BACTERIOLOGICAL, PHYSIOLOGICAL, ETC. 19 ORGANIC ANALYSIS. Methods for the Estimation of Mixtures of Four or More Carbohydrates involving Oxidation with Bromine. E. G. Wilson and W. R. G. Atkins. (Biochem. J . , 1916, 10, 504-521 .)-The five carbohydrates usually found in foliage leaves are starch, cane-sugar, dextrose, Iawulose, and maltose, with, in addition, sometimes small quantities of pentoses.For the estimation of starch, see Davis and Daish (ANALYST, 1914,39, 312) ; and for pentoses, Davis and Sawyer (ibid., 1915, 40,128) and Daish (ibid., 1914,39,555). The present work deals only‘with the four remaining sugars. Since all methods involving the use of Fehling’s solution for the quantitative estimation of reducing sugars are liable to error, owing to the solvent action of the hot sodium hydroxide on cuprous oxide, it was decided to employ Kendall’s method (ANALYST, 1912,37, 205).It was found that dextrose and mal- tose are quantitatively oxidised by bromine at room temperature when allowed to stand in a saturated solution; the liquid, however, still retains a small amount of its original reducing power towards albline copper solutions.Lzevulose is not oxidised at a11 by bromine under these conditions, or only to a, very small extent, while in the presence of -& sulphuric acid there is a slight loss of laevulose in sixty-six hours at room temperature .20 ABSTRACTS OF CHEMICAL PAPERS A mixture of cane-sugar, maltose, dextrose, and lzevulose may be a n a l y d as follows: (a) The amount of cane-sugar is determined by polarisation and reduc- tion before and after treatment with invertase.(6) The resulting mixture of re- ducing sugars is treated with bromine under standard conditions of acidity and temperature for a given time. The aldehydic sugars, dextrose (including the p r - tion derived from cane-sugar) and maltose, are oxidised, and the 'ketonic sugar kmdose, (including the portion derived from cane-sugar), remains behind.ma reducing power of the solution may then be taken as due to the lsvulose only. (c) The rotations due to the lzevulose and cane-sugar are then allowed for in the initially determined rotation. This gives the rotation due to dextrose and maltose- ( d ) The reduction due to laevulose is subtracted from tho initial reduction. This gives the reduction due to dextrose and maltose.From these data two equations may be constructed, which, when solved, give the amounts of the two remaining sugars. When only dextrose and lsvulose are present, the bromine method enables the latter to be estimated directly. The addition of dilute alkali (z or 2N sodium hydroxide) to a solution of dex- trose, laevulose, or maltose was found to produce a large temporary increase in its reducing power.From a study of the electrical conductivity of a mixture of sodium hydroxide and dextrose some indication was obtained of the formation of a com- pound between these two substances. An attempt to modify Barfoed's solution (cupric acetate and acetic acid) for quantitative work, by reducing the reaction to the first order instead of the fifth, was found unsuccessful (Bunzel, Amer.J . Phy8iol., 1908, 21, 23). H. F. E. H. Modification of the Pratt Method for Estimation of Citric Acid. J. J. Willaman. ( J . Amer. Ohem. Soc., 1916,38,2193-2199.)-Pectins are precipitated by adding twice the volume of 50 per cent. alcohol to the aqueous solution. The solution is filtered through paper on a Buchner funnel, and the precipitate washed with 65 per cent. alcohol.The filtrate is diluted until approximately of 30 per cent. alcoholic strength, and 5 C.C. of a 10 per cent. solution of barium acetate in 30 per cent. alcohol are then added. The barium citrate is filtered off on a Gooch crucible, washed once with 30 per cent. alcohol, dried, and dissolved in hot 6 per cent. phosphoric acid (60 c.c.), followed by about 40 C.C.hot water. The solution is transferred to a dia- Wing flask, the side tube of which is connected to a condenser, whilst the cork in the neck carries a tap funnel or other arrangement for running in a 0-05 per cent. solution of permanganate. Precautions must be taken against bumping. The adapter of the condenser dips into 40 C.C. of Denigh' mercuric reagent for acetone, made by heating 50 grms.of mercuric oxide with 500 C.C. water and 200 C.C. sulphuric acid, making up to 1,000 c.c., and filtering. The citric acid solution is heated to boiling, and permanganate is run in at the rate of about 5 C.C. a minute, the distillation pro- ceeding at a somewhat faster rate. When 'a deep pink colour has persisted for two minutes, the reaction is complete.The distillate is made up to 300 c.c., and if more bulky than this, 15 C.C. of Denigds' reagent are added for each additional 100 C.C. It is then boiled under a reflux condenser for forty-five minutes, after which the precipitate is filtered off and washed twice with hot water. The precipitate adheringORGANIC ANALYSIS 21 to the flask is dissolved in a small quantity of hot 5 per cent.hydrochloric acid, which is then used to dissolve the precipitate on the filter. The flask and filter are thoroughly washed with hot water, the solution nearly neutralised with 10 per cent. sodium hydroxide, made up to 100 c.c., and transferred to a burette. It is then titrated against a measured quantity of potassium iodide solution containing 28.02 grms.per litre, 1 C.C. of which equals 2 mgrms. of citric acid under the conditions of the experiment. G . C. J. Errors in the Estimation of Acid Values of Boiled Oils and Varnishes. E. E. Ware and R. E. Chrisfman. ( J . I d . and Eng. Chem., 1916,8, 996-997.)- The presence of metallic linoleates and resinates in a boiled oil or varnish will cam= the acid value t o appear too high, since the potassium hydroxide will not only neutralise the free fatty acids of the soap, but also hydrolyse the metallic soaps.In test experiments the linseed oil soaps of lead, manganese, cobalt, calcium, and zinc were prepared, dried in vacuo, dissolved in alcohol-ether, and their " acid values " estimated in the usual way, the molecular equivalent of the acid radicle being taken its 280.The values obtained differed by only a few tenths from the calculated values, except in the case of the calcium soap, the hydrolysis of which had only been partial, so that the value was 79.5 instead of 1864. In like manner the resin- ates of the same metals were partially hydrolysed in the test, the difference between the observed and calculated " acid values " indicating that from 81.3 to 94.0 per cent.were hydrolysed (calcium resinate, 49-1 per cent.). A sample of linseed oil with an acid value of 3.05 was heated with lead linoleat,e until a cloudy solution was obtained, and then contained 1.15 per cent. of ash and had an acid value of 10.20. The same oil heated alone under parallel conditions showed an acid value of 2.31 and contained 0.18 per cent.of ash, the decrease in the acid value being attributable to volatilisation of fatty acids. The difference between the amounts of ash indicated the presence of about 1 per cent. of lead oxide in solution, and this would cause an apparent acid value in the oil of 7.1, assuming that hydrolysis occurred to the same extent as in the experiments. C . A.M. Some Fatty Oils. S. Uchida ( J . SOC. Chem. Ind., 1916, 35, lQ89-1093.)- Paru rubber tree seed (Hevea braziliensis) yielded on expression 32.88 per cent. of a pale yellow oil with .weak drying properties. Shiromoji seed oil, from Lindera triloba (yield from kernels, 45.26 per cent.), is a pale yellow non-drying oil, with a, higher saponification value than any other known vegetable oil.Calophyllum Oil, from Calophyllum inophyllurn (yield from kernels, 33-83 per cent.), is a dark green, viscous, semi-drying oil, the high acid value of which is due to the presence of resin. H e r n d i a seed oil, from Hernandia peltata (yield from kernels, 33-89 per cent.), is a reddish-brown oil, which dries to a soft brittle film when heated for six hours at 100" C. It is suitable for soaps and rubber substitutes and as a, lamp oil.Hakuunboku seed oiE, from Styrax obussia (yield from kernels, 30-51 per cent.), is a semi-drying oil, containing a c6nsiderable amount of volatile fatty acids (butyrio acid). Akebi seed oil, from the Japanese shrub AEebia quinuta (yield frdm kernels, 17.69 per cent.), is a yellowish-brown non-drying oil, with an extremely high Reichert-Oil from OIL.Para rubber tree seed . . .. FATTY ACJDS. Shiromoji seed . . .. .. Calophyllum seed .. .. Hernandia seed . . .. .. Hakuunboku seed .. .. Akebi seed .. .. .. Kuromoji seed . , .. .. e Aburwhan seed . . .. .. Magnolia whole fruit . . .. ,) fruit flesh . . .. 9 ) 9 ) .. .. Tea seed .. .. .. Sp. Or. at 15' C. 0 *9239 0.9361 0 *9452 0.9380 0.9610 0 *9340 009401 0 a9348 0.9315 0 *9239 0.9288 0.9126 at 30"/30" C.: (3Oo/3O0 C.) - Refractive Index.La4720 (27.5" C.) 1.4732 (27.3" C.) 1.47925 (26.8" 0.1 1.47735 (27.2" C.] 1.48925 (27.5" C.: 1.46145 (27.5" C.: 1.4680 (27" C.) 1*4550 (27" C.) 1.4739 (26.8" C.) 1.4693 (27" C.) 1.4754 (27" C.) 1.4669 (27.6" C . ) 4.21 191.9 0.30'95*37 130.8 - 0.63 282.0 2-03 85-72 11.68 - 6-91 194.1 0.389341 95.49 37 7-39 195.7 1-77 93.17 126.1 - I 1.73 1814 16*45'92*94 115.4 - 25.45 246.4 39-76 85.80 18.78 255.6 2.53 86-22 66.29 - 2.60 2734 1.39 89.21 20.53 - 13-43 224.4 4-93 93-11 109.2 36- 35.5 13.59 205-0 4.67 91.83 89.53 36.5 78-38 3&3( 6.69 207.4 0.1719864 1246 - I 4-12 193.8 0*10(95*76 86-2 - I I---- ----I -- 27-0 185.0 1164 303.3 ti I4 14.0 287.1 12*19105-4 28-29 190.1 95.4 295.1 12-13 185-7 130.0 302-1 3 0~ c) 14-136 1664 114.7 337.8 31 191.7 77.8 292-7 P 9-5 262.0 37-0 214-1 ' td 134 277.2 18*282024 t4 28-284 201.9 99-662774 32-324 205-1 86-16 273.6 17.5-17-0 193.7 125.5 289.6 26.5 190.5 80.79 294.5 "!--.---- -3 IORGANIC ANALYSIS 23 Meissl value.It can be used for edible purposes and for the manufacture of soap and Turkey-red oil.Kuromoji seed oil, from Lindera serica (yield from kernels, 58.02 per cent.), has an aromatic odour. It is best suited for soap-making. Abura- chan seed oil, from Lindera prmox (yield from kernels, 18.17 per cent.), is a dark brown non-drying oil. Nagnolia fruit oil, from Magnolia hypoleuca (yield from seed and flesh, 31-51 per cent.), is a semi-drying oil suitable for burning and soap- making.The flesh alone yielded 35.34 per cent, of oil, and the seeds 7-74 per cent., but it would not be profita'ble to press them separately. Tea seed oil (Thea Chinemis) is a pale brown non-drying oil suitable for soap-making, lubrication, and burning. With strong sulphuric acid it gives an indigo-blue coloration, changing to greenish- brown on stirring. The following analytical values were obtained with these oils : (see p.22). C. A. M. Rapid Volumetric Estimation of Indigo. S. M. Jones and W. Spaans. ( J . Ind. a d Eng. Chern., 1916, 8, 1001-1002.)-This method, like that of Mtiller, is based on the reduction of indigo to indigo white. Instead, however, of working in presence of coal-gas, the authors use a current of hydrogen, which is found to be a great improvement.Formaldehyde sodium sulphoxylate also re- places the unstable sodium hydrosulphite as reducing agent; but as its action is slow, sodium bisulphite is added to set free sodium sulphoxylate and so accelerate the reaction. The sample (1 grm. of powder, proportionately more of paste) is mixed to an even paste with 15 C.C. sulphuric acid, a further 15 C.C.of acid is added, and the mixture heated to 55" to 60" C. for threehours, when sulphonation should be com- plete. The cooled mixture is diluted to 300 c.c., again cooled, diluted to exactly 1,000 c.c., and filtered. One grrn. of pure indigo is similarly treated to produce a standard for comparison. Fifty C.C. of the standard solution and 50 C.C. of 35 per cent. sodium bisulphite solution are mixed in a flask of 300 C.C.capacity, and the flask is closed with a rubber stopper with four holes. Through one of these a thermometer dips into the liquid, two provide for the admission and exit of hydrogen, whilst the fourth accommodates the tip of a burette. The air is displaced by hydrogen, and the mixture heated t o 75" C. Standard sodium formaldehyde sulphoxylate is then r m in from the burette until the blue colour disappears.This solution is made by dissolving 1 grm. of the solid reagent in 1,000 C.C. of water. The sample to be analysed is treated in a, similar manner. G. C. J. Estimation of Free and Combined Galactose. A. W. Van der Haar. (Chem. Weekblad, 1916, 13, 1204-L213.)-This method, which is a modification of that of Creydt (Dissert., Erlangen, 1888), is based on the oxidation of galactose to mucic acid by means of nitric acid.In the case of free d-galactose, from 0-25 to 1 grm. is heated with occasional agitation with 60 C.C. of nitric acid (sp. gr. 1.15 a t 15" C.) in a beaker 12 cm. in height and 60 mm. in diameter, which is placed in a boiling- water bath. When the contents of the beaker have been reduced to less than24 ABSTRACTS OF CHEMICAL PAPERS 20 grms., the mixture is cooled, made up to 20 grms.with water, and after the addi- tion of 0-5 grm. of pure dry mucic acid (purified by recrystallisation from alcohol), the beaker is allowed to stand for forty-eight hours a t a temperature of 15" C. The deposit of muck acid is then collected on an asbestos filter in a Gooch crucible,.washed with 5 C.C. of water, and dried in the water-oven until constant in weight, and the 0.5 grm. of mucic acid which w&s added to promote the crystallisation is deducted from the result. In the case of combined galactose 0-25 to 1 grm. of the anhydrous glucoside or polysaccharide is hydrolysed with 25 C.C. of 2 to 5 per cent. sulphuric acid. When insoluble products are formed they are filtered off, after twenty-f our hours, and the filtrate and washings concentrated.The resulting solution of galactose is rendered slightly alkaline with sodium hydroxide solution,. made up to 30 c.c., and after the addition of 30 C.C. of 50 per cent. nitric acid, and as much cane-sugar as corresponds to the non-sugar portion of the glucoside (e.g., sapongenins), the mixture is dxidised as described above.The following tables give- the quantities of galactose corresponding to the weights of mucic acid obtained: I. GALACTOSE ALONE. Muck Acid. Mgrms. -4 + 0-8 5.6 10.4 15.2 20 27 34 41 45 55 64 73 82 91 100 108.4 116.8 125.2 133-6 149.8 157.2 16443 172-4 180 142 Galac- tose. Mgrms. 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 Mucic Acid.Rlgr m s. 187 194 201 208 215 223.1 231-2 239.3 247 *4 255-5 263-6 27 1-7 279.8 287.9 296 303 3 10 317 324 331 338 345 352 359 366 374.9 Galactose. Mgrms . 260 270 280 290 300 310 320 330 340 350 360 370 380 390 400 410 420 430 440 450 460 470 480 490 500 510 Mucic Acid. Mgrms. 383.8 392.7 401.6 410.5 419.4 428.3 437.2 446.1 455 462 469 476 483 490 497 504 51 1 518 525 534 543 552 561 570 579 588 Galactose.Mgrms . 520 530 540 550 560 570 580 590 600 610 620 630 640 650 660 670 680 690 700 710 720 730 740 750 760 770 Mucic Acid. Mgrms . 597 606 615 623 631 639 647 655 663 671 679 688 695 703.5 712 720-5 729 73795 746 754.5 763 771.5 780 Galactose. Mgrms . 780 790 800 810 820 830 840 850 860 870 880 890 900 910 9201 930 940 930 960 970 980 990 1 ,(@oMuck Acid.Mgrms. - 4 + 2.4 8.8 15.2 21.6 28 34.9 41 43 48.7 55.6 62.5 70 7745 85 92.5 100 106.6 113.2 119.8 126.4 133 139.4 14543 152.2 158.6 165 I s;I%ms* i 181 i 189 ORGANIC ANALYSI8 25 II.-GALACTOSE MADE UP TO 1 GRM. WITH SACCHAROSE. Mgrms. 260 270 280 Mgrm s. 0 10 20 30 I 197 290 40 50 60 70 80 100 110 205 212 219 226 233 240 248-8 257.6 2664 275.2 292.2 3004 3084 316.8 284 325 332 339 346 I j ; j I 1 j i 300 310 320 330 340 350 360 370 380 390 400 410 420 430 440 450 460 470 480 140 150 160 170 180 353 490 360 500 368 510 1 1 , I Mgrms.376 384 392 400 408 416 424 432 440 447 454 461 468 475 483 491 499 507 515 523 531 539 547 555 564 573 Mgrms. 520 530 540 550 560 570 580 590 600 610 620 630 640 650 660 670 680 690 700 710 720 730 740 750 760 770 Mgrms.582 591 600 609 618 627 636 645 654 663 672 681 690 699 708 717 726 735 744 753 762 17 1 780 Galactose . Mgrms. 780 790 800 810 820 830 840 850 860 870 880 890 900 910 920 930 940 950 960 970 980 990 1,000 C. A. M. Extraction of Lactic Acid for its Estimation. E. Ohlsson. (Skand. Arch. Phydiol., 1916, 33, 231-234; through J.Chern. Soc., 1916, 1.10, ii., 542-543.)- In the estimtion of lactic acid according to Von Furth-Charnass, as modified by Embden, the extrwtion with ether is a troublesome and lengthy process. For the isolation in a pure state the author finds ethyl acetate a better solvent, and when the acid is merely to be estimated, amyl alcohol is much better. The liquid con- taining lactic acid is saturated with ammonium sulphate, filtered after twelve hours, and mixed with & volume of 50 per cent.sulphuric acid. It is then shaken with 2 volumes of amyl alcohol, which are then freed from the acid by shaking with sodium carbonate, and used similarly for four more successive extractions of the fluid, Traces of amyl alcohol are removed from the sodium carbonate extracts by three extractions with benzene, and the estimation is then carried out in the usual way.26 ABSTRACTS OF CHEMICAL PAPERS Differentiation of the Two Naphthols by Means of Titanic Acid dissolved in Sulphuric Acid. G. Denigis. (Ann.. Chim. anal., 1916, 21, 216-217.)-A solution of titanic acid in sulphuric acid (cf. ANALYST, 1916, 41, 343) gives a bright green coloration when mixed with a small quantity of a-naphthol, whilst a blood- red coloration is obtained with ,&naphthol. If the mixtures are diluted with wetic acid, the green coloration given by a-naphthol changes to red-violet, but in the ome of @naphthol the blood-red colour remains unchanged. The reaction may be applied as 8 ring test, the naphthol being dissolved previously in acetic acid. The esters of the two naphthols yield similar distinctive reactions with the reagent. W. P. S.