(j). Pipette 20 ml into a large test tube, add 2 ml of the standard basic Pb acetate solution, cork, and allow to stand for 2 hours.

Filter with suction on a 25-ml tared Gooch having an asbestos mat at least 3 mm thick. When nearly all the liquid has run through, fill the crucible with cold water. Repeat to a total of four washings, taking care to prevent formation of fissures in the precipitate_by keeping it covered with water and avoiding too great suction. Dry at 100° C, weigh, and multiply the weight by 20.

(2) WINTON LEAD NUMBER [10].

Reagent-Standard basic lead acetate solution. To a measured volume of the reagent prepared for determination of the Canadian lead number (1) add 4 volumes of water, and filter. A blank should be run with each set of determinations.

Determination of lead in the blank.-Transfer 25 ml of the standard basic Pb acetate to a 100-ml flask, add a few drops of glacial acetic acid, and make up to the mark with water. Shake, and determine PbSO, in 10 ml of the solution, as directed below. The use of acetic acid is imperative in order to retain all Pb in solution when the reagent is diluted with water.

Determination.-Transfer 25 g of the sample to a 100-ml flask by means of water. Add 25 ml of the standard basic Pb acetate solution and shake. Fill to the mark, shake, and allow to stand for at least 3 hours before filtering. Pipette 10 ml of the clear filtrate into a 250-ml beaker, add 40 ml of water and 1 ml of H2SO4, shake, and add 100 ml of 95-percent alcohol, dry in a water oven, and ignite in a muffle or over a Bunsen burner, applying the heat gradually at first and avoiding a reducing flame. Cool and weigh. Subtract the weight of PbSO, so found from the weight of PbSO, found in the blank, and multiply by the factor 27.33. The use of this factor gives the Pb number directly, without the various calculations otherwise required.

(1) APPARATUS.

(j) CONDUCTIVITY VALUE [11]

Conductivity cell. Should be made of resistance glass with platinized Pt electrodes firmly fixed and adequately protected from displacement. These electrodes may be sealed in a vessel into which the solution under examination may be run and subsequently drawn off (Zerban type), or attached to a support so that they can be lowered into a cylinder (or a 100-ml beaker) containing the solution (dipping type). The cell must be provided with a thermometer graduated in tenths of a degree over the range 20° to 30° C, and the bulb must be placed in the immediate vicinity of the electrodes. The cell constant should be approximately 0.15.

Galvanometer or a microphone hummer (or an induction coil) and a sensitive telephone receiver.

Suitable source of current.-Dry or storage cells if a hummer or induction coil is used; 110-volt alternating current if a galvanometer is used.

Resistances of 10 and 100 ohms. Should be fixed and accurate. Slide wire or Wheatstone bridge.

Device for control of the temperature of the cell to within ±0.1° C.This may consist of a thermostat or a vessel into which water of suitable temperature may be run to adjust the cell contents to 25° C.

(2) DETERMINATION OF THE CELL CONSTANT.-Prepare solutions of 0.3728 and 0.7456 g of dry KCl in water, which offers a resistance of at least 25,000 ohms in the cell, and make them to the mark at 20° to 25° C in 500-ml volumetric flasks. Fill the cell with the more dilute (0.01 M) solution, adjust to 25° ±0.1° C, measure the electrical resistance, and multiply the number of ohms by 141.2. Rinse with the stronger (0.02 M) solution, fill the cell with the solution, measure its resistance at 25° C, and multiply by 276.1. Average the two results.

(3) DETERMINATION.-Weigh out a quantity of sirup that contains 25 g of dry matter, transfer to a 100-ml volumetric flask with warm water of the same quality as that used in the determination of the cell constant, cool to 25° C, make to mark, and measure the resistance in the cell at 25° ±0.1° C. Divide the cell constant by the number of ohms found.

(k) MALIC-ACID VALUE (COWLES) [12]--TENTATIVE

Weigh 6.7 g of the sample into a 200-ml beaker; add 5 ml of water, then 2 ml of a 10-percent calcium acetate solution; and stir. Add, gradually and with constant stirring, 100 ml of 95-percent alcohol and agitate the solution until the precipitate settles, or let stand until the supernatant liquid is clear. Filter off the precipitate and wash with 75 ml of alcohol, 85 percent by volume. Dry the filter paper and ignite in a platinum dish. Add 10 ml of 0.1 N HCl, and warm gently until all of the lime dissolves. Cool, and titrate back with 0.1 N NaOH solution, using methyl orange indicator. The difference in milliliters divided by 10 represents the malic-acid value of the sample. Previous to use, the reagents should be tested by a blank determination and any necessary corrections applied.

3. DETERMINATION OF SUGARS IN MILK PRODUCTS

The estimation of lactose in fresh milk is, from the standpoint of sugar analysis, relatively simple, but the preparation of the sample for analysis involves procedures regarding which there is much dispute. The Association of Official Agricultural Chemists finds it satisfactory to clarify with copper sulfate in preparation for reducingsugar analysis or with mercuric nitrate for polariscopic analysis. On the other hand, the British Subcommittee on Milk Products recommends the use of zinc ferrocyanide, prepared in the presence of the sample, but permits the use of phosphotungstic acid. The volume of the precipitate is relatively large and must be calculated or determined.

An important commodity is condensed milk sweetened with sucrose. Occasionally the sucrose is partially inverted by the action of enzymes or microorganisms, and in some instances further action of microorganisms converts the levulose so formed into a nonreducing levan. In some cases, other sugars may be added as sweetening agents. The analysis of such sugar mixtures becomes an intricate problem.

Many of the methods for the sugar mixtures in milk products have been elaborated solely for the purposes of milk analysis, but appear to be capable of general application to other products.

(a) DETERMINATION OF THE VOLUME OF THE PRECIPITATE

If the percentages of fat and protein of a milk product are known, the volume of the clarification precipitate can be calculated with fair approximation by the method of the British Subcommittee on Milk Products [13]. The volume of the protein precipitate varies with the nature of the precipitant. For phosphotungstic acid clarification the specific volume of fat is 1.08 and for protein, 0.74. The volume then is fatX1.08+protein X0.74, to which is added a further empirical correction of 1.5 ml.

For the zinc ferrocyanide method the volume of precipitate is about double that for the phosphotungstic acid precipitate. The volume of precipitate is then fatX1.08+protein X1.55. The British subcommittee, however, recommends actual determination of the volume of the precipitate for each sample [13].

(1) Weigh 100 g of milk into a 200-ml flask. Add precipitating reagents, make up to the mark at 20° C, filter, and polarize at 20° C. Reading A.

(2) Weigh 100 g of the same milk and 16.8 g of sucrose into a 200-ml flask, dissolve the sugar, add precipitating reagents, make up to the mark at 20° C, filter, and polarize at 20° C. Reading=B.

(3) Weigh 140 g of the same milk into a 200-ml flask and add seven-fifths of the previous quantities of precipitating reagents. Make up to the mark at 20° C, and filter. Take 70 ml of the filtrate, make up to 100 ml, and polarize at 20° C. Reading C.

(4) Measure 70 ml of the filtrate from (3) into a 100-ml flask, add 8.4 g of sucrose, dissolve, make up to the mark at 20° C, and polariez at 20° C. Reading D.

The correction for volume of precipitate for 100 g of milk

-2001-milliliters.

(1) OPTICAL METHOD.

(b) LACTOSE IN MILK [14]

Reagents. (a) Acid mercuric nitrate solution.-Dissolve mercury in twice its weight of concentrated nitric acid and dilute with an equal volume of water.

(b) Mercuric iodide solution.-Dissolve 33.2 g of KI and 13.5 g of HgCl, in 200 ml of glacial acetic acid and 640 ml of water.

Determination.-Determine the specific gravity (20°/20°) of the milk and place in a flask graduated at 102.6 ml, the volume of milk indicated in table 31. Add 1 ml of solution (a) or 30 ml of solution (b), fill to the mark, shake frequently for at least 15 minutes, filter through a dry filter, and polarize. The volumes in the table are those of a double-normal weight of lactose (32.9 g per 100 ml); hence, if a 200-ml tube is used, divide the saccharimeter reading by 2 to obtain the percentage of lactose in the sample.

(2) CHEMICAL METHOD.-Dilute 25 g of the sample with 400 ml of water in a 500-ml volumetric flask and add 10 ml of CuSO, solution (Soxhlet solution 1) and about 7.5 ml of a KOH solution of such strength that 1 volume is just sufficient to precipitate completely the copper as hydroxide from 1 volume of the copper sulfate solution. (Instead, 8.8 ml of 0.5 N NaOH solution may be used). After the addition of the alkali solution, the mixture must still have an acid reaction and con

TABLE 31.-Volume of milk corresponding to lactose double-normal weight

tain copper in solution. Fill the flask to the 500-ml mark, mix, filter through a dry filter, and determine lactose in an aliquot of the filtrate by the Lane and Eynon or Munson and Walker method.

4. REFERENCES

[1] U. S. Dept. Agri., Bur. Chem. Bul. 110 (1908) and 154 (1912); Z. Nahr. Genussm. 18, 625 (1909).

[2] U. S. Dept. Agr., Bur. Chem. Bul. 110, 60 (1908).

[3] U. S. Dept. Agri., Bur. Chem. Bul. 110 (1908) and 154 (1912).

[4] U. S. Dept. Agri., Bur. Chem. Bul. 154, 15 (1912); J. Ind. Eng. Chem. 17, 612 (1925); J. Assn. Official Agr. Chem. 15, 78 (1932).

[5] Z. Nahr. Genussm. 22, 412 (1911).

[6] Z. Nahr. Genussm. 19, 72 (1910).

[7] J. Hortvet, J. Am. Chem. Soc. 26, 1523 (1904); J. Assn. Official Agr. Chem. 15, 79 (1932); 16, 79 (1933); 17, 73 (1934); 18, 83 (1935).

[8] A. E. Leach and A. L. Winton, Food Inspection and Analysis, 4th ed., p. 652 (John Wiley & Sons, Inc., N. Y., 1920).

[9] J. F. Snell, J. Assn. Official Agr. Chem. 4, 437 (1921); 16, 80 (1933); 17, 74 (1934).

[10] A. L. Winton and J. L. Kreider, J. Am. Chem. Soc. 28, 1204 (1906); J. Official Agr. Chem. 16, 80 (1933); 17, 74 (1934).

[11] C. A. Browne, J. Am. Chem. Soc. 28, 435 (1906); J. Assn. Official Agr. Chem. 16, 80 (1933); 17, 74 (1934).

[12] H. W. Cowles, Jr., J. Am. Chem. Soc. 30, 1285 (1908).

[13] Analyst 55, 116 (1930).

[14] Methods of Analysis of the Association of Official Agr. Chem., 5th ed. (1940).

XII. DETERMINATION OF PENTOSANS

1. DESCRIPTIVE

Pentosans are polysaccharides that yield pentoses upon hydrolysis. As their names imply, araban is thus related to arabinose, and xylan to xylose. The determination of pentosans depends upon their conversion into furfural and the subsequent determination of the furfural. The following reactions are assumed to take place:

In carrying out pentosan determinations, one is therefore concerned with the quantitative conversion of the pentosan into furfural and the subsequent determination of the furfural.

The method universally adopted for converting pentosans into furfural consists in boiling the substance with 12-percent hydro

chloric acid and collecting the distillate at the rate of 30 ml per 10 minutes. The quantity of distillate collected in some procedures is arbitrarily defined, while in others the distillation is continued until furfural ceases to come over, as indicated by the aniline acetate test." Methylfurfural gives only a faint-yellow color with aniline acetate. A solution of aniline in alcohol is used to test for this substance. Furfural and hydroxyfurfural derivatives are very sensitive to the aniline alcohol reagent, giving a bright-red color.

Jolles [1] employed steam distillation in order to remove the furfural from the acid solution more rapidly and reported quantitative conversion of arabinose and xylose into furfural. Pervier and Gortner [2] found that steam distillation from a 12-percent hydrochloric acid solution as used by Jolles, or from an 18- to 20-percent hydrochloric acid solution (the acid percentage being indicated by the boiling point), resulted in quantitative yields of furfural from arabinose and xylose. They also concluded that the rate of distillation did not affect the yield, and thus a slow current of steam was sufficient to sweep out the furfural as fast as it was formed from the pentose. When the usual method of distillation from 12-percent hydrochloric acid was used, these authors obtained a 95.3-percent yield of furfural from arabinose and 96.1-percent yield from xylose. These figures are approximately 7 percent higher than those generally obtained.

Kline and Acree [3] concluded that steam distillation gave no better yields of furfural from xylose than did the usual method of distillation. Kullgren and Tydén [4] converted pentosans to furfural by distilling from a 13.15-percent hydrochloric acid solution saturated with sodium chloride. They thus had a constant acid concentration in the distilling flask.

Hughes and Acree [5] reported that a rapid steam distillation from a 12-percent hydrochloric acid solution saturated with sodium chloride gave quantitative yields of furfural from xylose as compared with 96-percent yields when a slow current of steam was used, irrespective of whether or not the solution was saturated with sodium chloride.

The gravimetric methods for the determination of furfural include the use of phloroglucin [6], barbituric acid [7], thiobarbituric acid [8], and diphenyl thiobarbituric acid [9]. The phloroglucin method has been carefully studied and generally adopted.

The volumetric methods include the use of phenylhydrazine [10], potassium bisulfite [1], and potassium bromate-bromide solution. The latter was studied by Pervier and Gortner [2]. They used an electrometric titration method and found that under their experimental conditions the potassium bromate and furfural reacted in the molecular ratio 1:3. Powell and Whittaker [11] determined the amount of bromine absorbed by a back-titration method, using potassium iodide and sodium thiosulfate. Under their experimental conditions they found that 4 atoms of bromine reacted with 1 molecule of furfural. În 1933, Magistad [12] pointed out that the furfuralbromate reaction is influenced greatly by temperature. Hughes and Acree [13] carried out the reaction at 0° C. By carefully controlling the temperature and time of reaction, they obtained reproduc

The aniline acetate reagent is prepared by shaking equal volumes of aniline and water in a test tube and adding glacial acetic acid until the solution is clear. Place a drop of the reagent on a filter paper and allow a drop of the distillate to spread into the reagent. If furfural is present, a red color will appear where the circles intersect. Toward the end of the distillation the red line will appear only after drying.