UNIVERSITY OF CALIFORNIA agricultural experiment Station COLLEGE OF AGRICULTURE E - J - Wickson, d. rector BERKELEY, CALIFORNIA CIRCULAR No. 54 (August, 1910) Some Creamery Problems and Tests BY LEON M. DAVIS (Assistant in Dairy Husbandry) One of the ends most sought by the creameryman is the manufacture of a uniform product. If he would successfully market his butter, this is of great importance. While it is true that in many places the markets make little distinction, still there is an ever-increasing demand for good butter, and this is the demand which the good creameryman would supply. In no way can the uniformity of our creamery products be more surely obtained than through the keeping of creamery records. A study of the method blanks which have been sent in AAdth the Scoring Contest butter, and the scores made since the California Scoring Contest began in February, 1909, reveals the fact that there is a great lack of uniformity, not only in butter made at different creameries, but often in that made at the same creamery. It also appears, from these blanks as well as from correspondence, that a few makers are not familiar with some of the essential creamery calculations and tests. While it is necessary to keep constantly remind- ing the average man of what he knows, lest he forget it, this circular is not presented entirely for that purpose. Most creamerymen are already acquainted with everything herein recorded. It is written rather with the thought of presenting in a brief though intelligible manner, a few creamery problems and suggestions for all those who, for any reason, are unfamiliar with them. Greater uniformity in filling out method blanks is also urged. If these are incomplete, full criticism is impossible, and the chief educa- tional value of the Scoring Contest is lost. It is the mark of a good workman to be able to render a complete report. OVERRUN. There is perhaps no part of the creamery business which is watched more keenly than the matter of overrun. In this day of competition, the cry of the creamery everywhere is, ''We must have all the overrun possible." Yet, a buttermaker or manager is occasionally found who is not familiar with overrun, nor with the factors influencing it. To the creameryman who has not studied this subject carefully, the follow- ing should be helpful, as well as interesting. By overrun is meant the difference between the pounds of butter made and the original pounds of fat. It does not refer to the fat in the butter, but to the fat paid for in the milk or cream from which the butter is made. The amount of butter which can be made from a given amount of cream, and hence the overrun, is influenced by the following factors : (1) Test of the cream. This determines the amount of fat delivered. 1000 lbs. of cream testing 39% contains as must fat as 1500 lbs. of cream testing 26%. (2) Loss of fat in skim milk. This usually averages about .1%, but is increased by using too low temperatures in separating, by wrong treatment of separator, such as increasing or reducing speed, or by overfeeding, thereby not exposing the milk to centrifugal force for a sufficient length of time. (3) Loss of fat in buttermilk. This too varies under different conditions, from .04% to .3%. In churning a rich cream, there is less buttermilk ; hence, less loss than in thin cream under the same con- dition. High churning temperatures favor a heavy buttermilk loss. (4) Amount of water, salt and curd incorporated. The moisture content of butter made strictly for quality averages 12% to 13%, but in commercial butter it varies from 8% to 16%. Where a buttermaker is not careful, a variation of one to four per cent, in moisture content may be found in different churnings. The legal limit of 16% requires the buttermaker to keep below that mark, but it is to his interest to approach it as closely as possible. The salt content of our Contest butter is found to vary from .81% to 3.39%, but in every case where it has gone over 2.50% it has been criticized for heavy salting. An average may be taken as 2.25%, though salt content is always more or less variable. Curd incorporated is usually about 1%. (5) Mechanical losses. These constitute the heaviest fat losses which a creamery has to sustain, and include losses in weighing, slop- ping, irregularities of haulers, not rinsing cans or vats, and, butterfat washed into drain from churn. These losses amount to considerable at times, and have been calculated by some to be never less than 2%. If the amount of overrun equals the difference between the pounds of butter made and the pounds of fat paid for, then the _ , _ Butter made — Fat paid for w .,__ Per cent, of overrun = X 100 Example : Fat paid for Pounds of butter made = 1485 Pounds of fat paid for = 1240 Pounds of overrun = 245 Per cent, of overrun = 19.75 1485 — 1240 X 100 = 19.75 1240 The following figures show the possible overrun from 1000 pounds of butterfat under different percentages of loss, and when butter made therefrom contains different percentages of fat. {Butterfat 81% Moisture 16 Salt 2 Curd 1 100% Average Average No losses buttermilk loss total losses Possible pounds of butter 1234.5 1229.6 1204.9 Pounds overrun 234.5 229.6 204.9 Per cent, overrun 23.4 22.9 20.4 (Butterfat 83% Moisture 14 Salt 2 Curd 1 100% Possible pounds of butter Pounds overrun Per cent, overrun Composition of Butter No losses Average buttermilk loss Average total losses 1204.8 1200 1175.9 204.8 200 175.9 20.4 20 17.5 ( Butterfat 85% ter J Moisture 12 j Salt 2 I Curd 1 100% No losses Average buttermilk loss Average total losses 1176.4 1171.7 1148.2 176.4 171.7 148.2 17.6 17.1 14.8 Possible pounds of butter Pounds overrun Per cent, overrun In creameries throughout the state the overrun varies from 10% to 20%. An excessively high overrun may be due to : 1. Short weights. 2. Reading tests low. 3. Reducing mechanical and other losses to a minimum. 4. Incorporating too much water. Low overrun is the result of : 1. Overweighing. 2. Reading tests too high. 3. Excessive losses. 4. Making butter of a high fat content. STARTER MAKING. From the standpoint of the judge, there is nothing about butter which is criticized as is flavor. Next to having good raw material to work with, there is perhaps no other factor which is of more im- portance in producing good flavor than is the use of good starter. But, just as it is impossible to make good butter from poor cream, so, it is impossible to make good starter from poor milk. It is apparent, then, that the securing of good milk for starter making must first be given attention, and though such often becomes a problem, yet, in this day of gathered cream butter-making, it is usually a problem worth solving. Starter propagation is nothing more nor less than the growing of lactic acid bacteria, and all precautions observed in so doing are merely to furnish a favorable environment for such growth. The addition of a commercial culture to sterile milk introduces pure lactic acid bacteria in such numbers that, under ordinary methods of sterilization, their growth predominates any other type which may be present. In order to properly and successfully make starter, two points must be constantly kept in mind, namely, — sterilization, and freedom from contamination. For the ordinary person, it is hard to realize that the use of a dirty dipper may contaminate a whole can of starter, yet this is not only scientifically but also practically true. For the propagation of mother-starters, glass vessels should be used, as they are easily cleaned, are transparent, and are not subject to rust as are metal vessels. An ordinary quart milk bottle is satisfactory, and may be made sterile by inverting over a steam jet for about fifteen minutes, or by placing in the sterilizer cabinet mentioned later. In either case, steam must be turned on carefully, for these bottles are not of uniform thickness, and uneven expansion, due to heat, may cause them to break. Preparation of milk for mother starter. After the sterilized bottle has cooled, fill it about three-fourths full of good skim milk, and cap. This is ordinary milk, and at the best, will contain organisms which will be of no aid in starter making. Hence, it too must be sterilized. There are various ways of doing this. One of the most satisfactory is the use of a sterilizer cabinet, which is merely a covered can with perforated bottom, so that it can be placed over a steam jet, allowing the steam to circulate inside. A thermometer may be placed through the top of this can. If such a sterilizer is not obtainable, the milk may be heated by placing bottles in a bucket of water, into which a steam hose can be introduced. If the latter method is followed, it will be necessary to fasten the bottles in some way, as there will be more or less motion in the w 7 ater when the steam is turned on. Heat this water to 190° F.-200 F. for thirty minutes, then gradually cool to 70°, by slowly running cold water into the bucket. Let the bottles remain in this water until the milk has had time to cool thoroughly. Never put a thermometer into sterile milk, as it may carry contamination. If poor milk has been used, it will be necessary to repeat the heating on the second daj^, as the resistant forms of bacteria, or spores, are not killed the first day, and will germinate, spoiling the milk. But if good milk is used, cooled immediately, and inoculated, there is found to be little danger from these spores, as lactic acid development takes place so rapidly as to be unfavorable for their development. Making the inoculation. Pure lactic acid cultures, as sent out by commercial laboratories, exist in two forms — dry and liquid — and it is a matter of personal choice on the part of the buttermaker which of these forms he will use. The first inoculation is made by adding the entire contents of a bottle or package of culture to a bottle of sterile milk. Danger of contamination at this time, may be prevented by naming the necks of both bottles with an alcohol flame, previous to the transfer. The bacteria may be distributed throughout the milk by shaking, and this is especially important if the culture was in dry form. Inoculated milk wall usually curdle in twenty-four hours, or less, if kept at a temperature of 70° -85° F. The bottle of starter resulting from the first inoculation will have a powdery, sweet taste, and should not be used in cream ripening, but this will usually dis- appear after the second or third inoculation, and instead there will be a sharp pleasant flavor, and a curd of good granular texture. To successfully propagate mother-starter, it is necessary to go through the before-mentioned process of sterilization daily, inoculating a fresh bottle of sterile milk from the bottle of starter made the pre- vious day. It is well to examine this starter, and such can be done just before making the inoculation. Pour a small amount into a cup, taste, smell, and test for acidity. Do not pour the portion examined back into bottle, but throw it away, as it has likely become con- taminated. If the examination shows the starter to be all right, pour a small amount into the bottle of sterile milk, and shake the latter well to distribute the bacteria throughout. In preparing starter for cream ripening, the work is done on a much larger scale. There are several styles of starter cans, each differently designed, but all answering the same purpose. Nothing more than a general outline can be given for using any of them, as many conditions must be considered. The same care must be exerted as in handling mother starters, but, even under the most careful treat- ment, it is impossible to totally prevent contamination. Thus is seen the necessity of the mother starter. One day 's starter making : 1. Pasteurize freshly separated skim milk in starter can, at a temperature of 185° F. for twenty to thirty minutes, then cool to 70° -85° F., stirring continually. 2. Sterilize and cool bottle of skim milk. 3. Examine mother starter made previous day, and if all right inoculate from it the bottle of sterile milk and starter can of pasteurized milk. 4. Keep these at a temperature of 70°-85° F., until ripened, then cool to 50° -60° F., if they are to be held any length of time before using. Do not stir or shake after first signs of curdling appear. The matter of temperature control will be seen to be of importance. The construction of a starter can is such that trouble from that source is reduced to a minimum, but, the bottles containing the mother starter require attention, unless special apparatus is at hand. In many parts of California, ordinary room temperature is favorable much of the year, but in cool weather, some part of the boiler room is more suitable. When the bottle of starter is ripe, it can be placed in a cold storage room, for, at a temperature of 50° F., or below, bacterial growth is retarded, and the organisms remain in dormant state. It often happens that a starter develops a bad flavor, or becomes contaminated with gas producing bacteria. In such a case, a butter- maker realizes the value of carrying two mother starters. Many do carry two. starters, as so doing requires but little more time. The period a starter will last can not be foretold, so the only rule to follow is, — use while good, and when it "goes off," throw it away. Good starter has a clean mild acid flavor, a delicate aroma, and a smooth granular body. The acidity should be about .7%, for at that point the greatest number of bacteria are present. If a higher acidity is developed, there is a tendency for the starter to become hard and lumpy. COLOEING. No creameryman needs an explanation of the terms "mottled" or "wavy or streaky color." They refer to a difference or unevenness of color, and appear in butter as irregular, lighter and darker portions, and often as spots. The general opinion regarding these defects seems to be that they are the result of uneven salting. This is in part the cause, but another factor enters, and that is the presence of buttermilk or casein compounds. Salt as put into butter should be dissolved by the water present, and a brine solution result. If butter contains casein com- pounds, left in by failure to remove all the buttermilk, they are acted upon and hardened by this brine solution. Consequently, when the butter is worked, streaks and spots result. Where these streaks and spots occur, the lighter portions are due to the presence of casein compounds. The yellow and clear portions are free from these, and the fat is surrounded by clear brine. Well washed butter very seldom shows mottles, unless the salting is done unevenly. Mottled or streaky butter, then, may be prevented by churning at a low temperature to keep the butter in fine granules, washing thoroughly, and working sufficiently to insure equal distribution of salt. Even under these conditions, a certain amount of buttermilk will be retained within the granules, but not sufficient to cause these defects. Mottles or streaks do not necessarily detract from the palatability and wholesomeness of butter, but the fact that present day markets are governed, to a great extent, by appearances, makes it the more important that every buttermaker guard against them. ACIDITY. The best flavor in butter is produced when cream has reached the proper acidity. Although many buttermakers receive poor cream, it is evident also that many ripen it improperly before churning; hence the importance of the acid test. A few experienced makers are able, by taste and smell, to ascertain approximately when the proper degree of acidity has been reached, but a special test which will measure the exact amount of acid present, is more to be relied upon, as the flavor of ripened cream varies some- what with different degrees of richness. Before a man can intelligently test milk or cream, he must know why he does certain things. It may be helpful, therefore, to those creamerymen who are not familiar with the common and practical acid tests in use, to outline briefly the principle and the manner of making these tests. The acid tests most commonly used in creameries are : Marschall 's, Manns', and Farrington's Alkaline Tablet test. All of these are based on the principle that a definite amount of alkali solution of known strength will neutralize a definite amount of lactic acid. By chemistry we are able to determine that one cubic centimeter of one-tenth normal sodium-hydroxide solution, which is the alkaline neutralizer used generally, will combine with exactly .009 gram of lactic acid. In order that we may determine with the eye just when all of the lactic acid in the milk or cream being tested has been neutralized by this alkaline solution, an indicator is used which shows by a change of color when this point is reached. In making a test, therefore, as long as any free lactic acid remains in the milk or cream the color remains white, but when all the lactic acid has been neutralized by the alkaline solution, it changes to pink, and if an excess of alkaline solution is put in, it becomes red. In order to get an accurate test, it is necessary that the proper pinkish shade to attain be well in mind, and this may be determined in the following way: Place two samples of milk or cream in white dishes, side by side, and dilute each with its own volume of distilled water. Dishes with wide tops will expose more surface to view, and the addition of dis- tilled water will thin the sample, and aid in noting any color changes. Add a few drops of indicator to one, then stir with a rod and run neutralizer in slowly, until the first suggestion of pink color. appears, which will indicate the turning point. By comparing a sample being tested, with the original, it is much easier to notice the first change of color. Using the neutralized sample as a color standard, color a small amount of skim milk to the same shade by the use of a few drops of diluted carmine ink. Such a sample may be kept from curdling by the use of uncolored preservative, which must, however, be added before the ink. This skim milk, of standard color for acid titration, may then be bottled and sealed, and kept with the acid test for com- parison as each test is made. The alkaline solution in general use for acid testing, is a one-tenth normal sodium-hydroxide solution. By a one-tenth normal solution is meant one containing just one-tenth as much sodium-hydroxide as a standard solution. Again by chemistry Ave learn that a standard sodium-hydroxide solution contains exactly 40 grams of sodium- hydroxide in 1000 c.c. of distilled water. Therefore, a one-tenth normal solution contains 4 grams of sodium-hydroxide in 1000 c.c. of distilled water. It is not possible for the average person to make these solutions, as one properly standardized requires the work of a chemist. They may be obtained at all dairy supply houses and at some drug stores. MAESCHALL'S ACID TEST. Apparatus Combined bottle and burette, the latter graduated to c.c. 9 c.c. pipette Bottle of indicator White tea cup Making tlic test. — Measure into the cup, 9 c.c. of sample to be tested. The pipette should be rinsed with distilled water, and the rinsings put in cup also, for this does not affect the amount of acid present, inasmuch as the distilled water is pure. Caution is given against using ordinary water, as it may contain a large amount of alkali. Condensed steam from a steam pipe, or rain water can be used in place of distilled water, as these are free from impurities. To the sample, now add a few drops of indicator. Fill the graduated burette to zero with alkaline solution, and run the latter into cup slowly until a pink color remains after shaking. The sample may be kept well mixed by giving it a rotary motion while the test is being made. When 10 the pink color, as determined by the standard, is attained, read from the burette the amount of neutralizer used. Each c.c. of neutralizer is equivalent to one-tenth per cent, acid, for 1 c.c. of one-tenth normal sodium-hydroxide neutralizes .009 grams lactic acid, and 9 c.c. of milk or cream are used. 1 c.c. neutralizer X .009 grams lactic acid X J 00 — .1 /o 9 c.c. milk or cream Thus, if 6 c.c. of neutralizer are used, the per cent of acid equals .6%, or if 4.2 c.c, the per cent, of acid equals .42%. MANNS' ACID TEST. C Burette graduated to 1/10 c.c. Apparatus J Pip^te measuring any known amount J Bottle of indicator v White tea cup Making the test. Measure into cup with pipette, any known amount of sample, rinse and add indicator as in the previous test. Fill the graduated burette to zero with alkaline solution, and run latter into cup until proper shade is attained, using the precautions before men- tioned. From the c.c of neutralizer, and the c.c. of sample used, the per cent, of acid is calculated by the following formula : _ . „ ., c.c. neutralizer X .009 , A „ Per cent, of acid = X 100 c.c. sample used Thus, if 17.6 c.c. of cream require 7 c.c. of neutralizer, 7 X - 009 X 100 = .35% acid 17.6 or, if 50 c.c. cream requires 20.4 c.c. neutralizer, 20 - 4X - 009 X100 = .36%acid 50 or, if 9 c.c. cream require 4 c.c. neutralizer, 4 X .009 9 X 100 = .4% acid FARRINGTON'S ALKALINE TABLET TEST. f 100 c.c. graduated cylinder Apparatus -j 17.6 c.c. pipette (, White tea cup Solution. For this test the solution is made up from alkali tablets, which already contain the indicator. Each tablet contains a definite 11 amount of alkali, so that a solution for testing 17.6 c.c. of cream is made by placing five of them in the graduated cylinder, and filling the latter with distilled water to the 97 c.c. mark. The tablets must be thoroughly dissolved before the solution is ready for use, and this may be accomplished more quickly if a cylinder is such that it may be corked and laid horizontally. Making the test. Measure into the cup, 17.6 c.c. of sample, and rinse as in previous tests. Add the tablet solution carefully until, after shaking or stirring, a pink shade remains, indicating that all the acid in the sample has been neutralized by the alkaline solution. Each c.c. of tablet solution required for neutralization is equivalent to .01% acid. Thus, if 21 c.c. are required, the per cent, of acid is .21%, or if 47 c.c. are required, the per cent, of acid is .47%. A few buttermakers report the acidity in terms other than per cents., and it is impossible to tell what acidity they are using for churning. From the simple method of calculation given under Manns ' test, it is very easy to determine the per cent., and such would be of value in making criticisms. Acidity reported in terms of degrees, or, cubic centimeters, has no meaning to one who does not know exactly how the test was made, whereas, a definite per cent, of acid has a universal meaning. Always report acidity in per cent. MOISTUEE TESTING. Anyone who is familiar with testing of butter for moisture is well aware of the fact that an accurate test is not possible unless the sample taken for testing is a representative one. In view of the heavy penalties imposed because of excessive moisture, no buttermaker can afford to do the work ignorantly or carelessly. The matter of proper ways of taking samples and of testing is as yet more or less unsettled, but the follow- ing suggestions and precautions are generally recognized as being worthy of attention. (1) In taking a sample from the churn, remove a portion of the surface of the butter at various places of the churn, and by means of a spatula take out small pieces. Butter in the churn contains many water pockets and these must be avoided, as they are worked out in packing. In sampling from a cube, take with a trier from several different places; one in the middle, and the others between this and the* edges. The trier should extend the full depth of the cube. Unless the trier is used carefully, the free water appearing on the surface of the butter will be lost, but this should be placed with the butter in 12 the sample jar. A wire or thread is recommended for taking samples from a print, as butter can be cut easily in this way. Several small slices from different parts of the print are sufficient. An ordinary fruit jar is a very satisfactory container for moisture samples and, after they are placed in it, the cap should be screwed down air-tight. (2) Samples taken as above are approximate representatives only, so in order that the portions taken may become a homogeneous mixture, it is necessary that they be melted at as low a temperature as possible (not above 120° F.) in order that none of the volatile substance pass off as vapor, and then cooled until solid, shaking often to insure the even distribution of constituents. (3) Special scales for moisture testing are on the market, and are sensitive enough to give very satisfactory results. They must be kept free from dampness, and should be balanced and kept free from draughts while in use. The aluminum cups used in most moisture tests are capable of taking moisture from the air, if allowed to get thoroughly cool before weighing. In order that these cups, and also the container, may be dry and ready for use at any time, they may be cleaned and inverted over a steam pipe after each test. (4) If a quick or direct flame test is employed, there is danger of burning the sample, unless constant attention is given to it. Regardless of the special test used, all samples should be evaporated to constant weight ; that is, they should be reheated and re weighed until the per cent, of moisture, as determined, remains constant. (5) Every buttermaker should occasionally make duplicate tests to determine the accuracy of his work. If the moisture test of butter sent to the Scoring Contest, as determined at the creamery, is to be compared with the test made by the Division of Dairy Industry, the sample for testing at the creamery should not be taken from the churn, but from a cube or print. A sample from the churn will usually show one per cent, more moisture than a sample of the same butter from a cube or print, and, after the cube has been shipped and held in storage from one to four days, as is often the case with Contest butter, the difference will be still greater. The special points to be emphasized in moisture testing are : 1. An accurate, representative sample. 2. A homogeneous mixture of constituents. 3. Good scales. 4. Accurate weighing. 5. Reheating and reweighing to constant weight. 13 SALTING. As stated before, the salt content of entries sent in has varied from .81% to 3.39%, bnt that, invariably, criticism for high salting was made when it went over 2.5%. The most general criticisms regarding salt are, "too much salt," " grittiness, " or "uneven salt- ing." The first of these is due to an excessive amount of salt being put into the butter, either from a desire of the buttermaker to cover up undesirable flavors, or to a lack of system in measuring salt. There is no excuse for this, as salt is one of the most easily governed con- stituents of butter. Again is emphasized the need of uniformity, especially as salt in butter is becoming of more general interest. The object of salting is not to increase overrun. Quality is not to be sacrificed for quantity, for the consumer's rights must be regarded. Grittiness and uneven salting are related to each other, and quite often the former is due to the latter. Uneven salting occurs when butter is not worked sufficiently to distribute the salt evenly, and also, to a limited extent, if the maker is not careful when putting salt in the churn to get it well scattered. Grittiness may be due to uneven salting, whereby there exists in one portion of the butter an excess of salt and more than can be dissolved by the water present. It is also the result of using too much salt, or of using impure salt, which contains insoluble substances. Defects in salt can no doubt be overcome by wet salting, which was recommended in a former report, and which was outlined as follows : For each 10 pounds of butter, mix 1 pound of salt and 2 pounds of water. Bring salt and water together thirty minutes before using. After butter is drained, add the brine, revolve ten times on slow gear, and allow to stand ten minutes before working. The amount and time, however, can be varied somewhat to suit individual conditions. Any serious defects in salt may be determined by taste, but, in determining the uniformity of salt in butter from day to day, a salt test is necessary as the taste is to be relied upon only in determining extremes. Practical salt tests may be obtained at dairy supply houses, and the use of one enables a maker to determine accurately the per cent, of salt retained in his butter. The principle of these tests is the same, and may be given as follows : A silver nitrate solution of known strength neutralizes a definite amount of sodium chlorid, or common salt. By titration, in the presence of an indicator (potassium chromate solution) it is possible to determine, by the color of the 14 precipitate, when all the salt has been neutralized, and from the amount of neutralizer or silver nitrate solution used, the amount, and hence the percentage of salt in the butter analyzed can be calculated. Full directions for making these tests are sent with each set of apparatus.