12 THE ANALYST. ORGANIC ANALYSIS. (Chem. Zeit. 1901 xxv. 610.) -The sample is dried in the air or over quicklime ground or beaten to powder and 6 grammes of it are weighed into a 200 or 300 C.C. porcelain basin. Here it is moist-ened and well rubbed up with 10 C.C. of 20 per cent. sodium hydroxide and plaster of Paris is added by degrees till the whole forms a dry powdery mass. (This is better than adding plaster to only a paste-like consistency as recommended by Neumatnn.) The powder is brought into a stout 25 x 5 centimetre cylinder where it is repeatedly agitated for an hour with 100 C.C. of a mixture of ether and petroleum spirit. Twenty-five C.C. of the clear liquid are drawn off run into a glass basin treated with 40 or 50 C.C. of water and 1 drop of iodeosin solution; then an excess of decinormal acid is introduced and the solution titrated back with alkali of equal strength.The results are accurate. Experiments on the influence of ammonia on this process show (1) that the 25 C.C. of ethereal extract may contain 0.5 milligramme of ammonia as a maximum, but that the quantity is usually less ; and (2) that if the sample contained ammonium salts 98.5 per cent. of the corresponding ammonia would be retained in the gypsum. The method proposed by Keller (ANALYST 1898 xxiii. 235) for removing ammonia from the ethereal solution by a current of air is not trustworthy since the loss in alkalinity of the liquid and the alkaline vapours in the escaping air are due to slight volatilization of nicotine itself. Moreover in Keller’s process nicotine is retained by the aqueous liquid when it is extracted with the mixed solvents.Estimation of Nicotine in Tobacco. J. Toth. F. H. 5. Formaldehyde and Sulphuric Acid 8s a test for Alkaloids. H. Linke. (D. Pharm. Ges. Ber. 1901 xi. 258; through Chem. Zeit. Rep. 1901 184.)-The above reagent gives the following colour tests with alkaloids-Colchicine golden yellow gradually disappearing. Hydrastinine greenish yellow gradually disappear-ing. Strychnine green-brown on warming. Atropine brownish on warming ‘‘ dirty greenish-grey-brown.” Romatropine brown on warming. Scopolamine golden yellow to orange brown-black on warming. Veratrine yellow-brown reddish-brown on warming. Digitaline brick to wine-red dark wine-red on warming. Morphine : cherry-red then violet grey-black on warming.This reaction is very fine and sharp. Apomorphine violet then rust-red and black-blue greenish-black on warming. Codeine violet-blue brown-black on warming-a fine and very sharp reaction. Cocaine pilocarpine cantharidine eserine and caffeine give no reactions either hot or cold. F. H. L. A Reaction of Urotropine and Piperazine. Manseau. (BUZZ. de la soc. Pham. Bordeaux June 1901 ; Ann. de Chim. and. 1901 vi. 339 340.)-The author’s tests are based upon the reaction used by Jorissen for the detection of formaldehyde (ANALYST xxiii. 41). Urotropine or hexamethylene-tetramine is decomposed on treatment with acids into formaldehyde and ammonia and can thus be detected by the purple coloration, changing to blue which it then gives with morphine THE ANALYST.13 By using other opium alkaloids instead of morphine the following colorations are obtained codeine blue changing to dark - green ; apomorphine violet-blue ; narcelne saffron-yellow ; thebaine greenish-yellow ; and narcotine golden-yellow. Piperazine or diethylene-diamine also gives colorations when treated with opium alkaloids in the presence of sulphuric acid. Thus with morphine and codeine there is a slight violet tint whilst with papaverine narcelne narcotine and thebaine the colours vary from bright yellow to orange and deep brown. C. A. M. Action of Reducing Sugars on Nickel Salts in Alkaline Solution. M. Duyk. (BUZZ. de Z'Ass. belge. 1901 xv. 267 268.)-The author's reagent consists of 25 C.C. of a 20 per cent. solution of nickel sulphate with 25 C.C.of sodium hydroxide solution (specific gravity 1-33) 3 grammes of tartaric acid and 50 C.C. o€ water. On boiling this reagent with the smallest trace of a reducing sugar a pronounced turbidity is produced which eventually subsides as a black precipitate of probably, a sub-oxide of nickel. The advantages claimed for this reagent over Fehlings solution are that spon-taneous reduction does not occur and that the reaction is more distinct in doubtful cases. Moreover it is not affected by normal urine which often decolorizes Fehling's solution and its modifications. The author has found that cobalt salts are not reduced under the same conditions, and points to the possibility of basing a method of separating nickel from cobalt on this fact.C. A. M. Isolation of Amido-Sugars. H. Steudel. (Zeits. physiol. Chem. 1901., xxxiii. 223 ; through Chem. Zeit. Rep. 1901 275.)-In alkaline solution phenyl isocyanate combines with glucosamine to form a substance which is very insoluble in water and which separates from its solution in dilute acetic acid as a heavy crystalline powder having a sharp rnelting-point. When 2.25 grammes of glucosamine hydrochlorate were dissolved in 30 C.C. of water and 10 C.C. of normal potrtssium hydroxide and 1.19 grammes of phenyl isocyanate dropped in with constant agitation and cooling the liquid quickly solidified to a thick paste. The solid matter was re-moved and dissolved in hot acetic acid. On cooling large rhombic crystals separated, which were recrystallized from water and dried at 110" C.On analysis they gave figures corresponding with C,,H,,N,05. This body begins t o turn brown at 200° and melts sharply at 210" C. Isoglucosamine yields a similar product and the author believes that other amido-sugars may be isolated by the same process. F. H. L. Detection of Mineral Oil in Rosin Oil. D. Holde. (Mittheil. KgZ. Techn. Verswhsanstalten xu BerZin 1901 xix. 39 ; through Chern. Zeit. Rep. 1901 232.)-Ten C.C. of the oil are dissolved in 90 C.C. of 96 per cent. alcohol at the laboratory temperature shaking if necessary. If a large quantity of oil remains undissolved it may be considered &s a sign of the presence of much mineral oil. This may be confirmed by collecting the oil rinsing it with a little alcohol and determining it 14 THE ANALYST.refractive index. If however nearly all the original oil dissolves in the spirit any undissolved matter is neglected. The whole is cautiously diluted with water till a distinct milkiness is produced and the mixture is set aside (all night if so required). The clear alcoholic layer is poured off from the residue which should not exceed 1 c.c. and tha oil is rinsed with a few C.C. of alcohol. It is then dissolved as before in 20 C.C. of 96 per cent. alcohol and treated with water till not exceeding 3 drops of oil separate. These are again rinsed with spirit washed with hot absolute alcohol, dried in a basin and examined for their refractive index. An index of below 1.5330 at a temperature between 1 5 O and 18" C. shows the presence of mineral oil.F. H. L. Linseed Oil and its Adulterants. P. C. McIlhiney. (Report to the Com-missioner of Agriculture of New York State 1901 1-31.)-The author states that the character of linseed oil now made in New York State does not correspond with that given in even recent text-books for it is prepared by heating metallic oxides with a small quantity of the oil and adding this 4 d drier " to the main body which is kept at a much lower temperature than in the old-fashioned method. The figures given by the oil treated in the new way are practically identical with those for raw linseed oil. A table is also given showing the corrections for temperature to be added to or subtracted from the readings of a glass hydrometer correct at 15.5" C. when immersed in linseed oil. According to this the correction for 1" F.averages 0*0001361 and for 1" C. 0.000650. Hehner and Mitchell's test based on the amount of insoluble bromine derivatives (ANALYST xxiii. 310) is regarded as a valuable one for the detection of the adultera-tion of linseed oil. The author has obtained results substantially in agreement with theirs. C. A. M. The Detection of Drying and Marine Animal Oils. G. Halphen. (JOZ~T?~. Phrm. Chim. 1901 xiv. 359-365 and 391-397.)-The author's test is based upon the fact recorded by Behner and Mitchell (ANALYST xxiii. 315) that these oils (i.e. the glycerides) yield insoluble derivates on treatment with bromine. His reagent is prepared by mixing twenty-eight parts by volume of glacial acetic acid with four parts of nitro-benzene and one part of bromine; 0.5 gramme of the oil under examination is shaken in a stoppered tube with 10 C.C.of this reagent, until the liquid is homogeneous ahd is then allowed to stand. In this way it was found that no turbidity resulted after an hour in the case of olive almond castor cotton-seed poppy lard and neatsfoot oil. With the following oils no precipitate was formed but only a slight turbidity : Jaffa sesame oil and certain samples of the foot oil of the horse and sheep. A marked turbidity and eventually a heavy precipitate were given by the following nut oil (precipitate after five to ten minutes) hempseed linseed seal cod-liver whale spermaceti and Japanese fish oils. Colza oil was distinguished by giving a turbidity which on standing separated into two layers of liquid THE ANALYST.15 The precipitates thus obtained with the different classes of oils vary in their properties. Thus on washing the insoluble bromides of linseed oil and of fish oils with sulphuric ether to remove the excess of the reagent and then pressing them between filter-paper and drying them in the air heavy substances are left which differ in their behaviour towards carbon tetrachloride. The solutions obtained by treating the precipitates with this solvent in the pro-portion of about 2.5 C.C. to each 0.1 gramme at a temperature near the boiling-point and then filtering are very different. The linseed oil precipitate yields a perfectly clear solution which on cooling deposits little by little a gelatinous precipitate. With fish oils on the other hand the solution is opalescent and only deposits a trace of a crystalline precipitate on cooling.The bromides of other drying oils behave like linseed in this respect but in the case of hemp-seed oil several hours are required before the gelatinous precipitate is deposited. The other marine animal oils mentioned above all yielded opalescent solutions in this test. I n the case of the foot oil of terrestrial animals a clear solution which does not yield a gelatinous precipitate is obtained. The author describes a series of test experiments to show that by this method it is possible to detect 10 per cent. of linseed hempseed whale or fish oil in castor, cotton-seed or olive oils. C. A. M. Occurrence of Mixed Glycerides in Cacao Butter. J. Klimont. (Berichte, 1901 xxxiv.2636.)-Cacao butter in the author's experiments was separated into three fractions by crystallization from acetone. The first of these melted at 64" C., had an iodine value of 0 after repeated recrystallization and finally yielded crystals melting at 70" C. The second fraction consisting of granular crystals melted between 31" and 32" C., and had an iodine value of 28.9 a saponification value of 196.4 and a composition agreeing with the formula C55H10406. The author concluded that it was a triglyceride containing the three radicles of palmitic oleic and stearic acid. I t melted at 26" to 27" C. had an iodine value of 31.7 and a saponification value of 210.5 and from an elementary analysis was concluded to have the composition C,,H,GO,. It was found to consist of a mixture of tripalmitin and tristearin.The third fraction was also regarded as a mixed glyceride. The author was unable to identify triolein in cacao butter. C. A. M. The Use of Iodine Monobromide in t h e Analysis of Fats and Oils J. Hanus. (Zeit. fiir Untersuch. der Nahr. und Genussmittel 1901 xx. 913-920.)-For determining the iodine absorption of fats and oils the author uses iodine monobromide. This is pre-pared by slowly dropping 13 grammes of bromine into 20 grarnmes of finely powdered iodine contained in a beaker. During the addition of the bromine (which should take about ten minutes) the contents of the beaker are well stirred and kept below 8" C. Each drop of bromine causes a lump to form which must be broken up. Whe 16 THE ANALYST.all the bromine has been added a rapid stream of carbon dioxide is passed through the beaker to remove excess of bromine. The iodine monobromide BO obtained is a dark-grey substance having a metallic lustre. I t keeps well in closed vessels and is soluble in ethyl alcohol and glacial acetic acid. The reagent is prepared by dissolving 10 grammes of the monobromide in 500 C.C. of glacial acetic acid. The exact strength of this solution is determined by titrating 10 C.C. after the addition of 15 C.C. of 10 per cent. potassium iodide solution with standard thiosulphate solution. To obtain the iodine number of an oil or fat from 0.1 to 0.7 gramme of the same is dissolved in 10 C.C. of chloroform; 25 C.C. of the above reagent are added and the mixture is allowed to stand for fifteen minutes.Fifteen C.C. of 10 per cent. potassium iodide solution are then added and the titration carried out a8 usual with thio-sulphate solution. The use of starch solution at the end of the titrations may be dispensed with. With oils and fats having an iodine absorption below 100 the reaction is complete in ten minutes. The strength of the reagent should be checked in each set of determinations as it slightly decreases. The results obtained agree well with those of the Hiibl method. w. P. s. Quantitative Separation of Cholesterin (Phytosterin) from Fats. E. Ritter. (Chern. Zeit. 1901 xxv. 872.)-About 50 grammes of the fat are weighed into a 1,500 C.C. porcelain basin using if necemary 100 C.C. of alcohol to rinse it out of the weighing vessel.I t is then heated to the boiling-point on the water-bath and mixed, with continual stirring with an alcoholic solution of sodium ethoxide made by dissolving 8 grammes of sodium (freed from petroleum and oily matter) in 160 C.C. of 99 per cent. alcohol without cooling. The whole is heated on the water-bath till the alcohol has evaporated when 75 grammes (one and a half times the weight of the fat taken) of common salt are added together with enough water to dissolve most or all the contents of the basin. With constant stirring the liquid is evaporated to dryness, first over a naked flame next on the water-bath then in a drying oven at 80" C. and finally in the desiccator over sulphuric acid beginning to reduce the mass to powder as soon as possible. The residue is extracted with ether in a large Soxhlet employing a fat-free paper thimble with a plug of cotton-wool on top.This operation is generally finished in about nine hours. The ether in the flask is usually turbid at first owing to traces of glycerin but these gradually adhere to the walls and remain attached when the ether is poured off into another vessel to make room for a fresh quantity. Finally all the ethereal extract is brought into a 750 or 1,000 C.C. flask, the solvent distilled away the residue taken up in the minimum of alcohol and water is added nearly to fill the flask. The precipitated substance is collected on a paper washed with water dried on the filter in an oven at 60" C. scraped off into a tared basin the remainder rinsed off the paper with ether and the whole is distilled or evaporated dried at 100' or 120' C.and weighed. F. H. L. Characteristics of Neroli Oil from different Sources. E. Thsulier. (Bull. Xoc. Chim. 1901 xxv. 762-764.)-The author has examined thirty-three specimen THE ANALYST. 17 of Neroli oil distilled by the method in use at Grasse from orange-flowers collected between May 16 and June 2 of the present year. I t was found that the specific gravity varied but little whatever the origin of the oil the extreme values being 0.869 and 0.8726 and the mean 0.8709. The differences in the rotatory power were somewhat greater ranging from + 2" 50' to + 7" 2 0 in a 100 millimetre tube at 23" C. The proportion of methyl anthranilate was determined by the method of A. Hesse and 0. Zeitschel in which this ester is precipitated in the form of a sulphate which is only soluble with difficulty in ether but which can be dissolved in water and titrated.The average quantity in the samples of oil examined was 0.7 per cent. The esters determined as linalyl acetate varied between 8.08 and 14.7 per cent., the mean being 11.27 per cent. This maximum was lower than in preceding years, in which the highest proportion was usually 17 to 18 per cent. The yield of oil obtained from the flowers varied considerably with their origin. It ranged from 0.91 to 1.32 per cent. The mean was + 4' 48'. C. A. M. Composition of Cannes Geranium Oil. Jertncard and Satie. (Bid. SoC. Chim. 1901 xxv. 516-519.) - I n continuation of their research on geranium oil (ANALYST xxv. 160) the authors have made a complete examination of thirty-one samples of Cannes oil distilled at different periods between September 3 and October 6.In the subjoined table their results are summarized together with those of the united residual oils after neutralization saponification and determination of the alcohols : Geranium Oil. Ordinary . Neutralized Acetylated . . . Saponified . . . Specific Gravity at 25" C. 0.8874 to 0.8906 0-8777 0.9070 0.8802 Rotatory Power at 15" C. (I = 100) -9'38' to - IO"55' - 8'50' - 2'40 - 3"O' 3uperficial Tension. 2-939 to 2.970 2.910 3.042 2.907 Specific Viacoeity . 1'18" to 1'25" 1'23" 0'39" 1'42" Saponification Value. r Hot. 49 to 54 16.8 211.4 0 -Cold.27 to 36 0 0 0 Solubility in 70 per cent. Alcohol at 15" C. 1 vol. in 1-7 to 1.9 1.0 Insol. in 20 1.5 From these results the authors conclude that (1) Neutralization diminishes the specific gravity rotatory power and surface tension but increases the solubility ; (2) acetylation increases the specific gravity and surface tension but diminishes the specific viscosity and solubility ; (3) saponification increases the specific viscosity and the solubility, The fact that neutralization increases the solubility explains why Cannes geranium oil is somewhat more soluble than geranium oils from other sources for it contains less free acids than the others. C. A. M 18 THE ANALYST. Myrcenol. Barbier. (Conzptes Rend. cxxxii. 1048 ; through Pharm.Journ., 1901 lxvii. 89.)-The alcohol obtained by the hydration of myrcene is not identical with licareol as was stated by Power and Kleber but is a new alcohol-myrcenol-having the constitutional formula : It is a colourless strongly-smelling liquid boiling at 99" to 101" C. at 10-millimetre pressure and having a specific gravity of 0.9072 at 14.5" C. The formula given above is identical with that of Tiemann for licareol (linalool) and consequently the formula of the latter needs reviewing. The acetate-a pungent oily liquid-has an entirely different odour from that of linalyl acetate. By oxidation with chromic acid an aldehyde C,,H,,O is obtained which is not citral. I t boils at 110" C. at 10-millimetre pressure gives an oxime boiling at 148" to 150" C.under the same (CH,),C CHCH,*C(CH,)*OH.CH*CHp pressure and a semi-carbazone melting at 195" to 196" C. w. P. s. On the Bromine and Iodine Values of Proteids. W. Vaubel. (Zeit. anal. Chem. 1901 xl. 47@474.)-It was shown by Blum and the author (J. prakt. Chern., 1898 lvi. 393; and lvii. 365) that undecomposed albuminous bodies such as egg-albumin and casein were capable of absorbing the following amounts of halogens at the maximum Iodine 6 to 7 ; bromine 4 to 5 ; chlorine 2 to 3 ; and fluorine about 1 per cent. I t was also found that in these experiments a greater quantity of hydro-chloric acid. etc. was liberated than corresponded to the degree of substitution. The author has therefore made experiments to determine the total amount of halogen used both in the substitution and in the formation of hydrobromic or hydriodic acid by the withdrawal of hydrogen.Using a solution of bromate for the titration and deducting 9.0 grammes of bromine for the substitution and the corresponding hydro-bromic acid liberated the following results were obtained calculated on 100 grammes of the anhydrous and ash-free material : Bromine. Egg-albumin . . . . 35.04 grammes. Blood-albumin . . . . 40.76 ,, Casein . . . . . 27.00 ,, On dissolving these proteids in alkali and boiling them for an equal length of time the sulphur split off as hydrogen pulphide and reacting with the bromine was eliminated and less bromine absorbed. This sulphur as hydrogen sulphide was calculated from the bromine results to be as follows Egg-albumin 1.12; blood-albumin 0.72; and casein 0.45 per cent.One gramme of the air-dried albumin etc. was treated with 50 C.C. of water in a stoppered litre-flask and after standing over night mixed with 20 C.C. of & iodine solution, and left for three days ; 500 C.C. of water were then added and the liquid titrated with & thiosulphate solution of which not more than 11 C.C. or less than 6-5 C.C. was required. In this way it was found that the anhydrous and ash-free substances consumed In determining the iodine values Dieterich's method was first used THE ANALYST. 19 the following amounts of iodine Egg-albumin 21.345 ; blood-albumin 22.285 ; and casein 20.844 per cent, Deducting 6-5 x 2 grammes for the iodine absorbed by substitution and the corre-sponding hydriodic acid liberated the following values were obtained for the hydrogen withdrawn by the iodine without substitution : Iodine.Hydrogen. Bromine. Hydrogen. Egg-albumin . . . 8-345 0.0657 . 35.04 0-4380 Blood-albumin . . . 9.285 0.0731 . 40.76 0.5095 Casein . . 7.644 0.0601 - 27-00 0.3450 From a comparison with the bromination figures these iodine values appear to be incomplete. By treating the solution of the proteid with 10 grammes of sodium bicarbonate, and adding 100 C.C. of the iodine solution considerably higher results were obtained, thus Egg-albumin 56-41 ; blood-albumin 60.87 ; and casein 51.88 per cent. De-ducting from these values 13 gramrnes of iodine for the substitution process the following figures were obtained in which the amounts of hydrogen are in much closer agreement with the bromination results : Iodine.Hydrogen . Per cent. Per cent. Egg-albumin . . . . . 43-41 0.3418 Blood-albumin . . 47.87 0.377 Casein . . . . 38-88 0,3061 C. A. M. Gas - Volumetric Estimation of Chlorides and Phosphates in Urine. E. Riegler. (Wieney mecl. BZ. 1901 xxiv. 527; through Chem. Zeit. Rep. 1901, 246.)-This process depends on the liberation of nitrogen from hydraeine sulphate by means of silver chloride according to the equation 4AgCl+ N,H,.H,SO + 6NaOH = 4Ag + 4NaCl+ Na,SO + 6H,O + N,, the nitrogen being measured in any convenient instrument and converted into weight by Baumann’s tables. From the chlorides of the urine the silver chloride is prepared in the usual fashion ; the phosphoric acid is thrown down with magnesia mixture, the precipitate filtered and washed and dissolved in nitric acid.The solution is mixed with silver nitrate and ammonia to a faintly alkaline reaction and boiled for a few minutes. The silver phosphate is then decomposed with hydrochloric acid in nitric acid solution. One milligramme of nitrogen corresponds with 8.23 milli-gramnies of NaC1 or with 3-34 rnilligrammes of P,O,. F. H. L. The Determination of Traces of Sugar in Urine. E. Raimann. (Zeit. anal. Chm. 1901 xl. 390-402.)-1n this paper the author first shows that a singleprecipi-tation of the sugar with phenylhydrazine gives unreliable results and then describes the following modification 500 C.C. of the urine are fermented with fresh yeast at 34O C. for twenty-four hours and then treated at the same time as 500 C.C.of the non-fermented urine in the following manner After the addition of 100 C.C. of a 25 per cent. solution of lead acetate the liquid is filtered and the excess of lea 20 THE ANALYST. removed from the filtrate by means of hydrogen sulphide; 450 C.C. of the second filtrate are evaporated on the water-bath to about 100 c.c. then made up to 120 C.C. and filtered. Of this filtrate 100 C.C. are heated on the water-bath for one and a half hours with 5 C.C. of glacial acetic acid and 3 C.C. of phenylhydrazine and after standing for twenty-four hours the precipitate collected on a dried and weighed filter-paper and dried first in the air and subsequently at 110' C. until constant in weight. The difference between the weight of osazone thus obtained and that given by the unfermented urine is taken as pure phenyl-glucosazone.I n test experiments in which known quantities of sugar were added to the fermented urine the amount of osazone calculated into glucose only averaged 49 per cent. of the theoretical quantity and in order to obtain the percentage of sugar in the urine the difference in weight of the osazone precipitates was multiplied by the factor 0.329. In this way the following results among others were obtained : Sugar Added. Sugar Found. Per cent. Per cent. 0.005 0.006 0*100 0.095 0.206 0.213 Error. Per cent. + 0.001 - 0.005 + 0.011 The fermented urine invariably yields with phenylhydrazine a non-crystalline precipitate which adheres firmly to the bottom of the beaker.C. A. M. The Detection of Acetanilide in Urine. A. Petermann. (Ann. de Chim. anal., 1901 vi. 165.)-About 10 C.C. of the urine are boiled for a few minutes with 25 C.C. of hydrochloric acid. After cooling 1 C.C. of a 3 per cent. aqueous solution of phenol is added followed by 2 or 3 drops of a 10 per cent. solution of calcium chloride the liquid being shaken after the addition of each drop. In the presence of para-amido-phenol the decomposition-product of acetanilide the urine acquires a red colour, which however it is difficult to distinguish from the coloration which normal urine gives with hydrochloric acid. Concentrated ammonium hydroxide is next poured down the side of the glass, this reagent changing the red colour to blue. The reaction is certain in the presence of acetanilide but may be doubtful when it is not present.As a confirmatory test 100 or 200 C.C. of the suspected urine are mixed with a fourth of its volume of hydrochloric acid and boiled for several minutes. When cool the liquid is neutralized with calcium carbonate and extracted several times with ether. The latter is decanted and shaken with water containing one-fourth of its volume of hydrochloric acid. The lower layer is then withdrawn gently warmed to expel the last traces of ether and tested for acetanilide as above. C. A. M. Estimation of Ammonia in Urine. 0. Folin. (Zeits. physiol. Chem. 1901, xxxii. 515; through Chem. Zeit. Rep. 1901 239.)-The urine or the solution of urea, is diluted with 400 or 500 C.C. of water and distilled with magnesia or lime-wate THE ANALYST.21 for forty-five minutes (counting only the time of actual ebullition) collecting the distillate in decinormal acid Without stopping the operation a volume of water, roughly equal to that which has distilled over is then added to the retort and the process is continued for a second period of forty-five minutes using a different receiver. On the assumption that the decomposition of the urea proceeds at a uniform speed the amount of ammonia in the second distillate (as determined by titration) deducted from that in the first gives the ammonia which existed as such in the urine. F. H. L. On the Estimation of Urea in Urine. J. H. Long. (Jouriz. Amer. Chem. SOL, xxiii. 633.)-In the opinion of the authors the old Liebig method in which the urea is titrated with mercuric nitrate solution is capable of rendering good service where a number of quick comparative determinations are to be made.The method was discarded because other substances besides urea present in urine react with mercuric nitrate and consequently the authors have made numerous experiments to determine the influence which these bodies exercise. The most important of these after the chloride-the effect of which is well known and easily corrected-are ammonia uric acid and creatinin. The average correction for these threelbodies was found to be 2.0 C.C. mercury solution which may be used in the case of fresh urine with only a small margin of error whilst if the disturbing bodies themselves are estimated the corresponding correction may be applied with considerable accuracy.A. G. L. Estimation of Urea in Urine 0. Folin. (Zeits. physiol. Chem. 1901 xxxii., 504 ; through Chem. Zeit. Rep. 1901 239.)-Urea is quantitatively decomposed into ammonia and carbon dioxide within half an hour when it is heated with magnesium chloride boiling in its own water of crystallization (at 160" C.). Accordingly 3 C.C. of the sample 20 grammes of magnesium chloride and 2 C.C. of strong hydrochloric acid are boiled in a 200 C.C. Erlenmeyer flask under a 200 x 10 millimetre reflux tube till the drops of condensed liquid fall with a hissing noise into the flask. The boiling is continued for another twenty-five or thirty minutes when the whole is cautiously diluted with water rinsed into a 1-litre flask and distilled with 7 C.C.of 20 per cent. sodium hydroxide till about 350 C.C. have passed over. The distillate is boiled up cooled and titrated every 1 C.C. of decinormal ammonia in the liquid corresponding with 3 milligrammes of urea. Corrections must be made for any ammonia in the magnesium chloride and for pre-existing ammonia in the urine. F. H. L. The Analysis of Shellac. E. J. Parry. (Chemist and Druggist 1901 689.)-The writer states that samples of shellac absolutely free from rosin are the exception and not the rule For detecting this adulteration with rosin he relies on the deter-mination of the acid number the ester value and the iodine absorption. Pure samples of shellac gave acid values between 55 and 65 with an average of 60 whilst rosins gave an average of 160. The ester values of shellacs lie between 155 and 175, those of rosins generally below 20 averaging about 10.The iodine numbers o 22 THE ANALYST. different samples of shellac examined lay between 4 and 10. Rosin gives an average iodine absorption of 110. w. P. s. Detection of Chrome Yellow in Dyed Cotton. P. Cazeneuve. (Bull. SOC. Chim. 1901 xxv. 7 61-762.)-Diphenyl carbaeide is an extremely sensitive reagent for chromic acid the limit of sensibility exceeding 1 part in 1,000,000. The cotton fibre is treated with 1 C.C. of a 10 per cent. solution of potassium hydroxide an immediate decolorization being effected in the presence of lead chromate. The liquid is then rendered strongly acid with acetic acid and shaken with a pinch of diphenyl carbazide or its acetate when a pronounced violet colour confirms the presence of chromic acid.C. A. M. The Use of Certain Indicators by Artificial Light. A. Kufferath. (Zeits. f. angew. Chem. 1901 916.) -Experiments were carried out to ascertain whether the indicators could be relied upon for use with different sources of artificial light. The indicators used were Methyl orange fluorescein cochineal coralline p-nitro-phenol alizarine green B resazurine and luteol. These were all tried by daylight, with electric glow-lamp incandescent gas and aoetylene. From the results it appears that acetylene is the best artificial light to use where it is necessary to distinguish between two different colours as with methyl orange cochineal coral-line alizarine green B and resazurine. The sort of light is of less importance where the change is of a different nature as with p-nitrophenol and luteol which change from colourless to light yellow and fluorescein which changes from light yellow to a yellowish-green fluorescence.A. M. The Hubl Iodine Solution. M. Kitt. (Chem. Zeit. 1901 xxv. 540.)-The gradual disappearance of active iodine from the ordinary Hiibl iodine solution depends on the progress of a chemical reaction which continues until equilibrium is attained. By suitable treatment of the liquid the author finds it possible to hasten that progress, and to produce a reagent that is far more permanent in strength. Thirty grammes of iodine and 25 grammes of mercuric chloride are dissolved separately in 500 C.C. of 98 per cent. (by weight) alcohol and the mixed solution is boiled on the water-bath under an inverted condenser for one hour.Before ebullition the liquid contained 0-7656 gramme of iodine per 25 C.C. ; after 049321 gramme. In four days the boiled reagent only lost 0.75 per cent. in strength; in thirty-nine days 8.12 per cent. Ordinary Hubl iodine loses 2-12 per cent. in one day and 12.75 per cent. in twenty-seven days. F. H. L. The Estimation of Hydrogen Sulphide in Coal Gas. A. Muller. (Joum. f. Gasbel. xliii. ; Chm. News lxxxiii. (2163) 217.)-The reagents required are a solution containing 25 grammes of cadmium acetate and 200 C.C. of acetic acid per litre and a second containing $0 grammes of copper sulphate and 175 C.C. of concen-trated sulphuric acid per litre. The gas is passed at a moderate rate through 25 C.C THE ANALYST.23 of the first solution contained in an Erlenmeyer flask the volume being accurately measured. The contents of the flask are afterwards heated to about 50" to 60" C., and the resulting precipitate (CuO) is filtered washed with warm water calcined and weighed. Since 1 mol. of CuO corresponds in this reaction. to 1 mol. of HzS the volume of the latter is found by multiplying the weight of CuO by 34 and dividing the product by 79-the weight of a litre of hydrogen sulphide at the temperature of the experiment. c. s. Determination of the Heating Power of Gases. W. HempeL (Zeds. f. angew. Chem. 1901 713.)-The general arrangement of the apparatus is shown in Fig. 1. It consists of a burner A which is so arranged that it can be fed with the gas under examination and at the same time with oxygen.It is thus possible to so far reduce the volume of the products of combustion that they are completely cooled in a copper tube B only 18 centimetres long and 3-3 centimetres wide. The copper tube is closed above and is fixed by means of the rubber cork d into the wide glass tube E in which 500 C.C. water can be placed. The thermometer can be read to 0.02" C. The gas is contained in the reservoir B which is made as flat a 24 THE ANALYST. possible. in an ordinary generator L. The gas is ignited by means of a quite small flame of hydrogen generated In order to render it possible to supply exactly the same quantities of hydrogen and oxygen in different determinations, the gases are led through the capillary tubes m and n and their flow is regulated so that no bubbles rise out of the cylinders p and q.The gas under examination enters by the tube a hydrogen by the tube b and oxygen by c. The tubes are all provided with porcelain tips d e to prevent conduction of heat. The gas reservoir H i s connected with the Woulffe bottle & so arranged that the pressure of the gas remains practically the same throughout. The end of the combustion is indicated by the cessation of the bubbles from t. The apparatus is calibrated by burning a reservoir full of pure hydrogen. The small ignition flame is kept burning exactly the same time in each determination ; thirty seconds should be long enough. The heating effect of this is determined separately. For roughly comparative purposes the relative heating power of gases may be ascertained by observing the length of flame they give under identical conditions. For this purpose a one-hole burner is provided with a glass chimney on which graduations are marked. The pressure and temperature of the gases must of course be in each case the same. The results are somewhat affected by the specific gravity of the gas for light gases under the same pressure pass out more rapidly than heavy ones. A. M. The oxygen is contained in the holder 0. The construction of the burner is shown in Fig. 2