation of Chromium in Iron or Steel in presence of Phosphorus. By J. O. ARNOLD and H. J. HARDY (Chem. News., 57, 153-155). As the method of analysis previously described (Abstr., 1880, 646) is not accurate in the presence of much phosphorus, the authors have modified it. They observed that under suitable conditions chromium is readily precipitated as a basic phosphate of constant composition, and they base the improved method on that fact. They take 24 grams of metal, and after the fusion make the precipitate and solution together up to 301 c.c., then 250 c.c. of the clear solution (containing the chromium from 2 grams of metal) is acidified with hydrochloric acid and boiled (no alcohol is added, the necessary reduction being effected by the nitrous gases evolved), excess of sodium phosphate is now added, then dilute ammonia in slight excess, and the whole heated until the solution is clear and colourless. The precipitate is dissolved in hydrochloric acid, carefully evaporated to dryness, redissolved by boiling with a few c.c. of hydrochloric acid and filtered. The chromium phosphate is now reprecipitated from the diluted solution by a slight excess of dilute ammonia, washed carefully, dried, and weighed as Cr.P4019. This method has given good results in the authors' hands. Titrating the chromium in the precipitated phosphate by means of permanganate gives low results.

D. A. L. Volumetric Determination of Molybdenum and Lead. By C. SCHINDLER (Zeit. anal. Chem., 27, 137-142).-Ammonium

molybdate mixed with lead acetate gives a precipitate of lead molybdate, PbMoO., which is perfectly insoluble in acetic acid. An excess of the molybdate is detected by bringing a drop in contact with a solution of tannin (1-300). One part of the molybdate in 400,000 of water gives a visible yellow colour with tannin; strong solutions a blood-red. The lead acetate solution contains 40 to 50 grams (with some acetic acid) in the litre. The molybdate solution is made by dissolving 20 grams in 800 of water, adding ammonia until the turbidity disappears, and diluting until it corresponds with the lead solution in strength. To make a titration, the molybdate solution is acidified with acetic acid and diluted to 300 or 400 c.c. with hot water; a small excess of the lead solution is added, and then molybdate until a large drop of the clear, upper liquor brought in contact with a drop of the tannin solution on a porcelain plate gives a distinct orange colour. The exact strength of the lead solution should be determined by means of weighed quantities of pure ammonium molybdate, [(NH1),Mo-О1⁄2 + 4H2O], or, if otherwise ascertained, a deduction of 0.1 c.c. of molybdate solution must be made to allow for the excess required (207Pb to 144MoO3). The tannin solution must be freshly prepared. M. J. S.

Determination of the Amount of Soda and Lime requisite for Purifying Water. By O. BINDER (Zeit. anal. Chem., 27, 176).— A measured quantity of the water is mixed with an excess of saturated lime-water of known strength. The mixture is warmed to between 50° and 80°, cooled, made up with boiled distilled water, filtered, and the excess of lime in the filtrate titrated with sulphuric acid, using phenolphthalein as indicator. The amount of lime required by the water is greater than that which corresponds to the temporary hardness, since magnesia as well as magnesium carbonate is precipitated both from the hydrogen carbonate and from the sulphate and chloride, and because a certain quantity of lime is consumed by free carbonic acid and by organic matters.

For the quantity of soda required, 250 c.c.of the water is evaporated to dryness with 5 c.c. of normal sodium carbonate solution. The residue is dissolved in water, filtered, and the excess of soda titrated by an acid and methyl-orange. It is advisable in practice to use 10 grams per cubic metre above the quantity thus indicated.

M. J. S. Ash Determination. By L. REESE (Zeit. anal. Chem., 27, 133— 136). The incineration is carried on in a tube through which a rapid current of air is being drawn. Substances which are liable to intumesce and leave a difficulty combustible coal can thus be burnt in one-third to one-half of the time and at a lower temperature than would be required in a crucible. The substance is placed in a porcelain boat which is inserted into a short piece of combustion tube, narrowed at one end, and containing at that end a plug of platinum gauze to arrest any particles of ash carried off by the air current. The tube has a spiral of platinum wire round it to prevent adhesion to the outer tube in which it is placed to be heated.

M. J. S.