UNIVERSITY OF CALIFORNIA 
 
 COLLEGE OF AGRICULTURE 
 
 AGRICULTURAL EXPERIMENT STATION 
 
 BERKELEY, CALIFORNIA 
 
 CIRCULAR 230 
 Revised June, 1933 
 
 TESTING MILK, CREAM, AND SKIM MILK 
 FOR BUTTERFAT 
 
 Revision by D. H. NELSONi 
 
 This circular describes an approved and practical method of using 
 the Babcock test that is suitable for the dairyman, the manufacturer 
 of dairy products, and the official test supervisor. The directions given 
 below are expressed in such a manner that the tests can be made by 
 persons not familiar with chemical analysis or laboratory practices. 
 However, persons not familiar with such procedure would find it 
 helpful to observe the work being done at a local creamery or other place 
 where tests are being made before attempting to operate the test 
 independently. 
 
 TAKING THE SAMPLE 
 
 A proper sample is of first importance in any analysis. If the sample 
 is improperly taken, the results will be of no value, regardless of the 
 operator's knowledge or skill. If the milk is mixed by pouring from one 
 container to another a properly representative sample can be obtained. 
 This method is used in connection with cow-testing work. The milk 
 from a complete milking is poured into another pail and sampled at 
 once. It is essential that the sample be taken immediately after mixing 
 because the fat soon begins to rise again. In large vats the mixing may 
 be successfully accomplished by the aid of mechanical agitators. 
 
 When samples of milk are taken directly from the can, a milk stirrer 
 (fig. 1) may be used to thoroughly mix the top and bottom milk immedi- 
 ately before the sample is secured. 
 
 Composite Sampling. — A composite sample is composed of propor- 
 tional quantities of different lots of milk. In the official test work, com- 
 posites are taken with a graduated pipette and a certain quantity of 
 
 1 Associate in Dairy Industry. 
 
2 University of California — Experiment Station 
 
 milk is taken for each pound produced by the animal at each milking. 
 The common practice is to take 3 cubic centimeters per pound. This 
 enables the superintendent of dairy tests to check the work of the test 
 supervisor by comparing the test of the composite sample with the 
 computed average test. Composite samples should be kept in a clean, 
 well-sealed container to prevent the evaporation of moisture. Approved 
 types of containers, bottles with a capacity of 4 to 16 ounces, are shown 
 in figure 1. Those most commonly used in milk plants have a capacity 
 of 6 to 8 ounces. 
 
 The composite method of sampling is practiced by most large whole- 
 milk plants, in order to reduce the labor and expense of testing. The 
 type of samplers commonly used in sampling large quantities of milk 
 are also shown in figure 1. 
 
 Churned Samples. — Samples containing butter particles ("churned 
 samples") are unsatisfactory for testing. However, if a test must be 
 made, warm the sample to 110° Fahrenheit and hold it there until all 
 the lumps of fat are melted. Then mix and shake violently, taking the 
 sample quickly before the melted fat has a chance to rise. 
 
 Curdled Samples. — Curdled samples should never be tested in official 
 work ; but when necessary to test such samples add sufficient powdered 
 caustic soda or potash to dissolve the coagulated substance. The test is 
 then made in the usual manner. The volume of the sample is not changed 
 and therefore the direct measurement of the fat column gives the cor- 
 rect reading. 
 
 THE BABCOCK TEST 
 
 The Babcock test is based upon the fact that strong sulfuric acid 
 will dissolve the milk solids other than fat, and produce heat which 
 causes the fat globules to come together more easily. By centrif uging in 
 specially constructed bottles, the milk fat is completely separated from 
 the serum which has a higher specific gravity. 
 
 Test Bottles. — The bottles used for testing milk by the Babcock 
 method are marked with divisions on the neck which read from zero 
 to 8 per cent when 18 grams (delivered by the 17.6 cc pipette 2 ) is used 
 for the test. Each of the smaller divisions on the neck of the bottle 
 represents one-tenth per cent (0.1%). 
 
 Milk test bottles marked with divisions on the neck reading from 
 zero to 10 per cent are sometimes used. Each of the smaller divisions on 
 
 2 Sometimes the letters "ml" are used instead of the letters "cc". The letters 
 "ml" stand for "milliliter" which is practically equal to "cubic centimeter" in 
 volume. 
 
Cm. 230] Testing Milk and Cream for Butterfat 
 
 (J4 
 
 5 
 
 Fig. 1. — Sampling equipment: 1, type of rod commonly used for stirring milk; 
 2, dipper type of sampler now in general use for taking milk samples ; 3, the McKay 
 sampler occasionally used for milk sampling but designed for cream sampling; 
 4, thief sampler which is efficiently used for milk sampling ; 5, two types of bottles 
 satisfactory for preserving composite samples. 
 
4 University of California — Experiment Station 
 
 the neck of these bottles represents two-tenths per cent (0.2%). The 
 use of the 10 per cent bottle is not legal in California, and consequently, 
 the 8 per cent bottle must be used for official testing. 
 
 Milk Pipettes. — The glass milk pipette shown in figure 2 is used for 
 measuring the quantity of milk required for the test. A mark surround- 
 ing the neck of the pipette indicates the measure of milk to be used. 
 When the pipette is filled with milk to this mark it contains 17.6 cubic 
 
 <^> 
 
 a 
 
 i 
 
 33 
 
 a 
 
 <s 
 
 C~2 
 
 =£1 
 
 <fcl 
 
 cM 
 
 =fcf 
 
 3 
 
 £rz 
 
 ld-f 
 
 sfri 
 
 <U^ 
 
 • 75 
 c c 
 
 4 
 
 S 
 
 s 
 
 Fig. 2. — Glassware for testing: 1, milk test bottle, 8 per cent; 2, milk pipette, 
 17.6 cubic centimeters capacity; 3, acid measure, 17.5 cubic centimeters capacity; 
 4, acidometer. 
 
 centimeters, and 17.5 cubic centimeters or 18 grams of milk is delivered 
 into the test bottle, 0.1 cubic centimeter adhering to the inner surface 
 of the glass. 
 
 The Tester. — The centrifugal machines used for the Babcock test 
 vary in size from 2-bottle to 48-bottle machines. These machines may 
 be operated either by hand, by steam, or by electricity (fig. 3). In all 
 cases, it is essential to have the centrifuge securely fastened to its base. 
 
 The steam machines are preferred to the other types owing to the 
 uniformity of speed and also to the maintenance of an even temperature 
 which they make possible. 
 
Cm. 230] Testing Milk and Cream for Butterfat 5 
 
 Acid, — Commercial sulfuric acid with a specific gravity of 1.82-1.83 
 is used for the test. It is essential to have acid of this strength in order 
 to obtain clear butterfat readings. The acid should always be kept in 
 a tightly closed container as it weakens when exposed to the air, owing 
 to its property of absorbing moisture from the atmosphere. Sulfuric 
 acid is extremely dangerous owing to its corrosive action. When it is 
 spilled, it is essential to remove it immediately with water and then 
 check the corrosive action with an alkali, such as ammonia or washing 
 powder. Lead sheeting is used to protect testing tables as it is not 
 attacked by this acid. 
 
 Fig. 3. — Left, steam turbine tester. Center, electric centrifuge. Eight, hand 
 
 centrifuge. 
 
 The acidometer illustrated in figure 2 is the instrument commonly 
 used for determining the strength of commercial sulfuric acid. It is 
 standardized to read directly the specific gravity at 60 °F. 
 
 Convenient apparatus for use in disposing of waste acid is shown in 
 figure 4. 
 
 DIRECTIONS FOR MEASURING MILK INTO THE TEST BOTTLE 
 
 Pour the milk sample from one container into another several times 
 to insure thorough mixing. This should be done by pouring down the 
 side of the container and not into the center of it to avoid the introduc- 
 tion of air into the milk. Then immediately suck the milk into the 
 pipette some distance above the mark, holding it there by placing a 
 forefinger over the mouth of the pipette. By a slight release of the finger, 
 the milk is lowered directly to the mark. The stem of the pipette is then 
 inserted into the neck of the test bottle and the milk allowed to run in. 
 By blowing into the pipette, the last drop of milk is forced into the 
 bottle. In cases in which the results of testing are of great importance, 
 as in the testing of cows for advanced registry, all tests should be made 
 in duplicate. The milk bottles should be marked so that the samples 
 can be identified. 
 
6 University of California — Experiment Station 
 
 Adding Acid. — The milk and acid should be at a temperature of from 
 55° to 70° F, at the time the acid is added to the sample. In adding the 
 17.5 cubic centimeters of acid to the milk, it is essential to slope the test 
 bottle so the acid will pass down the side of the neck of the bottle, and 
 at the same time rotate the bottle so the acid will wash down all the 
 milk that adheres to the inside of the neck. Adding acid in the central 
 region of the bottle would cause a charring of the sample and might 
 cause the spilling of acid. The acid and milk should be mixed by a 
 rotary motion of the bottle. The test bottle should be held at such an 
 angle that if the reaction is violent enough to cause spattering, the 
 operator's face will not be in danger. 
 
 Fig. 4. — Convenient jar for handling waste acid. 
 
 Centrifuging the Samples— The samples should be placed in the 
 tester and centrifuged for a period of 5 minutes. If all the pockets in 
 the tester are not filled with test bottles, the bottles should be evenly 
 distributed around the tester so that the centrifuge will be evenly 
 balanced. The speed of the centrifuge is determined by the diameter 
 of the wheel, which is measured from the bottom of the cups when they 
 are in a horizontal position (see table 1). 
 
 Fill the bottles to the base of the neck with water after the 5-minute 
 period. To insure clear tests, the water used should be either soft or 
 distilled, and when adding it to the tests it should be at a temperature 
 of from 140° to 180° F. Centrifuge the bottles for 2 minutes after this 
 
Cm. 230] 
 
 Testing Milk and Cream for Butterfat 
 
 addition of water. Then stop the tester and add water to bring the fat 
 column within the graduated portion of the neck of the bottle. Centri- 
 fuge 1 minute longer, and remove the bottles from the tester for reading. 
 Reading the Tests. — The samples should be placed in a water bath 
 (fig. 5) at a temperature of 130° to 140° F for a period of 10 minutes 
 or more. Care should be taken that the fat in the neck of the bottle is 
 completely below the level of the water in the bath. 
 
 Fig. 5. — Type of water bath used for milk and cream samples. 
 
 In reading, the fat column is conveniently measured with a pair of 
 dividers. The measurement should include the lower and the upper 
 curve of the fat column (see fig. 7). Then transfer one point of the 
 divider to the zero mark on the neck of the bottle. The other divider 
 point will indicate directly the per cent of fat in the sample. 
 
 TABLE 1 
 
 Showing Speed of Centrifuge for Testers of Different Diameters 
 
 Diameter Revolutions 
 
 of wheel of wheel 
 
 in inches per minute 
 
 10 1000-1100 
 
 12 900-1000 
 
 14 875- 925 
 
 16 825- 875 
 
 Diameter 
 of wheel 
 in inches 
 
 18 
 
 20 
 
 22 
 
 24 
 
 Revolutions 
 
 of wheel 
 per minute 
 
 775- 825 
 725- 775 
 700- 750 
 675- 725 
 
 Abnormal Tests. — Charred samples may result from using too much 
 acid, using warm acid or milk, adding acid directly to the milk instead 
 of permitting it to run down the side of the neck by sloping the bottle, 
 
8 University of California — Experiment Station 
 
 allowing samples to stand too long before centrif uging, using acid of a 
 strength greater than that specified, or allowing milk and acid to stand 
 before mixing. The formation of curd in the tests is due either to using 
 weak acid, too little acid, insufficient mixing, or to using milk or acid 
 that is too cold. Using hard water causes the formation of a white foam, 
 or bubbles. 
 
 TESTING CREAM FOR BUTTERFAT 
 
 The range in specific gravity of cream of various fat contents is very 
 great. Hence, an 18-gram sample of cream cannot be obtained by means 
 of a 17.6 cubic centimeter measurement. Furtheremore, as cream of 
 high test has a thick consistency, much of it would adhere to the walls 
 of the pipette in case it were measured into the test bottle. Since separa- 
 tor cream also contains more or less air it is essential to weigh it in 
 order to obtain a proper sample. 
 
 Cream Test Bottles. — In California practice there are two bottles as 
 shown in figure 6, used for testing cream by the Babcock method. They 
 are 9 inches high and marked with divisions on the neck of the bottles 
 to read from zero to 50 per cent, the smaller divisions representing five- 
 tenths per cent (0.5%) each. One bottle is designed to read directly the 
 percentage of butterfat when 18 grams of cream is used; the other 
 when 9 grams of cream is used. 
 
 Cream Test Balances. — The balances used in weighing cream are 
 specially designed and range in capacity from 1 to 12 bottles. The 
 4-bottle torsion balance illustrated in figure 6 is very satisfactory. 
 Cream balances should be carefully checked for accuracy. Careful 
 handling of the cream balance is imperative for good results. It should 
 be set on a solid, level foundation and protected from the slightest 
 draft. When using it, the operator should stand squarely in front. 
 The beam should be kept locked at all times except the brief moment 
 necessary to observe whether it is in balance. When properly balanced, 
 the pointer will swing one or two divisions equally on each side of the 
 center, that is, the pointer should never stand still. 
 
 Preparation of the Cream Sample for Testing. — A sample of cream 
 can be prepared by pouring it from one container to another in a manner 
 similar to that described for milk. Most cream samples, however, are 
 clotted when they are presented to the operator for analysis. These 
 samples are placed in a bath of water at 90° to 110° F. This treatment 
 restores the liquid condition of the cream and makes it possible to 
 follow the procedure described for preparing the milk samples for 
 testing. 
 
Cir. 230] 
 
 Testing Milk and Cream for Butterfat 
 
 Weighing the Cream for Analysis. — The bottles are placed in the 
 sockets of the balance pan and balanced. By means of a pipette, a charge 
 of cream is passed into the cream test bottle. Considerable skill is re- 
 quired by the operator in order to check the flow of cream when the 
 pointer of the balance indicates that the correct charge has been deliv- 
 ered. If too much cream is added, a small portion can be poured from 
 the test bottle and the bottle again placed upon the scales and cream 
 
 Fig. 6. — Cream balance and test bottles: a, cream test balance; h, 18-gram, 50 per 
 cent, 9-inch cream test bottle; c, 18-gram weight; d, 9-gram, 50 per cent, 9-inch 
 cream test bottle ; e, 9-gram weight. 
 
 added from the pipette until it is accurately balanced. Care must be 
 used to get all the cream into the bottle and not on the outside or on 
 the balance pan. 
 
 Adding the Acid. — Cream contains less solids-not-fat than milk and 
 for this reason less acid is required in making the cream test. The 
 amount of acid necessary to be used will decrease as the richness of the 
 cream increases. The development of a coffee-brown color after mixing 
 the acid and cream is an indication that sufficient acid has been added. 
 The development of a darker color is a sign of too much acid. Under 
 normal conditions, with the cream and acid at a temperature of 70 °F, 
 
10 
 
 University of California — Experiment Station 
 
 approximately 4 to 8 cubic centimeters of acid are required for a 9-gram 
 sample, and 8 to 12 cubic centimeters would be required for an 18-gram 
 sample of average cream. 
 
 While adding the acid, it is necessary to revolve the cream bottle so 
 as to allow the acid to wash the cream into the bulb of the bottle. In 
 mixing, the bottle should be rotated evenly so that none of the mixture 
 is expelled from the bottle. 
 
 j*» aw^tj 
 
 / 
 
 VMS' 
 
 X) 
 Z5 
 20 
 15 
 
 
 
 5 
 
 \ 
 
 26% 
 
 THKKBIJ 
 1B15 
 26% 
 
 THI5MIUC 
 TK15 
 4% 
 
 Fig. 7. — Measuring milk and cream fat columns: arrows A and B indicate the 
 points for the correct measurement. The California Dairy Law requires the use of 
 glymol on all cream tests; consequently, the columns should be measured as 
 indicated on the center bottle. When glymol is not used, the reading should be 
 made as shown on the bottle on the left. It is incorrect to measure the fat column 
 from arrows A and C. 
 
 Centrifuging the Samples. — The samples should be placed in the 
 tester and centrifuged in the same manner as described for milk 
 samples. 
 
 Reading the Tests. — Upon removal from the tester, the bottles should 
 be placed in a bath of water at 130° to 140°F, for a period of at least 
 10 minutes. Care should be taken that the fat in the neck of the bottle 
 is completely below the level of the water in the bath. Just before reading 
 the tests, a few drops of glymol (1 quart of a white mineral oil called 
 hydrol colored with 1 ounce alkanet root or with oil-soluble aniline dye) 
 
Cm. 230] Testing Milk and Cream for Butterfat 11 
 
 are added to the tests. The addition of the glymol straightens the upper 
 meniscus or crescent-shaped curve of the fat, making possible an 
 easier and more accurate reading. The tests are then read by the aid 
 of a pair of dividers. The reading is taken from the lower part of the 
 lower meniscus to the straight line formed by the upper meniscus and 
 the glymol (fig. 7). 
 
 Abnormal tests are attributed to practically the same causes described 
 for abnormal milk tests. The readings of a fat column that appears 
 abnormal should never be recorded. 
 
 TESTING SKIM MILK, BUTTERMILK, AND WHEY FOR BUTTERFAT 
 
 The Babcock method of testing whey and skim milk is recommended 
 for general use as the results are easily obtained and sufficiently accu- 
 rate. However, in testing buttermilk the Babcock method should be 
 used only when it is not economical or feasible to use the Minnesota 
 Reagent method. The latter method is preferred since it affords results 
 that are easily duplicated and fat columns that are free from sediment. 
 Fat columns obtained by the Babcock method are frequently supported 
 by a dense black or gray substance which is charred material and not 
 easily recognized as such. This charred material makes the results 
 questionable. 
 
 Owing to the small amount of fat contained in skim milk, butter- 
 milk, and whey, a bottle of special construction is required for testing 
 them by the Babcock method. The bottles, known as double-neck skim- 
 milk bottles, have one large funnel neck for introducing the charge. 
 The other neck has a fine bore and has graduations sufficiently small to 
 record 0.01 per cent fat. The neck is graduated to record a fat content 
 of 0.50 per cent. 
 
 In testing these products the directions outlined for milk testing are 
 used. The mixing is equally as important as in milk or in cream testing. 
 
 Using the 17.6 cc pipette, the 17.5 cubic centimeter charge is delivered 
 into the funnel-like neck. In whirling it is necessary to place the bottles 
 in the tester so that the funnel neck faces the center of the centrifuge. 
 The double-necked bottles are of delicate construction and care should 
 be taken in centrif uging to maintain an even speed so as to avoid break- 
 age. Centrifuge for 10, 3, and 2 minutes, adding water as in testing 
 whole milk. 
 
 The percentage of solids-other-than-fat in the serum of skim milk, 
 buttermilk, and whey vary widely, hence the rule in regard to the 
 amount of acid to use is the same as in the testing of cream. The amount 
 
12 University of California — Experiment Station 
 
 of acid required for complete solution of the solids-not-fat is indicated 
 by the development of a coffee-brown color in the sample, following the 
 mixing of the acid with the sample. For normal skim milk and butter- 
 milk the amount of acid required for successful operation of the test 
 is about the same as required for whole milk. Mixing the acid and skim 
 milk is facilitated if only about half the acid is added and thoroughly 
 mixed with the milk before adding and mixing the remaining half of 
 the acid. 
 
 Since whey contains a smaller amount of solids than skim milk and 
 buttermilk, a smaller amount of acid is required. Ordinarily 12 cubic 
 centimeters of sulfuric acid or less is used for an 18-gram sample 
 of whey. 
 
 Whey, buttermilk, and skim milk contain a certain amount of residual 
 fat which will not separate from the serum upon centrifuging. Never- 
 theless, accurate testing of these products will indicate whether or not 
 efficient results are being obtained in separating, churning, or in cheese- 
 making. 
 
 The Minnesota Reagent method may be used for testing buttermilk, 
 but it gives questionable results on other dairy products. Its main 
 advantages are, that the fat columns are free of insoluble material, and 
 that the reagent is noncorrosive in action. 
 
 The procedure with this method is as follows : 
 
 Place 9 cc (approximately 9 grams) of the buttermilk, which has been 
 prepared as in the directions for preparing milk samples for testing, 
 into the double-necked test bottle used in the Babcock method. 
 
 Add 10 cc of the Minnesota Reagent (prepared in accordance with 
 the directions supplied by the manufacturer). 
 
 Thoroughly mix until all particles of curd are dissolved. 
 
 Place the test bottle in a water bath, temperature 180°-200° F, for 5 
 minutes, shaking once during this time. 
 
 Place the test bottle in the centrifuge so that the large neck is toward 
 the center and counter-balanced with another bottle of the same weight. 
 
 Centrifuge for one-half minute at the speed used in the Babcock test 
 on whole milk. 
 
 Add soft water, temperature approximately 180° F, to bring the fat 
 column within the graduated portion of the neck of the bottle. 
 
 Centrifuge for one-half minute at the above speed. 
 
 If the centrifuge is heated the reading may be made as soon as 
 removed from the centrifuge ; otherwise place in a water bath as used 
 for whole milk. 
 
Cm. 230] Testing Milk and Cream for Butterfat 13 
 
 Take the reading with the aid of dividers and in the manner pre- 
 scribed for the Babcock test of whole milk. 
 
 Multiply the reading by 2 to get the percentage of fat in the sample. 
 This is necessary because the 9-gram sample is only one-half of the 
 18-gram sample for which the scale on the neck of the bottle is made. 
 
 CLEANING THE GLASSWARE 
 
 The test bottles, while still warm, should be inverted over the waste 
 acid jar (see fig. 4) and shaken to completely remove the white calcium 
 sulfate deposited on the bottom of the bottle during the test. When 
 empty the bottles are rinsed with hot water and thoroughly washed with 
 a hot dilute solution of some alkali, such as sal soda, lye, or any of the 
 special cleaning preparations on the market. About an ounce or a table- 
 spoon of powder in 6 or 8 quarts of water is sufficient. When completely 
 cleaned they should be rinsed at least twice with clean hot water. 
 
 8m-7,'33