THE ANALYST. 37 ORGANIC ANALYSIS. Detection of Iodoform in Aqueous Fluids. L. von Stubenrauch. (.hit. f i i ~ Uiztersmh. der ATalir. z~7icl Geizr~ssmittel, 1898, 737.)-The method proposed by the author rests upon the following experiment. Two equal portions (one or two crystals) of iodoform are mixed into an emulsion with equal quantities (3 to 5 c.c.) of water. One portion is reduced by warming with zinc dust and a drop of acetic acid and afterwards filtered, whilst the other portion is left unreduced. On now adding a single drop of nitric acid and a little starch solution to each portion, a blue colour is produced in the reduced portion, but not in the other, It is necessary to limit the nitric acid added to a single drop. If a large quantity be used, the iodoform is decomposed and the blue colour produced without the reduction with zinc dust and acetic acid.On applying the test, if a blue colour be produced on addition of a drop of strong nitric acid and some starch, potassium iodide, hydriodic acid, or a soluble, easily decomposable organic compound of iodine may be present ; but if, on the contrary, the blue colour appear only after reduction with zinc dust and acetic acid, the presence of iodoform is indicated. A solution of iodoform in water behaves in the same way as an emulsion. A difficulty, however, arises when hydriodic acid, an iodide of an alkali metal, or a soluble organic compound of iodine, is present in conjunction with iodoform. In this case the test is only available when the proportion of the other bodies to the iodoform is relatively small, so that the difference in the depth of the blue colour produced before and after reduction is distinctly apparent.The presence of albumin, except in very small quantities, also interferes with the test. The author's experiments with the method show that iodoform taken internally is not excreted as such in the urine, but is converted into an organic iodine compound or into an iodide of an alkali metal. H. H. B. S. Preparation of a Standard Solution of Ethyl Aldehyde. X. Rocques. (Aiznal. de C'lzim. ilncd., 1898, iii., 365.)-The preparation of an alcoholic solution of aldehyde for use as a standard in the colorirnetrical determination of aldehyde is not an easy matter, owing to the readiness with which polymerization occurs and the difficulty of obtaining a pure substance.The author obviates this difficulty by using aldehyde .ammonia CH, - CH,NHs P O38 THE ANALYST. as the starting-point. This substance, as met with in commerce, is in the form of rhombohedra1 crystals melting at 70" to 80". It is very soluble in alcohol, is insoluble in ether, and is decomposed by dilute acids with the re-formation of aldehyde. It is purified by trituration in a mortar with successive portions of anhydrous ether, the liquid being decanted each time. The final residue is dried in the air, and afterwards in wacuo over sulphuric acid. 1,386 grammes of the dry compound are diisolved in about 50 C.C. of pure alcohol (96 per cent.), and 22.7 C.C. of normal sulphuric acid (mixed with more of the same alcohol) added, an immediate precipitate of ammonium sulphate being formed.The liquid is made up to 100 C.C. with the alcohol, and an additional 0-8 C.C. added to compensate for the volume of the ammonium sulphate. It is then Elhaken, allowed to stand overnight, and filtered. The filtrate contains 1 per cent. of aldehyde, and is diluted with the necessary quantities of water and 50 per cent. alcohol, so as to obtain a solution containing 50 inilligrammes in a litre of 50 per cent. alcohol. C. A. M. Reaction between Ferric Chloride and Phenol in presence of Alcohol. F. Peters. (Zeits. angew. Chem., 1898, l078.)-It is generally believed that the violet colour produced when ferric chloride is mixed with an aqueous solution of phenol does not appear if the latter be dissolved in alcohol.The author finds that on dropping a 10 per cent. aqueous solution of ferric chloride into a solution of phenol in dilute spirit, the colour is does not exceed 3.19 v/v, or reaction becomes indistinct, not prevented from appearing if the proportion of alcohol 2.53%; but as the amount rises to 3-44 v/v, or 2*73%, the and above that limit it does not occur at all. F. H. L. - ~- Reactions of Some Common Phenols. G. Deniges. (Bull. Soc. Pharm. Bordeaux, 1898, 241 ; through Ann, de Chim. Anal., 1898, iii., 381, 382.) Reactions with Mercuric Szdphate. --Five centigrammes of the polyphenol are dissolved in 2 C.C. of water, 3 C.C. of the author's mercuric reagent (ANALYST, xxiii., 216) added, and the whole shaken. h lemon-yellQw precipitate is obtained .. . ... . . . Pyrogallol. No precipitate. The liquid becomes yellow, changing to No precipitate. A yellowish-white ,, 9 , ... ... . . - Phloroglucinol. yellowish-red and to reddish-brown . . . ... . . . Pyrocatechin. H ydroquinone. The liquid remains colourless, or has a slight greenish-yellow tint . . . ... I.. ... On boiling becomes yellowish-red . . . . _ . ... On boiling no perceptible change . . . ... ... After immersion of the tube in cold water and .prolonged standing a yellowish precipitate ... ... Reactions with Soda.-From 15 to 20 centigrammes of the substance are dis- solved in 3 to 4 c . ~ . of alcohol, 1 c . ~ . of a solution of soda allowed to flow down the side of the tube and the zone of contact observed.THE ANALYST. 39 (a), A coloured zone is produced immediately.surmounted by a green ring. surmounted by a white ring. On shaking the liquid On shaking the liquid Red { becomes green ... ... ... ... Hydroquinone. ... becoiiies brown ... ... ... Pyrogallol. Yellow, with a yellow coloration of the supernatant alcohol ( b ) No immediate coloration, but after some time a green ring Pyrocatechol. White ... a. ... ... ... ... Phloroglucinol. appears ... ... ... ... ... ... Resorcin. (Gy. ANALYST, xxi., 295.) Humble-Bees' Wax. E. E. Sundvik. C. A. M. (Zeits. Ph?~sio$. Chenz., 1898, xxvi., 56 ; through Clzem. Zeit. li'ep., 1898, 321.)-From 130 grammes of comb the author obtained, by extraction with ether or chloroform, about 30 grammes of crude wax, melting between 35" and 40" C. The odour was pleasant, resembling honey, and not at all rancid.The wax bleached almost perfectly when exposed to daylight in thin layers ; but became very rancid in smell. Treated with weak caustic potash, and crystallized six or seven times from alcohol, it melted at 69"-70"; finally it melted at 74" or 75" C. The intermediate product possessed the sticky nature of ordinary beeswax ; the purest did not. The crystals form fine, soft, woolly needles. Even in the cold it is more or less soluble in the regular fat-solvents. Its empirical formula is C,,H7,0. Heated to 150" or 160" with benzoic anhydride, it gives a substance easily soluble in hot .or cold alcohol, which, after several recrystallizations, melts at 55" C. F. H. L. (&dI. de $'Ass. Belge, 1898, xii., 143-151.)-This paper gives an account of the authors' investpigation on the different methods of determining pentoses, and describes various modifica- tions which they have adopted.For the estimation of pentoses, after conversion into furfural, two methods are employed : that of Tollens, in which the furfural is precipitated by means of phenyl- hydrazine ; and that of Councler, in which phloroglucol is used as the precipitant. Several chemists have been unable to obtain satisfactory results by the second method, and hence the precipitation of the pentoses as hydrazones is the one in general use : The Estimation of Pentoses. A. Grbgoire and E. Carpiaux. C,H,O,+ C,H,NH.NH, = C,H,NH.NH.C,H,O + H,O. The hydrazone was formerly determined volumetrically (Tollens and Giinther) or gravimetrically (Tollens and Chalmot), but Flint and Tollens' recent researches (Landzu.Vers. Stat., dii., 395) have shown that in the volumetric method the distil- lation of the substance with hydrochloric acid yields small quantities of substances (acetone, etc.), which combine with the phenylhydrazine, although they do not form an insoluble compound, and hence the results are too high. In the gravimetric method the distillate containing the furfural is diluted to 400 C.C. with hydrochloric acid, neutralized with sodium carbonate, slightly acidified with acetic acid, the furfural precipitated with an acetic acid solution of phenyl- hydrazine, the solution brought to 500 c.c., and, after being mechanically stirred for thirty minutes, the precipitate collected on a filter of glass wool, washed with water,40 THE ANALYST.dried in a current of air, and weighed. The weight of the hydrazone multiplied by 0-538 gives the furfural. The authors find that there is a slight source of error in this method, owing to the hydrazone dissolving to a slight extent in the water required to wash the pre- cipitate. To obviate this they have devised a gas-volumetric process, in which the nitrogen in the phenylhydrazine is determined before and after the precipitation, and the difference calculated into the amount taken up by the furfural. The original phenylhydrazine is decomposed by means of copper sulphate : but this is not possible in the presence of sodium chloride, as in the filtrate from the precipitated hydrazone, unless the process is modified.Pheny lhydrazine acetate, like ammonium acetate, decomposes on boiling, but the authors' experiments show that by making the liquid strongly acid with hydrochloric acid, the phenylhydrazine can be boiled without alteration. In the gas-volumetric process finally adopted the nitrogen in the phenylhydrazine reagent was determined in the following manner : From 4 to 4.5 grammes' were dissolved in 250 c c. of water, 25 C.C. of the solution were mixed with 20 C.C. of con- centrated hydrochloric acid, boiled for several minutes to remove all air, and intro- duced while hot into a Schloesing's (nitric acid) apparatus containing 25 C.C. of zt 20 per cent. solution of copper sulphate and several C.C. of hydrochloric acid, from both of which all air had been expelled by boiling.The reaction took place instan- taneously, in accordance with the equation : The liberated nitrogen was eollected and measured in the usual manner, corrections made for the vapour tension of water and chlorbenzene at the temperature of obser- vation, and the calculated weight of nitrogen multiplied by 3.857. Thus in a typical case 94.8 C.C. of gas were obtained at a temperature of 19" C. and a pressure of 748 mm., which corrected for temperature, pressure, tension of the water vapour, and chlorbenzene vapour was 721 m.m., corresponding with 88.51 per cent, of the phenyl- hydrazine taken. For the precipitation of the furfural, 81.5 grammes of sodium chloride (the quantity corresponding to the amount of hydrochloric acid which distils over in the estimation of pentoses) were dissolved in about 400 C.C.of water, 2 drops of acetic acid added, and a quantity of furfural not exceeding 0.6 gramme. After the addition of an aqueous solution of phenylhydrazine (4 or 4.5 grammes in 250 c.c.), the liquid was made up to 500 c.c., and shaken for at least an hour in a Witt agitator. The filtrate was at once mixed with hydrochloric acid, which was found to prevent the decomposition of phenylhydrsaine acetate. It was then boiled to expel the air, and 200 C.C. containing 20 C.C. of hydrochloric acid were introduced into Schloesing's apparatus, and the nitrogen determined as before and deducted from the amount found in the first determination. The result multiplied by the factor 3,429 gave the amount of furfural. In seven analyses of furfural by this method the percentages found varied from 99.4 t o 102.4 per cent., the mean being 100.5 per cent.C,H,NH.NH, + 2CuS0, = H,SO, + C,H, + N, + Cu,SO,, C,H,N.NH, + 4CuS0, + HCl= 2H2S0, + C,H,Cl + ~CU,SO, + N,. C . A.M.THE ANALYST. ing matters. 41 alone. Non-reducible colouring matters. A Method of Analysing Natural and Artiflcial Organic Colouring Matters. A. R. Rota. I. Identification of Imhidual Colouriiag Mutters. - I n this paper €he author draws up a scheme of analysis based upon the recent views of Nietzki, Witt, Armstrong, and others, as to the relationship which exists between the constitution and colour of these substances. Regarding organic colouring matters as quinone derivatives (Armstrong, Nietski), those which are based upon mono- and di- amido-quinone are reducible by stannous chloride; while those which are regarded as derivatives of a quinone, in which an oxygen atom is replaced by a di-valent hydrocarbon group in the quinone ring, are not reducible, Thus, if O = R = 0 represent an ortho- or para- quinone, the nitroso- axo- and imido- quinone colour derivatives are reducible : (Chem.Zeit., 1898, 437-442.) O = R = N - or - N = R = N - oximido-quinone di-imido-quinone, but not such derivatives as oxyquinone- and triphenyl- methane Golouring matters : O = R = C = and - N = R = C = oxycarbo-quinone imido-carbo-quinone. The reduced colouring matters can be subdivided into two groups, according as to whether the colour is restored on oxidation with ferric chloride or atmospheric oxygen ; and the unreduced colouring matters can also be subdivided according to their behaviour on treatment with caustic potash.Four main groups are thus obtained, as shown in the subjoined scheme. The aqueous or alcoholic solution of the substance is diluted to about 1 : 10,000, and 5 C.C. treated with 4 to 5 drops of concentrated hydrochloric acid and about the same quantity of a 10 per cent. solution of stannous chloride. The mixture is shaken, and, if necessary, warmed to the boiling point. If no decolorization. occurs, the solution of the colouring matter should be again tested with stannous chloride after still further dilution, A. ClussiJicatioiz of Organic Colozcri?zg Matters. CLASS I. Nitro-, nitroso-, and azo- colouring matters, i n c l u d i n g azoxy- and hydrazo- colours.Indogenide- a n d imido- quinone colour- ing matters. CLASS IV. Aniido-derivatives of di- and tri- phenylme- Non-amide dipheiiyl- thane, auramines, acri- methane colouring mat- dines, quinolines, and ters, oxy-ketone colour- colour deriratilTes of ing matters (most of the thiobenzenil. I natural organic colour- ing matters).42 THE ANALYST. The individual colouring matters in the four main groups may often be identified by reference to one of the published tables of their physical, chemical, and tinctorial properties, which, however, only describe a part of the dyes now met with in commerce, A differ- entiation can often be made in the case of halogen derivatives of similar phthaleins, for instance, by determining the halogen after igniting the colouring matter with lime.The detection of sulphur by fusing the dye with potassium nitrate, and testing the melt for sulphuric acid, often enables one to differentiate between two substances, as, for instance, between the thiazines and oxazines. When the colouring matter was reduced by stannous chloride, the decomposition product may be further examined after removing the tin with sulphuretted hydrogen. Picric acid, for example, gives the colourless tri-amido-phenol, which on treatment with ferric chloride gives the blue amido-di-imido-phenol. The azo colouring matters give on reduction with stannous chloride at least two primary amines in accordance with the equation : I n doubtful cases resort must be made to the spectroscope. R - N = N - R, + 2H2 = R - NH, + R - NH,.These amines can often be separated by means of ether. The reduced solution, from which the tin has been removed by means of sulphuretted hydrogen, is treated with caustic potash and the liquid shaken with ether, which dissolves the non-sulphonated arnine, and leaves the sulphonated amine in the aqueous layer. The latter can be identified by the characteristic azo-compound which it forms with certain diazo- derivatives. Sulphanilic acid, for instance, obtained by the reduction of naphthol orange, combines with diazo-benzidine-chloride to form a yellow tetrazo-colouring matter. Naphthionic acid yields Congo-red ; and certain oxy-sulphonated amines, such as C,,H,-OH(8) Since the NH, group resulting from the azo groups must be in the para-position to another amido group in the radicle, a para-diamine is produced, and this can be easily recognised by the thiazin reaction (treatment of the solution freed from tin with hydrochloric acid and ferric chloride in the presence of sulphuretted hydrogen).A para-diamine is also obtained from those colouring matters which contain not the ctmido group, but two diazo groups, since the middle radicle contains the two azo groups in the para-position. Thus, Sudan 111. (A), on reduction with stannous chloride, gives : /NH2(2) \ SO,H(6), give a violet colour (diamine black R (C), etc.). By this reaction it is possible in the absence of an amido substance to determine A scheme of the characteristics of members of the four groups is shown in the whether a mono-azo or a diazo colouring matter is present.following tables :THE ANALYST. 43 water. Wool and silk dyed directly, but not cotton. The HC1. With HC1+ SnClz par- in presence of‘ tially reduced, giving red KOH. nitro-amido derivatives (nitra- mines) or nitro-phenols turn- aqueous solution shows ten- dency to decolorization with soluble in ether NITRO-PHENOLS. In- B. Class I.-Redwecl by HCI +- SnCl, awl not Reoxidixuble. / Noit - sidphonated. Sol- //O Victoria \H yellow. Sulphonded. Insoluble Naphthol in ether. yellow. presence of acetic acid. uble in ether in O=R=N NITRO-COLOURING MATTERS : ,NY?~A~!S. Soluble in ether in presence -N,R=N(O e.g., R - NO.,. OH Aurantia. Colourless solu- tion. Yields nothing t o ACID COLOUR- ING MATTERS. acetic acid.‘ a t e d. E x - OXYAZO-COLOUR- I n d i r e c t f o r Diamond yellow t r a c t e d b y ING MATTERS cotton wool.ether f r o rn w I T H C A R- Direct for cotton Chrysamine, dilute solution BOXYL GROUP. wool. in acetic acid. .N o N - A M I D O I n d i r e c t f o r Bordeaux B SuZp honated. COMPOUNDS. cotton wool. (4 - Not extracted Unaltered by Direct for cotton Azo-blue (A). solution i n \ A x I D o COM- I n d i r e c t f o r Solid yellow dilute acetic 1 P o u N D s. I cotton wool. by ether from( HNO,. N (P). I I wool. AZO-COLOURING MATTERS : Their aqueous so- l u t i o n decom- posed with KOH a n d extracted with ether, gives an ethereal ex- tract with the annexed charac- R -N=N - R. 1 acid. \ ( &poged by Direct for cotton Congo red (A). c wool. Class II,-Eerlucecl by HCl+ SILCI, awl Beoxidixable.0 x Y A z I N E s (no N/ ”>’ T H I A Z I N E S (sul- N/’R1\S tion is coloured readily or colourless, and ’ by \R = N I acid. I phur). =N- - The ethereal soh- ,The solution yields the ori- 1 in the cold. ginal colour to 5 per cent. acetic I Nile blue A (13). I NR / Methylene blue. i BASIC COLOURING MATTERS. Fixed on wool in dkaline bath. The coloured solu- tion is reduced but slowly and i n c o m p l e t e l y , I even on warming, j and with the ad- dition of much [ SnCl,+HCl. ’INDULINES. Blue colour with cone. H,SO,. Blue on dilution. SAFRANINES. Green c o l o u r w i t h H,SO,. On dilu- tion blue, then violet. Induline. Soluble in alcohol. Safranine T. Extra (A).Coloured. Does not Blue colouringmat- yield the colour ters changed by INDOPHENOLS. HC1 on warming.ING MATTERS. i Red or blue colour- to acetic acid. NEUTRAL COLOUR- ing matters. Un- With HNO, yield altered by HCl. INDoGENIDES. - - fibres in bath. 1 isatin. Non - sulphonated. Soluble in ether oXAZONES. in presence of acetic acid. acid. MATTERS. Fixed on wool acid bath. Class III.- Colozcriizn The ethereal so- lution is co- l o u r l e s s o r coloured. The c o l o u r i s y i e l d e d t o 5 per cent. acetic acid. BASIC COLOUR- ING MATTERS. Fixed on wool in alkaline bath (NH,). '"I stances* Matters Yzot '<; = 0 Indophenol. I1 Fluorescent blue, orcein. dR\0 \"Lo - SULPHONATED IN- DOGENIDES, Indigo carmine. SULPHONATED Thiocarmine R THIAZINES. (C>* Not reduced by -I SnC1,fHCl. SULPHONATED IN- DULINES. Containing t h Soluble nigrosin.Iqnido- quinoq~ Carbon Chrorrzophore - N = R = C = . Colourless, non-fluorescent, solution. Yellow colour y i ~ l ? c Y ~ ~ \ acetic acid non - fluorescent. aqueous solution is decolorized by KOH and decomposed by HC1. Colourless, ethereal solution. Green fluorescence. Aqueous solution pre- cipitated by KOH, hardly altered by 1 yielded to acetic acid - reddish- violet, blue, and green without fluo- decolorized on warming with KOH, HC1. Turns red with HNO,. Colourless, or coloured ethereal solu- tion. Non - fluorescent. Colour rescence. Aqueous solution usually and coloured yellow by HC1 (ex- cepting fuchsin). Ethereal solution colourless and non-fluorescent. Acetic acid coloured rose and fluoresces. Aqueous solution ,decolorized with KOH. The ).AURAMINES. I I (non-sulphon- ated)' I PYRONINES (coloured altered bv HC1./R\ \R/ -C-N /R C-R \R=N- R -c/ ' 0 \ R L ~ _ - ey.9 Auramine 0 (B). Phosphine. Fuchsine. Pyronine (G). Rhomamine S (BY). The coloured ethereal solution does not yield its colour to acetic acid. Quinoline yel- \ R = N Z uble in alco- low A (sol- hol) . NEUTRAL COLOURING MATTERS. non-fluorescent, and Insoluble in water. Soluble in 1 unaltered by aqueous alcohol. Ethereal solu- tion colour- less. Yields n o t h i n g t o acetic acid. ACID COLOUR- ING MATTERS. Soluble in water. Fixed on wool in acid bath (HC1). L \ acids and alkaiies. I in water. and alkalies. Yellow colouring matters. No fluorescence Reddish-violet, blue, or green colouring matters. Usually decolorized by KOH, little changed by HC1. Red or violet colouring matters.Soluble tated by HCl. Changed but little, or Unaltered by aqueous acids 1" I ITHIAzoLEs* in water with fluorescence. Precipi-1 S not at all, by KOH. colouring matters. Aqueous solution f fluorescent. Brownish - yellow or orange l U L P H 0 N ATED Q U I N 0 N - E - PHTHALONES. U L P H 0 N ATED FUCHSINXS. ULPHONATED RHODAMINES. - C = N I I S - R Quinoline yel- u b l e i n water). F u c l i s i n e S low A (sol- (B). Violamine R (MI. Primdin (B) .THE ANALYST. 45 :lass IV.--Colourinq Matters not Reduced by Snc'l, -I- HCI. Coiztaiizing the 0x9-qasinone Carbon C hromop hore 0 = R = C = . C-R \R, = 0 Remains unaltered. 15 2 -8 Not directly fixed on 2 ~1 r;l wool. Most of them insoluble in water. a g' PHENYL - METHANE without f l u o r e a - COLOURING MAT- cence.i Fixed directly on wool. $ Most of them soluble in water and alcohoL Fluorescence. NON-AMIDO TRI- Soluble in alcohol z s 0 OXY-KETONE d m COLOURING MATTERS. Most of them insoluble :in water, and indirect for fibres . D 7s' s o 1 v e s with yellow or reddish- y e l l o w colour. 15 O N O K E - TONES. -2H D i s s o l v e s QO with r e d , 2% r e d d i s h - a violet, vio- gz letgreen, or -c is blue colour. 43 DIKETONES u (quinones j. I /I{ \R i * Inclined to de- ~3 colorization, 2 X 3 "O warming(with $ decomposi- 2 2 tion). especially on ,-B E N z 0 P H E - co I .r( The free colour-5 for fibres. Colouring mat- ter remains in d i r e c t l y on wool. Auriii. Eosin. Alizarin yel- low A (B). Quercetin. Alizarin. co<$Jco Sulp hona t e d alizarin (ali- zarin-red).11. The Separation of Co~ozcri?zq Matters in a Mixture.-It is sometimes possible to effect a separation by treating the mixture with water or alcohol at the ordinary temperature, or with the aid of heat ; but as a rule extraction with ether or by fixing the dye on wool or other fibre is the most promising method. The behaviour of ether and wool is very similar, both extracting the free colouring matters, but not their salts. At the same time, not all colours soluble in ether can be fixed upon wool. Extraction with Ether.-It is possible to separate basic from acid colouring matters by adding potash to a dilute aqueous solution of the dye and shaking with ether, when the free bases dissolve in the ether, leaving the acids in the aqueous layer.The details of the process are as follow : 100 C.C. of the aqueous solution of the colouring matter are decomposed with 1 C.C. of a 20 per cent. solution of potash, and shaken with three times the volume of ether. The aqueous alkaline solution of the acid colouring matter is neutralized with acetic acid and examined subsequently. The ethereal solution of the colour base is washed with faintly alkaline water, and then shaken with one-third of its volume of 5 per cent. acetic acid. The acid layer is separated, and on evaporation on the water-bath leaves the colouring matter as a,46 THE ANALYST. residue. In those cases in which the colour remains in the ether, the latter is evaporated. In the presence of potash some few acid or neutral colouring matters are also extracted by the ether, as, for example, quinoline yellow, indophenol blue (soluble in alcohol), the various sudans, etc.These are all insoluble in water, but soluble in alcohol. I n the extraction of the colour bases different alkalies have different liberating powers. Thus, safranine requires caustic potash, while for fuchsine ordinary ammonia 1s sufficient ; others, again, such as the indulines, oxyazines and acridines, have their colour bases set free by very dilute ammonia; and others, such as chrysoidine, Bismarck brown, rhodamine S(By), Victoria blue, etc., dissociate in dilute aqueous solution. Thus, separation can often be made by successively shaking the aqueous solution with ether, first alone, then together with 1 per cent. ammonia, then with concentrated ammonia, and finally with 20 per cent.potash. A further separation of the bases taken up by the ether can sometimes be effected by shaking the ethereal solution with an equal volume of water, some being taken up by the water, others remaining in the ether. In this way it is possible to separate acridine yellow from the very similar phosphine. The colour bases remainingin the ether differ in their behaviour towards 5 per cent. acetic acid, some combining with it, others remaining unaltered. The acid colouring matters not extracted by ether from an alkaline aqueous solution can be separated by methods similar to those used with the colour bases. By successive extractions with ether, they can be separated into three groups: (1) those extracted by ether in the presence of acetic acid of 1 per cent.strength ; (2) those soluble in ether in the presence of hydrochloric or sulphuric acid; and (3) those insoluble in ether. Erythrosin can thus be separated from roccellin and from Bordeaux B, and direct yellow (A) from Congo brown R (A) and from Congo red (A). By treating the ethereal solutims with water and dilute ammonia, as in the case of the basic colouring matters, a further separation can often be made, as, for instance, picric acid from Martius yellow. Separation by means of Wool.-When a separation cannot be made with. ether, it is often possible by means of wool. An aqueous solution of the colouring matter (1 : 1,000) is rendered faintly alkaline by the addition of four or five drops of ammonia per 100 c.c., some wool added, and the liquid heated to boiling with constant stirring.This is repeated with fresh supplies of wool so long as the fibres are dyed. The wool is washed with boiling ammoniacal water, then with pure water, and extracted with 5 per cent. hot acetic acid. On evaporating this extract on the water-bath, the basic colouring matters are left, and can then be further separated. The use of wool is more effective in the separation of acid colouring matfers, some of which are directly fixed by it. A 0.1 per cent. solution of the colour mixture is acidified with hydrochloric acid (three to four drops per 100 c.c.), brought to the boil, and wool immersed for from three to five minutes with continual stirring. This treatment is repeated as with the basic colours.The dyed wool is washed with acidified water, and then with pure water, and the colour extracted with 5 per cent. ammonia. By boiling the liquid until the ammonia is expelled, the direct colours are obtained in neutral solution. Since Borne of the indirect colours are taken up toTHE ANALYST. 47 a slight extent by the wool fibre, this solution should also be treated with the wool in order to effect complete separation. I n this way the following colours can be separated from each other : Direct. Indirect.{ anocyanin. {Cochineal. (Saffron. Bordeaux B (A). Biebrich scarlet.JAcid yellow (A). The direct colours have a great difference in their affinity for wool, and with some it is dyed readily in a strongly acid solution (e.y., those with oxysulpho- groups), while others, of both acid and basic character, also dye in a neutral solution.The following colouring matters can thus be separated : Fixed in neutral bath. Alkali violet (B). Acid violet 4BN. (Orseille ,, acid ,, {Ponceau 6RB (A).{New coccine (A). \Bordeaux B (M). By-using a strongly acid solution (1 C.C. HC1 to 200 c.c.) the following separations ,? slightly 9 > {Orange G (A). {Methyl orange. can be made : Fixed in strongly acid bath. Bordeaux S (A). Bordeaux B (A). When a separation of the constituents of a mixture cannot be effected by means I t has the property of fixing some of the direct of wool, cotton-wool must be tried. dyes for wool, leaving others in the bath. Thus : Direct for cotton-wool. Carbazol (B). Cotton yellow R (B). Indirect ,, ,, (Diamond yellow R (By).{ Phloxin B (B). With cotton also some of the dyes are more readily fixed than others, especially under varying conditions of the acidity and concentration of the bath. Thus, in slightly acid bath (HC1) brilliant Congo (A) is readily fixed, but brilliant yellow (A) only with difficulty. If none of these means have effected a separation, other solvents, such as petroleum spirit, arnylic alcohol, chloroform, etc., should be tried. With petroleum spirit, for example, eosin can be separated from Martius yellow. C. A. M. Use of Basic Lead Acetate in the Polarimetry of Sugar Solutions. Prinsen- Geerligs. (D. Zzdcerind, 1898, xxiii., 1753 ; through Chem. Zeit. Rej?., 1898, 320.) -Although basic lead acetate does not precipitate pure lmdose, yet from impure solutions, and notably such as contain much saline matter, it carries down more or less sugar with it. On preparing artificial liquids from honey and common salt to imitate natural juices (13.44 per cent.of invert sugar and 7-68 per cent. of sodium chloride), and adding increasing proportions of basic lead acetate, the amount of laevulose precipitated varied between 343 and 23.14 per cent., thus causing a corre- sponding increment in the dextro-rotatory power of the solution. When sufficient of the reagent is introduced to decompose all the salt's present in the juice, further additions have no effect on the sugar. As lead nitrate, normal acetate, bone-charcoal, and zinc dust are all unsatisfactory, it is necessary to employ basic acetate ; and in order to obtain comparable results, the author suggests that the quantity of lead required to render molasses fit for optical examination shall be determined, and the same quantity used for all other kinds of syrup and crude sugars.The precipitated48 THE ANALYST. lzevulose also contains glucose, but the proportion of the latter never exceeds 25 per cent. of the total sugar thrown down. F. H. L. The Estimation of Indigo on Fabrics. A. Binz and F. Rung. (Xeit. angezo. Chem., 1898, 904-905.)-Renard (Bull. Xoc. Chim., xlvii., 41, 1887) proposes to heat a weighed portion of the fabric with a measured quantity of hydrosulphite solution, and determine the indigo white in an aliquot portion of this, His niethod is objection- able, since indigo white is obstinately retained by the tissue.The method of extracting the fabric with aniline in a Soxhlet apparatus, as proposed by Honig (Zeit. any. Chenz., 1889, ZSO), is much simpler. As, however, boiling aniline exercises a destructive action on indigotin, glacial acetic acid, as proposed by Brylinski (Rev. gBn6.p.. mat. color., 1898, 52), forms a better solvent, The paper contains an account of the authors' comparative experiments with Brylinski's method slightly altered, and with a modification of the hydrosulphite method. In the latter the fabric was treated with the hydrosulphite solution on the water-bath until completely decolorized, and the indigo-white removed from the tissue by repeated washing out with hot water, of which from 2.5 to 3 litres were required to remove only a few decigrammes of the dye.The whole of the solution of indigo white thus obtained was oxidized by means of a current of air, and the indigo collected on a tared filter-paper, dried and weighed. In the experiments on the glacial acetic acid method a suitable quantity (10 grammes) of dyed cotton was heated for three or four hours over a naked flame with 150 C.C. of glacial acetic acid, and then poured into 300 c.c, of water. Instead of now filteringoff with a suction-pump, as Brylinski directs, the authors found it preferable to add 150 C.C. of ether. The whole of the indigotin remained in suspen- sion in the ethereal layer, and this was separated from the diluted acetic acid by means of a separatory funnel. The former was filtered off, the precipitate washed with alcohol and ether, and dried at 110" C.The results obtained by the acetic acid method were somewhat higher than those of the hydrosulphite process, and this the authors considered might be due to over-reduction by the hydrosulphite, or to an acetyl derivative of cellulose dissolving in the acetic acid. Brylinski states that the latter body is soluble in ether, and the authors found that this was the case ; for, on extracting two equal weights of a dyed fabric, to one of which about forty times its weight of bleached cotton-wool had been added, the amounts of indigotin obtained were practically the same. C. A. N. The Determination of Sulphur in Asphalt. E. H. Hodgson. {Jozir. Amer. Cl~om. Xoc., vol. xx. 1113, pp. 882-889.)-Various samples were examined by the methods of Carius (heating with strong nitric acid in sealed tubes), Peckham (defla- gration), Eschka (ignition with calcined magnesia), and by fusion with sodium peroxide.From the results the sealed tube method appears the most accurate, but is a lengthy operation, and the tubes are liable to explode. Of the others, that of Eschka is probably the best, giving good results and requiring least time and attention. c. s.THE ANALYST. 49 Use of Persulphates to Detect Albumin in Urine. C. Strzyowski. (Schweiz. Wochs. Chem., 1898 [48] ; through Deutsche Chem. Zeit., 1899, xiv., 2.)-A 10 per cent. aqueous solution of potassium, sodium, or preferably ammonium persulphate, pre- cipitates albumin from either acid or alkaline urine, and does not throw down peptones or urates.By means of a pipette the reagent is brought underneath a layer of the suspected urine contained in a test-tube, when even in dilutions of 1: 100,000 a grayish-white turbidity is produced at the line of contact. In presence of bile pigments the colour is bright green. F. H. L. Estimation of Chlorides in Urine, Wine, Beer, and Cider. Loubiou. (Rep. Pharm., 1898 [3], x., 493; through Chem. Zeit. Rep., 1898, 319.)-Lead peroxide oxidizes in the cold those constituents of urine which act on silver nitrate, and decolorizes it sufficiently to enable potassium chromate to be used as an indicator in the titration of chlorides. 20 C.C. are shaken with 2 or 3 grarnmes of the peroxide and filtered; 10 C.C. of the filtrate are mixed with 5 drops of saturated potassium chromate, diluted to 50 c.c., and titrated as usual.Similarly, employing Blarez’s process, 50 C.C. of white or red wine (beer, cider, or vinegar) are treated with 3 C.C. of chromate, 50 granimes of salt, 100 C.C. of water, and 5 grammes of lead peroxide; the whole is shaken, filtered, and an aliquot portion titrated. F. H. L. Separation of Albumoses from Peptones. P. Muller. (Zeits. Shysiol. Chem. , 1898, xxvi., 48 ; through Clzern. Zeit. Bcp., 1898, 320.)-After experiments with the salts of various heavy metals, and especially with uranium acetate, the author prefers ferric hydrate for this purpose. The liquid to be treated is mixed with an equal volume of 30 per cept. ferric chloride solution, and alkali is added till it is only just acid. The precipitate is removed, two or three pinches of zinc carbonate are thrown into the filtrate, which is shaken thoroughly and again filtered.It will be clear, colourless, and free from albumoses, giving no turbidity (or only a very slight one visible against a dark background) when saturated with ammonium sulphate. The method has proved successful in all cases where it was tried except with Witte’s peptone, which involved a concentration of the filtrate to one-fourth or one-fifth of its volume, followed by a second treatment with a few drops of ferric chloride and zinc carbonate as before. 3’. H. L. Sicilian Sumach and its Adulterants. F. Andreasch. (Gerber, 1898 ; through &?its. nngezo. Clzem., 1898, 1154.)-True sumach is the product of Rhus coriarin, an Arabian plant ; but it is blended with the leaves of the Sommacco fimeneddu, which contain less tannin.Inferior specimens are mixed with stalk, earth, sand, and already extracted sumach. As adulterants there are added leaves of the carobbe, OC CistzLs salz;ifolius (very frequently in Sicily), of the fig, vine, of Ailanthzcs gla~~duloscc, Pistacia Zentisczu (Stinco sondro, Lentisco), and of Turnarix Africana (Bruco, albero di Giuda) : the two latter being the most common. The amount of tannin and non- tannin in sumach and its chief adulterants is shown in the annexed table. The average proportion of tannin in pure sumach may be taken at 23 or 24, seldom falling50 THE ANALYST. below 22 : a yield of under 20 per cent. points either to the presence of other parts of the plant or to sophistication.Large additions of stalk, etc., can usually be detected by the eye, if not by examination of the aqueous extract ; for the bulk of the red colouring matter in the woody fibre passes into solution, and on acidification with acetic acid yields a weak but distinct red tint. Tannins. Non - tannins. Silician sumach. . . . 21--27-5 per cent. . . . 16-22 per cent. Pist acia ... ... 13-17 ,, ... 20-26.6 ,, Tamarix ... ... 8.3-9 7 ,, ... 23-26-5 ,, Ailanthus . . . ... 10 ,) ... 17-5 ,, When a little formaldehyde is dropped into a neutral decoction of pistacia, a pale-yellow precipitate is produced. Even if the material is chiefly genuine sumach, a yellowish-coloured cloud forms, which appears as a precipitate in time, but only settles after several days. The deposit is gelatinous and coheres on the filter, so that it can scarcely be washed. The substance is insoluble in cold water ; but it does not represent a, quantitative separation of the pistacia. If the test be carried out on very old samples of sumach, which perhaps have fermented, or if much tamarix be present, the formalin may produce a precipitate even in the absence of pistacia; but the latter cannot be mistaken for the real deposit : it is only small in amount, never gelatinous, and settles in twelve hours. When solid potassium cyanide, or its strong solution, is stirred up with a decoc- tion of tamarix, or sumach adulterated with tamarix, a flocculent dark-yellow pre- cipitate is formed which settles rapidly. Pure sumach gives no deposit or a mere trace, and the reaction is characteristic of tamarix alone among the possible adul- terants. The reagents in both tests must be pure, and the solution examined should be perfectly clear, and must not exceed industrial strength, viz.,. about 0.75 gramme of tannin per 100 C.C. F. H. L. Estimation of Oil of Bergamot. A. Soldaini and C. Berte. (BOLL. chim. farm., 1898, xxxvii., 577 ; through Chenz. Zeit. Rep., 1898, 311.)-The usual adul- terants added to bergamot oil are turpentine and (or) lemon oil. To detect them, 15 C.C. of the sample should be distilled at a pressure of 20 or 30 mrn. till 5 C.C. have passed over, and the opticity of the oil, the distillate, and the residue determined. The annexed table indicates the influence of both or either of the impurities men- t ioned : Temperature. Oil. Residue. Dis tillste. Pure bergamot oil ... Y . ... 14" C. +14" 50' -0" 56" +41" 2 0 + 5% of lemon oil . . . ... 14" C. +17" 11' +3" 20' +42" 28' +2*5% of turpentine and 2.5% of lemon oil ... ... 13.5" C. + 14" 55' + 2" 40' + 40" 20' +5% of turpentine ... ... 14" C. +12" 36' -0" + 35" 28' F. a. L.