24 THE ANALYST. ORGANIC ANALYSIS. The Detection of Formaldehyde. Thevenon. (BUZZ. des Scieizces Phurnz. ; through A m . cle Chim. anal., 1905, vol. 10, p. 433.)-On adding a few crystals of metol (methyl-paramidophenol sulphate) to a solution containing formaldehyde, a garnet- red coloration is produced. Heat accelerates the reaction, but the temperature must not exceed 75" C., or the colour will become yellowish-brown. Acetic or lactic acid or sodium or magnesium sulphate do not interfere with the test, but alkalies change the colour to reddish-brown. The reaction is capable of detecting 1 part of commercial formalin in 10,000. For the detection of formaldehyde in milk, the casein is precipitated by the addition of a few drops of dilute (1 : 5) acetic acid, and the test applied to the filtered serum kept for thirty minutes at a temperature not exceeding 75" C.C. A. M, Determination of Glycerin according to Shukoff and SchestakoR. (Zeits. angew. Chem., 1905, xviii., 1656.)-From a number of determinations made in the laboratory of the Schlebusch Dynamite Factory, the authors conclude that this method should, for the present, at any rate, be only used a8 a check on other methods, as, whilst values found by it are generally in good accord with those given by other methods, occasionally it yields results for crude glycerins differing by about 1 per cent. They also state that the extraction with acetone should be continued for six hours, not for four; and they dry their glycerin at go", or even 96' to 97" C., instead of at 75" to 80" C., at which last temperature constant weight is obtained only after a very long time. They also omit, as being unnecessary, the initial filtration after adding acid to an alkaline liquid, or alkali to an acid one.A. G. L. The Identification of Lactones by Means of Hydrazine. E. E. Blaise and A. Luttringer. (BUZZ. SOC. Chim., 1905, xxxiii., 1095-1104.)-Lactones, if not too impure, can be identified by means of hydrazine, which combines with them to form crystalline compounds with well-marked melting-points. I n applying the test the lactone is mixed with an equi-molecular proportion of hydrazine hydrate, and heated on the water-bath until the whole of the water has been expelled. As a rule thirty minutes is sufficient for this when the quantity of lactone does not exceed 2 to 3 grams, but in the case of certain lactones it is necessary to repeat the evaporation once or twice.The dry crystalline product left on cooling the substance is purified by crystallization from absolute ethyl acetate, and its melting-point determined. These compounds are extremely soluble in water and absolute alcohol, but are only slightly soluble in absolute ether. In the author's opinion they are hydrazino- lactones formed by the direct combination of 1 molecule of each constituent, and have the general formula :THE ANALYST, 25 c- -... .. . -c I C-NH-NH, I I 0 OH. Hydrazino-13-y-dimethylbutyrolactone melts at 96" to 97" C., hydrazino-y- methyl butyrolactone at 61" to 62' C.? and hydrazino-a-methyl-y-amylbutyrolactone at 116" C. C. A, M.The Polarimetric Determination of Starch. E. Ewers. (Seeits. ofeszt1. Che?n., 1905, xi., 40?-415.)-The starch is brought into solution by successively digest - ing it with acetic acid, hydrochloric acid, and water, the solution obtained being after- wards clarified and polarized. The details of the method proposed are as follows : 10 grams of the starch or meal are placed in a 200 C.C. flask together with 50 C.C. of glacial acetic acid, and heated in a boiling-water bath for twenty minutea. About 130 C.C. of cold water are then added, and the digestion continued for one hour at a temperature of 45" C., with frequent shaking. After cooling, 2 to 3 C.C. of saturated potassium ferrocyanide solution are added, the volume is made up to 200 c.c., the mixture filtered, and the filtrate polarized in a 200 mm.tube at a temperature of 20" C. The reading obtained gives the amount of soluble carbohydrates, etc., in the sample, and is usually about +0.2" on the Ventzke scale. A second portion of 10 grams of the sample is then digested with 50 C.C. of glacial acetic acid for ten minutes in a boiling-water bath; 10 c.c of dilute hydrochloric acid (1 : 10) are now added, and the heating continued for exactly six minutes. Hot water is then added to make the volume up to 180 ex., and the mixture heated for a further fifteen minutes. The solution is finally cooled, clarified by the addition of 2 C.C. of the ferrocyanide solution, filtered and polarized as described previously. I n the case of potato starch the digestion after diluting with water is omitted.The following readings on the Ventzke scale were obtained, after correction for the soluble carbo- hydrates, with a number of commercial starches : Wheat starch, + 44.7' to + 45.4' ; rice starch, + 45.5" ; maize starch, + 45.0" ; potato starch, + 44.5". The quantity of water, ash, proteids, etc,, having been determined in the samples, it is calculated that 10 grams of pure starch dissolved to 200 C.C. and polarized in a 200 mm. tube would give the following readings: Wheat, +52-7" to +52%"; rice, +52*6"; maize, + 52.4" ; and potato, + 53.3". TI-. P. s. The Determination of Dextrose or Invert Sugar. J. Wolff. (Aiz?~. de Chim. anal., 1.905, vol. 10, p. 427-431.)-The method is based on dissolving the reduced cuprous oxide in a solution of ferric sulphate containing sulphuric acid, and titrating the resulting ferrous sulphate with standard potassium permanganate solution.The reagents required are : (1) Fehliizg's soZz~tio?z, consisting of (a) 40 grams of copper sulphate per litre, and ( b ) 200 grams of Rochelle salt and 150 grams of sodium hydroxide per litre. (2) Fewic szdphate soluzctioiz-viz., 50 grams of ferric sulphate and 200 grams of sulphuric acid per litre. (3) Potassizcm perrnanganate soZuthz- 5 grams per litre, standardized on +& oxalic acid solution. In the determination26 THE ANALYST. from 5 to 15 C.C. of the 0.5 per cent. sugar solution are treated with 40 C.C. of the mixed Fehling's solution in a conical flask, and the liquid diluted to 60 C.C. and boiled for exactly three minutes, after which it is filtered rapidly through a Soxhlet tube or Gooch's filter.The cuprous oxide is washed by decantation with hot water, and then dissolved in 10 C.C. of the ferric sulphate solution- Cu,O + Fe,(SO,), + H,SO, = 2FeSO,+ 2CuS0, + H,O. The solution is passed through the filter, the flask and filter washed with an additional 10 C.C. of the ferric sulphate solution, and finally with hot water, and the filtrate and washings titrated with permanganate solution. C. A. M. New Method of Determining Sugars by Means of the Immersion Refractometer. B. Wagner. ( Z e k ofentl. Chem., 1905, xi., 404-407.)-The amount of cuprous oxide precipitated from Fehling's solution by a sugar may be determined by tbe immersion refractometer (ANALYST, 1903, xxviii., 91), and the correspondir?g quantity of sugar thus ascertained. The cuprous oxide is collected in an asbestos filter-tube in the usual way and washed.I t is then dissolved in a few cubic centimeters of concentrated nitric acid, and the solution received as it drops from the tube in a small basin, After washing the filter with a little water, the copper solution is evaporated to dryness, the residue is taken up with exactly 5 C.C. of 2 per cent. nitric acid, which has a refraction of 91.6, and then diluted with water to a volume of 10 C.C. The refraction of the solution is now taken. I t is necessary to graduate the instrument by taking readings with solutions of copper produced by dissolving the cuprous oxide given by known quantities of sugar, as the refrac tion is not exactly proportional to the rise in concentration of the sugar solutions 0.002 per cent.of grape-sugar gives a reading of about 19" on the scale of the instrument, whilst 1 per cent. reads 73". For a difference therefore of 0.998 per cent. of grape-sugar, there is a range of observation of So, permitting of considerable accuracy in the the determinations. w. P. s. The Hydrolysis of Sodium Palmitate. R. Cohn. (Berichte, 1905, xxxviii., 3781-3784.)-According to Schwarz (Zeit. f. of. Chem., 1905, vi., 301) it is impossible to use aqueous acid for the back-titration of the excess of alkali in the determination of acid and saponification values, unless sugcient alcohol be used to prevent the hydrolysis of the soap. It is shown by the author, however, that Schwarz must have failed to observe the neutralization point, for when a solution of palmitic acid in an excess of hot sodium hydroxide solution is titrated with aqueous hydrochloric acid, neutrality is reached when the liquid changes its colour from deep red to faint pink, and with practice the results thus obtained agree within 0.5 per cent.of theoretical amounts. If more acid be then added to the neutralized liquid, the pink colour does not disappear until the whole of the palmitic acid has separated out. I t is impossible, with phenol-phthalein as indicator, to obtain quantitative titrations of this hydrolytic dissociation ; but this can be done with methyl-orange as indicator, the reaction then taking place in accordance with the equation : C,,H,,.COONa + HC1= C,,H,,.COOH + NaCl.THE ANALYST4 27 The behaviour of sodium palmitate on titration is thus exactly like that of sodium carbonate, if methyl-orange be used as indicator.The reaction continues alkaline, notwithstanding the addition of acid, so long as free palmitic acid ions from the hydrolysis of the palmitate remain in solution; but as soon as the whole of the palmitic acid has separated out, the free hydrogen ions enter into solution, and the colour of the methyl-orange is immediately changed to red. C. A. M. Remarks on Lubricants. G. Blars. (Chem. Zeit. Rep., 1905, xxix., 325.)- I n a reply to an article by Schreiber, the author declares that in many cases such' exacting tests are unnecessary for lubricating oils. The author discusses the flash- point and asphalt content.He proposes for the contract supply of cylinder oils, statements of the specific gravity, and of the viscosity at 150" (after Engler's method). The amount of free acids reckoned as SO, must not be more than 0.01 per cent. ; free fatty acids must not be formed in a current of steam. A content of asphalt up to 0.4 per cent. only is permissible; cylinder oil must form a clear solution in petroleum benzine of 0.700 specific gravity ; for machine oils the requirements must be different. For this case mixtures of mineral and fatty oils give the best results, especially for heavily-loaded axles. The specific gravity must lie between 0.900 and 0.915. The demand for a freezing-point of .- 15" the author thinks unnecessary, and tbe flash-point is tolerably beside the question. For dynamo oils, where the axles are not too heavily loaded, the viscosity at 20" must not be more than 15 to 16, and at 50" not more than 34 to 4. I n a few cases, such as in cart oils and fats, one need only require the lubri- cant to be free from resin, resin products, and substances increasing the difficulty of working, and have a flash-point not under 100'.E. K. H. On American Colophony. Paul Levy. (Zeits. aizgezo. Chem., 1905, xviii., 1739.)-From the analysis of a number of salts and esters of abietic acid, the author concludes that the acid has the formula C:L'oH3002, and not CI9Hz8O2, as given by Mach (~~olzcltsheft, 1893, xiv., 186; 1894, xv., 627). A. G. L. The Determination of Free Hydrocyanic Acid in Aqueous Solution. G. Guerin. (Joz~wz.Phawn. Chiiiz., 1905, xxii., 433.)-The author recommends an addition of borax to the hydrocyanic acid solution prior to the titration with standard silver nitrate or iodine solution. Modification of Liebig's Meethod.-Ten C.C. of the hydrocyanic acid solution are mixed with 10 C.C. of a 3 per cent. solution of sodium diborate and titrated with a standard solution of silver nitrate (3,148 grams per litre ; 1 C.C. = O-OOlHCN), added drop by drop with continual shaking until there is a, permanent turbidity. The reactions involved are as follows : (1) 2HCN + Na,B,O, = H,B,O, + 2NaCN. (2) 2NaCN + AiNO, = AgCN.NaCN + NaNO,. (3) AgCN.NaCN + AgNO,= 2AgCN + NaNO,. The solution under examination should be free from ammonium salts, though28 THE ANALYST, if necessary they can be converted into borate by the addition of a slight excess of a saturated solution of boric acid.Modification of the Method of Forclos and G8Zis.-Ten C.C. of the hydrocyanic acid solution are mixed with 10 C.C. of the 3 per cent. solution of borax, and titrated with a standard solution of iodine (9.407 grams of iodine, and 16 to 18 grams of potassium iodide per litre ; 1 C.C. = 0.001 gram HCN.) until there is a persistent yellow tint. (1) 2HCN + 21, = 2CNI + 2HI. (2) 2HI +- Na,B,O, = H,B,07 + 2NaI. The benzaldehyde in cherry-laurel water does not interfere with the accuracy of the results. This second method is preferable to the modified Liebig's method for cherry-laurel water and bitter-almond water, since both contain ammonium com- pounds, which necessitates the use of boric acid if the first method is employed.C. A. M. The Detection of Free Hydrochloric Aeid in Gastric Juice. Cipollina. (Rqorma Ned.; throiigh any^ cle Chim. anal., 1905, vol. 10, p. 446.)-A reagent con- sisting of a mixture of aniline water with sodium or calcium hypochlorite has a violet coloration, which is modified by small quantities of hydrochloric acid (above 0.2 per cent.) to a bluish shade, and by still smaller quantities (0.025 to 0.2 per cent.) to violet-red, subsequently changing to yellow. The presence of free lactic acid has no influence upon the colour. C. A. M. Analysis of Gambier. Greshoff. (Pharm. WeekbZad., xlii., 599 ; through Pharm. Journ., 1905, vol. 75, p. 657.)-The following method is described for the examination of a gambier, an astringent extracted from a cinchonaceous plant, Uncaria gambier.Water Extract.-Five grams of the powdered gambier are placed in a litre flask which is nearly filled with previously boiled and cooled water ; the whole is gently warmed until the gambier is dissolved, and then set aside in the dark for twenty-four hours. After diluting to the mark, the solution is filtered and the extract deter- mined by evaporating 100 C.C. of the filtrate, the residue obtained being dried at 105" C. for three hours. Tannin.-To 125 C.C. of the solution 2.5 grams of Merck's voluminous alumina are added, the mixture well shaken, allowed to stand for twenty-four hours, then filtered and 100 C.C. of the filtrate evaporated. The difference in the weight of the extract before and after treatment with alumina is taken as tannin.Crude Catechin.-Two grams of the ganibir are powdered with an equal quantity of pumice, transferred t o a flask with 50 C.C. of ethyl acetate, and placed aside for twenty-four hours, with occasional shaking. The mixture is then filtered, the residue washed twice with 5 C.C. of ethyl acetate, and the solvent and washings evaporated, The residue is dried at 105" C. Crystalline Catechin.-The residue of crude catechin is dissolved in 10 C.C. of warm water, filtered, the filter washed with 5 C.C. of water, and the filtrate allowed to crystallize, inoculating if necessary with a crystal of catechin. The crystals obtained are collected on a small filter, washed with 5 C.C. of water, and dried at 105" C. Good gambir should contain from 30 to 35 per cent.of this constituent.THE ANALYST. 3 w g 5.91 4.57 7.36 6.82 13.00 6.16 4-20 16.82 15-15 14.17 11.56 9.22 29 d 25 q 34.80 34.19 16.06 17.31 17.00 8.19 7.81 8.12 8.06 9.06 13.62 14.00 Moisture.-Two grams of the gambier are powdered with 2 grams of dry pumice and the mixture dried s t 105" C . The norms1 amount of moisture in gambier is about 15 per cent. Ask-The ash varies from 2 to 4 per cent., and should not exceed 5 per cent. w. P. s. 3.37 3.79 2.85 3-41 3.35 6.66 8-12 8.79 .0-87 7.37 5-33 6-40 The Composition of Certain Marine Algze (Seaweeds) and of the Products obtained from them. J. Konig and J. Bettels. (Zed. Unterszcckc. Nahr. GenusmitteZ, 1905, vol. 10, p. 457-473.)-The composition of various Eastern Asiatic seaweeds is shown in the following table, the results expressing percentages on the air-dried substance : 0.15 2-52 0.30 2.5C 0.71 13.21 1-13 13.39 0.91 12.90 1-12 11.27 0.84 12.33 1.33 25.91 1.37 26.16 16.52 5.3C 1-06 14-08 0.25 9.23 Porphyra ...... ... ... Porphyra tenera ... ... Gelidium cartilagineum . . . Laminaria japonica . . . ... Cystophyllum . . . ... ... Cystophyllum fusiforme ... Ecclonia Licyclis .. ... Undaria pinnatifida . . . ... Gelidium raw . . ... ... Gelidium bleached . . . ... Laminaria ... ... ... ... Enteromorpha compressa . . . Origin. Nitrogen- Extractives. Water. Fat. free Protcid. 31.94 21-75 7-87 7-37 7.37 5.68 5.44 3.13 4-25 5-50 7.50 5-31 0.87 0.59 0.98 0.73 0.80 0.50 0.39 0.50 0.43 0-20 0.28 0.65 L7.87 L6.49 L6.34 i0.47 16.04 17-02 .6-35 .7.43 16.28 15.35 5 1 2 :0*16 c c 2 7.51 7-57 12.49 5 *74 11.88 j0.06 39.29 i1.18 30.53 12.12 18.72 35.13 Porphyra and gelidium were found to contain i-galactose and d-galactose, the respective products of these two seaweeds-namely, nori, or vegetable isinglass, and agar-agar, also yielding the same sugars.The quantity of the seaweeds at the author's disposal did not permit of the detection of other hexoses, such as mannose, which Tollens has stated to be present in these a l p . In the case of enteromorpha compressa, however, rhamnose was certainly present. I t will be noticed that the specimens containing notable amounts of pentosans also yielded methylpentosans. The authors have also analysed two samples of edible birds'-nests, with the following percentage results : East Coast of Java ...1042 57.37 0.09 21.98 Hongkong ... ... 16-33 54.72 I I 1 - 1 -30 THE ANALYST. The composition is quite ciifferent to that of other seaweed products (agar-agar and nori). Whilst the quantity of proteid is very large, the amount of carbohydrates is sinall-15 to 20 per cent.--and there is no reason to doubt the conclusion that edible birds'-nests are formed from the vomit of sea-swallows. w. P. s. A Sensitive Reaction of Formaldehyde and Nitric or Nitrous Compounds with Proteids. E. Voisenet. (BaZZ. Soc. Chim., 1905, xxxiii., 1198-1214.)-1f a pinch of egg albumin be mixed with 2 to 3 C.C. of water and a drop of formaldehyde solution (5 per cent. solution of commercial formalin), and the liquid treated with three times its volume of hydrochloric acid containing a trace of potassium nitrite (acid of specific gravity 1.18 to which has been added 0.5 C.C.of a 3.6 per cent. solution of potassium nitrite per litre), there is an immediate rose coloration which gradually changes to deep violet-blue. The reaction is produced in the cold, but is favoured by heat, a temperature of about 50" (3. being the most suitable for the detection of formaldehyde, of which as little as 1 part in 10,000,000 will produce a rose coloration. For fixed quantities of albumin and nitrous acid the intensity of the coloration increases with the amount of formaldehyde added until it reaches a, maximum, after which it decreases, until finally the colour disappears when the aldehyde is in large excess. The reaction is equally sensitive for albumin, and is not prevented by the albumin being present in excess.Sulphuric acid behaves like hydrochloric acid in the test, which can be used advantagously for the detection of nitrous compounds in the former. The reaction is to be attributed to the production of small quantities of scatol and indol formed by the action of concentrated acids on proteids, the oxidation of these compounds by the nitrous acid and the condensa- tion of the oxidation products by the formaldehyde with the formation of the colour- ing-matter. Most proteids give the reaction, keratin, gelatin, and pure peptones being exceptions. The majority of aldehydes are incapable of replacing form- aldehyde in the test, but salicylic and other phenolic aldehydes give the same reaction, whilst acrolein and benzaldehyde give deep blue or bluish-green colorations quite distinct from the formaldehyde violet colour.The reaction is not character- istic of nitrous acid, being also produced by other oxidizing agents, such as chlorine, bromine, iodine, hydrogen peroxide, nitric acid, ferric salts, etc. I t is particularly sensitive with nitric acid, of which it i8 capable of detecting 1 part in 20,000,000. The colour is not produced, or is rapidly destroyed, by reducing agents, including formaldahyde in excess. The reaction can be used for the detection or colorimetric determination of formaldehyde in milk, and for the determination of albumin in urine, in which case the flocculent deposit obtained on coagulating the acidified liquid is dried and examined. C. A. M. Toxicity of Saponin. Bourcet and Chevalier. (Bzdl. Xci. Plzam., vii. , 262 ; through Pharm. Jozmz., 1905, vol. 75, p. 691.)-The saponin of commerce is usually not a simple body, but a mixture of acid saponins, which are innocuous, with neutral sapotoxins, which are decidedly poisonous. The latter, when applied externally to the body, produce local antesthesiu and erysipelatous inflammation. Injected into the blood, they are fatal in doses of 0.0002 to 0.0005 gram per kilo of animal, causingTHE ANALYST. 31 paralysis of the central nervous system, and arresting the cardiac and respiratory actions. I n less than fatal doses, the gastro-intestinal organs are strongly inflamed, ending nearly always in perforation. When administered internally, the sapotoxins are less toxic, but cause all the symptoms of catarrh of themucous membrane. The harmless saponins can be separated from the sapotoxins by making the powdered drug into a paste with calcined magnesia, drying the mixture on a water-bath, and extracting the mass with boiling ethyl acetate. The extract is filtered, evaporated to one-half its volume, and mixed with anhydrous ether, when absolutely innocuous saponin is precipitated. The saponins form a series of bodies of varying composition, and split up when hydrolysed into sapogenin and a sugar, which may be galac- tose, dextrose, or rhamnose. I n certain cases crotonic aldehyde is a product of the hydrolysis. The saponins are not dialyzable. w. P. s.