f Jlgpgl»ll^ iiiiiiiiw^^ ■iilfiivllliiilii 'IK CI MlfjIC .RRINCITON : WOLI. LIBRARY QF CONGRESS. oCt^ZJ^ Chap.?..—. Copyright No. Shell/ ^"^ 51^^ TESTING MILK AND ITS PRODUCTS. A MANUAL FOR DAIRY STUDENTS, CREAMERY AND CHEESE- FACTORY OPERATORS AND DAIRY FARMERS. BY / E. H. FARRINGTON and F. W. WOLL Professor in ^Charge of Dairy School Asst. Prof, of Agrl. Chemistry Of the Uqiversity of Wisconsiq. ©Uith ^ilu^tvatianft. FIRST EDITION. madison, wis. Mendota Book Company 1897. ALL RIGHTS RESERVE^. L WO COPIES-RECEIVED Copyright, 1897 By E. H. FARRINGTON and F. W. WOLL M. J. Cantwell, Printer Madison, Wis. PREFACE. The present volume is intended for the use of dairj' students, creamery and cheese-factory operators, practical dairymen, and others interested in the testing or analysis of milk and its prod- ucts. The subject has been largely treated in a popular manner; accuracy and clearness of statement, and systematic arrangement of the subject matter has, however, been constantly kept in mind. The aim has been to make the presentation intelligible to students with no further training than a common-school education, but their work will naturally be greatly lightened by the aid and guidance of an able teacher. Complete directions for making tests of milk and other dairy products are given ; the difficulties which the beginner may meet with, are considered in detail, and suggestions offered for avoiding them. It is expected that a factory operator or practical dairy- man, by exercising ordinary common sense and care, can obtain a sufficient knowledge of the subject through a study of the various chapters of this book to make tests of milk, cream, etc., even if he has had no previous experience in this line. For the benefit of advanced dairy students who are somewhat familiar with chemistry and chemical operations. Chapter XIV has been added giving detailed instruction for the complete chem- ical analysis of milk and other dairy products. The detection of preservatives and of artificial butter or filled cheese has also been treated in this connection. As the subject of milk testing is intimately connected with the payment for the milk delivered at butter- and cheese factories, and with factory dividends, a chapter has been devoted to a discus- sion of the various systems of factory book-keeping, and tables iv Testing Milk and Its Products. greatly facilitating the work of the factory secretary or book keeper have been prepared and are included in the Appendix. Acknowledgment is due to the following parties for the use of electrotypes, viz: Vermont Farm Machine Co., Bellows Falls, Vt.^ Cornish, Curtis and Greene Mfg. Co., Fort Atkinson, Wis.; Elgin Mfg. Co., Elgin, 111.; F.B. Fargo & Co., Lake Mills, Wis.; DeLaval Separator Co., N. Y. City; Henry Troemner, Philadelphia, Pa.; Springer Torsion Balance Co., N. Y. City; J. H. Monrad, Win- netka. 111.; Borden & Selleck Co., Chicago, 111.; Dairymen's Sup- ply Co., Philadelphia, Pa.; and the agricultural experiment sta- tions at New Haven, Conn., and Madison, Wis. University of Wisconsin, Madison, Wis., Oct. 1, 1897. E. H. FARRINGTON. F. W. WOLL. TABLE OF CONTENTS. Introduction. Pages^ The need of a practical milk- and cream test. Introduction of milk tests. Short's test. Other milk tests. The Babcock test. Foreign methods. Gerber's method. DeLaval butyrometer. Fjord's centrifugal cream-test, ----- 1-10* Chap. I. Composition of milk and its products. Water. Fat. Casein and albumen. Milk sugar (lactose). Ash. Other components. Colostrum milk, 11-19^ Chap. II. Sampling milk. Sweet milk. Partially churned milk. Sour milk. Frozen milk, 20-24 Chap. III. The Babcock test. Directions for making the test: Sampling. Add- ing acid. Mixing milk and acid. Whirling bottles. Add- ing water. Measuring the fat. Discussion of the details of the test: 1. Glass- ware. Test bottles. Marking test bottles. Cleaning test bottles. Pipettes. " Fool pipettes." Acid measures. The Swedish acid bottle. Calibration of glassware. Cal- ibration with mercury. Cleaning mercury. Calibration with water, a. Measuring the water, b. Weighing the water. 2. Centrifugal machines. Speed required for the complete separation of the fat. Ascertaining the neces- sary speed of testers. Hand testers. Power testers. 3. Sulfuric acid. Testing the strength of the acid. The Swedish acid tester. The color of the fat column an index to the strength of the acid used. Influence of tem- perature on the separation of fat. 4. Water to be used in the Babcock test. Reservoir for water. 5. Modifica- tions of the Babcock test. The Russian milk test. Bart- lett's modification, ..-.----- 25-63'- vi Testing Milk and Its Products. Chap. IV. Cream testing. Pages. Errors of measuring cream. Avoiding errors of measuring cream. Cream test bottles. The bulb-necked cream bottles. The Winton cream bottle. Use of milk test bottle. Use of 5 cc. pipette. Weighing the cream, 64-73 Chap. T. Babcock test for other milk products. Skim milk, butter milk, and whey. The double- necked test bottle. The double-sized skim milk bottle. Cheese. Condensed milk, 74-79 Chap. YI. The lactometer and its application. The Quevenne lactometer. Influence of tempera- ture. N. Y. Board of Health lactometer. Reading the lactometer. Time of taking lactometer readings. Cal- culation of milk solids. Adulteration of milk. Calcula- tion of extent of adulteration. Skimming. Watering. Watering and skimming. Other methods of adulteration 80-93 Chap. YII. Testing the acidity of milk and cream. Cause of acidity in milk. Methods of testing acidity. Manns' test. Devarda's acidimeter. The alka- line tablet test. Acidity of cream. Determination of acidity in sour cream. The standard solution used. Spill- man's cylinder. Rapid estimation of the acidity of ap- parently sweet milk and cream. Detecting preser valine in milk. "Alkaline tabs," - 94-110 Chap. YIII. Testing the purity of milk. The Wisconsin curd test. The fermentation test, 111-115 Chap. IX. Testing milk on the farm. Variations in milk of single cow^s. Cause of vari- ation in fat content. Number of tests required during a period of lactation in testing cows. When to test a cow. a. As to quality of milk produced, b. As to quantity of milk produced. Curler's method. Record of tests. Sampling milk of single cows. Size of test sample. Variations in herd milk. Ranges in variations of herd milk. Influence of heavy grain feeding on the quality of milk. Influence of pasture on the quality of milk. Method of improving the quality of milk, 116-133 Table of Contents. vii Chap. X. Composite samples of milk. Pages. Methods of taking composite samples, a. Use of tin dipper, b. Drip sample, c. Scovell sampling tube. One-third sample pipette. Accuracy of the described methods of sampling. Preservatives for composite sam- ples. Bi-chromate of potash. Other preservatives. Care of composite samples. Fallacy of averaging percent- ages. A patron's dilemma, 134-14-9 Chap. XT. Cream testing at cream-gathering creameries. The space system. The oil test churn. The Bab- cock test lor cream. Sampling tube. Sampling cream for composite testing, - - 150-159 Chap. XII. Calculation of butter and cheese yields. Calculation of yield oi butter: Butter fat test and yield of butter. Variations in composition of but- ter. Overrun of churn over test. Factors influencing the overrun. Calculation of overrun. Conversion factor for butter fat. Butter yield from milk of different richness. a. Use of butter chart, b. Use of overrun table. Calcu- tion of yield of cheese: a. From fat. b. From solids not fat and fat. c. From casein and fat, . . - . 160-173 Chap. XIII. Calculating dividends. Calculating dividends at creameries: Proprie- tary creameries. Co-operative creameries. Illustrations of calculations of dividends. Other systems of payments. Paying for butter delivered. Relative value tables. Cal- culating dividends at cheese lactones: Proprietary fac- tories. Co-operative factories. Illustrations of calcula- tions of dividends, 174-185 Chap. XIY. Chemical analysis of milk and its products. A. Milk: Specific gravity. Water. Alternate method. Fat. Casein and albumen. Van Slyke's method. Milk sugar. Ash. Acidity of milk. Detection of preser- vatives in milk. Boracic acid. Bi-carbonate of soda. Fluorids. Salicylic acid. Formalin. B. Skim milk, but- viii Testing Milk and Its Products. ter milk and whey. C. Butter. Sampling. Determina- Pages. tion of water. Fat. Casein. Ash. Complete analysis of butter in the same sample. Detection of artificial but- ter. Filtering the butter fat. Specific gravity. Reichert- Wollny method (Volatile acids). D. Cheese. Water. Fat. Casein. Ash. Other constituents. Detection of oleomargarine cheese ("Filled" cheese) - - - - 186-203 Appendix. Table I. Composition of milk and its products. Table II. Milk standards. Table III. Ouevenne lactometer degrees corres- ponding to the scale of N. Y. Board of Health lactome- ters. Table IV. Correction table for specific gravity of milk. Table V. Percent of solids not fat, correspond- ing to to 6 percent of fat and lactometer readings of 26 to 36. Directions for the use of tables VI, VII and IX. Table VI. Pounds of fat in 1 to 10,000 pounds of milk testing 3 to 5.35 percent. Table VII. Amount due for butter fat, in dol- lars and cents, at 12 to 25 cents per pound. Table VIII. Relative value tables. Table IX. Butter chart, showing calculated yield of butter, in pounds, from 1 to 10,000 pounds of milk testing 3.0 to 5.3 percent of fat. Table X. Overrun table, showing pounds of but- ter from 100 pounds of milk. Table XI. Yield of cheese, corresponding to 2.5 to 6 percent of fat, with lactometer readings of 26 to 36. Table XII. Comparisons of Fahrenheit and Centigrade (Celcius) thermometer scales. Table XIII. Comparison of metric and custom- ary weights and measures. Suggestions regarding the organization of co- operative creameries and cheese factories, . - - 205-232 Index, 233-236 TESTING MILK AND ITS PRODUCTS. INTRODUCTION. The need of a rapid, accurate and inexpensive method of determining the amount of butter fat in milk and other dairy products became more and more apparent, in this country and abroad, with the progress of the dairy industry, and especially with the growth of the factory system of but- ter and cheese making during the last few decades. So long as each farmer made his own butter and sold it to pri- vate customers or at the village grocery, it was not a matter of much importance to others whether the milk produced by his cows was rich or poor. But as creameries and cheese factories multiplied, and farmers in the dairy sections of our country became to a large extent patrons of one or the other of these, a system of equitable payment for the milk or cream delivered became a vital question. 1. The need of a practical milk=and cream test. The creameries in existence in this country up to within ten years were nearly all conducted on the cream-gathering plan: the different patrons set their milk, and cream gatherers hauled the cream to the creamery, usually twice or three times a week, where the mixed lots of cream were then ripened and churned. The patrons were paid per inch of cream furnished; a creamery inch is a quantity of cream which fills a can twelve inches in diameter, one inch high, or 113 cubic inches. This quantity of cream is sup- 2 Testing Milk and Its Products. posed to make a pound of butter, but cream from different sources, or even from the same sources at different times, varies greatly in butter-producing capacity, as will be shown under the subject of cream testing (140*). The system of paying for the number of creamery inches delivered could not therefore long give satisfaction. The proposition to take out a small portion, a pint or half a pint of the cream furnished by each patron, and deter- mine the amount of butter which these samples would make on being churned in so-called test churns, found but a very limited acceptance, on account of the labor involved and the difficulty of producing a first-class article of all the small batches of butter thus obtained. 2. The introduction of the so-called oil test churn (187) in creameries, which followed the creamery inch system, marked a decided step in advance, and it soon came into general use in cream-gathering districts. In this test, glass tubes of about f inch internal diameter and nine inches long, are filled with cream to a depth of 5 inches, and the cream churned; the tubes are then placed in hot water, and the column of melted butter formed at the top is read off by means of a scale showing the number of pounds of but- ter per creamery inch corresponding to different depths of melted butter. While the oil test is capable of showing the difference between good and poor cream, it can not, accord- ing to investigations conducted at the Wisconsin experiment station, make strictly accurate distinctions between different grades of good and of poor cream. As a result, perfect justice can not be done to different patrons of creameries where payments for cream delivered are made on the basis of this test. Refers to paragraph number. Introduction. 3 3. In cheese factories, and since the introduction of the centrifugal cream separator, in separator creameries, the problem of just payment for the milk furnished by different patrons was no less perplexing than in case of gathered- cream factories. By the pooling system generally adopted, each patron received payment in proportion to the number of pounds of milk delivered, irrespective of its quality. Patrons delivering rich milk naturally will not be satisfied with this system when they find out the quality of their milk as compared with that of their neighbors. The temp- tation to fraudulently increase the amount of milk delivered, by watering, or to lower its quality by skimming, will fur- thermore prove too strong for some patrons; the fact that it was difficult to prove any fraud committed, from lack of a reliable and practical method of milk analysis, rendered this pooling system still more objectionable. 4. As another instance in which the need of a simple test for determining the fat content of different kinds of milk was strongly felt may be mentioned the case of private dairymen and breeders of dairy cattle, who desired to follow up the butter-producing capacities of the individual cows in their herds. The only manner in which this could be done, was by the cumbersome method of trial churnings: by sav- ing the milk of the cow to be tested, for a day or a week, and churning separately the cream obtained. This re- quires a large amount of work when a number of cows are to be tested, and can not therefore be done except in com- paratively few cases, with cows of great excellence, or by farmers having plenty of hired help. 5. Introduction of milk tests. The first method which fulfilled all reasonable demands of a practical and 4 Testing Milk and Its Products. reliable milk and cream test was the Babcock test, invented by Dr. S. M. Babcock, chemist to the Wisconsin experiment station; a description of the test was first published in July, 1890, as bulletin No. 24 of Wisconsin experiment station, entitled: A new method for the estimation of fat in milk, esi^ecially adapted to creameries and cheese factories. This test which is now known and adopted in all parts of the world where dairying is an important industry, was not, however, the first method proposed for this purpose which could be successfully operated outside of chemical laboratories, It was preceded by a number of diflferent methods, the first one published in this country being Short's method, invented by Mr. F. G. Short, and described in bulletin No. 16 of Wisconsin experiment station in July, 1888. 6. Short's test. In this ingenious method, a certain quantity of milk (20 cc.*) was boiled with an alkali solution, and afterwards with a mixture of sulfuric and acetic acids; a layer of insoluble fatty acids separated on top of the liquid, and was brought into the graduated neck of the test bottles b}^ addition of hot water; the reading gave the per cent, of fat in the sample of milk tested. Short's method did not find very wide application, both because it was rather lengthy and its manipulations some- what difficult for non-chemists, and because several other methods were published shortly after it had been given to the public. 7. Other milk tests. Of these may be mentioned, besides the Babcock test, already spoken of, the Failyer and Willard method,! Parson's method,t Cochran's test,§ * See 44, footnote. f Kansas experiment station report, 1888, p. 149. % N. H. experiment station report, 1888, p. 69. ? Journal of Anal. Chem., Ill (1889), p. 3S1. Introduction. 5 the Patrick or Iowa station test,^ and the Beimling (Lefif- mann and Beam) test.t Of foreign methods published at about the same time, or previously, the lactocrite,+ Lieber- mann's method, || the Schmid,^ Thoerner,*[ and Roese- Gottlieb** methods may be noted. 8. AH these tests were similar in principle, the solids not fat of the milk being in all cases dissolved by the action of one or more chemicals, and the fat either measured as such in a narrow graduated tube, or brought into solution with ether, gasoline, etc., and a portion thereof weighed on evaporation of the solvent. While this principle is an old one, having been employed in chemical laboratories for man}' years past, the adaptation of it to practical conditions, and the details as to apparatus, and chemicals used were of course new and different in each case. The American tests given were adopted to a limited extent within the states in which they originated, and even outside of them, as in case of the Short, Patrick and Beimling methods. The Babcock test soon, however, nearly everywhere replaced the different methods mentioned, and during the past five or six years it has been in practically exclusive use in creameries and cheese factories in this country where payments are made on basis of the qualit}" of the milk delivered, as well as in * la. exp. sta., bull. No. 8, February, 1890; Iowa Homestead, June 14, 1889. t Vermont exp. sta., bull. No. 21, September, 1890. For description of these and other volumetric methods of milk analysis, see Wiley, Agricultural Analysis, Vol. Ill, p. 490, et seq ; Wing, Milk and its Products, p. 33, et seq; and Snyder, Chemistry of Dairying, pp. 112-113. X Analyst, 1887, p. 6. II Fresenius' Zeitschr. 22, 383. gibid. 27, 464. \ Chem. Centralbl., 1892, 429. ** Landw. Vers. Stat., 40. 1. 6 Testing Milk and Its Products. the routine work in experiment station laboratories, and among milk inspectors and private dairymen. , 9. The Babcock test. An examination of the causes of the present general adoption of the Babcock test will show the strong points of the test, and the requirements made of a practical milk test. The main causes why this test has replaced all competitors are doubtless to be sought in its simplicity and its cheapness: Its manipulations are few and easily learned, and it is cheap, both in first cost and as regards running expenses. The test is speedj^, and accurate within one or two-tenths of one per cent.* Only one chemical is used, and no training in or knowl- edge of chemistry is required on part of the operator. The percentages of fat in the samples tested are shown directly from the readings of the fat column, without refer- ence to a scale or table. Only a small quantity of milk is used (about two-thirds of an ounce, 17.6 cc.) The apparatus is easily kept in order, and the chances of accidents in operating the test, with properl}^ made machines, are very slight indeed. A small and a large number of samples may furthermore be tested with equal facility at the same time, up to the capacity of the tester. The results obtained may be easily verified b}' renewed tests in the same or another machine. The test bottles when charged with the samples of milk * For a summary of comparative analyses made by the Babcock test and gravimetric analysis up to 1892, see Hoard's Dairyman, Oct. 7, 1892, p. 2560; also Schrott, Milchzeitung, 1896, p. 183, et seq. Int7'oduction. ' 7 or other dair}' products may be left for months, if desired, before the test is completed, and correct results still be obtained. The completed tests will keep indefinite!}- in the bottles, so that the results may be verified at any future time, if desired. Sour milk may be analyzed with perfect • assurance of accurate results, provided it can be properly sampled. The test is finally applicable, besides to full milk, to cream, skim milk, butter milk, whey, condensed milk and (if a small scale for weighing out the sample is available) to cheese. 10. With all these advantages, the Babcock milk test is not in every respect an ideal test. The handling of the very corrosive sulfuric acid requires constant care and attention; the speed of the tester, the strength of the acid, the temperature of the milk to be tested, and other points, always require watching, lest the results obtained be too low, or otherwise unsatisfactory. The test is, however, for general purposes, in the opinion of the writers, the very best at our disposal, and in the hands of careful intelligent operators, will easil}- give most satisfactory results. 11. Foreign methods. In European countries three practical milk and cream tests, besides the Babcock test, are in use at the present time, viz : Gerhers acid-hutyrome- ter, De LavaVs huiyrometer. and Fjord's cejitrifugal cream test.^ Of these, the last test given has never been introduced into this country, to our knowledge, and the former two, only on a small scale. * The Lister-Babcock milk test advertised in English papers and known as such in England, is the regular Babcock test, to which the English manufacturers have prefixed their name. 8 Testing Milk and Its Products. 12. The Qerber method* (fig. 1) is essentially the old Beimling method worked out independently by the Swiss chemist, Dr. N. Grerber. In this test sulfuric acid of the same strength is used as in the Babcock test, and a small quantity of amyl alcohol is added. The amyl alcohol facili- tates the separation of the fat, but introduces a source of error which may become serious, and especially so, where the results obtained with a Fig. 1. The Gerber acld-butyrometer. ^gw lot of amyl alcohol can not be compared with gravimetric analysis or with tests made with amyl alcohol known to give correct results. 13. In the De Laval butyrometer (fig. 2) the same acid is used as in the Babcock test, but the tubes employed, and the manipulations of the method differ materially from this test; a smaller sample of milk is taken (only 2 cc.) and a correspondingly small quantity of acid used. The results obtained are correct. Where a large number of milk samples are tested every day, as is the case, for instance, in European milk control stations, the butyrometer may be preferable to the Babcock test; but it requires more skill of the operator and is more difficult to work satisfactorily in case of milk which cannot be easily sampled, as sour, lop- pered, or partially churned milk. The machines placed on * Gerber, Die Praktische Milch-Pruefung. Introduction. " 9 the market both by Dr. Gerber and the De Laval Company are more expensive than the Babcock testers sold in this country; the De Laval test requires a high speed, 5-6000 Fig. 2. De Laval's butyrometer. revolutions per minute, and therefore places greater de- mands for solidity in the machine than does the Babcock test. 14. Fjord's centrifugal cream tester* (fig. 3) is ex- tensively used in Denmark and is mentioned in this connec- tion as it furnishes a fairly reliable method for compar- ing the quality of different lots of milk. The method was published in 1878, b}^ the late N. J. Fjord, director of the state experiment sta- tion in Copenhagen, through Fig. 3. Fjord's centrifugal cream tester. whoSC excrtioUS and On * State Danish experiment station, Copenhagen, sixth and ninth reports, 1885-7. lo • Testing Milk and Its Pj'oducts. whose authority it was generally introduced into Danish creameries in the middle of the eighties. No chemicals are added in this test, the milk being simply placed in glass tubes, seven inches long and about 2-3 of an inch in diame- ter, and whirled for twenty minutes at a rate of 2000 revo- lutions per minute at 55° C. (131° F.) The reading of the cream layer thus obtained gives the per cent, of cream, and not of butter fat, in the sample tested. 192 samples of milk can be tested simultaneously. Within the limits of normal Danish herd milk, the results obtained correspond to the percents of fat present in the samples, one per cent, of cream being equal to about 0.7 per cent, of fat; outside of these limits the test is, however, unreliable, especially in case of very rich milk and strippers' milk. Only sweet milk can be tested by this method. The recent introduction of milk tests proper into Denmark, like the Grerber, Babcock and De Laval tests, will, however, in all probability in time force the Fjord cream test out of Danish creameries, for similar reasons that relegated to obscurity the gravity cream tests (creamometers.) Before going over to the main part of the present work, the discussion of the Babcock milk test, a brief description of the chemistry of milk and its products is given, so that the student may understand what are the components of dairy products, and the relation of these to each other. Only such points as have a direct bearing on the subject of milk testing and the use of milk tests in butter and cheese fac- tories or private dairies will be treated in this chapter, and the reader is referred to standard works on dairying for further information in regard to the composition of dairy products. Composition of Milk and Its Products. 1 1 CHAPTER I. COMPOSITION OF MILK AND ITS PRODUCTS. 15. Milk is composed of the following substances: water, fat, casein, albumen, milk sugar and ash. A few other substances are present in small quantities, but are of no practical importance and will not be considered here. The components of the milk less the water are known, collect- ively, as milk solids, or total solids, and the total solids less the fat, i. e. casein, albumen, milk sugar, and ash, are often spoken of as solids not fat, or the non-fatty milk solids. The milk serum includes all components of the milk less the fat; the serum solids are therefore another name for the solids not fat; when given, they are, however, generally calculated to per cent, of milk serum, not of milk. If, e. g., a sample of milk contains 9 per cent, of solids not fat, and 3 per cent, of fat, the milk serum will make up 97 per cent, of the milk, and the serum solids, .^j = 9.28 per cent, of the milk serum. IG. Water. The amount of water contained in milk ranges from 80 to 90 per cent. Normal cows' milk will not as a rule contain more than 88 per cent, of water, nor less than 84 per cent. In states where there are laws regulating the sale of milk, as is the case m eighteen states in the Union (see Appendix), the maximum limit for water in milk in all instances but one (South Carolina) is 88 per cent. ; the state mentioned allows 88.5 per cent, of water in milk offered 12 Testing Milk and Its Products. for sale within her borders. The effect of fraudulently in- creasing the water content of milk by watering is considered under Adulteration of Milk. 17. Fat. The fat in milk is not in solution, but sus- pended as very minute globules, which form an emulsion with the milk serum; the globules are present in immense numbers, viz: on the average about one hundred million in a single drop of milk; a quart of milk will contain about two thousand billions of fat globules, a number written with thirteen ciphers. The size of the globules in the milk from the same cows varies according to the stage of the period of lactation, the globules being largest at the beginning of the lactation period, and gradually decreasing in size with its progress. Different breeds of cows have fat globules of different average sizes; the Channel Island cows are thus noted for the relatively large fat globules of their milk, while the Lowland breeds, the Ayrshire, and other breeds have uniformly smaller globules. The diameter of average sized fat globules in fresh milkers is about 0.004 millimeter, or one-six thousandth of an inch; that is, it takes about six thousand such globules placed side by side to cover one inch in length. The globules in any sample of milk vary greatly in size; the largest globules are recovered in the cream when the milk is set, or run through a cream sepa- rator, and the smallest ones remain in the skim milk; prop- erly skimmed separator skim milk will contain only a small number of very minute fat globules. Milk fat is composed of so-called glycerides of the fatty acids, i. e. compounds of the latter with gl3'cerine; some of the fatty acids are insoluble in water, viz: palmitic, stearic, and oleic acids, while others are soluble and volatile, the Comj>osition of Milk and Its Products. 13 chief ones among the latter being butyric, caprylic, and caproic acids. The glycerides of the insoluble fatty acids make up about 92 per cent, of the pure milk fat, and about 8 per cent, of the glycerides of volatile fatty acids are there- fore found in natural milk- (and butter-) fat. The distinc- tion between natural and artificial butter lies mainly in this point, since artificial butter (butterine, oleomargarine) as well as other solid animal fats contain only a very small quantity of volatile fatty acids. The glycerides of the vola- tile fatt}^ acids are unstable compounds, easily decomposed through the action of bacteria or light; the volatile fatty acids thus set free, mainly butyric acid, are the cause of the unpleasant odor met with in rancid butter. The per cent, of fat in cows' milk is generally between three and six per cent. American milk contains on the average toward four per cent, of fat. The milk from single cows in perfect health will occasionally go below or above the limits given, but the mixed milk from a whole herd rarely falls outside of these limits. The legal standard for fat in milk in most states of the Union is 3 per cent.; Rhode Island allows milk containing 2.5 per cent, of fat to be sold as pure, while Georgia and Minnesota require it to contain 3.5 per cent, and Massachusetts 3.7 per cent, (in the months of May and June; see Appendix.) 18. Casein and albumen. These belong to the so- called nitrogenous substances, distinguished from the other components of the milk by the fact that they contain the element nitrogen. Another name is albuminoids or protein compounds. Casein is precipitated by rennet in the presence of soluble calcium salts, and by dilute acids and certain chemicals; albumen is not acted upon by these 14 Testing Milk and Its Products. agents, but is coagulated b3^ heat, a temperature of 170° F. being sufficient to effect a perfect coagulation. The casein, with fat and water, form the main components of most kinds of cheese; in the manufacture of cheddar and most other solid cheeses, casein is coagulated by rennet, and the curd thus formed holds fat and whey mechanically, the latter containing in solution small quantities of non-fatty milk solids. The albumen goes into the whey, and in some countries is also made into cheese by evaporating the whey under constant stirring; usually whole milk of cows or goats is added and incorporated into such cheese {primost^ goat cheese). Casein is present in milk partly in solution, in the same way as milk sugar, soluble ash-materials and albumen, and partly in suspension, in an extremel}^ fine colloidal condi- tion, mixed or combined with insoluble calcium phosphates. The casein and calcium phosphates in suspension in milk may be retained on a filter made of porous clay (so-called Chamherland filters). About 80 per cent, of the nitrogenous compounds of nor- mal cows' milk are made up of casein; the rest is largely albumen. If the amount of casein in milk be determined by precipitation with rennet or dilute acids, and the albu- men by boiling the filtrate from the casein precipitate, it will be found that the sum of these two compounds does not make up the total quantity of nitrogenous constituents in the milk. The small remaining portion (about five per cent, of the total nitrogenous constituents) is called by vari- ous authors, globulin, albumose, hemi-albumose, nuclein, proteose, etc. The nitrogenous constituents of milk are very unstable compounds, and their study presents many and great difficulties; as a result we find that no two scien- Composition of Milk and Its Products. 15 tists who have made a special study of these compounds agree as to their properties, aside from those of casein and albumen, or their relation to the nitrogenous substances found elsewhere in the animal body. For our purpose we may, however, consider the nitrogen compounds of milk as made up of casein and albumen, and the term casein and albumen used in this book is meant to include the total nitrogenous constituents of milk, as obtained by multiplying the total nitrogen content of the milk by 6.25.* The quantity of casein in normal cows' milk will vary from 2 to 3.5 per cent., and of albumen from .5 to .8 per cent. The total content of casein and albumen will range between 2.5 and 4.2 per cent, the average being about 3.5 per cent. Milk with a low fat content will contain more casein and albumen, than fat, while the reverse is generally true in case of milk containing more than 3.5 per cent, of fat. 19. Milk sugar or lactose belongs to the group of or- ganic compounds known as carbohydrates. It is a commer- cial product manufactured from whe}^, and is obtained in this process as pale white crystals, of less sweet taste, and less soluble in water than ordinary- sugar (cane sugar, sucrose). About 70 per cent, of the solids in the whey, and 33 per cent, of the milk solids, are composed of milk sugar. When milk is left standing for some time, viz: from one to several days, according to the temperature of the surround- ing medium, it will turn sour, and soon become thick and loppered. This change in the composition and the appear- ance of the milk is brought about througrh the action of * The factor 6.2.5 is generally used for obtaining the casein and albumen from the total nitrogen in the millc, although 6.37 would be more correct, since these substances, .according to our best authorities, contain on the average 15.7 per cent, of nitrogen. 1 6 Testing Milk and Its Products, d acid-forming bacteria on the milk sugar; these are present in ordinary milk in immense numbers, and under favorable conditions of temperature multiply rapidly, feeding on the milk sugar as they grow, and decomposing it into lactic acid. When this change alone occurs, there is not neces- sarily a loss in the nutritive value of the milk, since the milk sugar breaks up directly into lactic acid; this is shown by the following chemical formula: Ci„H,„0„. HoO {lactose) =-4: Q,Yi,Oz (lactic acid.)'' Ordinarily the souring of milk is, however, more com- plicated, and other organic bodies, like butyric acid, alcohol, etc., and gases like carbonic acid and hydrogen are formed, resulting in a loss in the feeding value of the milk. While sour milk may therefore contain a somewhat smaller proportion of food elements than sweet milk, the feeding of it to farm animals, especially pigs, will generally produce better results than is obtained in feeding similar milk in a sweet condition. The cause of this may lie in the stimulating effect of the lactic acid of the sour milk on the appetites of the animals, or in its aiding digestion by in- creasing the acidity of the stomach juices. That the souring of milk is due to the activities of bac- teria present therein is shown clearly by the fact that steril milk, i. e., milk in which all germ life has been killed, will remain sweet for any length of time when kept free from infection. The amount of milk sugar found in normal cows' milk varies from 3 to 6 per cent, the average content being about * One molecule of milk sugar is composed of 12 atoms of carbon (C), 22 atoms of hydrogen (H), 11 atoms of oxygen (0), and one molecule of water (HjO). In the same way, the lactic acid molecule consists of :> atoms of carbon, 6 atoms of hydrogen, and 3 atoms of oxygen. Co7npositwn of Milk and Its Products. 17 5 per cent.; in sour milk this content will be decreased to toward 4 per cent. 20. Ash. The ash or mineral substances of milk are largely composed of chlorids and phosphates of sodium, potassium, magnesium and calcium; iron oxid, and sulfuric and citric acids are also present in small quantities among the normal mineral milk constituents. The amounts of the different bases and acids found in milk ash have been deter- mined by a number of chemists; the average figures ob- tained are given in the following table, calculated per 100 parts of milk (containing .75 per cent, of ash) and per 100 parts of milk ash. Mineral Components of Milk. In per cent, of Milk. In per cent, of Ash. Potassium oxid (K^O) 19 per ct. 25.64 per ct. Sodium oxid (NajO) 09 12.45 Lime(CaO) 18 24.58 Magnesia (MgO) 02 3.09 Iron oxid (FcjOg) 002 .34 Phosphoric anhydrid (P2O5) 16 21.24 Chlorin (CI) 12 16.34 .762 per ct. 103.68 per ct. Less oxygen, corresponding to chlorin 012 3.68 .75perct. 100.00 per ct. The combinations in which the preceding bases and acids are contained in the milk are not known with certainty; the following scheme is, however, given on the best authority and is believed to be substantially correct. 2 1 8 Testing Milk and Its Products. Percentage Composition of Milk Ash {Soeldner). Sodium chlorid (common salt) 10.62 per ct. Potassium chlorid — 9 16 Mono-potassium phosphate 12.77 Di-potassium phosphate 9.22 Potassium citrate 5.47 Di-magnesium phosphate 3.71 Magnesium citrate 4.05 Di-calcium phosphate 7.42 Tri-calcium phosphate 8.90 Calcium citrate 23.55 Lime combined with casein 5.13 100.00 According to the same author, 36 to 56 per cent, of the phosphoric acid found in milk, and from 53 to 72 per cent, of the lime, are present in suspension in the milk, as di- and tri-calcium phosphates, and may be filtered out by means of Chamberland filters (18), or by long continued centrifuging (Babcock *.) The rest of the ash constituents are dissolved in the milk serum. The ash content of normal cows' milk varies but little, as a rule only between .6 and .8 per cent, with an average of .7 per cent. Milk with a high fat content generally con- tains about .8 per cent, of ash; strippers' milk always has a high ash content, at times even exceeding one per cent. Ordinarily, the mineral constituents of milk are, however, the components least liable to variations. 21. Other components. Besides the milk constituents enumerated and described in the preceding pages, normal milk contains a number of substances which are only pres- ent in small quantities and have only scientific interest, such * Wisconsin experiment station, twelfth report, p. 93. Comfositioyi of Milk and Its Products, 19 as the milk gases (carboaic acid, oxygen, nitrogen), citric acid, lecithin, cholesterin, urea, h3^poxanthin, lactochrome, etc. The percentage composition of cows' milk will be seen from the tables given in the Appendix. Tables are also given showing the average composition of milk products, like skim milk, butter milk, whey, cream, butter, cheese and condensed milk. 22. Colostrum milk. The liquid secreted directly after parturition is known as colostrum milk or biestings. It is a thick, yellowish, viscous liquid; its high content of albumen and ash is characteristic, and also its low content of milk sugar. Owing to the large quantit}' of albumen which colostrum contains it will coagulate on being heated to boiling. The secretion of the udder gradually changes from colostrum to normal milk in the course of four to five days; the milk is considered fit for direct consumption, or for the manufacture of cheese and butter when it does not coagu- lated on boiling, and is of normal appearance as regards color, taste, and other properties. For composition of colos- trum milk, see Appendix. 20 Testing Milk and lis Products. CHAPTER II SAMPLING MILK. 23. The butter fat in milk is not in solution, like sugar dissolved in water, but the minute fat globules or drops, in which form it occurs, are held in suspension in the milk serum (17). Being lighter than the serum, the fat globules have a tendency to rise to the surface of the milk. If, there- fore, a sample of milk is left undisturbed for even a short time, the upper layer will contain more fat than the lower portion. This fact should always be borne in mind when milk is sampled. The rapidity with which fat rises in milk can be easily demonstrated by leaving a quantity of sweet milk undisturbed in a cylinder or milk can for a few minutes, and testing separately the top, middle and bottom layer of this milk. Experiment. Fill the cylinder used for making the lactometer test (100) with milk, thoroughly mixed by pouring; measure a pip- etteful of milk immediately into test bottle A. Allow the milk in the cylinder to remain undisturbed for ten minutes, and then measure a pipetteful of milk from the top of that in the cylinder, into test bottle B. Next pour out most of the milk from the cylinder, and measure into test bottle C, a pipetteful of the last portion of the milk in the cylinder. After completing the tests of A, B, and C, in the usual manner (32), record the results of each test in the note book. 24 . The amount of mixing necessary to evenly distribute the constituents of milk throughout its mass, can also be Sampling Milk. 21 demonstrated by adding a few drops of cheese color to a quart of milk. The yellow streaks through the milk will be noticed until it has been poured several times from one vessel to another, when the milk will have a uniform pale yellow color. Stirring with a stick or a dipper will not produce an even mixture so quickly or so completely as pouring the milk a few times from one vessel to another, and in sampling milk for testing it should always be mixed by pouring just before the milk is measured into the bottle; if several tests are made of a sample, the milk should be poured before each sampling. 25. Partially churned milk. A second difficulty some- times met with in sampling whole milk arises from the fact that a part of the butter fat may be separated in the form of small batter granules by too zealous mixing, or by reckless shaking in preparing the sample for testing. This will happen most readily in case of milk from fresh cows, or with milk containing exceptionally large fat globules. When some of the butter granules are thus churned out, they very quickly rise to the surface of the milk after pouring, and cannot again be incorporated in the milk by simple mixing; it is, therefore, impossible to obtain a fair sample of such milk for testing, without taking special measures which will be explained in the following. The granules of butter may be so small as to pass into the pipette with the milk, and the quantity measured thus contain a fair proportion of them, but they will be found sticking to the inside of the pipette when this is emptied, and so fail to be carried into the test bottle with the milk. A similar partial churning of the milk will sometimes take place in the transportation cans. When such milk is received 22 Testing Milk and Its Products, at the factory, the butter granules are caught by the strainer cloth through which the milk is poured, and thus lost both to the factory and to the farmer. This separated fat cannot be put into the cream, or added to the granular butter, with- out running the risk of making mottled butter, and it will not enter into the sample of milk taken for testing purposes. When milk samples are sent in small bottles by mail or express, or carried to the place of testing, they very often arrive with lumps of butter floating in the milk or sticking to the glass. This churning of the milk can be easily pre- vented by filling the bottle or the can completely with milk. If there is no space left for the milk in which to splash around, the fat will not be churned out in transit. 26. Approximately accurate results may generally be obtained with a partially churned sample of milk, if a tea- spoonful of ether is added to it. After adding the ether, cork the bottle and shake it until the lumps of butter are dissolved in the ether. This ether solution of the butter will mix with the milk, and from the mixture a uniform sample may generally be taken without difficult}^ The dilution of milk by the ether introduces an error in the test- ing, and only the smallest quantity of ether necessar}^ to dissolve the lumps of butter should be used. If desired, a definite quantity of ether, say 5 or 10 per cent, of the volume of the sample of milk to be tested, may be added; in such cases the result of the test must be increased b}' the per cent, of ether added. Example. To a 4 oz. sample (120 cc.) of partially churned milk, 5 per cent., or 6 cc, of common ether are added: the mixture gives an average test of 4.2 per cent. The test must be increased by f§gX4.2=.21, and the original milk, therefore contained 4.2 + .21=4.41 per cent, of fat. Sam fling Milk. 23 Instead of adding ether to partiall}' churned samples, it has been suggested to warm the milk to 110° F. for a suffi- ciently long time to melt the butter granules; the sample is now shaken vigorously until a uniform mixture of milk and melted butter is obtained, and a pipetteful then drawn from the sample. 27. Sampling sour milk. When milk becomes sour, the casein is coagulated and the mechanical condition of the milk thereby changed so as to render a proper sampling very diffi- cult. The butter fat is not, however, changed in the process of souring: this has been shown by one of us in a series of tests which were measured from one sample of sweet milk into six test bottles. A test of the milk in one of these test bottles was made every month for six months, and ap- proximately the same amount of fat was obtained in the tests throughout the series, as was found originally in the milk when tested in a sweet condition.* If the milk is in condition to be sampled, the souring of it does not therefore interfere with its being tested b}^ the Babcock test, or with the accuracy of the results obtained. In order to facilitate the sampling of sour or loppered milk, some chemical is added which will re-dissolve the coagulated casein and produce a uniform mixture, that can be readil}' measured with a pipette. An}- alkali (powdered potash or soda, or liquid ammonia) will produce this effect. Only a very small quantit}^ of powdered alkali is necessary for this purpose. The complete action of the alkali on sour milk requires a little time, and the operator should not try to hasten the solution by adding too much alkali. An ex- *See Hoard's Dairyman, April 8, 1892. The same holds true for cream, as shown by Winton, (U. S. Dept. Agr., Div. of Chemistry, bull. 43, p. 112.) 24 Testing Milk and Its Products, cess of alkali will often cause such a violent action of the sulfuric acid on the milk to which the acid is added, (on ac- count of the heat generated or the presence of carbonates in the alkali) that the mixture will spurt out of the neck of the test bottle, when it is shaken in mixing the milk and the acid. When powdered alkali is added to the milk, it should be allowed to stand for a while, with frequent stirring, until the curd is all dissolved, and an even translucent liquid is obtained. Such milk may become dark colored by the action of the alkali, but this color does not interfere with the accuracy of the test. Instead of powdered soda or potash, these substances dis- solved in water (soda or potash lye), or strong ammonia water, may be used for the purpose of dissolving the coagu- lated casein in a sample of sour milk. In this case, a defi- nite proportion of alkali solution must, however, be taken, 5 per cent, of the volume of milk being usually sufficient, and the results obtained are increased accordingly. (See example cited on p. 22). 28. Sampling frozen milk. When milk freezes, it separates into two distinct portions: Milk crystals, largely made up of water, with a small admixture of fat and other solids, and a liquid portion, containing nearly all the solids of the milk. In sampling frozen milk it is there- fore essential that hotli the liquid and the frozen part be warmed and mixed thoroughly on the disappearance of the crystals, by pouring gently back and forth from one vessel into another; the sample is then taken and the test proceeded with in the ordinary manner (32). The Babcock Test. 25 CHAPTER III. THE BABCOCK TEST. 29. The Babcock test is founded on the fact that strong sulfuric acid will dissolve all non-fattj^ solid constituents of milk and other dairy products, and will set free the fat. This will separate on standing, but to effect a speedy and complete separa- tion, the bottles holding the mixture of milk and acid are placed in a centrifugal machine — a so-called tester^ and whirled for five min- utes; hot water is then added so as to bring the liquid fat into the graduated neck of the test bottles, and after a repeated whirling, the length of the col- umn of fat is read off, Fig. 4. The first Babcock tester made. showino" the per CCnt. of fat contained in the sample tested. 26 Testing Milk and Its Products. Sulfuric acid is preferable to other strong mineral acids for the purpose mentioned, on account of its affinity for water; when mixed with milk, the mixture heats greatly, thus keeping the fat liquid without the application of arti- ficial heat, and rendering possible a distinct reading of the column of fat brought into the neck of the test bottles. 30. So far as is known, any kind of milk can be tested by the Babcock test. Breed, period of lactation, quality or age of the milk is of no importance in using this method, so long as a fair sample of the milk can be secured. Sam- ples of milk or other dairy products, rich in solids, require a little more effort to perfect a thorough mixture with the acid than thin milk or other dairy products low in solids, like whey, which may be readily mixed with the acid. A — Directions for Making the Test. 31. The various steps in the manipulation of the Bab- cock test are discussed in the following pages; attention is drawn to the difficulties which the beginner and others may meet with in working the test, and the necessary precautions to be observed in order to obtain accurate and satisfactory results are explained in detail. The effort has been to treat the subject exhaustively, and from a practical point of view, so that persons as j^et unfamiliar with the test may turn to the pages of this book for help in any difficulties which they ma}' encounter in their work in this line. 32. Sampling. The sample to be tested is first mixed by pouring the milk from one vessel to another two or three times so that every portion thereof will contain a uniform amount of butter fat. The measuring pipette which has a capacity of 17.6 cubic centimeters, (see fig. 6), is filled with the milk immediately after the mixing is completed, by The Babcock Test. 27 dl sucking the milk into it until this rises a little above the mark around the stem of the pipette; the forefinger is then quickly placed over the end of the pipette before the milk runs down below the mark. By loosening a little the pressure of the finger on the end of the pipette, the milk is now al- lowed to run down until it just reaches the mark on the stem; the quantity of milk contained in the pipette will then, if this is cor- rectl}' made, be exactly 17.6 cc. The finger should be dry in meas- uring out the milk so that the de- livery of milk ma}^ be checked b}' gentle pressure on the upper end of the pipette. The point of the pipette is now placed in the neck of a Babcock test bottle (fig. 5) and the milk is allowed to flow slowly down the '^^^iS inside of the neck. Care must be taken that none of the milk meas- ured out is lost in this transfer. The portion of the milk remain in the point of the pipette is blown into the test bottle. The best and safest manner of holding the bottle and the pipette in this transfer is shown in fig. 7. Fig. 8 shows a position which should be avoided, since by holding Fig 5. Babcock Fig. 6 milk test bottle. 17.6 cc. pipette. 28 Testing Milk and Its Products. the bottle in this way, there is a danger that some of the milk may completely fill the neck of the bottle, and as a result, flow over the top of the neck. Fig. 7. The right way of emptying pipette into test bottle. 33. Adding acid. The acid cylinder, (fig. 9), holding 17.5 cc, is then filled to the mark with sulfuric acid of a specific gravity of 1,82-1.83. This amount of acid is care- fully poured into the test bottle containing the milk. In adding the acid, the test bottle is conveniently held at an The Babcock Test. 29 angle, (see fig. 7), so that the acid will follow the wall of the bottle and not run in a small stream into the center of the milk. By pouring the acid into the middle of the neck of the test bottle,, there is also a dan- ger of completel}^ filling this with acid, in which case the plug of acid formed will be pushed over the edge of the neck by the expansion of the air in the bottle, and may be spilled on the hands of the operator. The milk and the acid in the test bot- tle should be in two distinct layers, with- out anj^ black por- tion of partially mixed liquids be- tween them. Such pj^ g ^he wrong way of emptying pipette into a dark layer is often t^^t bottle. followed by an indistinct separation of the fat in the final reading. The cause of this is possibly that a partial mix- ture of acid and milk before the acid is diluted with the water of the milk may bring about too strong an action of the acid on the milk, and the fat in this small portion may 30 Testing Milk and lis Products. be slightly charred by the strong acid. The appearance of black flocculent matter in or below the column of fat which generally results, in either case renders a cor- rect measurement of fat difficult, and at times even impossible; if the black specks occur in the fat column itself, the readings are apt to be too high; if below it, the difficulty comes in deciding where the column of fat begins. 34. Mixing milk and acid. After ad- ding the acid, this is carefully mixed with the milk b}^ giving the test bottle a rotary motion. In doinoj this, care should be taken that none Fig. 9. 17 5 cc. => ' acid cylinder. of the liquid spurts into the neck of the test bottle. When once begun, the mixing should be continued until completed; a partial and interrupted mixing of the liquids will often cause more or less black material to separate with the fat when the test is finished. Clots of curd which separate at first by the action of the acid on the milk, must be entirely dissolved by per- sistent and careful shaking of the bottle. Beginners some- times fail to mix thoroughly the milk and the acid in the test bottle. As the acid is much heavier than the milk, a thin layer of it is apt to be left unnoticed at the bottom of the bottle, unless this is vigorously shaken toward the end of the operation. The mixture becomes hot by the action of the acid on the water in the milk and turns dark colored, owing to the effect of the strong sulfuric acid on the nitrogenous constituents and the sugar of the milk. Colostrum milk, or milk from fresh cows will form a violet colored mixture with the acid, owing to the action of The Babcock Test. 31 the latter on the albumen present in considerable quantities in such milk (22). When milk samples are preserved by means of potassium bichromate (172), and so much of this material has been added that the milk has a dark yellow or reddish color, the mix- ture of milk and acid will turn greenish black, and a com- plete solution is rendered extremely diflScult on account of the toughening effect of the bichromate on the precipitated casein. An indistinct separation of the fat is also some- times obtained in such samples, but this difficulty can gen- erally be overcome by using a little less than the regular quantity of acid. 35. Whirling bottles. After the milk and the acid have been completely mixed, the test bottle is at once placed in the centrifugal machine or tester and whirled for four to five minutes at a speed of 600 to 1,200 revolutions per min- ute, the proper speed being determined by the diameter of the tester (57). It is not absolutely necessary to whirl the test bottles in the centrifuge as soon as the milk and the acid are mixed; they may be left in this condition for any reasonable length of time (24 hours, if necessary) without the test be- ing spoiled. If left until the mixture becomes cold, the bottles should, however, be placed in warm water (of about 160° F.) for about 15 minutes before whirling. Four minutes at full speed is usually sufficient for the first whirling of the test bottles in the centrifuge; this will bring the fat to the surface of the liquid in the body of the bottle. 36. Adding water. Hot water is now added by means of a pipette, or some special device, until the bottles are filled up to the beginning of the neck. The bottles are 32 Testing Milk and Its Products. whirled again at full speed for one minute, and hot water added a second time, until the lower part of the column of fat comes within the scale on the neck of the test bottle, preferably to the 1 or 2 per cent, mark, so as to allow for the sinking of the column of fat, owing to the gradual cool- ing of the contents of the bottle. By dropping the water directly on the fat in the second filling, the column of fat will be washed free from light flocculent matter, which might otherwise be entangled therein and render the read- ing uncertain or even too high. A final whirling for one minute completes the separation of the fat. 37. Measuring the fat. The amount of fat in the neck of the bottle is measured by the scale or graduations on the neck. Each division of the scale represents two-tenths of one per cent, of fat, and the space which the fat occupies shows the per cent, of butter fat contained in the sample tested. The fat obtained should form a clear yellowish liquid dis- tinctly separated from the acid solution beneath it. There should be no black or white sediment in or below the col- umn of fat, and no bubbles or foam on its surface. The bottles should be kept warm until the readings are made, so that the column of fat will have a sharply defined upper and lower meniscus. The fat is measured from the lower line of separation be- tween the fat and the water, to the top of- the fat column, at the point 5, shown in the figure 10, the reading being thus taken from a to h^ and not to c or to d. Comparative grav- imetric analyses have shown that the readings obtained in this manner give correct results. While the lower line of the fat column is nearly straight, the upper one is curved, The Babcock Test. 33 a and errors in the reading of the column are therefore easily made, unless the preceding rule is observed. The readings should be made when the fat has a temperature of about 140° F., although the results obtained will not be appreciably affected if the temperature falls below 120°. The fat separated in the Babcock test solidifies at about 100^ F. No read- ings should be attempted if the fat is partly solidified, as it is impossible to get an accurate reading in this case.* A pair of dividers will be found convenient for measuring the fat, and the liability of error in reading is decreased by their use. The points of the dividers are placed at the up- per and lower limits of the fat column (from a to 6 in fig. 10). The dividers are now lowered, one point being placed at the zero mark of the scale, and the mark at which the other point touches the scale will then show the per cent, of fat in the sample tested. =— Fig. 10. Measuring the col- umn of fat in a Babcock test bottle. * The effect of differences in the temperature of the fat on the readings ob- tained will be seen from the following: If 110 and 1£0° F. be taken as the extreme temperatures, at which readings are made, this difference of 40° F. (22.3° C) would make a difference in the volume of the fat column obtained in case of 10 per cent, milk, of .00064 x 2 x 22.3 = .028544 cc. or .14 per cent., .0C064 being the expansion coetficient of pure butter fat per degree Centigrade between £0 and 100° C. {Zune, Analyse des Beurres, I, 87), and 2, the volume of the fat in cc. contained in 17.6 cc. of 10 per cent. milk. On 5 per cent, milk this extreme difference would therefore be about .07 per cent., or considerably less than one-tenth of one per cent. 3 34 Testing Milk and Its Products, B. — Discussion of the Details of the Babcock Test. 38. Although the manipulations of the Babcock test are few and comparatively simple, various difficulties ma}^ be met with in using it, particularl}' in the hands of beginners. The main points that have to be observed as to apparatus and testing materials in order to obtain correct and satisfac- tory results by this test will now be considered, and such suggestions and help offered, as has been found desirable from an extensive experience with a great variety of milk samples, apparatus, and accessories. 1. — Glassware. 39. Test bottles. When 17.6 cc, or 18 grams of milk, are measured into the Babcock test bottle, the scale on the neck of the bottles shows directh^ the percent of fat found in the milk. The scale is graduated from to 10 per cent. 10 per cent, of 18 grams is 1.8 grams. As the specific gravity of pure butter fat (i. e. its weight compared with that of an equal quantity of pure water) at the temperature at which the readings are made (about 120° F.), is 0.9, 1.8 grams of fat will occupy a volume of -g = 2 cubic centimeters. The space between the and 10 per cent, marks on the necks of the test bottles must therefore hold 2 cc, if correctly made. The scale is divided into 10 equal parts, each part repre- senting one per cent., and each of these are again sub-divid- ed into five equal parts. Each one of the latter divisions therefore represents two-tenths of one per cent, of fat when 17.6 cc. of milk is measured out. The small divisions are sufficiently far apart in most Babcock test bottles to make possible the estimation of one-tenth of one per cent, of fat in the samples tested. The Bahcock Test. o^ The figures and lines of the measuring scale become in- distinct by use; the black color may be restored by rubbing a soft lead pencil over the scale, or by the use of a piece of burnt cork after the scale has been rubbed with a little tallow. On wiping the necks with a cloth, or a piece of paper, the black color will show in the etchings of the glass, mak- ing these plainly visible. 40. The test bottles should have a capacity of about 50 cc, or less than two ounces; they should be made of well-annealed glass that will stand sudden changes of tempe- rature without breaking, and should be sufficiently heavy to withstand the maximum centrifugal force to which they are likely to be subjected in making tests. This force may not on the average be very far from 30.G5 lbs. (see 57) or the pressure exerted in whirling the bottles filled with milk and acid, in a centrifugal machine of IS inches diameter, at a speed of 800 revolutions per minute. Special forms of test bottles used in testing cream and skim milk are described under the heads of cream- and skim milk testinor. 41. Marking test bottles. Test bottles can now be bought with a small band or portion of their neck or body ground or "frosted," for numbering the bottles with a lead pencil. Bottles without this ground label can be roughened at any convenient spot by using a wet fine file to rub ofl^ the smooth surface of the glass. There is this objection to the latter method that unless carefully done, it is apt to weaken the bottles so that they will easily break, and to both meth- ods, that the lead pencil marks made on such ground labels are easily erased during the test, unless the bottles are carefully handled. Small strips of tin or copper with a 36 Testing Milk and Its Products. number stamped thereon are sometimes attached as a collar around the necks of the bottles. They are, however, easily lost, especially when the top of the bottle is slightly broken, or at any rate, are soon corroded so that the numbers can onl}^ be seen with difficulty. The best and most permanent label for test bottles is made by scratching a number with a marking diamond into the glass directlj^ above the scale on the neck of the bottles. In ordering an outfit, or test bottles alone, the operator may specify that the bottles are to be marked 1 to 2\^ or as many as are bought, and the dealer will then put the numbers on with a marking diamond. A careful record should be kept of the number of the bottle into which each particular sam- ple of milk is measured. Mistakes are often made when the operator trusts to his memory for locating the different bottles tested at the same time. 42. Cleaning test bottles. The fat in the neck of the test bottles must be liquid when these are cleaned. The bottle should be shaken in emptying the acid, in order to remove the white residue of sulfate of lime, etc., from the bottom ; if the acid is allowed to drain out of the bottle without this being shaken during the emptying, this residue will be found to stick very tenaciously to the bottom in the subsequent cleaning with water. A convenient method of emptying the test bottles Is shown in the illustration (fig. 11). After reading the fat, the bottles are taken from the tester and placed, neck down, in the \ inch holes of the board cover of a five-gallon stone- ware jar. An occasional shaking while the liquid is running from the bottles will rinse off the precipitate of sulfate of lime. A thorough rinsing with boiling hot water by means The Babcock Test. 37 of an apparatus, devised bj- one of us* (see fig. 12) is gen- erally sufficient to remove all grease and dirt, as well as acid solution, jili^WI^^^Ri from the inside of the bottles. When the bottles have been rinsed, the}' are placed in an inverted po- sition to drain, on a galvanized iron rack, as shown in fig. 13, where they are kept un- FiG. 11. v/aste acid jar. til needed. The outside of the bottles should occasionally be wiped clean and dr}'. 43. The amount of unseen fat that clings to glassware is generally not sufficient to be noticed in the results ob- tained in testing whole milk, but it plays an important part in testing samples of separator skim milk. It may be readily noticed by making a blank test with clean water in bottles which have been used for testing ordinary milk, and have been cleaned by simply draining the contents and rinsing once or twice with hot water; at the conclusion of the test the operator will often find that a few drops of fat — sometimes enough to condemn a separator — will collect in the neck of the bottles, although the water tested has not been near a separator. Boiling hot water will generally clean the grease from glassware for a time, but all test bottles should, in addition, * Farrington. 38 Testing Milk and Its Products. be given an occassional bath in some weak alkali, or other grease-dissolving solution. Persons doing considerable milk testing will find it of advantage to provide themselves with a small copper tank, which can be filled with a weak alkali-solution (figs. 14 and 15). After having been rinsed Fig. 12. Apparatus for cleaning test bottles. A, apparatus in position; the water flows from the reservoir through the iron pipe h into the inverted test bot- tle d through the brass tube c, screwed into the iron pipe. B shows construction of the rubber support on which the top of the test bottle rests; /, draining sink. with hot water, the test bottles are placed in the hot solu- tion kept in the tank, where they may be left completely cov- ered with the liquid. If the tank is provided with a small The Bahcock Test. 39 faucet at the bottom, the liquid can be drawn otf when the test bottles are wanted. A tablespoonful of Savogran to about two gallons of water will make a very satisfactory cleaning solution; sal soda, Gold Dust, Lewis lye or Babbitt's potash are equally efficient. The cleansing properties of solutions of an}^ of these substances are increased by warm- ing the liquid. The test bottles must be rinsed twice with hot water after they are taken from this bath. F^G. 13. Draining-rack for test bottles. The black stains that sometimes stick to the inside of test bottles after prolonged use, can be removed with a little muriatic acid. 44. Pipette. The difference in the weights of various samples of normal milk generally falls within comparatively narrow limits: if a given quantit}^ of water weighs 1 pound, the same quantity of the usual grades of normal milk will weigh from 1.029 to 1.033 pounds, or on the average 1.03 lbs. 18 grams* of water measures 18 cc; 18 grams of * Cubic centimeters (abbreviated: ce.) are the standard used for measuring volume in the metric system, similar to the quart or pint measure in our ordinary system of measures. 1 quart is equal to a little more than 1000 cubic centimeters. Tn the same way, grams represent weight, like pounds and ounces. 1 cc. of water at 4° Centigrade weighs 1 gram. 1000 grams (= 1 kilogram) is equal to 2.2 lbs. Avoir. (See Appendix, for Comparisons of Metric and Customary Weights and Measures). 40 Testing Milk and Its P7'oducts. milk will therefore take up a smaller volume (measure less) than 18 cc, viz: 18 divided by 1.03, which is very nearly 17.5. This is the quantity of milk taken in the Bab- cock test. A certain amount of milk will adhere to the walls of the pipette when it is emptied, and this thin film has been found to weigh about one-tenth of a gram; conse- quently 17.6 cc. has been adopted as the capacity of the pip- ette used for deliv- ering 18 grams of Fig. 14. Tank for cleaning test bottles. milk. For convenience in measuring the milk, the shape of the pipette is of importance. The mark on the stem should be two inches or more from the upper end of the pipette. The lower part should be small enough to fit loosely into the neck of the test bottle, and not contracted to a fine hole at the point; the point should be large enough to allow a quick empt3ing of the ' -r,. -, ox Fig. 16. Pip- pipette. (Fig. 16). ettepoints- 45. Fool pipettes. Soon after the Babcock test A, proper con- V , , 11 1 J • struction; B, began to be generally used at creameries as a means , irable of paying for the milk, a creamery supply house put construction. The Bahcock Test. 41 on the market a 20 cc. milk-measuring pipette, which was claimed to show the exact butter value of milk, instead of its content of butter fat, as in the case in using the ordinary 17.6 cc. pipette. A 20 cc. pipette will deliver 2.4 cc, or 13.6 per cent, more milk than a 17.6 cc. pipette, and it follows that the results obtained by measuring out milk for Babcock tests with these pipettes will be about 13.6 per cent, too high. In con- sidering the subject of Overrun it is noted that the excess of butter yield over the amount of fat contained in a certain quantity of milk will range from about 10 to 15 per cent., or on the average 12-13 per cent. The 20 cc. pipntes may, therefore, give approxi- FiG. 15. Rack for holding test bottles in tank shown in fig. 14. mately the yield of butter obtained from a quantity of milk, but as will be seen, this yield is variable, according to the skill of the butter maker, and according to conditions beyond his control, and cannot therefore be used as a standard in the same manner as the fat content of the milk. Similar 22 cc. pipettes were also sent out. These pipettes created a great deal of confusion during the short time they w^ere on the market, and were popularly termed ^'fool pipettes," as the tests obtained bj' them did not give, what they professed to do, an accurate and definite measure of the but- ter-producing qualities of different lots of milk. It is not known that any of these pipettes are on the market at the present time. 42 Testing Milk and Its Products. 46. Acid measures. A 17.5 cc. glass cylinder for measuring the acid is generally included in the outfit, when a Babcock tester is bought. This cylinder answers every purpose if only occasional tests are made; the acid is poured into the cylinder from the acid bottle, as needed, or a quan- tity of acid sufficient for the number of test bottles to be whirled at a time, is poured into a small glass beaker, pro- vided with a lip, or into a porcelain pitcher; these may be more easily handled than the heavy acid bottle, and the acid measure is then filled from such a vessel. Where a considerable number of tests are made regularly, the acid can be measured into the test bottles faster and^ with less danger of spilling, by using some one of the many devices proposed for this purpose. There is some objection to nearly all of these appliances, automatic pipettes, buret- tes, etc., although they will often give good satisfaction for a time while new. Sulfuric acid is so corrosive, and opera- tors as a rule take such poor care of such apparatus, that it is a very difficult matter to design a form which will remain in good wording order for any length of time. Automatic pipettes attached to acid bottles or reservoirs, to prove satisfactory, must be made entirely of glass, and strong, of simple construction, tightly closed and quickly operated. 47. The Swedish Acid Bottle answers these requirements- better than any other device for this purpose known to the writers at the present time. Its use is easily understood (see fig. 17) ; it gives good satisfaction if the hole in the glass stop cock through which the acid passes has a diameter of at least one- eighth of an inch, as is generally the case. We have used or inspected some half a dozen other devices, which have been placed on the market by various dealers The Bahcock Test. 43 for delivering the acid, but cannot recommend them for use in factories, or outside of chemical laboratories. 48. Instead of measuring out the acid, Bartlett* re- cently suggested adding it directly to the milk in the test bottles, till the mixture rises to a mark on the body of the bottle at the point where this will hold 37.5 cc, i. e., the total volume of milk and acid (74). This method of adding the acid is in the line of simplicity and may in time become generally adopted. The marks should, however, be put on by the man- ufacturers, as the operator in attemping to do so will be apt to weaken or break the bottles. Fig. 17. Swedish acid Calibration of glassware. 1. — Test bottle; the side-tube is made to hold 17.5 cc. of BOTTLES. The Babcock milk test bot- ^cid. ties are so constructed that the scale or graduation of the neck measures a volume of 2 cubic centimeters, between the zero and the 10 per cent, marks (39). The correctness of the graduation may be easily ascertained by one of the following methods: 49. Calibration with mercury. 27.18 grams of metallic mercury are weighed into the perfectly clean and dry test bottle; since the specific gravity of mercury is 13.59, double this quantity will occupy a volume of exactly 2 cubic centi- meters (44*). The neck of the test bottle is then closed with a small smooth and soft cork, or a wad of absorbent cotton, cut off square at one end, the stopper being pressed down to the first line of the graduation. The bottle is now Maine experiment station, bull. No. 31. 44 Testing Milk and Its Products. inverted so that the mercury will run into its neck. If the total space included between the and 10 marks is just filled with the two cubic centimeters of mercury, the grad- uation is correct. Bottles, the whole length of the scale of which vary more than two-tenths of one per cent, are inac- curate, and should not be used. The mercury may be conveniently transferred from one test bottle to another, by means of a thin rubber tube which is slipped over the end of the necks of both bottles, and one weighing of mercury will thus suffice for a number of cali- brations. In transferring the mercury, care must be taken that none of it is lost, and that small drops of mercury are not left sticking to the walls of the bottle emptied. A sharp tap on the bottle with a lead pencil will help to remove minute drops of mercury from the inside of it. Unless the bottles to be calibrated are thoroughly cleaned and dry, it is impossible to transfer all the mercury from one bottle to another. After several calibrations have been made, the mercury should be weighed again in order to make certain that none has been lost by the various manipulations. The scale, fig. 28, shown in (84) is sufficiently delicate for making these weighings. 50. Cleaning mercury. Even with the best of care, mer- cury used for calibration of glassware will gradually become dirty so that it will not flow freely over a clean surface of glass. It may be cleaned from mechanical impurities, dust, films of grease, water, etc., by filtration through heavy filter paper. This is folded the usual way, placed in an ordinary glass funnel and its point perforated with a couple of pin holes. The mercury will pass through in fine streams, leav- ing the impurities on the filter paper. Mercury may be The Bahcock Test. 45 freed from foreign metals, zinc, lead, etc., sometimes noticed as a grayish, thin film on its surface, by leaving it in contact with common nitric acid for a number of hours; the mercury is best placed in a shallow porcelain or granite ware dish and the nitric acid poured over it, the dish being cov- ered to keep out dust. The acid solution is then carefully poured off and the mercury washed with water; the latter is in turn poured off, and the last traces of water absorbed by means of heavy clean filter paper. The mercury to be used for calibration of glassware should be kept in a strong bottle, closed by an ordinary stopper. In handling mercury, care must be taken not to spill any portion of it; finger-rings should be removed when calibra- tions with mercury are to be made. Mercury forms the most satisfactory and accurate ma- terial for calibration of test bottles, on account of its heavy weight and the ease with which it may be manipulated. Equally correct results may, however, with proper care be obtained by using one of the following methods: Calibration with water. This may be done by means of a delicate pipette or burette, or by weighing in a somewhat similar manner, as explained in case of calibration with mercury. 51. Si, Measurivg the water. Fill the test bottle with water to the zero mark of the scale; remove any surplus water and dry the inside of the neck with a piece of filter paper or clean blotting paper; then measure into the bottle 2 cc. of water from an accurate pipette or a burette, divided to 20 of a cubic centimeter. If the graduation is correct, 2 cc. will fill the neck exactly to the 10 per cent, mark of the scale. 46 Testing Milk and Its Products. 52. b, Weighing the water. Fill the bottle with water to the zero mark of the scale and remove any surplus water in the neck, as before. Weigh the bottle with the water contained therein. Now fill the neck with water to the 10 per cent, mark, and weigh again. The difference be- tween these weights should be 2 grams. In all cases when calibrations are to be made, the test bottles, or other glassware to be calibrated, must be thoroughly cleaned with strong sulfuric acid, or soda lye, and washed repeatedl}^ with pure water, and dried. Grlass- ware is not clean unless water will run freely over its sur- face, without leaving any adhering drops. 53. Intermediate divisions. The space between and 10 on the scale of the Babcock test bottle is divided into 50 divisions, each five of which, as previously shown (39) representing 1 per cent. Since these intermediate divisions are generally paade with a dividing machine, they are as a rule correct, but it has happened that the divisions have been inaccurately placed, although the space between and 10 was correct. The accuracy of the intermediate divisions can be ascertained by sliding along the scale a strip of paper upon which has been marked the space occupied by one per cent., and compare the space with those of each per cent, on the scale. 54. 2. — Pipette and acid cylinder. The pipette and the acid cylinder used in the Babcock test ma}^ be calibrated by any of the methods already given. Sufficiently accurate results are obtained by weighing the quantity of water which each of these pieces of apparatus will hold, viz: 17.6 grams and 17.5 grams, respectively. The necessity of pre- vious thorough cleaning of the glassware is evident from The Babcock Test. 47 what has been said in the preceding. The pipette and the acid measure ma}' be weighed empt}' and then again when filled with pure water to the mark, or the measureful of water may be emptied into a small weighed vessel, and this weighed a second time, In either case the weight of the water contained in the pipette or the acid measure is ob- tained by difference.* Calibrations of the acid cylinder are generally not called for, except as a laborator}- exercise, since small variations in the amount of acid measured out do not affect the accu- racy of the test. 2. — Centrifugal Machines. 55. The capacity of the testing machine to be selected should be governed by the number of tests which are likely to be made at one time. For factory purposes a twentj^-five to thirty bottle tester is generally large enough, even if toward a hundred samples of milk are to be tested at a time. The operator can use his time more economically in running a machine of this size than one holding fift}' or sixt}' bottles; the work of filling or cleaning the bottles and measuring the fat can be done while the centrifuge is run- ning if a double suppl}^ of bottles are at hand. Large test- ers require more power than smaller ones, and when sixt}' tests are completed, man}' of the bottles will cool, and the fat column crystalize, before the operator has time to read them, unless special precautions are taken for keeping the bottles warm. * 1 cubic centimeter of distilled water weighs 1 gram, when weighed in a vacuum at the temperature of the maximum density of water (4° C); for the purposes of calibration of glassware used in the Babcock test, sufficiently accurate results are, however, obtained by weighing the water in the air and at a low room temperature (60° F.) 48 Testing Milk mid Its Products. 56. The tester should be securely fastened to a solid foundation and set so that the revolving wheel is level. The latter must be carefully balanced in order that the tester may run smoothly at full speed when empty. A machine that trembles and shakes when in motion is neither satis- factory nor safe, and the results obtained are apt to be too low. High -standing machines are more apt to cause trou- ble in this respect than low machines, and should therefore be subjected to a severe test before they are accepted. If all the sockets are not filled with bottles when a test is to be made, the bottles must be placed diametrically op- posite one another so that the machine will be balanced when run. The bearings should be kept cleaned and oiled with as much care as the bearings of a cream separator. The cover of the machine should always be kept on while the bottles are whirled, and should not be removed until the machine stops; the cover should be tight-fitting and may be fastened with hooks soldered on the side of the machine; test bottles have sometimes been broken while the machine has been running at full speed, and every possible precau- tion should be taken to protect the operator from any dan- ger from spilled acid or broken glass. 57. Speed required for the complete separation of the fat. There is a definite relation between the diameter of the Babcock testers and the speed required for a perfect separation of the fat. In the preliminary work with the Babcock test the inventor found that with the machine used, the wheel of which had a diameter of eighteen inches, it was necessary to turn the crank, so as to give the test bot- tles seven to eight hundred revolutions per minute, in order to effect a maximum separation of fat; later work has The Babcock Test. 49 shown that this speed is ample. Taking therefore this as a standard, the centrifugal force to which the con- tents of the test bottles are subjected when supported on an eighteen inch wheel and turned 800 revolutions per minute, can be calculated as follows : The centrifugal force, F, acting on the bottles is expressed by the formula F= ^-^^ a) 32.2r ^ in which w = the weight of the bottle with contents, in pounds ; Y = the velocity, in feet per second, and r = the radius of the wheel, in feet. When the wheel is turned 800 times a minute, a bottle supported on its rim will travel 2 TTr X '6%'' = 2X 3.1415 X A X «6'd' = 62.83 feet per second. The weight of a bottle, with milk and acid, is very near 3 ounces, or /e of a pound. Substituting these values for V and w, gives p^ h X 62.83^ =30.65 lbs. 32.2 X h The bottles are therefore, under conditions given, subjected to a pressure of 30.65 lbs. In order to calculate the speed required for obtaining this force in case of machines of other diameters, the value of v in formula (I) is found from ^ ^/ 32.2 F X r w^ Substituting the values for F and w, (11) v== |/32.2 X 30.65r _ ^^^^^ h In this equation the values r = 5, 6, 7, . . . 12 inches are substituted in each case (i\, ^^, ^^, • • • if feet), and the velo- city in feet per second then found at which the bottles are whirled when placed in wheels of diameters 10 to 24 inches, and subjected in each case to a centrifugal force of 30,65 lbs. As the number of 60 r revolutions per minute = -75 , v being as before the velo- 4 50 Testing Milk and Its Products. city in feet per second, and r the radius of the wheel, the speed at which the wheel must be turned, is found by substituting for v the values obtained in the preceding calculations in case of wheels of different diameters. The results of these calculations are given in the following table: Diameter )/ wlieel, d Velocity in feet per second, v. Number of revolutions of whet I, per minute, 10 46.84 1074 12 51.31 980 14 55.43 909 16 59.26 848 18 62.84 800 20 66.24 759 22 69.47 724 24 72.56 693 These figures show that a tester, for instance. 25 inches in diameter, requires less that 700 revolutions per minute for a per- fect separation of the fat in Babcock bottles, while a ten-inch tes- ter must have a speed of nearly 1100 revolutions, in order to obtain the same result. 58. To find the number of turns of the handle corres- ponding to the number of revolutions made by the wheel, the handle is given one full turn, and the number of times which a certain point or part of the wheel revolves, is noted. If the wheel has a diameter of 20 inches, and revolves 12 times for one turn of the handle, the latter should be turned -^2^=63 (see table), or about once every second, in order to effect a maximum separation of fat. By counting the num- ber of revolutions, watch in hand, and consulting the pre- ceding table, the operator will soon note the speed which must be maintained in case of his particular machine. It is vitally important that the required speed is always kept up; if through carelessness, worn-out or dry bearings, slipping belts, etc., the speed is slackened, the result will come too The Bahcock Test, 51 low; it may be a few tenths, or even more than one percent. Care as to this point is so much the more essential as the results obtained by too slow whirling may seem to be all right, a clear separation of fat being often obtained even when the fat is not completely separated. 59. Ascertaining the necessary speed of testers. In buying a tester the operator should first of all satisfy himself at what speed the machine must be run to give correct results ; the preceding table will serve as a guide on this point. He should measure out a dozen tests of the same sample of milk, and whirl half the number at the speed required for machines of the diameter of his tester. Whirl the other half at a somewhat higher speed. If the averages of the two sets of determinations are the same, within the probable error of the test (sa}^, less than one-tenth of one percent.) the first whirling was sufficient, as it is believed will gener- ally be the case. If the second set of determinations come higher than the first set, the first whirling was too slow, and a new series of tests of the same sample of milk should be made to ascertain that the second whirling was ample. This method will test not only the speed required with the particular machine at hand, but will also serve to indi- cate the correctness of the calibration of the bottles. A large number of tests of the same sample of milk made as directed (pouring the milk once or twice previous to taking out a pipetteful for each test) should not vary more than three-tenths of one percent, at the outside, and in the hands of a skilled operator will come within two-tenths of one per- cent. If greater discrepancies occur, the test bottles giving too high or too low results should be further examined, and <5alibrated according to the directions already given (49). 52 Testing; Milk and Its Products. 60. Hand testers. When only a few tests are made at a time, and at irregular intervals, as in case of dairymen who test single cows in their herds, a small hand tester answers every purpose. These may be had in sizes from two to twelve bottles. In selecting a particular make of tester the dairymen has the choice of a large number of difierent kinds of machines. It is a source of regret that most of the machines placed on the market for this purpose in the past have been so cheaply and poorly constructed as to prove very unsatisfactory after having been in use for a time. The sharp competition between manufacturers of dairy supplies, and the clamor of dairymen for something cheap, fully account for this condition of affairs. This ap- plies especially to the many machines made with belts or friction application of power. The main objection to these machines is the uncertainty of the speed obtained, when they have been in use for some time, and the belt or friction appliance begins to slip. Hand testers made with cog gear wheels are more to be depended on for giving the necessary speed than belt or friction machines; they are generally noisy, but the bottles are always whirled at the speed which the number of turns made by the crank would indicate. Fig. 18. " No-tin '' test. The Babcock Test. 53 The " No-tin " test (see fig. 18) is, in the opinion of the authors, worthy of special mention among the hand testers made at the present time; it is a six-bottle geared machine, durable of construction, and runs smoothly and without noise. 61. Power testers. For factory purposes, steam tur- bine machines (fig. 19) are most satisfactory when well made and well cared for. They should always be provided with a speed indicator and steam gauge, both for the purpose of knowing that sufficient speed is attained, and also to pre- vent what may be serious accidents from a general smash- up, if the turbine ''runs wild" by turning on too much steam. The revolving wheel of the tester should be made of wrought or malleable iron, or of wire, so that it will not be broken by the centrifu- gal force, thus avoiding seri- ous accidents. The swinging pockets which hold the test bottles in some machines, should be so made that the bottles will not strike the center of the revolving frame when in a horizontal position. Tests have often been lost by the end of the neck catching at the center, the bottles thus failing to take an upright position when the whirling stops. Fig. 19. Type of Babcock steam turbine testers. 54 Testing Milk and Its Products. 3. — Sulfuric Acid. 62. The sulfuric acid to be used in the Babcock test should have a specific gravity of 1.82-1.83.^ The com- mercial oil of vitriol which can be bought for about 2 cents a pound in carboy lots, is commonly used. One pound of acid is sufficient for fifteen tests. The acid should be kept in stoppered glass bottles, preferably glass or rub- ber stoppered ones, since a cork stopper is soon dissolved by the acid and rendered useless. If the bottle is left uur corked, the acid will absorb moisture from the air and will after a time become too weak for use in this test. Lead is the only common metal which is not dissolved by strong sulfuric acid; where considerable milk testing is done, it is therefore desirable to provide a table covered with sheet lead on which the acid may be handled. 63. The acid dissolves iron, tin, wood and cloth, and burns the skin. If acid is accidently spilled, plenty of water should be used at once to wash it off. Ashes, potash, soda, and ammonia neutralize the action of the acid, and a weak solution of any one of these alkalies can be used after the acid has been washed off with water. The red color caused by the action of the acid on clothing can be removed by wetting the spot with weak ammonia water; the ammonia must, however, b*e applied while the stain is fresh, and is in its turn washed off with water, 64. Testing the strength of the acid. The strength of the acid can be easily tested by the use of such a balance as shown in fig. 27 (84). A dry test bottle is weighed, and * A specific gravity of 1.82 means that any given volume of the acid weighs 1.82 times as much as the same volume of water at the same temperature (see also under Lactometer, 94), The Bahcock Test. 55 then filled with acid exactly to the zero mark, or to any other particular line of the scale. It is then again weighed accurate- ly; the difference between these two weights will give the weight of the acid in the bottle. Next empty the bottle and rinse it thoroughly with water (until the water has no longer an acid taste); fill the bottle with water to the same line as be- fore and weigh; the difference between this weight and that of the empty bottle gives the weight of the same volume of water as that of the acid weighed. Divide the weight ot the acid by the weight of the water; the quotient gives the spe- cific gravity of the acid. If this is between 1.82 and 1.83, the strength of the acid is correct. The outside of the test bottle should always be wiped dry before the liquids are weighed in it. Unless great care is taken in measuring out the acid and the water, and in weighing both these and the test bottle, the results obtained will not be trustworthy. 65. Too strong acid can sometimes be successfully used by taking less than the required amount of each test, e. g. about 15 cc. Operators are warned against reducing the strength of the acid by adding water to it, as accidents are very apt to occur when this is done. A too strong acid can, if desired, be weakened by si m plying leaving the bottle which holds it, uncorked for a time; or by pouring the acid into a bottle containing a small quantity of water; in the latter case, the first portions of acid should be added care- fully, a little at a time, shaking the bottle after each addi- tion, so as not to cause it to break from the great heat evolved in mixing the acid and the water. Never dilute acid hy pouring water into it. 66. If the acid is too weak, correct results may some- times be obtained by using more than the specified quantity, 56 Testing Milk and Its Products. say 20 cc. If a good test is not obtained with this quantity of acid, a new lot must be secured, as its specific gravity in such a case must be below 1.82. The observing operator will soon be able to judge of the strength of the acid by its action on milk in mixing the two liquids in the Babcock test bottles; it is indeed remarkable what slight differences in the specific gravity of the acid will make themselves apparent in working the test, as regards the rapidity with which both the curdled milk is dissolved, and the mixture of acid and milk turns black. 67. The relation between the strength of sulfuric acid and its specific gravity will be seen from the following table: Strength of Sulfuric Acid {Lunge and Isler, 1890). Sulfuric Acid 97 per cent. 96 95 94 93 92 91 90 89 88 Specific Gravity (15° C,u-aier4° C). 1.841 1.840 1.839 1.837 1.834 1.830 1.825 1.820 1.815 1.808 It will be noticed that the sulfuric acid to be used in the Babcock test should contain 90 to 92 per cent, of acid (H2SO4); slightly weaker or stronger acid than this may, as previously stated, be used by adjusting the quantity of acid taken for each test to the strength of the acid, but success- ful tests cannot be made with acid weaker than 89 per cent, or stronger than 95 per cent. H The Babcock Test. 57 68. The Swedish acid tester (fig. 20) is a small hy- drometer, intended to show whether the acid to be used in the Babcock test is of the correct strength. We have examined a number of these testers, and have found them practicall}^ useless for the purpose in- tended. The reason for this lies in insufficient •ij sensitiveness in the instrument; while the testers 1 examined were found to sink to the line marked Correct on the scale, when lowered into sulfuric acid of a specific gravity of 1.83, they would sink to a point much nearer the same mark when lowered into either too strong or too weak acid, than to the lines marked Too strong or Too weak^ respectively. An examination of the proportionate parts of the testers shows that such must be the case: The total weight of the testers varies between 7 and 8 grams ; the diameter of the stem is nearly 5 millimeters, and the dis- tance between the two lines marked Too strong and Too weak is 13.5 millimeter. A good hydrometer, such as used in chemical laboratories for determining the spe- Swedisli ^^^^ gravity of liquids of 1.8 to 2.0, weighs about 75 acid tester, grams; the diameter of the stem is 6 mm., and the dis- tance between the 1.82 and l.Si marks on the scale is 15.5 mm. ; these limits may be taken to represent too weak and too strong acid, respectively. Comparying now this hydrometer with the Swedish tester, the weight of the for- mer would make it ten times as sensitive as the latter, if the size of the stem was the same in either case; as it is, the tester has the advantage in point of thinness of stem (see 94), as the volumes of the same lengths of stem in the two instruments ate as the square of their diameters, i. e., as 25:36. This means that the Swedish tes- ters are onh^ =: _ as sensitive as the hydrometer, or a •^ 10 X 25 7 ^ •diiference on the scale of the latter amounting to 15.5 mm. (see ^8 Testing Milk and Its Products. above), would represent only 2.2 mm., on the scale of the Swedish tester. The line marked Too strong must therefore be only 1.1 mm. (2V of an inch) below the Correct line; and that marked Too^ weak the same distance above the line. But this is too small a distance to be differentiated by persons unfamiliar' with the use of delicate hydrometers, especially since the meniscus of the liquid formed around the stem of the tester renders an accurate reading somewhat difficult. The Swedish acid tester can be made more delicate by changes in one or two directions: by making the bulb larger, thus necessi- tating an increase in weight, or by making the stem thinner. By way of comparison it maybe stated that the hydrometers used for determining the specific gravity of the ether-fat solution in Sox- hlet's areometric method of milk analysis have a stem only 2 mm. in diameter, and the distance of the scale between .765 and .745 is 70 mm., or 2% inches. Even if these testers are changed as suggested, their practicabil- ity still remains an open question. The action of sulfuric acid of different strength is very characteristic (66), and in the hands of experienced operators, is as delicate an index to the strength of the acid as can be desired, making rather unnecessary a separate instrument for ascertaining the correctness of the strength of the acid used in milk testing. 69. The color of the fat column an index to the strength of the acid used. The strength of the acid is indicated to a certain extent by the color of the fat which separates in the neck of the test bottle when milk is tested. If the directions given for making the test are care- fully followed, the fat separated out will be of a golden yellow color. If the fat is light colored or whitish, it gen- erally indicates that the acid is too weak, and a dark colored fat, with a layer of black material beneath it, shows that the acid is too strong. The strength of the acid used in the test is not sufficient at ordinary temperatures of testing to appreciably dissolve The Babcock Test, 59 the fat, but a variation in the strength of the acid or in the temperature of the milk influences the intensity of the ac- tion of the acid on the fat, as shown in the color of the fat obtained. The following experiment shows the relation between the strength of the acid, the temperature of the milk, and the color of the fat. First: — From a sample of milk measure the usual quantity for testing into each of three bottles, A, B and C. Place A in ice water, and C in warm water, leaving bottle B at the ordinary temperature. After the bottles have been left for twenty minutes under these conditions, add the normal quantity of acid to each and proceed with the test in the ordinary manner. Second: — Measure some of the same milk into three other bot- tles, D, E and F. Into test bottle D pour the usual amount of rather weak acid; add the same amount of acid of normal strength (1 .82-1.93) to bottle E. and add 17.5 cc. of a still stronger acid, (concentrated sulfuric acid, sp. gr. 1.84) in test bottle F; complete these tests in the usual way. On the completion of the preceding six tests the operator will notice that the fat in the necks of test bottles A (cold milk) and D (weak add) is much lighter colored than that in C {warm milk) and F {strong acid), and that the color of the fat in B {normal temperature) and E {normal acid) is somewhere between that of these two series. 70. Influence of temperature on the separation of fat. The intensity of the action of the sulfuric acid on the milk is influenced by the temperature of either liquid; the higher the temperature, the more intense will be the action of the acid on the solids of the milk. It may be noticed that acid from the same carboy will act differently on milk in summer than in winter time, if the acid and the milk are not brought to the same temperature before testing during 6o Testing Milk and Its Products, both seasons, by cooling or heating, respectively. The temperature of the liquids may be as low as 40° F. in winter and as high as 80° F. in summer. This difference of forty degrees will often have considerable influence on the clear- ness of the fat separated, showing white curdy substances, and a light colored fat in winter, and black flocculent specks, with a dark colored column of fat in summer. Both these defects can be avoided when the acid is of the proper strength, by bringing the temperature of the milk and the acid to about 70° F. before the milk is tested. The operator should be particularly cautious against over- heating either milk or acid; so intense an action may be caused thereby as to force the hot acid out of the neck of the test bottle when it is added to the milk, thus spoiling the test and possibly causing an accident. 4. — Water to be Used in the Babcock Test. 71, Rain water, condensed steam, or soft water should be used for the purpose of bringing the fat into the neck of the test bottles. The surface of the fat column will then usually be clear and distinct. The foam or bubbles that sometimes obscure the upper line (meniscus) of the fat, making the point indistinct from which to measure it, is gen- erally caused by the action of the acid on the carbonates in hard water. The carbonic acid gas liberated from hard water by the sulfuric acid is more or less held by the viscid fat and produces a layer of foam on its surface. If clean soft water cannot be obtained for this purpose, hard water may be used by adding a few drops of sulfuric acid to the water before it is heated, thus causing the carbonic acid to be driven out of it. By simply boiling, most hard waters will be rendered soft, and adapted to use in the Babcock The Babcock Test. 6i test, as the carbonates which cause this foaming are thereby precipitated. If the test has been completed, and a layer of foam ap- pears over the fat, it may be removed by adding a drop or two of alcohol which will destroy the foam. If this is done, the fat column should be read as rapidly as possible after the alcohol is added, and before the latter unites with the fat and increases its volume. Fig. 21. The Russian test. 72. Reservoir for water. When only a few tests are made at one time, the hot water can be added with the 17.6 cc. pipette. If many tests are made, the water is more con- 62 Testing- Milk and lis Products. veniently and quickly filled into the test bottles by drawing it from a small copper reservoir or tin pail suspended over the testing machine.* The flow of water through a rubber tube connected with the reservoir, is regulated by means of a pinch cock. The water must be hot when added to the test bottles so as to keep the fat in a melted condition until the readings are taken. The use of zinc or steel oilers, or perfection oil cans has been suggested, as a handy and rapid method of adding hot water to the test bottles. 5. — Modifications of the Babcock Test. 73. The Russian milk test. The same chemical and mechanical principles applied in the regular Babcock test, yy are used in the Russian milk test, ex- // cept that in this case the machine in // which the bottles are whirled, and // the bottles themselves, are so con- tf^ // structed that the latter can be filled with hot water while the machine is running at full speed, thus saving time and trouble incident to the stopping of the tester and filling the bottles by means of a pipette. The milk-measuring pipette (fig. 22) and the acid measure used in the Russian test are one-half the ordinary size, and the test bottles are made in two pieces, with a detachable narrow graduated stem (see fig. 23). The machine is substantially made of cast iron; it is provided with a very satisfactory speed indi- cator which shows at any time the number of Fig. 22. Pip. revolutions at which the bottles are being turned. ette usedin the rrii • '^^ j. >• t_ ^u Russian test. ^^^ accompanying illustrations show the appa- * Ordinary tinware rusts very soon when water is left standing in it, and cop- per reservoirs are therefore more economical. The Babcock Test. 63 n ratus used in the Russian test. In our experience with this machine there seemed to be a tendency toward too low results. 74. Bartlett's modification. Bartlett* proposed a modification of the method of procedure in the Babcock test, which aims to simplif}^ the manipu- lations. 20 cc. of acid are added, instead of 17.5 cc, and the bottles filled with the milk-acid mixture are left standing for not less than five minutes and then filled to within the scale, with hot water; the bottles are then whirled for five minutes at the regular rate (48). In the experience of the authors the modification cannot always be depended upon to give satisfactory results. It was tried by each of the one hundred students in the Wisconsin Dairy School during 1896-'97; while some of these operators obtained a clear separation of fat, and results that compared favorably with those made by the regular Babcock test, others failed to obtain correct results with the method as modified. It is not known that the modification has proved superior to, or taken the place of the regular Babcock test to any extent, * Maine experiment station, bull. No. 31, (S. S.) Fig. 23. Test bottles used in the Russian test. 64 Testing Milk and Its Products. CHAPTER IV. CREAM TESTING. 75. Cream may be tested by the Babcock test in the same manner as milk, and the results obtained are accurate Fig. 24. Students'testing dairy products ^W=*^^^--=«^^ when the necessary care has been taken in sampling the cream and in measuring the fat. The composition of cream varies greatly according to the process of creaming, tem- perature of milk during the creaming, quality and composi- tion of the milk to be creamed, etc. The cream usually met with in the creameries or on the market will contain from Cream Testing. 65 15 to 30 per cent, of fat; during late years the practice in man}^ creameries has been to churn a very rich cream, <3ontaining from 30 to 40 per cent, of fat. Such cream is, however, rarely met with, and cream containing 25 per cent, of fat may be considered of average composition for a good quality of cream. ^ If 18 grams of such cream is measured into an ordinary Babcock test bottle, it follows that there will be 18 X .25 = 4.5 grams (or 1^^^ = 5 cc.) of pure butter fat in the bottle. It is shown, however, (p. 34), that the space from to 10 in the neck of these bottles holds exactly 2 cc. The neck of the milk test bottles will not therefore be large enough to show the per cent, of fat in a sample of cream if 18 grams are taken for testing, so that less cream must be measured out, or special forms of test bottles used (79). 76. Errors of measuring cream. Several factors tend to render inaccurate the measuring of cream for the Babcock test, and in exact work, it is recommended to weigh the amount taken for a test. If a 17.6 cc. pipette is used for measuring the cream, it will not deliver 18 grams of cream, as it will of milk, for the following reasons: 1. The specific gravitj^ of cream is lower than that of milk; if a certain quantity of milk weighs 1030 lbs., the same quantity of cream will weigh from 1020 lbs. to below 1000 lbs., the weight being determined by the richness of the cream; the more fat the cream contains, the less a certain quantity of it, e. g., a gallon, will weigh. f 2. Cream is thicker (more viscous) than milk at the same temperatures, and more of it will adhere to the sides of the measuring pipette than in case of milk. This is of special importance in testing very rich or sour cream. * For average quality of cream furnished at five Connecticut creameries see (194). t For specific quantity of cream of different richness, see table on p. 66. 5 C^ Testing Milk and Its Products* 3. In case of separator cream, more or less air will become incorporated with the cream during the process of separa- tion. In the ripening of cream, fermentation gases are de- veloped and held in the cream in the same way as bread dough holds the gases generated by the yeast. In either case the weight of a certain measure of cream is diminished.. 77. As an illustration of the effect of the preceding fac- tors on the amount of cream measured out by a Babcock 17.6 cc. pipette, the following weighings of separator cream are given. The cream was in all cases fresh from the sep- arator; it was weighed as delivered by the pipette into a Winton cream bottle (81), and the test proceeded with at once. The specific gravity of the cream measured out was determined by means of a picnometer. The data given are in all cases averages of several determinations; the samples^ of cream have been grouped according to their average fat contents.* Weight of cream delivered hy a 17.6 cc. pipette. Per ct. of fat in cream Specific gravity (17.5° C.) Weight of cream delivered, grams 10 1.023 17.9 15 1.012 17.7 20 1.008 17.3 25 1.002 17.2 30 .996 17.0 35 .980 16.4 40 .966 16.3 45 .950 16.2 50 .947 15.8 The data given in the table show plainly the variations irt the specific gravity of cream of different richness and the * For influence of condition of cream on the amount measured out with a 17.6 cc. pipette, see also Bartlett, Maine exp. sta., bull. 31 (s. s.) Cream Testing, 67 error of making tests of cream by measuring it with a 17.6 CO. pipette; if the cream to be sampled is fresh separator cream, testing over 30 per cent., less than 17.0 grams of cream will be delivered into the test bottle, and the results of the reading will be at least one- eighteenth too low, or about 1.4 per cent, on a 25 per cent, cream. If the cream is sour, the error will of course be still greater. 78. Avoiding errors of measuring cream. The pre- ceding table shows that a 17.6 cc. pipette in case of cream containing less than 25 per cent, of fat, and fresh from the separator, will deliver only about 17.2 grams of cream; it is therefore evident that the pipette to deliver cream for the Babcock test must be made larger than the 17.6 cc. pipette used in testing milk. Quite satifactory results may be ob- tained in testing such cream, by using a 18 cc. measuring pipette; to avoid the expense and trouble of using two dif- ferent pipettes, one for milk and one for cream, a pipette with two marks on the stem, at 17.6 cc. and at 18 cc, has been placed on the market, the former mark being used when milk is tested, and the latter for cream. In testing cream by the Babcock test, one of two methods may be followed: First, one of the special forms of cream test bottles which have been devised is used; or, Second, only sufficient cream to be tested in a regular Babcock milk test bottle is taken for a sample. 79. Cream test bottles. Three special forms of bot- tles have been devised for testing samples of cream by the Babcock test; two of these were suggested by Bartlett of Maine,* in 1892; one with a long detachable neck designed * Maine experiment station, bulletins 3 and 4 (second series). 68 Testing Milk and Its Products, m for testing very rich cream (up to 35 per cent, fat), and the other with a neck wid- ened into a bulb in the middle so as to allow a large quantity of fat to be meas- ured. The former kind of cream bottle has, so far as is known to the writers, found but a limited distribution; the neck is too long to be used in the ordinary centrifugal machines, and is not attached until the base portion, containing the cream, acid and first filling with water, has been whirled. This cream bottle is more difficult to handle and cannot be considered as practical as either of the two other forms of bottles devised for testing cream. 80. The hulh-necked cream bottles, (fig. 25), allow the testing of cream containing 23 or 25 per cent, of fat, the usual quantity of cream (18 grams) being measured out. The neck is graduated from to 23 per cent., and in some cases to 25 per cent, the graduation extending both below and above the bulb. This is sometimes an in- convenience, as the water must be added carefully so that the lower end of the col- umn of fat will always come below the bulb, in the graduated part of the neck, and not in the bulb itself. Especially in case of beginners, tests are often lost when this bottle is first used, for the reason given, until the operator learns to add the proper amount of hot Fig. 25. The bulb- uecked cream test bottle. Cream Testing, 69 water to float the fat to some point within the scale. It is recommended to fill these bottles with the first portion of hot water to just above the bulb, so that one can see how much water to add the second time in order to bring the fat within the scale. Each division of the scale on these cream bottles represents two-tenths of one per cent, of fat, as in case of the milk test bottles. 81. The Winton cream bottle. The cream test bottle devised by Winton,* (fig. 26), has a neck of the usual length, and suffi- ciently wide to measure 30 per cent, of fat. The scale of the neck is divided into one- half percents, but readings of a quarter of a percent can easily be estimated. De- terminations of fat in cream accurate to a quarter of a percent are sufficiently exact for most commercial purposes, e. g., in creameries, and this form of cream bottle will be found very convenient in making tests of composite samples of cream. 82. Use of milk test bottle. Cream may be tested by emptying a 17.6 cc. pip- ettef ul of the sample into two or more test bottles, dividing the amount about equally between the bottles, and filling the pipette with water once or twice, which is then in turn divided about equally between the test bottles; the * Connecticut experiment station (New Haven), bull. No. 117- report 1894 p. 224. Fig. 26. The Win ton cream bottle. *]0 Testing Milk and Its Products. per cent of fat in the cream is found by adding the readings obtained in each of the bottles. This method does away with the error incident to the adhesion of cream to the side of the pipette, but not that due to the low speci- fic gravity of the cream, and the results obtained will there- fore be somewhat too low. The dilution of the cream with water in the test bottles not only makes it possible to bring into the bottle all the cream measured out, but also insures a clear test. If ordinary cream is mixed with the usual quantity of sulfuric acid used in the Babcock test, a dark- colored fat will generally be obtained, while the cream diluted with an equal or twice its volume of water, when mixed with the ordinary amount of acid, will give a light yellow, clear column of fat, which will allow of a very dis- tinct and sharp reading. The number of bottles to be used for testing a sample of cream by this method must be regulated by the richness of the cream. If a sample probably contains 20 per cent, or more, one pipetteful should be divided nearly equally be- tween three milk test bottles, and two-thirds of a pipetteful of water is added to each bottle. If the cream contains less than 20 per cent, of fat, it will only be necessary to use two milk test bottles, dividing the pipetteful between these, and adding one-half of a pipetteful of water to each bottle. By using cream test bottles (79), more accurate tests may be obtained, in case of cream containing as much as 25 per cent, of fat, by dividing one pipetteful between two bottles, rinsing half a pipette of water into each one, than by add- ing all the cream to one bottle, without rinsing the pipette, for reasons apparent from what has been said in the pre- ceding. Cream Testing. 71 83. Use of 5 cc. pipette. When the cream is in good condition for sampling, satisfactory results can be obtained by the use of a 5 cc. pipette; 5 cc. of cream are measured into a milk test bottle, and two pipettefuls of water are added. In this way all the cream in the pipette is easily rinsed into the test bottle. The readings multiplied by 11 = 3.6 will give the per cent, of fat in the cream. If the specific gravity of the cream tested varies appreciably from 1, corrections should be made accordingly; e. g,. if the specific gravity is 1.02, the factor should read _1L_ = 3.63: 6xl.0'2 if .95, 3.79, etc. 84. Weighing the cream. For the reasons already given, it is always to be preferred to weigh the cream into the test bottles when accurate tests are required. When a small delicate balance is used, this can be done quite rap- idly. Either of the scales shown in the accompanying illus- tration (figs. 27 and 28), will be found useful and sufficiently accurate for this purpose; a small scale of this kind is also Fig. 27. Scale used for weighing cream, cheese, etc., in the Babcock Fig. 28. Torsion balance used for weighing cream, cheese, etc., la the Babcock test. convenient and helpful in testing cheese, butter and con- densed milk, and in determining the strength of sulfuric acid, and the accuracy of test bottles and pipettes (q. v.). In test- ing cream by weight, the test bottle is first weighed empty, 72 Testing Milk and Its Products, and again when 5 to 10 cc. of cream have been measured into it; the difference between the two weights gives the weight of cream taken for the test. If the cream contains less than 30 per cent, of fat, the regular milk test bottle can be used for testing the cream, if not much more than 5 grams are weighed out; if more cream is taken, or if this is richer than 30 per cent., it is advisable to use the bulb-necked cream bottle. The operator should be careful in weighing the cream not to spill it on the outside of the test bottle, as the bal- ance does not discriminate between cream inside and out- side of the bottle. Sufficient water is added to the bottle to make the total volume about 15 cc. The usual quantity of acid (17.5 cc.) is then added, and the test completed in the ordinary manner. The reading of the amount of fat in the neck of the test bottle in this case does not show the correct per cent, of fat in the cream, because less than 18 grams were weighed out. The per cent, of fat in the cream tested is obtained by multiplying the reading by 18, and dividing the product by the weight of the cream taken. Example: Weight of cream tested, 5.2 grams; reading of col- umns of fat 1^4.8, 2)4,7^ average 4.75; per cent, of fat in the cream ilZ^XlS =16.44. 5.2 The weighing of cream and the reading of the fat column must be made very carefully; a division of one-tenth on the neck of the test bottle has a value of over three-tenths of one per cent, of fat when 5 grams of cream are tested, and six-tenths of one per cent, if only 3 grams of cream are weighed out. The reading is rendered more accurate and certain if a number of tests of a sample are made, at least two or three, and the results averaged. Cream Testing. n The accompanying illustration, (fig. 29), shows the proper method of reading the fat column in cream tests; readings are made from a to 5, not to d or to c. 85. No special precautions other than those required in testing milk have been found necessary in testing cream, except that it is sometimes advisable not to whirl the test bottles in the centrifuge at once after mixing, but to let the cream- acid mixture stand for a while, until it — d turns dark colored. At first, the mixture of cream and acid is much lighter colored than that of milk and acid, owing to the smaller amount of solids not fat contained in the cream. The liquid beneath the fat in a completed test of cream is sometimes milky, and the fat appears white and cloudy, making an exact reading difficult. Such defects can usually be overcome by placing the test bottles in hot water for about 15 minutes previous to the whirling, or by allowing the fat"coiumrfln"he the fat to crystallizc (which is done by neck of a cream hot- i i i • • tie. Readings should cooliug the bottlcs in ICC water or cold water be made from a \o b, not to d or to c. after the last whirling) and remelting it by placing the bottles in hot water. 74 Testing Milk and Its Products. CHAPTER V. BABCOCK TEST FOR OTHER MILK PRODUCTS. 86. Skim milk, butter milk, and whey. Each di- vision on the scale of the neck of the regular Babcock test bottle represents two-tenths of one per cent. (39). When a, sample of skim milk or butter milk containing less than this per cent, of fat, is tested, the estimated amount is ex- pressed by diflerent operators as one-tenth, a trace, one- tenth trace, or one to five hundreths of one per cent. Gravi- metric chemical analyses of skim milk have shown that samples which give only a few small drops of fat floating on the water in the neck of the test bottle, or adhering to the side of the neck, generally contain one-tenth of one per cent, of fat, and often more. Samples of skim milk containing less than one-tenth of a per cent, of fat are very rare, and it is doubtful whether a sample of separator skim milk representing a full run of, say 5000 lbs. of milk has ever shown less than five -hundreths of one per cent, of fat. Under ordinary factory conditions, few separators will deliver skim milk containing under one-tenth of one per cent, of fat, when the sample is taken from a whole day's run. This must be considered a most satisfactory separation.* 87. The reason why the Babcock test fails to show all the fat present in skim milk must be sought in one or two " For comparative analyses of separator skim milk by the gravimetric method and by the Babcock test, see bull. 52, Wis. exp. station. Babcock Test for Other Milk Products, 75 causes: a trace of fat may be dissolved in the sulfuric acid, or owing to the minuteness of the fat globules of such milk they may not be brought together in the neck of the bottles at the speed used with Babcock test. If a drop of the dark liquid obtained in a Babcock bottle from a test of whole milk, be placed on a slide under the microscope, it will be seen that a fair number of very minute fat globules are found in the liquid. These globules are not brought into the column of fat in the neck of the bottle by the cen- trifugal force exerted in the Babcock test; the loss of the fat contained in these fine globules is compensated for, in the testing of whole milk, by a liberal reading of the column of fat separated out, the reading being taken from the lower meniscus of the fat to the top of the upper one (see p. 32); in some separator skim milk, on the other hand, not enough fat remains to completely fill the neck, and the apparent result of the reading must there- fore be increased by from five-hundredths to one-tenth of one per cent. It follows from what has been said that tests of skim milk showing no fat in the neck of the test bottles on completion of the test, generally indicate inefficient work of the centri- fugal tester, or the operator, or both. The test should be repeated in such cases, using more acid and whirling for full four minutes. In order to bring as much fat as possible into the neck of the bottles, in testing skim milk, it is advisable to add some- what more acid than when whole milk is tested, viz: about 20 cc, and to whirl the bottles at full speed for four to five minutes. The readings must be taken as soon as the whirl- ing is completed, as owing to the contraction of the liquid by cooling, the fat is otherwise spread over the inside of 76 Testing Milk and Its Products. the neck of the test bottle as a film of grease which cannot be measured by the scale. 88. The double-necked test bottle, (fig. 30), suggested by one of us,* is made especially for measuring small quanti- ties of fat and gives most satisfactory re- sults in testing skim milk and butter milk. Each division of the scale in these bottles represents five-hundredths of one per cent., and the marks are so far apart that the small fat column can be easily estimated to single hundredths of one per cent. In the first forms, now not in common use, the neck was graduated to hundredths of one per cent. The value of the divisions of the scale on the double-necked test bottles has been dis- cussed of late in dairy papers, and various opinions have been expressed whether they show one-tenth or one-twentieth (.05) of one per cent, of fat. By calibration with mercury the value of the divisions will be found to be .05 or one-twentieth of one per cent., but, as shown above, the results obtained in using Fig. 30. The double-necked skim milk bot- tle, (sometimes called the Ohls- son or B. & W. bottle). the bottles for thin separator skim milk come .05 to .1 per cent, too low, so that, practically speaking, each division may be taken to show one-tenth of one per cent., if the column of fat obtained fills only one or two divisions of the scale. The double-necked bottle is very convenient for the testing of separator skim milk, thin butter milk and whey. The milk, acid, and water are added to the bottle * Farrington, and constructed by Mr. J. J. Nussbaumer, of Illinois. Babcock Test for Other Milk Products. 77 through the large side-tube; the mixing of milk and acid, and the addition of hot water, must be done with great care so that none of the contents are forced into the fine measur- ing tube and lost. When the fat is in the lower end of the measuring tube, it can be forced up into the scale by press- ing tightly with the finger on the top of the side tube. This test bottle is more fragile and expensive than the ordinary Babcock bottles, and unless carefull}^ handled, will not prove a good investment; the bottle has recently been made of heavier glass and this form is to be highly recommended. 89. The double-sized skim milk bottle is of no particular value. It is difficult to obtain a thorough mixture of the milk and the acid in these bottles, and the tests invariably €ome too low, more so than in case of the regular Babcock bottles, or the double-necked skim milk bottles. 90. The testing of butter milk or whey by the Babcock test offers no special difficulties, and what has been said in regard to tests of separator skim milk is equally true in case of these by-products. Whey contains only a small quantit}^ of solids not fat (less than 7 per cent), and the mixing with acid, and the solution of the whey solids therein, is therefore readily accomplished; the acid solution is of a light reddish color, turning black but very slowly. 91. Cheese. Cheese can be easily tested by the Bab- cock test if a small scale (fig. 27-8) is at hand for weighing the sample; the results obtained will furnish accurate information as to the amount of fat in the cheese, provided good judgment and exactness is used in sampling and weighing. The following method of sampling the cheese is recommended: * * U. S. Dept. of Agriculture, Chemical Division, bull. No. 46, p. 37. ^8 Testing Milk and Its Products. " Where the cheese can be cut, a narrow wedge reaching from the edge to the center of the cheese will more nearly represent the average composition of the cheese than any other sample. This may be cut quite fine, with care to avoid evaporation of water, and the portion for analysis taken from the mixed mass. When the sample is taken with a cheese trier, a plug taken perpendicular to the sur- face, one-third of the distance from the edge to the center of the cheese, should more nearly represent the average composition than any other. The plug should either reach entirely through or only half through the cheese. " For inspection purposes the rind may be rejected, but for investigations, where the absolute quantity of fat in the cheese is required, the rind should be included in the sam- ple. It is well, when admissible, to take two or three plugs on different sides of the cheese and after splitting them lengthwise with a sharp knife, take portions of each for the test." 92. When a satisfactory sample of the cheese has been obtained, about 5 grams are weighed into a milk test bottle, or a larger quantity may be used with a cream test bottle. The test bottle is first weighed empty, and again after the pieces of cheese have been added. About 15 cc. of hot water is added to the cheese in the test bottle, and this is shaken occasionally until the cheese softens and forms a creamy emulsion with the water. A few drops of strong ammonia will aid in this mixing and disintegration, the pro- cess being hastened by placing the test bottles in hot water. When all lumps of cheese have disappeared in the liquid, the test bottles are cooled to about 70° F., acid is added, and the test completed in the ordinary manner. Babcock Test for Other Milk Products. 79 The per cent, of fat m the cheese is obtained by multiply- ing the reading of the fat column by 18 and dividing the product by the weight of cheese added to the test bottle. The weighing of the cheese and the reading of the fat must be done with great care, since any error introduced is more than trebled in calculating the per cent, of fat in the cheese, 93. Condensed milk. The per cent, of fat in con- densed milk can be obtained by weighing about 8 grams into a test bottle and proceeding in exactly the same way as given under testing of cheese. It is not necessary to add ammonia or to warm the condensed milk in the test bottles, since the solution of this in water is readily effected with- out any outside agency. Enough water should be added to make the total volume of liquid in the bottles about 17.6 cc. If a scale is not available for weighing the sample, fairly accurate results may be obtained by diluting the condensed milk with water (1:1), and completing the test in the ordi- nary manner. When this is done, the results must be cor- rected for the dilution which the sample received. So Testing Milk and Its Products, CHAPTER YI. THE LACTOMETER AND ITS APPLICATION. 94. The Quevenne lactometer. This instrument, (see fig. 31), consists of a hollow glass cylinder weighted down by means of mercury or fine shot so as to fioat in milk in an upright position, and provided with a narrow stem at its upper end, inside of which is found a graduated paper scale. In the better forms, like the Quevenne lactometer shown in the figure, a thermometer is melted into the cylinder, with its bulb at the lower end of the lactometer, and its stem rising above the lactometer scale. The lactometer is used for the determination of the spe- cific gravity of milk. The term specific gravity, means the weight of a certain volume of a solid or a liquid substance compared with the weight of the same volume of water at 4° C (39.2° Fahr.); for gases the standard of comparison is air or hydrogen. If the milk which a can will hold, weighs exactly 103.2 lbs., this can will hold a smaller weight of water, say, 100 lbs., as milk is heavier than water; the spe- cific gravity of this milk will then be 1^ = 1.032. 95. The lactometer enables us to determine rapidly the relative weight of milk and water. Its application rests on well-known laws of physics: When a body floats in a liquid, the weight of the amount of liquid which it replaces, is equal to the weight of the body. It will sink further into a light liquid than into a heavy one, because a larger vol- The Lactometer and Its A-ppltcation. Fig. 31. Quevenne lactometer floating in milk in a tin cylinder <100). 6 ume of the former will be required to equal the weight of the bod}'. A lac- tometer will therefore sink deeper into milk of a low specific gravity than into milk of a high specific gravity. The scale of the Quevenne lactome- ter is marked at 15 and 40, and divided into 25 equal parts, with figures at each five divisions of the scale. The single divisions are called degrees. The 15 de- gree mark is placed at the point to which the lactometer will sink when lowered into a liquid of a specific gravity of 1.015, and the 40 degree mark, at the point to which it will sink when placed in a liquid of a specific gravity of 1.040. The specific gravity therefore is changed to lactometer degrees by multiplying by 1000 and substracting 1000 from the product. Example: Given, specific gravity 1.0345; corresponding lactometer degree, 1.0345 X 1000 — 1000 = 34.5. 96. Influence of temperature. Like most liquids, milk will expand on being warmed, and the same volume will therefoie weigh less when warmed than before; that is, its specific gravity will be decreased. It follows then that a lactometer is onl}' correct for the tem- perature at which it is standardized. If 82 Testing Milk and Its Products. a lactometer sinks to the 32 mark in a sample of milk of a temperature of 60" F., it will sink below this mark if the temperature of the milk is 50° F., and will not sink so far down as 32, if the temperature is 70° F. Lactometers in the market at present are generally standardized at 60° F., and to show the correct specific gravity, the milk to be tested should first be warmed (or cooled, as the case may be) to exactly 60° F. As this is a somewhat slow process, tables have been constructed for correcting the results for errors due to differences in temperature (see Appendix). 97. As the fat content of a sample of milk has a marked influence on its specific gravity at different temperatures, the coefficient of expansion of fat differing greatly from that of the milk serum, the table cannot give absolutely accurate corrections for all kinds of milk, whether rich or poor. But the errors introduced by the use of one table for any kind of whole milk within a comparatively small range of tem- perature, like ten degrees above or below 60°, are too small to have any importance outside of exact scientific work, and in such, the specific gravity is always determined by means of a picnometer, or a specific gravity bottle, at the tempera- ture at which the calibration has been made. In taking the specific gravity of a sample of milk by means of a lacto- meter, the milk is always warmed or cooled so that its tem- perature does not vary ten degrees either way from 60° F. 98. The temperature correction table for whole milk^ given in the Appendix shows that if, e. g., the specific grav- ity of a sample of milk taken at 68° F. was found to be 1.034, its specific gravity would be 1.0352 if the milk was cooled down to 60°. If the specific gravity given was found at a temperature of 50°, the correct specific gravity of the milk would be 1.033. The Lactometer and Its Afph'cation. 83 In practical work in factories or at the farm, sufficiently accurate temperature. corrections may generally be made by adding .1 to the lactometer reading for each degree above 60° F., and by subtracting .1 for each degree below 60°; e. g.^ if the reading at 64° is 29.5, it will be about 29.5 f .4 = 29.9 at 60° F.; and 34.0 at 52° F. will be about 84.0 — .8 = 33.2 at 60° F. By reference to the table in the AppendiXy we find it to be 33.0 in either case. The scale of the thermometer in the lactometer should be placed ahove the lactometer scale so that the temperature may be read without taking the lactometer out of the milk; this will give more correct results, will facilitate the reading and save time. 99. N. Y. Board of Health lactometer. In the East, and among city milk inspectors generally, the so-called New York Board of Health lactometer is often used. This doe& not give the specific gravity of the milk directly, as is the case with the Quevenne lactometer, but the scale is divided into 120 equal parts, known as Board of Health degrees, the mark 100 being placed at the point to which the lactometer sinks when lowered into milk of a Specific gravity of 1.029 (at 60° F.); this is considered the lowest limit for specific grav- ity of normal cow's milk. The zero-mark on the scale shows the point to which the lactometer will sink in water; the distance between these two marks is divided into 100 equal parts, and the scale is continued below the 100 mark to 120. As 100° on the Board of Health lactometer corresponds to 29° on the Quevenne lactometer, the zero mark showing in either case a specific gravity of 1, the degrees on the former lactometer may easily be changed into Quevenne lactometer degrees by multiplying by .29. To further aid in this 84 Testing Milk and Its Products. transposition, a table is given in the Appendix, showing the readings of the two scales, between 60° and ]20° on the Board of Health lactometer. 100. Reading the lactometer. For determining the specific gravity of milk in factories or private dairies, tin cylinders, 1^ inches in diameter, and 10 inches high, with a base about 4 inches in diameter, are recommended, (see fig. 31); another form of specific gravity cylinders, in use in chemical laboratories, is shown in fig. 32. The cylinder is filled with milk of a temperature ranging between 50° and 70° F. to within an inch of the top, and the lactometer is slowly lowered therein until it floats; it is left in the milk for about half a minute before the lactometer- and the thermometer read- ings are taken, both to allow the escape of air which has been mixed with the milk in pouring it pre- paratory to the specific gravity determination, and to allow the thermometer to adjust itself to the temperature of the milk. The lactometer should not be left in the milk much more than a minute before the reading is taken, as cream will very soon Fig. 32. )3e2in to rise on the milk, and the reading, if taken Specific ^ gravity later, will be too high, as the bulb of the lactometer cylinder. ^.|^ ^^ floating in partially skimmed milk. In read- ing the lactometer degree, the mark on the scale plainly visi- ble through the upper portion of the surface meniscus of the milk should be noted. Owing to surface tension the milk in immediate contact with the lactometer stem will'^rise above the level of the surface in the cylinder, and this must be taken into consideration in reading the degrees. There is no need of reading closer than one-half of a lactometer degree in the practical work of a factory or dairy. The Lactometer and Its Af plication, 85 101. Time of taking lactometer readings. The specific gravity of milk should not be determined until six to twelve hours after the milk has been drawn from the udder, as too low results are otherwise obtained.* The cause of this phenomenon is not definitely understood; it may lay in the escape of the gases in the milk, or in changes in the mechanical condition of the nitrogenous components of the milk, (most likely of the casein), occurring on stand- ing. The results obtained after twelve hours will as a rule come about one degree higher than when the milk is cooled down directly after milking, and its specific gravity then determined. Calculation of Milk Solids. 102. A number of chemists have prepared formulas for the calculation of milk solids when the fat content and the specific gravity (lactometer reading) of the milk are known. By careful work with milk tester and lactometer it is possible by means of these formulas to determine the com- position of samples of milk with considerable accuracy out- side of, as well as in, chemical laboratories. As the com- plete formulas given by various chemists (Behrend and Morgen, Clausnitzer and Mayer, Fleischmann, Hehner and Richmond, Richmond, Babcock,)t are very involved, and require rather lengthy calculations, tables facilitating the figuring have been prepared. The formulas in use at the present time, in this country or abroad, are those proposed by Fleischmann, Hehner and Richmond, and Babcock. Babcock's formula forms the foundation of the tables for solids not fat given in the Appendix, as it is generally taught in American dairy schools. * Bulletin No. 43, Chem. Div., U. S. Dept. of Agriculture, p. 191. t Agricultural Science, vol. Ill, p. 139. 86 Testing Milk and Its Products, By the use of these tables the percent of solids not fat corresponding to lactometer readings from 26 to 36, and to fat contents from 2 to 6 percent may be found. The for- mula, as amended in 1895, is as follows,* S being specific gravit}^, and / the percent of fat in the milk. vSolids not fat = / 100 S - Sf , _ \ ^ 00 - f ) 2.5 VlOO — 1.0753 Sf / ^ 103. The derivation of this formula is as explained in the following: Milk is considered a mechanical mixture of butter fat and milk serum, the latter being an aqueous solution (or in part semi- solution) of all the solid components of the milk, except the fat. If/ therefore designate the percent, of fat in a sample of milk, then 1 00 — f = per cent, serum in the milk. Let furthermore the specific gravity of the serum be called x, and the increase in specific gravity of the milk caused by one per ceat. of serum solids = a. The difference between the specific gravity of milk serum and that of water is nearly directly pro- portional to the amount of serum solids present, and the per cent, of serum solids will therefore be obtained if this difference be divided by the constant factor a, representing the increase in specific gravity caused by one per cent, serum solids. If this per cent, be multiplied by the per cent, of serum in the milk, and the product divided by 100, the result will give the per cent, solids not fat in the milk, viz. : Per cent, solids not fat = X ... . (I) a 100 ^ ^ To find the value of x in terms of 5, the specific gravity, and /, the fat of the milk, it is necessary to determine the value of a from a large number of analyses. The volume of a substance in cc. is equal to its weight in grams divided by its specific gravity ; therefore Volume in cc. of 100 grams of milk = * Wisconsin experiment station, twelfth report, p. 120. The Lactometer and Its Amplication. 87 Voli.ime in cc. of serum in 100 grams of milk = X The specific gravity of pure butter fat being ,93, therefore Volume in cc. of fat in 100 grams of milk = = 1.0753 f .c/o Since milk is made up of fat and serum, we have 100 ^ 100 -f ^ 1.0^53 f S X By solving this equation we find that x= lOQS-sf .... (II) 100 — 1.0753 Sf To obtain the value a, the last equation is solved for a large number of analyses of different kinds of milk. 1 is subtracted from the average value obtained for x (equation II) and the differ- ence is divided by the per cent, of serum solids. Such calculations have shown that the value of a with normal cow's milk will come very near .004; substituting therefore this value and that of x (equation II) in equation I, and reducing, we have Solids not fat= / lOQS-Sf _ \ ^^^^ _ ^ ^.s V 100 — 1.0753 Sf / 101. The tables made up from this formula, giving the percentages of solids not fat corresponding to certain per- cents of fat and lactometer readings, are given in the Ap- pendix. A careful examination of the same discloses the fact that the percent of solids not fat increases uniformly at the rate of .25 percent for each lactometer degree, and .02 percent for each percent of fat. This relation is ex- pressed by the following simple formulas: ^' ">^ Solids not fat = i L X .2 f ' '^^ t '^\ Total solids = :^ L x 1.2 f , L being the lactometer reading at 60° F. (specific gravity X 1000 — 1000), and/ the percent of fat in the milk. Rule : a, To find the percent of solids not fat in milk, add two-tenths of the percent of fat to one-fourth of the lactometer reading, and 88 Testing Milk and Its Products. b, To find percent of total solids in milk, add one and two- tenths times the percent of fat to one-fourth of the lactometer reading. These formulas and rules are easily remembered, and can be quickly applied without the use of tables. The results obtained by using them do not differ more than .04 percent from those of the complete formula for milks containing up to 6 percent of fat, and may be safely relied upon in prac- tical work. Adulteration of Milk. 105. The problem of determining whether or not a sam- ple of milk is adulterated, becomes an important one in the work of milk inspectors, and dairy and food chemists. Managers of creameries and cheese factories are also some- times interested in ascertaining possible adulterations in case of some patron's milk, although at present, since the general introduction of the Babcock test in factories, and the payment for the milk on basis of the amount of butter fat delivered, the temptation to water or skim the milk has been largely removed. In the city milk trade, especially in our larger cities, watered or skimmed milk is still fre- quently met with, in spite of the vigilance of their milk inspectors or the officers of the city boards of health. When the origin of a suspected sample of milk is known, a second sample should always be taken on the premises by, or in the presence of, the inspector, and the composition of the two samples compared. If the suspected sample is con- siderably lower in fat content, than the second, so-called control-sample, with a normal percent of solids not fat, it i& skimmed; if the solids not fat are below normal, it is watered; and if both these percentages are abnormally low, the sam- ple is most likely both watered and skimmed. The Lactometer and Its Amplication. 89 106. In order to determine whether or not a sample ot milk is skimmed, or watered, or both skimmed and watered, the percents of fat and of solids not fat in the sample must be ascertained, and if a control-sample can be secured, these determinations for both samples compared. The proper latitude to be allowed for the natural variation in the com- position of milk differs accordinoj to the origin of the milk; in case of milk from single cows, the variations in fat con- tent from day to day may exceed one percent, although under ordinary conditions the percent of fat in most cows' milk will not vary that much. The content of solids not fat is more constant, and rarely exceeds one-half of a per- cent from day to day with single cows. Cows in heat or sick cows may give milk differing considerably in composi- tion from normal milk.* 107. Mixed herd milk is of comparatively uniform com- position on consecutive days, and as most milk offered for sale or delivered to factories is of this kind, the task of the milk inspector is made considerably easier and more certain on this account. Daily variations in herd milk beyond one percent of fat, and one-half percent of solids not fat, are sus- picious and may be taken as fairly conclusive evidence of adulteration. This is especially true in case the control- sample shows a comparatively low content of fat or solids not fat. 108. Where a control-sample cannot be taken, the legal standards for fat or solids in milk, in force in the various states, are used as a basis for calculating the extent of adulteration of a sample of milk. A list of legal standards for milk in this country and abroad is given in the Appendix. These standards determine the limits below which the milk * Blyth, Foods, their Composition and Analysis, London, 1882, p. 246 et seq. po Testing Milk and lis Products. offered for sale must not fall within the respective states. Legally it matters not whether a sample of milk offered for sale has been skimmed or watered by the dealer or by the cow; in the latter case, the cows producing the milk are of a breed or strain that has been bred persistently for quantity of milk, without regard to its quality. In most states the legal standards for the fat content of milk is 3 per cent., and for solids not fat, 9 or 9.5 per cent. There are, however, cows that normally will produce milk containing only 2.5 to 2.8 per cent, of fat and less than 8.5 per cent, solids not fat. Such milk cannot therefore be legally sold in most states in the Union, and the farmer offering such milk for sale, even if he does not know the composition of the milk produced by his cows, is as liable to prosecution as if he had directly watered the milk. By mixing the milk of several cows, the chances are that the mixed milk will contain more fat and solids not fat, than called for by the legal standard; if such should not be the case, cows producing richer milk must be added to the herd so as to raise the quality of the herd milk up to the legal standard. 109. Calculation of extent of adulteration.* In the following formulas, the percents found in the control- samples, if such are at hand, are always substituted for the legal standards. a. Shimming. — 1. If a sample of milk has been skimmed, the following formula will give the number of pounds of fat abstracted from 100 lbs. of milk: Fat abstracted = (x) = legal standard for fat — f, . (I) / being the percent of fat in the suspected sample. 2. The following formula will give the percent of fat ab- * Woll, Handbook for Farmers and Dairymen, New York, 1897, pp. 207-8. The Lactometer and Its Afflication. 91 stracted, calculated on the total quantity of fat, originally found in the milk: x=100-, , !^^}^l, ,, .... (II) legal standard for fat b. Watering. — 1. If a sample is watered, the calculations are most conveniently based on the percentage of solids not fat in the milk. Percent, of foreign ('' extraneous ") water in adulterated milk=100— , , , ^ P-^ -^. . . . (Ill) leg. stand, for solids not fat S being the percent, of solids not fat in the suspected sample. Example: — A sample of milk contains 7.5 percent, solids not fat; if the legal standard for solids not fat is 9 percent., 100 — 7 5 X 100 — — 1-^ = 16.7, shows the percent, of extraneous water in the milk. 2. Watering of milk may also be expressed in per cent. of water added to the original milk, by formula IV: Percent, water added to original milk _ . s _ 100 X leg. stand, for sol. not fat ^^^ .yYN 100 X 9 In the example given above, — ^^— - — — 100 = 12 percent, of 7.5 water was added to the original milk. c. Watering and shimming. — If a sample has been both watered and skimmed, the extent of watering is ascertained by means of formula (III) or (IV), and the fat abstracted found according to the following formula: Percent, fat abstracted = (x)=leg. stand, for f ^t _ l£ii^^?5ii|li£L^^ii^' Xf .(V) 92 Testing Milk and Its Products. 'Example : — A sample of milk contains 2.4 per cent, of fat, and 8.1 per cent, solids not fat; then Extraneous water in milk = 100 — 8.1 X 100 10 per cent. Fat abstracted = 3 9 X 2.4 8.1 = .33 per cent. 100 lbs. of the milk contained 10 lbs. of extraneous water and .33 lbs. of fat had been skimmed from it.* 110. other methods of adulteration. Milk which has been watered, or skimmed, or both, is sometimes further adulterated by unscrupulous milk peddlers through the addition of a small quantity of cheese color; this will mix thoroughly with the milk, and, if added judiciously, will im- part a rich cream color to it. The presence of foreign col- oring matter in milk is easily shown by shaking 10 cc. of the milk with an equal quantity of ether; on standing, a clear ether solution will rise to the surface; the solution will be yellow colored, if artificial coloring matter has been added to the milk, the intensity of the color indicating the quantity added; natural fresh milk will give a colorless ether solution. *The following blank will be found convenient for work in the laboratory or testing room : REPORT BY. DAIRY SCHOOL. DATE 189. MILK TEST. No. of sample. Lactome- ter read- ing. Tempera- ^Zltt^^ ture reading ^^^^- at 60° Per cent, of fat. Solids not fat. Adultera- tion, kind and amt. Remarks. The Lactometer and Its Application. 93 111, The following method given by Wallace"^ is claimed to detect one part of coloring matter in 100,000 of milk. About 30 cc. of milk are coagulated with a few drops of acetic acid by the aid of heat. The coloring matter of annatto being insoluble in acids, is precipitated with the casein. The coagulated mass is separated from the whey by straining through a coarse cotton cloth, and the excess of liquid pressed out. The casein is placed in a mortar and rubbed with ether. The ether is then poured into a separatory funnel and about 10 cc. of a dilute soda solution (1 : 100) are added. After thorough shaking, the funnel is set aside to allow the liquids to separate. The lower la^-er which consists of soda solution in which the annatto is dissolved is drawn off into a porcellain dish, and in it are placed two discs of filter-paper, and the liquid is greatly evaporated. If annatto is present, the discs will be dyed orange to buff color. One of the well-washed discs is moistened with a dilute sodium carbonate solution to fix the color; the other is touched with a drop of stan- nous chlorid, and becomes instantly changed in color to a rich pink. "This test may be used in any condition of the milk, and with as small a quantity as 10 cc." It is not known that other methods of adulterating milk than those mentioned are practiced at the present time. For methods of detection of preservatives in milk, see Chapter X. N. J. Dairy Commissioner Report, 1896, p. 94 Testing Milk and Its Products. CHAPTER N\\. TESTING THE ACIDITY OF MILK AND CREAM. 112. Cause of acidity in milk. Very soon after milk is drawn from the udder, it will be found to have an acid reaction, when phenolphtaleiu is used as an indicator.^ The acidity in fresh milk is not due to the presence of free or- ganic acids in the milk, like lactic or citric acid, but to acid phosphates, and possibly also in part to free carbonic acid gas in the milk, and to the acid reaction of casein. Even in case of so-called sweet milk, nearly fresh from the cow, a certain amount of acidity, viz: on the average .07 per- cent, is therefore found. When the milk is received at the factory it will rarely test less than .10 percent, of acid, calculated as lactic acid; some patrons bring milk day after day that does not test over .15 percent, of acid; others bring milk that tests from .20 to .25 percent, and some lots, although very rarely, will test as high as .3 of one per- cent, of acid. It has been found that milk will not usually smell or taste sour, or "turned," until it contains .3 to .35 percent, of acid. 113. The acidity in excess of that found normally in milk as drawn from the udder, is due to other causes than those described, viz: the presence of acid milk-components. Bacteriological examinations of milk from various sources * Freshly drawn milk shows an amphoteric reation to litmus, i. e., it colors blue litmus paper red, and red litmus paper blue. Testing the Acidity of Afilk and Cream. 95 and of different age have shown that there is a direct rela- tion between the bacteria found in normal milk, and its acidity; the larger the number of bacteria per unit of milky the higher the acidity of the milk. The increase in the acidity of milk on standing is caused by the breaking-down of milk sugar into lactic acid through the influence of acid- forming bacteria. Since the bacteria get into the milk through lack of cleanliness during the milking, and careless handling of the milk after the milking, this being kept under conditions that favor the multiplication of the bac- teria contained thereiu, it follows that an acidity test of fresh milk give a very accurate measure of the care bestowed in handling the milk; such a test will show which patrons take good care of their milk and those who do not wash their cans clean, or their hands and the udders of the cows before milking, and have dirty ways generally in milking and caring for the milk. The acidity test will always be high in case of milk kept for more than a day (Monday milk) or delivered after a warm sultry day or night. The bacteria have had a chance to multiply enormously in such milk, even if it be kept cooled down to 40°-50° F., and as a result considerable quantities of lactic acid have been formed. The determination of the acidity of fresh milk is explained in detail below (128). 114. Methods of testing acidity. Methods of meas- uring the acidity or alkalinity of liquids by means of certain chemicals giving characteristic color reactions in the pres- ence of acid or alkaline solutions (so-called volumetric meth- ods of analysis) have been in use for many years in chemical laboratories. They were applied to milk as early as 1872 by Soxhlet,* and the method worked out by Soxhlet and * Jour. f. prakt. Chemie, 1872, p. 6, 19. 96 Testing Milk and Its Products. Henkel has since been in general use by European chemists. They measured out 50 cc. of milk to which was added 2 cc. of a 2 percent alcoholic solution of phenolphtalein, and this was titrated with a one-fourth normal soda solution ^ (see below). In this country, Dr. A. Gr. Manns,t in 1890, pub- lished the results of work done in the line of testing the acidity of milk and cream, and the method of procedure and apparatus proposed by him has become known under the name of Manns test, and has been advertised as such by dealers in dairy supplies. 115. Manns' test. The acid in milk or cream is measured by using an alkali solution of a strength, together with an indicator, which shows by a change of color in the milk when all its acid has been neutralized. Any of the alkalies, soda, potash, ammonia, lime or barium can be used for making the standard solution, but it requires the skill, and apparatus of a chemist to prepare it of the proper strength. A one-tenth normal solution % of caustic soda is the alkali solution used most frequently in determining the acidity of milk, and is the solution labeled Neutralizer of Manns' test. * Fleischmann, Lehrb. d. Milchwirtschaft, p. 23. t Illinois experiment station, bulletin No. 9. X A normal solution of a chemical is known by the chemist as a solution contain- ing as many grams of the chemical in a liter (1000 cc.) as the figure representing its molecular weight. Caustic soda is made up of an atom each of sodium (Na), oxygen (O), and hydrogen (H); its molecular weight is therefore 23 + 16 + 1 = 40 Na H A normal soda solution then is made by dissolving 40 grams of soda in water, making up the volume to 1000 cc; a one-tenth normal solution will contain one- tenth of this amount of soda, or 4 gr. dissolved in one liter. One cubic centimeter of the latter solution will contain .004 grams of soda, and will neutralize .009 grams of lactic acid. The formula for lactic acid is CsHgOs (see p. 16) and its molecu- lar weight therefore 3 X 12 + 6 X 1 + 3 X 16 = 90. A one-tenth normal solution therefore contains 9 grams per liter, and .009 grams per cubic centimeter. Testing the Acidity of Milk and Cream. 97 The indicator used for this purpose is phenolpJitalein^ a yellowish light powder; its compounds with alkalies are red, in weak alkaline solutions, pink colored; while its acid compounds are colorless. The phenolphtalein solution used is prepared by dissolving 10 grams in 300 cc. of 90 percent, alcohol (Mohr). 116. In testing the acidity of either milk or cream it is necessary to measure out with exactness the quantity of liquid to be tested; Manns recommended using a 50 cc. pipette. This amount of milk or cream is measured into a <;lean tin, porcellain or glass cup, a few drops of the phenol- phtalein solution are added, and the "Neutralizer" (or alkali solution) is cautiously dropped in from a burette, the point at which the solution stands before it is drawn out being noted. By constant stirring during this operation it will be noticed that the pink color formed by the addition of even a drop of alkali solution will at first entirely disappear, but as more and more of the acid in the sample becomes neu- tralized, the color will disappear more slowly, until finally a point is reached when the pink color remains permanent for a time. No more alkali should be added after the first ap- pearance of a permanent and uniform pink color in the sample. This color will fade and gradually disappear again on standing, owing to the efi'ect of the carbonic acid of the air, to which acid, phenolphlalein is very sensitive. The amount of the alkali solution used for the test is then ob- tained from the reading on the scale of the burette, and this shows the degree of acidity in the sample. The per cent, of acid in the sample is calculated by multiplying by .009 the number of cc. of alkali solution used, and dividing the product by the number of cc. of the sample tested, the quotient being multiplied by 100. 98 Testing Milk and Its Products. c. c. alkali x .009 Percent acidity = -. — - — 7—5 X 100 •^ c. c. sample tested If 50 cc. of cream required 32 cc. of alkali solution to produce a permanent pink color, the percent of acid in the 32 X 009 cream would be ^- — X 100 =.58 percent. A part of this calculation may be saved by using a factor for multi- plying the number of cc. of alkali added in each test. This factor is obtained by dividing ,009 (the number of grams of lactic acid neutralized by one cc. of alkali solution) by the number of cc. of sample tested, and multiplying the quotient by 100. If a 50 cc. pipette is used for measuring the sample to be tested, the factor will be (.009 -^ 50) X 100 = .018; if a 25 cc. pipette is used, the factor will be (.009 -f- 25) x 100 = .036; and if a 20 cc. pipette is used, (.009 -^ 20) X 100 = .045 will be the factor to be applied in calculating the percent of acidity, the number of cc. of alkali used being in all cases multiplied by the particular factor corresponding to the volume of the sample tested. 117. A table showing the percent of acid corresponding to different quantities of alkali solution for a given quantity of milk or cream can be easily prepared by the use of the factors given, and the necessity of calculating the result in percent thus saved. If a 50 cc. pipette is used for measur^ ing out milk or cream, the table will thus be as follows: cc. alkali Corresponding CC. alkali Corresponding olution used acidity solution used acidity 1 CC .018 percent. 6cc .108 per cent 2cc .036 7cc .126 3cc .054 8cc .144 4 cc .072 9 cc .162 5 cc .090 10 cc .180 Testing the Acidity of Milk and Creajn. 99 With a little practice the operator will easily be able to determine the acidity of different portions of the same milk or cream to within 1 cc. of alkali solution (= .02 per cent, of acid, when a 50 cc. sample is taken). 118. Manns s testing outfit. The apparatus (see fig. 33) and chemicals needed for testing the acidity of milk or cream by the so-called Manns' test include 1 gal. one-tenth normal alkali solution; 4 oz. of an alcoholic solu- tion of phenolphtalein, one 50 cc. glass burette with stop-cock, one bur- ette stand, and a pipette for measur- ing out the sample. This outfit will make about 100 tests and is sold for $5.00.* 119. The alkaline tablet test. Solid alkaline tablets were proposed by Farrington in 1894, as a substitute for the liquid used in Manns' test.f It was found possible to mix a solid alkali and coloring matter, and com- press the mixture into a small tablet, which would contain an exact amount of alkali. The advantage of the tab- lets lies in the fact that they will keep . J J 11 1- ^^^- ^^' Apparatus used in tar better than a standard alkali Manns' test. * Devarda^i acidimeter (Milchzeiting, 1896, p. 785) is built on the same principle as Manns' test; one-tenth soda solution is added to 100 cc. of milk in a glass-stop- pered graduated flask, 2 cc. of a 4 per cent, phenolphtalein solution being used as an indicator. The graduations on the neck of the flask give the "degrees acidity" directly. t Illinois experiment station, bulletin No. 32. lOO Testing Milk and Its Products. solution, and they can be easily and safely sent by mail; they also require less apparatus and are considerably cheaper than standard alkali solutions; 1000 of these tab- lets, costing $2.00, will make about 400 tests * Similar alka- line tablets were placed on the market in Europe at about the same time, viz: Stokes' Acidity Pellets in 1893, and Eichler's Sceurepillen (acid-pills) in 1895.t Two methods of using the tablets have been proposed, one, for the titration (determination of acidity) of ripening cream, in the manufacture of sour-cream butter; and the other, for determining the approximate acidity of different lots of apparently sweet milk or cream. 120. Determination of acidity in sour cream. The method is equally applicable for the determination of the acidity of sour cream, sour milk and butter milk, but is most frequently employed in testing the acidity of ripening cream, to examine whether or not the ripening process has reached the proper stage for churning the cream. The ap- paratus used (see fig. 34) are as follows: 1 Babcock 17.6 cc. pipette. 100 cc. graduated cylinders; it is well to provide two or three of these, although only one is strictly necessary. 1 white cup. 121. Preparation of the solution. The tablet solution used during the past two years was prepared by dissolving five tablets in 50 cc. of water; with 20 cc. of cream each cubic centimeter of this solution represents .017 percent of acidity (lactic acid) in the sample tested. The amount of acid in a given sample was then obtained by multiplying * The tablets are sold by dealers in dairy supplies, t Milchzeiting, 1895, pp. 513-16. Testing the Acidity of Milk and Creaju. loi the number of cubic centimeters of the tablet solution used by .017. 122. According to a suggestion recently made* the strength of the solution has been changed in such a manner that the percentages of acidity are indicated directly by the number of cubic centimeter of tablet solution used in each test. The solution may be made up in two ways, viz: by use of a 20 cc. or a 17.6 cc. pipette. i Fig. 34. Apparatus used for determining the acidity of cream or millc. a. Use of 20 cc. ^pipette. When a 20 cc. pipette is used for measuring the sample to be tested, the tablet solution is prepared by dissolving one tablet for every 17 cc. of water; for five tablets 85 cc. of water are therefore taken. When made in this way, each cubic centimeter of solution repre- * By Mr. C. L. Fitcb, of Hoard's Creameries, (Hoard's Dairyman, Sept. 3, 1897). I02 Testing Milk and Its Products. sents .01 percent of acid in the sample tested, 20 cc. of cream being taken; the number of cubic centimeters required to produce a pink color in the sample tested as read off directly from the graduations of the cylinder used for mak- ing the tablet solution gives the percent of acid in the sam- ple, 10 cc. being equal to .10 percent acid, 32 cc. to .32 per- cent, 65 cc. to .65 percent, etc. b. Use of 17.6 cc. pipette. The 17.6 cc. milk pipette of the Babcock test may be used for measuring the sample for acidity testing, and the results read directly from the grad- uated cylinder, if the tablet solution is prepared by taking one tablet for every 19.5 cc. of water; five tablets are there- fore dissolved in 97 cc. of water. 123. As cream during its ripening process should gen- erally have from .5 to .6 percent of acid, before it is ready to churn, a 50 cc. cylinderful of tablet solution of this strength will not be sufficient to make a test of cream con- taining over .5 percent of acid, although it is enough for testing the cream up to this point during the ripening pro- cess. The acid-testing outfit should therefore contain a 100 cc. graduated cylinder, instead of one of 50 cc. capacity, so that cream of any amount of acidity up to one percent can be tested. A tablet solution of the strength given has not only the advantage over the solution previously recom- mended* (5 tablets to 50 cc. of water) of showing the per- cent of acidity directly, without tables or calculations, but being weaker, the unavoidable errors of determination are decreased by its use. Equally accurate results may be obtained by using solu- tions made up according to method a or method 5, explained * 111. exp. sta., bull. 32 ; Wis. exp. sta., bull. 52. Testing the Acidity of Milk and Cream. 103 in the preceding. Thie latter method (17.6 cc. cream, 5 tablets per 97 cc. of water) has, however, the advantage in point of economy of apparatus, since a 17.6 cc. pipette is found in creameries and dairies with the Babcock test out- fit and is therefore most likely already available for use in testing the acidity of cream. This method is therefore con- sidered preferable and referred to as 124. The standard solution. The preparation of this solution is as follows: Five tablets are added to the 100 cc. cylinder which is filled to the 07 cc. mark with clean soft water, tightly corked, shaken and laid on its side, as the tablets will dissolve more quickly when the cylinder is placed in this position than when left in an upright position with the tablets at the bottom. Several cylinders con- taining the tablet solution may be prepared; as soon as one is emptied, tablets and water are again added, and the cylinder is corked and placed in a horizontal posi- tion. In this way fresh solutions ready for testing are always at hand. The cylinder is kept tightly corked, while the tablets are dissolving, so that none of the liquid is lost by the shaking. It is well to put the tablets in the cylinder with water at night; the solution will then be ready for use in the morning. Excepting a flocculent residue of inert matter, " settlings," which will not dissolve, the tablets must all disappear in the solution before this is used. The strength of the tablet solution does not change perceptibly by standing for twenty-four hours; but a change takes place in solutions more than a day old. The solid tablets will not change if kept dry. The only precaution necessary is to use a fresh solution when acidity tests are made. 125. Making the test. The cream to be tested is thor- oughly mixed, and 17.6 cc. is measured into the cup. The I04 Testing Milk and Its Products. pipette is rinsed once with water, and the rinsings added to the cream in the cup. A few cc. of the tablet solution pre- pared as given above are now poured from the cylinder into the cream, and mixed thoroughly with it by giving the cup a gentle rotary motion. The tablet solution is added in small quantities until a permanent pink color appears in the sam- ple. The number of cc. of tablet solution which have been used to color the cream is now found from the scale of the cylinder. In comparing the results of one test with another, the same shade of color should always be adopted. The most delicate point is the first change to a uniform pink color which the sample shows when the acid contained there- in has just been neutralized. This shade of color is easily recognized with a little practice. The pink color is not per- manent unless a large excess of alkaline solution has been added, on account of the influence of the carbonic acid of the air (116), and the operator should not therefore be lead to believe by the reappearance of the white color after a time, that the point of neutralization was not alread}^ reached when the first uniform shade of pink was observed. 126. Acidity of cream. 17.6 cc. of sweet cream is gen- erally neutralized by 15-20 cc. of this tablet solution, rep- resenting from .15 to .20 percent, of acid. A mild sour cream is colored by 35 cc. tablet solution, and a sour cream ready for churning, by about 50 cc. tablet solution. As the cream ripens, its acidity increases. The rate of ripening depends largely on the temperature at which the cream is kept. Cream containing .5 to. 6 per cent, of acid will make such butter as our American market demands at the present time. Cream showing an acid test of .55 per cent, may not Testing the Acidity of Milk and Creafu. 105 ^z be too sour, but .65 per cent, of acid is very near, if not on the danger line, since such cream is likely to make strong flavored, almost rancid butter. Each lot of cream should be tested as soon as it is read}^ for ripening, and the result of the test will show whether the cream should be warmed or cooled in order to have it ready for churning at the time desired. Later tests will show the rate at which the ripen- ing is progressing, and the time when the cream has reached the proper acidity for churning. 127. Spillman's cylinder. The graduated cylinder, shown in fig. 35, was devised by Prof. Spillman of Wash- ington experiment station, for use in testing the acidity of milk and cream with Farring- ton's alkaline tablets. The following direc- tions are given * for making tests with this piece of apparatus. " All that is needed in addition to the acid test graduate shown in the accompanying illustration is a common prescription bottle of six or eight ounce capacity, and a package of Farringtons alkaline tablets. Fill the bottle with water and add one tablet for each ounce of water in the bottle. Shake the bot- FiG 35 s iii-^ ^^^ frequently to aid in dissolving the tablets, man's cylinder, " Maldiig the test. In making the test, the used in determin- .,j, t../>iit^^i i ing the acidity of ^^^^ ^cst graduate IS filled to the zero mark cream or miiic. ^i^\^ the milk or crcam to be tested. The tab- let solution is then added, a little at a time, and the graduate shaken after each addition, in order to thoroughly mix the milk and the tablet solution. In shaking the * Wasliington experiment station, bulletin No. 24. io6 Testino Milk and Its Products. "ti graduate, give it a rotary motion to prevent spilling any of the liquid. Continue adding the tablet solution until a per- manent pink color can be detected in the milk. The level of the liquid in the graduate, measured by the scale on the graduate, will then be the percent, of acidity of the milk. It is best to stand the graduate on a piece of white paper, so that the first pink coloration of the milk may be easily detected." 128. Rapid estimation of the acidity of appar= ently sweet milli or cream, a. Milk. The alkaline tab- let method offers a ready means of estimating the acidity of fresh milk or cream that is still apparently sweet. The selection of the best kinds of milk is especially important in pasteurizing milk or cream. Investigations have shown that milk which gives the highest acid test contains, as a rule, a larger number of bacteria and spores, not destroyed by pasteurization, than does milk giving a low acid test; the acidity test may therefore be used to advantage for the purpose of selecting milk best adapted for pasteurization, as well as such as is to be retailed or used in the manufacture of high-grade butter and cheese. In distinguishing milk fit for pasteurization purposes from that which is doubtful, an arbitrary standard of two- tenths of one percent of acid may be taken as the upper limit for milk of the former kind. The apparatus used in making this test is shown in the accompanying illustra- tion, (fig. 36), and consists of a white tea cup; either a four, six or eight ounce bottle, and a No. 10 brass cartridge shell, or a similar measure. A solution of the tablets in water is first prepared, one tablet being always added for each ounce of water: four tablets in a four ounce bottle; six, in a six Testing the Acidity of Milk mid Cremn. 107 ounce bottle, etc., the amount of tablet solution prepared depending on the number of tests to be made at one time. The bottle is filled up to its neck with clean soft water, and the solution prepared in the manner previously given (124). The manner of operating the test is as follows: SOunceBoUle. "Measure Fig. 36. Apparatus used for rapid estimation of the acidity of apparently sweet milk or cream. 129. Operating the test. As each lot of milk is brought to the creamery in the morning and poured into the weigh car, it is weighed, and the cartridge-shell dipper filled with the milk; this is then poured into the white cup. The same or another No. 10 shell is now filled twice with the tablet solution, and emptied into the milk in the cup. Instead of dipping twice with one measure or a No. 10 shell, a tin measure can be made holding as much as two No. 10 shells. The liquids are then mixed in the cup by giving this a quick rotary motion. The color of the mixture is now io8 Testing Milk and Its Products. noticed. If the milk remains white it contains more than two-tenths of one percent of acid, and should not be used for pasteurization. If it is colored after having been thor- oughly mixed with two measures of tablet solution, it con- tains less than this amount of acid and may be safely used, as far as acidity goes, for pasteurization or any other pur- pose for which it is necessary to have thoroughly sweet milk. The shade of color obtained will vary with different lots of milk ; the sweetest milk will be most highly colored, but a milk retaining even a faint pink color with two measures of tablet solution to one measure of milk contains less than .2 percent acid. By proceeding in the manner described, the man at the factory weigh can is able to test the acidity of the milk de- livered nearly as quickly as he can weigh the milk; and ac- cording to the results of the test he will send the milk to the general delivery vat, or to the pasteurization vat, as the weighing can may be provided with two conductor spouts. 130. Size of measure necessary. It is not necessary to use a No. 10 shell for a measure in working the preceding method; one of any convenient size that can be filled ac- curately and quickly, will answer the purpose equally well,. if a measure of the same size is used for- both the sample and the tablet solution. When this is done, each measure- ful of tablet solution made up as directed, will represent one-tenth percent of acid in the sample tested. 131. b. Cream. Cream can be tested in the way already described for testing the acidity of fresh milk, by adding ta one measureful of cream in the cup, as many measures of tablet solution as are necessary to change the color of the cream when the two liquids are thoroughly mixed. If one Testing the Acidity of Milk and Creaijt. 109 measure of tablet solution colors one measure of cream, this contains less than .1 percent acid; if five measures of tablet solution are required, the cream contains about .5 percent acid, etc. By proceeding in the manner described, the ope- rator can estimate the acidity to within .05 per cent, of acid, if half measures of tablet solution used are observed. The results thus obtained are sufficiently delicate for all practical purposes. 132. Detecting preservaline * in milk. The tablet solution furnishes a simple method of detecting preservaline in milk. The application of the alkaline tablets for this purpose was first dis- cussed in bulletin No. 52 of Wisconsin experiment station. The acidity of the milk is increased by the addition of preservaline, but neither the odor nor taste of the milk is affected thereby. By ad- ding to sweet milk the amount of preservaline which the manu- facturers claim will keep it sweet for 36 hours, its acidity may be increased to .35 per cent., in a sample of milk which before adding the preservaline tested perhaps only .15 per cent. acid. As before stated, unadulterated milk will usually smell or taste sour, or "turned", when it contains .30-.35 per cent, acid; milk testing as high as this limit, which neither smells nor tastes sour in any way, is therefore in all probability adulterated with pre- servaline or some other preparation containing boracic acid, or a similar compound. 133. "Alkaline Tabs." These are not alkaline tab- lets, but a substitute which was put on the market by a New York firm. The outfit furnished consisted of four packages of paper discs made of filter paper, each of about the size of an old-style copper cent; two packages of square paper; one glass of about 10 cc. capacity, and one small * Preservaline is the trade name of an antiseptic extensively advertized for preserving milk and cream. It consists essentially of boracic acid and borax, the use of which in milk and other dairy products oflfered for sale, is prohibited by law in many states. no Testing Milk and Its Products. glass bottle. The directions stated that each paper disc represented .1 per cent, acidity when added to the small glassful of milk or cream, with two of the square papers, the whole to be well shaken in the long glass bottle. The acidity of the sample of milk or cream was claimed to be measured by counting the number of round papers required to produce a pink color in the sample tested. An investigation of the reliability of these " Tabs " soon disclosed the fact that they were entirely inaccurate, and that no dependence could therefore be put on the results obtained by their use. A report of the comparative work done in testing the acidity of milk or cream by a one-tenth normal alkali solution and these " Alkaline Tabs " was pub- lished in the dairy press in 1895, to which reference is here made as to the details sf the results obtained.* Hoard's Dairyman, Sept. 6, 1895. Testing the Purity of Milk. Ill Fig. 37. Students operating the Wisconsin curd test. CHAPTER YIII. TESTING THE PURITY OF MILK. 134. The Wisconsin curd test. Cheese makers are often troubled with so-called floating or gassy curds which produce cheese defective in flavor and texture. These faults are usuall}^ caused by some particular lot of milk containing 112 Testing Milk and Its Products. impurities that cannot be detected by ordinary means of in- spection. The Wisconsin curd test is used to detect the ^source of these defects and thus enable the clieese maker to exclude the milk from the particular farm or cow to which the trouble is traced. This test is similar in princi- ple to tests that have for many years past been in use in cheese-making districts in Europe, notably in Switzerland,* but was worked out independently at the Wisconsin Dairy School in 1895 and has become generally known as the Wis- consin curd test from the description of it in the report of Wis. experiment station for 1895.t The method of operat- ing this test is as follows (see fig. 37). 135. Method of making the test. Pint glass jars, thor- oughly cleaned and sterilized with live steam, are provided; they are plainly numbered or tagged, one jar being pro- vided for each lot of milk to be tested. The jars are filled about two-thirds full with the milk from the various sources (it is not necessary to take any exact quantity); they are then placed in a tank or vat containing water, which is heated until the milk in the jars has a temperature of 98° F. The thermometer used must not be transferred from one sample to another, unless special precautions are taken, for fear of contaminating the pure lots of milk by impure ones. When the milk has reached a temperature of 98°, add to each sample, 10 drops of rennet extract, and mix by giving the jar a rotary motion. The milk is thus curdled, and the ourd allowed to stand for about twenty minutes until it is firm. It is then cut fine with a case knife, and after settling, the whey is poured oflf. The best tests are made when the separation of the whey is most complete. By allowing the * Herz, Unters. d. Kuhmilch, Berlin, 1889, p. 87. t Twelfth report, p. 148. Testing the Purity of Milk. 113 samples to stand for a short time, more whey can be poured off, and the curd thereb}^ rendered firmer. The water around the jars is kept at a temperature of 98°, the vat is covered, and the curds allowed to ferment in the sample jars for six to twelve hours. During this time the impurities in any particular sample will cause gases to be developed in the curds so that by examining the same carefull}^, by smelling of them, and cutting them with a sharp knife, those having a bad flavor or a spong}' or in any way abnormal texture may be easil}^ detected, and the lot of milk from which it was made, thereby picked out. 136. By proceeding in the same way with the milk from the different cows in a herd, the mixed milk of which pro- duced abnormal curds, the source of contamination in the herd may be located. Very often the trouble will be found to come from the cows' drinking foul stagnant water or from fermenting matter in the stable. In the former case the pond or marsh must be fenced off, or the cows kept away from it in other ways; in the latter, a thorough cleaning and disinfection of the premises are required. If the milk of a single cow is the source of contamination, it must be kept by itself, until the milk is again normal; under such conditions the milk from the healthy cows may of course safel}^ be sent to the factory. 137. The fermentation test. The Gerber fermenta- tion test (see fig. 38) furnishes a convenient method for discovering the cause of abnormal fermentations which show themselves in tainted, pin-holey, gassy, or floating curds, and is also useful in examining the purity of different lots of milk. The test consists of a tin tank which can be heated 114 Testing Milk and Its Products. by means of a small lamp, and into which a rack fits, hold- ing a certain number of cylindrical glass tubes; these are all numbered and provided with a mark and a tin cover. In making the test, the tubes are filled to the mark with milk, the number of each tube being recorded in a note book, opposite the name of the particular patron whose milk was placed therein. The tubes in the rack are put in the tank, which is two-thirds full of water; the temperature of the Fig. The Gerber fermentation test. water is kept at 104-106° F. for six hours, when the rack is taken out, the tubes gently shaken, and the appearance of the milk, its odor, taste, etc., carefully noted in each case. The tubes are then again heated in the tank at the same temperature as before, for another six hours, when observa- tions are once more taken of the appearance of the milk in each tube. The tainted milk may then easily be discovered, by the abnormal coagulation of the sample. Testing the Purity of Milk. 115 According to Gerber,* good and properly handled milk should not coagulate in less than twelve hours, when kept under the conditions described, nor show anj^thing abnor- mal when coagulated. Milk from sick cows and from cows in heat, or with diseased udders will always coagulate in less than twelve hours. If the milk does not curdle inside of a day or two, it should be tested for preservatives (240). * Die praktische Milch-Pruefung; Woll, Handbook f. Farmers and Dairymen^ pp. 253-5. ii6 Test in or Milk and Its Products. c> CHAPTER IX. TESTING MILK ON THE FARM. 139. Variations in milk of single cows. The varia- tions in the tests of milk of single cows from milking to milking, or from day to day, are greater than many cow- owners suspect. There seems to be no uniformity in this variation, except that the quality of the milk produced generally improves with the progress of the period of lacta- tion; even this may not be noticeable, however, except when the averages of a number of tests made at different stages ■during the lactation period are compared with each other. When a cow gives her maximum quantity of milk, shortly after calving, the quality of her milk is generally poorer (by one percent of fat, or less) than when she is drjdng off". Strippers' milk is therefore as a rule richer in fat than the milk of fresh cows. 140. By testing separately every milking of a number of cows through their whole periods of lactation, the results obtained have seemed to warrant the following conclusions in regard to the variations in the tests of the milk from single cows, and it is believed that these conclusions allow of generalization.* 1. Some cows' milk tests about the same at every milking. Such cows generally give a uniform quantit}^ of milk from day to day. * Illinois experiment station, bulletin 24. Testing Milk on the Far^n. 117 2. Other cows give milk that varies in an unexplainable way from one milking to another. Neither the morning nor the evening milking is alwa5's the richer, and even if the interval between the two milkings is exactly the same, the quality as well as the quantity of milk produced will vary considerably. Such cows are mostly of a nervous, excitable temperament, and are easily affected by changes in feed, drink, or surrounding conditions. 3. The milk of a sick cow, or of a cow in heat, generally tests higher than when the cow is in a normal condition; the milk yield generally decreases under such condition; marked exceptions to this rule have, however, been observed. 4. Starved or underfed cows may give milk testing higher than when the cows are properly nourished, probably on account of an accompanying feverish condition of the ani- mal. The milk is, however, more generally of an abnor- mally low fat content, which may be readily increased to the normal percent of fat in the milk by liberal feeding. 5. Fat is the most variable constituent of milk, while the solids not fat vary within comparatively narrow limits. The summary of analysis of 2400 American samples of milk cal- culated by Cooke * shows that while the percentage of fat varies from 3.07 to 6.00 percent, or nearly three percent^ that of casein and albumen varies only from 2.92 to 4.30 percent, or less than one and one-half percent, and the milk sugar and ash content increases but little (about .69 percent), as the milk grows richer, within the range given. 6. A test of only one milking may give a very erroneous impression of the average quality of a certain cow's milk. A composite sample (see below) taken from two or more * WoU, Handbook for Farmers and Dairymen, p. 195. ii8 Testing Milk and Its Products. successive milkings will more nearl}^ represent the quality of the milk which a cow produces at the time of the samp- ling. 141. The variations that may occur in testing the milk of single cows, are illustrated by the following figures ob- tained in an experiment made at the Illinois experi- ment station,* in which the milk of each of six cows was weighed and analyzed daily during the whole periods of lactation. Among the cows were pure-bred Jerseys, Shorthorns, and Holsteins, the cows being from 3 to 8 years of age, and varying in weight from 850 to 1350 lbs. During a period of two months of the year, the cows were fed a heavy grain ration consisting of 12 lbs. of corn and cob meal, 6 lbs. of wheat bran, and 6 lbs. of linseed meal, per day per head. This course of feeding was tried for the purpose of increasing, if possible, the richness of the milk. The influence of this heavy grain feed, as well as that of the first pasture grass feed, on the qualit}^ and the quantity of the milk produced is shown in the following table which gives the complete average data for one of the cows (No. 3). The records of the other cows are given in the publication re- ferred to; they were similar to the one here given, in so far as variations in quality are concerned. Bulletin No. 24. Testing Milk on the Farm. no Average results obtained in weighing and testing a cow's milk daily during one i^eriod of lactation. '53 ^ . la 3 Daily mi yield. Ik Tests of one day's milk. Yield of fat per day. Month. 1. < ^ CO 1^ go bo 2 h 920 927 1035 1047 1054 1079 1105 1180 1130 12.1 ItJ.O 16.1 14.3 13.8 14.5 12.1 9.3 6.4 16.0 17.7 17.7 16.0 16.5 17.2 14.0 12.2 9.3 10.0 14.0 13.5 12.5 11.5 10.0 9.2 6.0 3.5 3.8 3.7 3.6 3.8 4.0 3.8 3.9 4.2 4.7 4.9 4.6 5.8 4.7 5.8 4.6 4.6 6.2 7.9 3.0 2.7 3.2 3.4 3.0 3.4 3.2 2.8 2.9 .46 .59 .58 .54 .55 .55 .47 .39 .30 .60 .76 .84 .61 .72 .70 .57 .60 .50 .34 January .44 .51 Marc h .50 April .46 May .44 June .35 July .27 August .16 14-i. The average test of this cow's milk for her whole period of lactation was 3.8 per cent, of fat (i. e. the total quantity of fat produced -- total milk yield, X 100) ; the milk of the cow twice during this time tested as high as 5.8 per cent., and once as low as 2.7 per cent., while tests of 3.0 and 4.6 per cent, were obtained a number of times. The aver- age weight of milk produced per day by the cow was 1-1 lbs. ; this multiplied by her average test, 3.8, shows that she pro- duced on the average .53 lbs., or about one-half a pound of butter fat per day during her lactation period. If, how- ever, her butter-producing capacity had been judged by the test of her milk for one day only, this test might have been made either on the day when her milk tested 5.8 per cent., or when it was as low as 2.7 per cent. Both of these tests were made in midwinter when the cow gave about 16 lbs. of milk a day. Multiplying this quantity by 5.8 gives .93 lbs. of fat, and by 2.7 gives .43 lbs.' of fat. Either result might show the butter fat produced by the cow on certain days, but neither gives a correct record of her actual average daily performance for this lactation period. I20 Testing Milk and lis Products. A sufficient number and variety of tests of the milk of many cows have been made to prove that there is no defi- nite regularity in the daily variations in the richness of the milk of single cows. The only change in the quality of milk common to all cows is, as stated, the natural increase in fat content as the cows are drying off, and even in this case the improvement in the quality of the milk sometimes does not occur until the milk yield has dwindled down very materially. 143. Causes of variations in fat content. The quality of a cow's milk is as a rule decidedly influenced by the fol- lowing conditions: Rough treatment. Exposure to rain or rough weather. Change of feed. Change of milkers. Rapidity of milking. Length of interval between milkings. Unusual excitement or sickness. 144. Disturbances like those enumerated frequently in- crease the richness of the milk for one, and sometimes for several milkings, but a decrease in quality follows during the reaction or the gradual return to normal conditions, and taken as a whole, there is a considerable falling oflf in the total production of milk and butter fat by the cow, on ac- count of the nervous excitement which she has gone through. Aside from changes due to well-definable causes, like those given above, the quality of some cow's milk will often change very considerably without any apparent cause. The dairyman who is in the habit of making tests of the milk of his individual cows at regular intervals, will have abundant material for study in the results obtained, and he will soon Testing Milk on the Farm. 121 be able to tell from the tests made, if these are continued for several days, whether or not the cows are in a normal healthy condition, or have been subjected to excitement or abuse in any way. 145. Number of tests required during a period of lactation in testing cows. The daily records of the six cows referred to on p. 118 give data for comparing their total production of milk and butter fat during one period of lac- tation, as found from the daily weights and tests of their milk, with the total amount calculated from weights and tests made at intervals of 7, 10, 15 or 30 days. The aver- ages of all results obtained with each of the six cows show that weighing and testing the milk of a cow every seventh day, gave 98 per cent, of the total milk and butter fat, which according to her daily record was the total product. Tests made once in two weeks gave 97.6 per cent, of the total milk, and 98.5 per cent, of the total butter fat, and tests made once a month, or only ten times during the period of lactation, gave 96.4 per cent, of the total milk, and 97 per cent, of the total production of butter fat. 146. The record of one of the cows will show how these calculations are made: It was found from the daily weights and tests that cow No. 1, in one lactation period of 307 days, gave 5,04-1 lbs. of milk which contained 254 lbs. of butter fat. Selecting every thirtieth day of her record as testing day, the total production of milk and fat is shown to be as follows: 122 Testing Milk and Its Pi'oducts, Production of milk and butter fat per day. Testing day. Weight of milk. Test of milk. Yield of butter fat. Nov. 4 lbs. 20.5 18.7 17.7 20.0 18.2 19.5 17.7 13.1 12.2 3.2 per cent. 4.7 4.6 4.9 4.5 4.7 4.4 4.8 5.5 6.2 7.2 lbs. .96 Dec. 4 Tan. 3 .86 .86 Feb 2 .90 Mar. 3 .86 April 2 .81 May 2 June 1 .85 .72 July 1 July 31 .76 .23 Total, 159.7 lbs. 15.97 lbs. 7.81 lbs. Average per day. 4.89 .78 lbs. The average daily production of the cow, according to the figures given in the preceding table, was nearly 16 lbs. of milk, containing .78 lbs. of butter fat. Multiplying these figures by 307, the number of days during which the cow was milked, gives 4903 lbs. of milk and 240 lbs. of fat. This is 141 lbs. of milk and 14 lbs. of fat less than the total weights of milk and butter fat, as found by the daily weights and tests, or 2.8 and 5.5 percent less, for milk and fat pro- duction, respectively. This is, however, calculated from only ten single weights and tests, while it required over 600 weighings and 300 tests of the milk to obtain the exact amount, Similar calculations from the records of the other cows gave fully as close results, showing that quite satisfactory records of the total production of milk and butter of a cow may be obtained by making correct weighings and tests of her full day's milk once every thirty days. Testing Milk on the Farm. 123 147. When to test a cow. The Vermont experiment station for several j^ears made a special stud}^ of the ques- tion when a cow should be tested in order to give a correct idea of the whole 3'ear's performance, when only one or two tests are to be made during the lactation period.* The results obtained may be brief!}- summarized as follows: a. As to quality of milk j^roduced. If two tests of each cow's milk are to be made during the same lactation period, it is recommended to take composite samples at the inter- vals given below. First sample. Second sample. For spring cows, 6 weeks after calving, 6V2-7V^ mos. after calving summer " 8 " " "16 -7 mos. " fall "8-10" " " 151/2-7 mos. If only one test is to be made, approximately correct re- sults may be obtained by testing the milk during the sixth month from calving, in case of spring cows; during the third to fifth month in case of summer calving cows, and during the fifth to seventh month for fall calving cows. In all cases composite samples of the milk for at least four da3'S should be taken (162). " The test of a single sample, drawn from a single milking or day, will not of necessity, or indeed, usually, give trustworthy results." This method of obtaining the average composition of the milk produced during a lactation period is naturally more correct in case of mixed herd milk, than when single cows are tested. b. As to quantity of milk produced. The milk may be weighed for four days in the middle of the month, and the entire month's yield obtained with considerable accuracy (barring sickness and drying off), by multiplying the sum by * Sixth report, 1892, p. 106; Ninth report, 1895, p. 1' 124 Testing Milk and Its Products. 7, 7|- or 7f , according to the number of days in the different months. The weighing is most readily done by means of a spring balance, the hand of which is set back so that the empty pail brings it to zero (fig. 39). If several pails are to be used, they should first be made to weigh the same by putting a little ■THE DAIRYMENi'S SUPPLY C0i4^ PHIL A. PA. ''^'^^^' Fig. 39. Milk scale. Fig. 40. Automatic milk scale for weighing and registering the milk of individual cows. solder on the lighter pails. Milk scales which weigh and automatically register the yield of milk from twenty cows have recently been placed on the market (see fig. 40); the pail is hung on the hook, and by pressing the button show- ing the number of the particular cow, the weight is recorded on the milk sheet. Testing Milk on the Farm. 125 148. H. B. Gurler, the well-known Illinois dairyman, suggests* a method of determining the total production of a cow during one lactation period from the test of her milk for one week onl3\ This plan is, however, only recom- mended to those who cannot or do not care to take the neces- sary time to make a more reliable test. The test should be made after the cow has been in milk for about three months. It is necessary that the cow should be producing a normal quantity and qualit}'- of milk at the time when the the test is made; if, e. g., a cow is fresh March 1, and tested June 1, she is probably on good pasture feed, and produces more milk and butter fat than at any previous or subsequent time. She should, therefore, be tested either before she is put on pasture or after the luxuriance of the pasture is gone, Mr. Gurler suggests that the milk be weighed for a week, and a composite sample taken of the milkings. At the end of this time it may be found that the cow gave 154 lbs. of milk during the week, and the com- posite sample tested 4.3 percent fat; the cow, therefore, pro- duced ^.^ lbs. of butter fat during the week, or on the average, .94 lbs. per day. This average yield is now multi- plied by 252, the number of days in 8.4 months, and a yield- of 237 lbs. of butter fat is obtained. This will very nearly represent the total production of fat by the cow during the whole period of lactation. It is assumed in this calculation that the cow gives milk more than 8.4 months, and that what is produced beyond this time will bring the production during the last 2.4 months up to the same average per month as in the first six months. American Dairyiog.p. 18. 126 Test in o- Milk and Its Products, l-il). Record of tests. Where tests of the single cows in a herd are made regularly, a complete record should be kept in a note book arranged in about the manner shown on the following blank. RECORD OF TESTS OF COWS. MONTH , 189 Date. Name of cow. Pounds of milk. ! No. of 1 test bottle. Percent. of butter fat Yield of butter fat pounds. Remarks. 150. Sampling milk of single cows. In sampling the milk of single cows, all the milk obtained at the milking must be thoroughly mixed, b}^ pouring it from one vessel to another for a few times, or stirring it thoroughly by means of a dipper moved up and down, as well as horizon- tally in the pail or can in which it is held; the sample for testing purposes is then taken at once. A correct sample of a cow's milk cannot be obtained by milking directly into a small bottle from one teat, or by filling the bottle with a little milk from each teat, or by taking some of the first, middle, and last milk drawn from the udder. Such samples cannot possibly represent the quality of the milk of one entire milking, since there is almost as much diiference be- tween the first and the last portions of a milking, as between skim milk and cream.* Lack of care in taking a fair sample is the cause of many surprising results obtained in testing milk of single cows. 151. When a cow is to be tested, she should be milked dry the last milking previous to the day when the test is to * Woll, Handbook for Farmers and Dairymen, p. 194; Agricultural Science, pp. 540-42. Testifig Milk on the Farm. 127 be made. The entire quantity of milk obtained at each milking is mixed and sampled separatel3^ On account of the variations in the composition of the milk, a number of tests of successive milkings must be made. As this involves considerable labor, the plan of taking composite samples is preferable; the method of composite sampling and testing is explained in detail under the second subdivision of Chapter X (162) ; suffice it here to say that the method followed in case of single cows' or herd milk, is to take about an ounce of the thoroughly mixed milk of each milking; this is placed in a pint or quart fruit jar containing a small quantity of some preservative, preferably about one-half a gram of powdered potassium bi-chromate. If a number of composite samples of the milk of single cows are taken, each jar should be labeled with the number or name of the particular cows. Composite tests are generally taken for four days or for a week. If continued for a week, the jars will contain at the end of this time a mixture of the milk of fourteen milk- ings. The composite sample is then carefully mixed by pouring it gently a few times from one jar to another, and is tested in the ordinary manner. The result of this test shows the average quality of the milk produced by the cow during the time the milk was sampled, 152. As the amount as well as the quality of the milk produced by single cows vary somewhat from day to day, and from milking to milking, it is quite important in testing single cows, especially when the test includes only a few days, to take a proportionate part (an aliquot) of each milk- ing for the composite test sample. This is easily done by means of a Scovell sampling tube, the use of which is ex- plained in another place (166). I2{ Testino Milk and Its Products. 153. 5ize of the testing sample. Four ounces is a sufficient quantity for a sample of milk if it is desired to determine its percent of fat only; if the milk is to be tested with a lactometer, when adulteration is suspected, as much as a pint is needed for a sample. If this sample of milk is put into a bottle and carried or sent away from the farm to be tested, the bottle should be completely filled with milk, to prevent a partial churning of butter in the sample during transportation (25). 154. Variations in herd milk. While considerable variations in the quality of milk of single cows are often met with, a mixture of the milk of several cows, or of a whole herd, is comparatively uniform from day to day; the individual differences tend to balance one another so that variations when they do occur, are less marked than in case of milk of single cows. There are, however, at times marked variations also in the test of herd milk on successive days; the following figures from the dairy tests conducted at the World's Columbian Exposition in Chicago in 1893 will show the correctness of this statement. The test included twenty- five cows each of the Jersey and the Guernsey breeds, and twenty-four of the Shorthorn breed. Tests of herd milk on successive days. Date. Jersey. Guernsey. Shorthorn. Tulv 16. 1893 4.8 percent. 5.0 4.7 4.6 5.0 4.6 percent. 4.5 4.4 4.6 4.5 3.8 percent. Tulv 17 1893 3.8 July 18 1893 3.8 July 19, 1893 July 20, 1893 3.7 " 3.8 Testing Milk on the Farm. 129 On July 17, 1893, the mixed milk of the Jersey cows tested two-tenths of one percent higher than on the preceding day; the Guernsey herd milk tested one-tenth of one per- cent lower, while the Shorthorn milk did not change in composition; comparing the tests on July 19 and 20, we find that the Jersey and Shorthorn milk tested four- tenths and one-tenth of one percent higher, respectively, on the latter day than on the former, and the Guernsey milk tested one- tenth of one percent lower. 155. Ranges in variations of herd milk. According to Fleischmann,^ herd milk on single days may vary from the average values for the year, as follows, expressed in per- cent of the latter: The specific gravity (expressed in degrees) may go above or below the yearly average by more than 10 percent. The percent of fat content may go above or below the yearly average by more than 30 percent. The percent "of total solids ^may go above or below the yearly average by more than 14 percent. The percent of solids not fat may go above or below the yearly average by more than 10 percent. To illustrate, if the average test of a herd during a whole period of lactation is 4.0 percent, the test on a single day may exceed 4.0 + i^^o'o X 4.0 = 5.2, or may go below 2.8 per- cent, (viz. 4.0 — n)"o X 40); if the average specific gravity is 1.031 (lactometer degrees 31),t the specific gravity of the milk on a single daj' may var}- between 1.0279 and 1.0341 (31 + iJo"o X 31 = 34.1 ; 31 — iV'o X 3.1 = 27.9. * Book of the Dairy, p. 32. t See page 81, 9 130 Testing Milk and Its Products. 156. Influence of heavy grain feeding on the quality of milk. If cows are not starved or underfed, an incrase in the feeding ration will not materially change the richness of the tnilk produced, as has been shown by careful feeding experiments, conducted under a great variety of conditions and in many countries. Cows that are fairly well fed will almost invariably give more milk when their rations are in- creased, but the milk will remain of about the same quality after the first few days are passed as before this time, pro- vided the cows are in good health and under normal condi- tions. Any change in the feed of cows will usually bring about an immediate change in the fat content of the milk, as a rule increasing it to some extent, but in a couple of days wh^n the cows have been accustomed to their new feed, the fat content of the milk will again return to its normal amount. 157. The records of the cows included in the feeding experiment at the Illinois station, to which reference has been made on p. 118, furnish illustrations as to the effect of heavy feeding on the quality of milk. The feed as well as the milk of the cows were weighed each day of the experi- ment; during the month of December each cow was fed a daily ration consisting of 10 lbs. of timothy hay, 20 lbs. of corn silage and 2 lbs. of oil meal; the table on p. 119 shows that cow No. 3 produced on this feed, on the average, 12.1 lbs. of milk, testing 3.8 percent of fat. In January the grain feed was gradually increased until the ration consisted of 12 lbs. ot timothy hay, 8 lbs. of corn and cob meal, 4 lbs. of wheat bran, and 4 lbs. of oil meal. All cows gained in milk on this feed ; cow No. 3 thus gave an average of 4 lbs. more milk per daj^ in January than December, but the average test of her milk was 3.7 percent, or one-tenth of one percent Testing Milk on the Farm. 131 lower than during the preceding month. The heavy grain feeding was continued through February and March, when it reached 12 lbs. of timothy hay, 12 lbs. of corn and cob meal, 6 lbs. of wheat bran and 6 lbs. of oil meal per day. The records show that the flow of milk kept up to 16 lbs. per day in February, in case of this cow, but fell to 14 lbs. in March and April, the average test of the milk being, in February, 3.6; in March 3.8, and in April, 4.0 percent. The milk was, therefore, somewhat richer in April than in December, but not more so than is found normally, owing to the progress of the period of lactation. 158. Influence of pasture on the quality of milk. On May 1, the cows were given luxuriant pasture feed and no grain; a slight increase in the average amount of milk produced per day followed, with a reduction in the test, this being 3.8 percent, — the same as in December. During all these changes of feed, there was, therefore, not much change in the richness of the milk, while the flow of milk was increased by the heavy grain feeding for several months, as well as by the change from grain feeding in the barn to pasture feed with no grain.* 159. The increase in the amount of butter produced by a cow, which has often been observed as a result of a change in feed, doubtless as a rule, comes from the fact that more, but not richer milk is produced. The quality of milk which a cow produces is as natural to her as is the color of her hair, and is not materially changed by any special system of normal feeding.f ♦For further data on this point, see Cornell (N. Y.) exp. sta., bulletins 13, 22, 36 and 49; N. D. exp. sta., bull. 16; Kansas exp. sta., report 1888; Hoard's Dairy- man, 1896, pp. 921-5. f On this point almost endless discussions have in recent years taken place in the agricultural press of this and foreign countries, and the subject has been under 132 Testing Milk and Its Products. 160. Method of improving the quality of rnXWu. The quality of the milk produced by a herd can generally be im- proved by selection and breeding, i. e., by disposing of the cows giving poor milk, say below 3 percent of fat, and by breediog to pure-bred or high-grade bulls of a strain that is known to produce rich milk. This method cannot work wonders in a day, or even in a year, but it is the only cer- tain way which we have to improve the quality of the milk produced by our cows. It may be well in this connection to call attention to the fact that the quality of the milk which a cow produces is only one side of the question; the quantity is another, and equally important one. Much dissatisfaction and grumbling about low tests among patrons of creameries and cheese factories would be stopped if this fact was more generally borne in mind. A cow giving 3 percent milk should not be condemned because her milk does not test 5 percent; she may give twice as much milk per day as a 5 percent cow, and will therefore produce considerabl}^ more butter fat. debate at nearly every gathering of farmers where feeding problems have been con- sidered. Many farmers are firm in their belief that butter fat can be •' fed into" the milk of a cow, and would take exception to the conclusion drawn in the pre- ceding. The results of careful investigations by our best dairy authorities point conclusively, however, in the direction stated, and the evidence on this point is overwhelmingly against the opinion that the fat content of the milk can be mate- rially and for any length of time increased by changes in the system of feeding. The most conclusive evidence in this line is perhaps the Danish co-operative cow feeding experiments, conducted during the p'sst ten years, with over 2,000 cows in all. The conclusion arrived at by the director of the Copenhagen experiment sta- tion, und ,'r whose supe-'vision and direction the experiments have been conducted has been stated over and over again in the published reports of the station: that the changes of feed made in thediffdrent lots of cows included in the experiments have had practically no influenc j on the chemical composition (the fat content) of the milk produced. In these experiments grain feeds have been fed against roots, against oil cake, and against wheat bran or shorts; grain and oil cakes have further- more been fed against roots, and roots have been given as an additional feed to the standard rations tried, — in all cases with tl\e same negative results as far aa changes iu the fat contents of the milk produced are concerned. Testing Milk on the Farm. 133 The point whether or not a cow is a persistent milker is also of primary importance; a production of 300 lbs. of butter fat during a whole period of lactation is a rather high dairy standard, but one reached by many herds, even as the aver- age for all mature cows in the herd. Dairymen should re- member that a high production of butter fat in the course of the whole period of lactation is of more importance than a very high test. 134 Testing Milk and Its Products. CHAPTER X. COMPOSITE SAMPLES OF MILK. 161. Shortly after milk testing bad been introduced to some extent in creameries and cheese factories, it was sug- gested by Patrick, then of Iowa experiment station,* that a great saving in labor, without a coincident diminution in the accuracy of the results, could be obtained by mixing the daily samples of milk from one source, and testing this mix- ture, instead of each sample contributing thereto. Such a mixture is called a composite sample. The usual methods of taking such samples at creameries and cheese factories dur- ing the past few years have been as follows: 162. Methods of taking composite samples, a. Use of tin dipper. Either pint or quart Mason fruit jars, or milk bottles provided with a cover, are used for receiving the daily samples. One of these jars is supplied for each patron of the factory and is labeled with his name or number. A small quantity of preservative (bichromate of potash, bichlorid of mercury, etc., (see 172) is added to each jar; these are placed on shelves, or somewhere within easy reach of the operator inspecting and weighing the milk as it is re- ceived at the factory. When all the milk delivered by a patron is poured into the weighing can and weighed, a small portion thereof, usually about an ounce, is put into the jar labeled with the name or number of the patron. The samples * Bulletin No. 9, May 1890. Composite Samples of Milk. 135 are conveniently taken by means of a small tin dipper holding about an ounce. This sampling is continued for a certain number of days, a week, ten days, or sometimes two weeks, a portion of each patron's milk being added to his particular jar every time he delivered milk. Each of these composite samples are then tested; this test takes the place of separate daily tests, and gives accurate information regarding the average quality of the milk delivered by each patron during the period of sampling. The weight of butter fat which each patron brought to the factory in his milk during this time, is obtained by multiplying the total weight of milk delivered during the sampling period b}" the test of the composite sample. 163. This method of taking composite samples has been proved to be practically correct. It is absolutely correct onl}^ when the same weight of milk is delivered daily by the particular patron. If this is not the case, the size of the various small samples should bear a definite relation to the milk delivered ; one-sixteen hundredth, or one-two thousandth of the amount of milk furnished should, for instance, be taken for the composite sample from each lot of milk. This can easily be done by means of special sampling devices (see 165). As the quantities of the milk delivered from day to day by each patron vary but little, perhaps not exceed- ing 10 percent of the milk delivered, the error introduced by taking a uniform sample, e. g., an ounce of milk, each time is, however, too small to be worth considering in fac- tory work, and the method of composite sampling described is generally adopted in separator creameries, and in cheese factories, where the payment of the milk is based on its quality. 136 Testing Milk mid Its Products. 164. By this method of composite sampling each lot of rich, medium or thin milk receives credit for the amount of butter fat which it contains, and complications that might arise from testing only one day's milk at irregular intervals are avoided. In order to obtain reliable results by com- posite sampling it is essential that each lot of milk sampled shall be sweet and in good condition, containing no lumps of curdled milk or small butter granules churned out. The milk is of course always evenly mixed before the sample is taken. 165. b. Drip sample. Composite samples are sometimes taken at creameries and cheese factories by collecting the milk that drips through a small hole or tube placed in the conductor spout through which the milk runs from the weighing can to the receiving vat or tank. A small portion of the drip is then each day placed in the composite sample jar, or the quantity of drip obtained is regulated so that all of it may be taken. In the latter case the quantity of milk delivered will enter into the composite sampling as well as its quality, and the sample from, say 1000 lbs. of milk will be twice as large as the sample from 500 lbs. of milk. Where it is desired to vary the size of the sub-samples, according to the quantity of milk delivered from day to day, it is necessary to adopt the method of collecting drip sam- ples, just explained, or to make use of special sampling de- vices, like the " milk thief," a Scovell sampling tube, etc. The principle of both these tubes is the same, and it will be suflScient to describe here only one. 166. c. The Scovell sampling tuhe.^ This convenient de- vice for sampling milk (fig. 41), consists of a drawn copper Kentucky experiment station, 8th report, pp. xxvi-xxxii. Composite Samples of Milk. 137 or brass tube, one-half to one inch in diameter; it is open at both ends, the lower end sliding snugly in a cap provided with three elliptical openings at the side, through which the milk is admitted. The milk to be sampled is poured into a C3^1indrical pail, or the factory weighing can, and the tube, with the cap set so that the apertures are left open, is lowered into the milk until it touches the bottom of the can. The tube will be filled instantly to the level of the milk in the can, and is then pushed down, thereby closing the apertures of the cap and confining within the tube a column of milk representing exactl}^ the quality of the milk in the can, and forming an aliquot part thereof. The milk in the sampling tube is then emptied into the composite sample jar by turning the tube upside down. 167. If the diameter of the sampling pail used is 8 inches, and that of the sampling tube ^ inch (these dimensions will be found convenient Fig 41 ^^ sampling milk from single cows), then the sampH^tJg'^ube quantity of milk secured in the tube will always stand in the ratio to that of the milk in the pail, of {\) ■ to 8 '^, * that is, very nearly 1:256; no matter how much or how little milk there is in the pail, the sample will repre- sent 256 part of the milk. For composite sampling of the milk of single cows, this proposition will prove about right; if more milk is wanted for a sub-sample, the milk to be *The contents of a cylinder are represented by the formula Trr^h, r being the radius of the cylinder, and h its height. The relation between two cylinders of the same height, the radii of which are R and r, is, therefore, as TfR^h to ;Tr'h, or as R^ to r-. 138 Testing Milk and Its Products. sampled may be poured into a can of smaller diameter. If the mixed milk from a number of cows is to be sampled, a wider sampling can is used. By adjusting the diameters of the tube and the can, any desired proportion of milk can be obtained in the sample. For factory sampling, with a 26-inch weighing can, a tube three-quarters of an inch in diameter will be found of proper dimensions. 168. The sampling tube will furnish a correct sample of the milk in the can, even if this has been left standing for some time; it is better, however, to take out the sample soon after the milk has been poured into the can, as the possible error of cream adhering to the side of the sampling tube is then avoided. 109. The accuracy of the sampling of milk by means of the Scovell tube was proved beyond dispute in the breed tests conducted at the World's Columbian Exposition in 1893, in which tests this method of sampling the milk pro- duced- by the single cows, and the different herds was adopted.* The data obtained in these breed tests also fur- nish abundant material proving the accuracy of the Babcock test. In using any one of these tubes, the size of the sample is regulated by the amount of milk in the sampling can, as the milk always rises to the same height in the tube as in the can. In all cases cylindrical sampling cans must be used. 170. Composite sampling with a " one=third sam= pie pipette." Milk is sometimes sampled directly from the weighing ca,n into the Babcock test bottle by means of a * Kentucky experiment station, 8th report, pp. xxx-xxxi. Another form of a milk sampling tube in use at the Iowa experiment station is described and illus- trated by Mr. Eckles, in Breeder's Gazette, May 19, 1897. Composite Samples of Milk. 130 pipette holding 5.87 cc, which is one-third the size of the regular pipette. This quantity is measured into the test bottle from three successive lots of milk, and the test then made in the ordinary manner. In this way one test shows the average composition of the milk delivered during three successive days, or deliveries. When this method is adopted, as many test bottles are provided as there are patrons; there is no need of using any preservatives for the milk in this case. Fig. 42 shows a convenient rack for holding the test bottles used in composite sampling with a "one-third sample pipette." Accurate results can be obtained by this method of sam- pling, if care is taken in measuring out the milk, and if it is not frozen or contains lumps of cream. It is doubtful if the method has any advantage over the usual method of composite samp- ling. If milk is delivered daily and each lot is sampled with the one-third pipette, twice or three times the num- ber of tests are required as when composite sam- ples are taken and tested once every week, or every ten days. This method fur- thermore takes more time in the daily sampling than the latter, as the ;i|fiw|Mjjiw liiiliijiiliiiiilii MMMMWS Fig. 42. Test bottle rack for use in creameries and cheese factories. 140 Testing Milk and Its Products, quantity of milk must be measured out accurately each time. If the test bottle is accidentally broken, or some milk spilled, the opportunity of ascertaining the fat con- tent of the milk delivered during the three days is lost; if a similar accident should occur in testing composite samples collected in jars, another test can readily be made. 171. Accuracy of the described methods of sam= pling. An experiment made at the Wisconsin Dairy School may here be cited, showing that concordant results will be obtained by the use of the drip sampling method and the Scovell tube. Two composite samples were taken from fifty different lots of milk, amounting to about 6000 lbs. in the aggregate. One sample was taken of the drip from a hole in the conductor spout through which the milk passed from the weighing can; the other was taken in the weighing can by means of a Scovell sampling tube. The following percentages of fat were found in each of these samples: Bahcock test. Gravimetric analysis. Drip composite vsaiiiple 4.0 percent. 4.04 percent. Scovell tube composite sample.. 4.0 " 4.06 " Preservatives tor Composite Samples. When milk is kept for any length of time under ordin- ary conditions, it will soon turn sour and become lop- pered, and further decomposition shortly sets in, which renders the sampling of the milk both difficult and unsatis- factory (19). The changes which occur are due to the forma- tion of lactic acid through the action of bacteria on milk sugar: the acid in turn coagulates the casein of the milk, but does not destroy or attack the butter fat (27). The period dur- ing which milk will remain in an apparently sweet, or fresh condition varies, with the temperature at which it is kept. Coni-posite Samples of Milk. 141 and with the cleanliness of the milk, from less than a day to a week or more; milk will not generally remain sweet longer than two days at the outside, at ordinary summer or room temperature. In order to preserve composite samples of milk in a proper condition for testing, some chemical which will check or prevent the fermentation of the milk must be added to it. A number of substances have been proposed for this purpose. 172. Bi=chromate of potash. Of these, bichromate of potash is to be preferred, in the opinion of the authors, on account of its relative harmlessness, its cheapness and efficiency. The bi-chromate method for preserving samples of milk was proposed by Mr. J. A. Alen, city chemist of Gothenbuig, Sweden, in 1892,* and has been generally adopted in dairy regions in this countr}" and abroad. While not perfectly harmless, the bi-chromate is not a violent poison like other chemicals proposed for this purpose, and no accidents are liable to result from its use; at least none have been known to the writers to occur during the years that it has been used in creameries or dairies as a preserv- ing agent. 173. The quantity of bi-chromate necessary for preserv- ing half a pint to a pint of milk for a period of one or two weeks is about one-half gram (nearly 8 grains). As there are about 900 half-grams in a pound, this quantity will suffice for nine weeks for a creamery having one hundred patrons, if tests are made once a week, or for three months (90 days), if tests are made every ten da3^s. According to Winton and Ogden,t a .22-inch pistol cart- * Biedermann's Centralblatt, 1892, p. 549. t Connecticut experiment station report 1884, p. 222. 1^2 Testing' Milk and Its P^'odticts. ridge shell cut to \ inch long, or a .32-inch calibre shell cut to \ inch long will hold, when loosely filled, enough of pow- dered bi-chromate to preserve \ pint, a .32-inch calibre shell cut to \ inch long will hold enough to preserve one pint. These shells may be conveniently handled by soldering on to them a piece of stifl wire to serve as a handle. The amount of bi-chromate placed in each composite sample jar would fill about half the space representing one percent in the neck of the Babcock milk test bottle. 174. The first portions of milk added to the composite sample jars containing the specified amount of bi chromate will be colored almost red, but as more milk is added, day by day, its color will become lighter yellow. The complete sample should have a light straw color; such samples are most easily mixed with acid when tested. If more bi-chromate is used, the solution of the casein in the acid is rendered diflScult and calls for more persistent shaking. Bi-chromate can be bought at drug stores or from dairy supply dealers at about 30 cents a pound, and will cost about 25 cents a pound at wholesale. Powdered bi-chromate of potash should be ordered, and not crystals, as the latter dissolve only slowly in the milk. Farrington's bi-chromate tablets contain the correct quantity of preservative for a quart sample, and will be found convenient; price per 1,000 tablets, $2.00. 175. Other preservatives for composite samples. Among other substances recommended for use in butter or cheese factories as milk preservatives for composite samples are boracic acid compounds, formalin, chloroform, carbon bi- sulfid,* copper-ammonium sulfate, sodium fluorid, ammonia- *Delaware experiment station, eighth report, 1896, which also see for trials with a large number of different preservatives. Composite Samples of Milk. 143 glycerin (sp. gr. 1.031), and mixtures containing mercuric chlorid (corrosive sublimate) with anilin color (rosanilin).* The coloring matter in the latter compounds is added to give a rose color to the sample preserved, thus showing that the milk is not fit for consumption; the bi-chromate giving naturally a yellow color to the milk, renders the addition of any special coloring matter unnecessary. None of the substances mentioned are as cheap as bi- chromate or more effective for the purpose for which they are used, when milk is to be kept not to exceed two or three weeks. The compounds containing corrosive sublimate are violent poisons, and must always be handled with the greatest care, lest they get into the hands of children or persons un- familiar with their poisonous properties; they will preserve the milk longer than bi-chromate when applied in sufficient quantities, but for factory use the latter is amply effective, and has, as already stated, the advantage in several other respects. 176. Care of composite samples. The composite sample jars should be kept covered to prevent loss by evaporation, and in a cool place. They should be kept in the dark, in a special closed closet, or at least out of direct sun- light; the chromic acid formed by the reducing influence of light on chromate solutions produces a leathery cream which is very difficultly dissolved in sulfuric acid. A coating of white shellac will protect the labels of the composite sample jars so that they may be used for a long time, allowing the jars to be washed and cleaned after each period of testing. The shellac is applied after the names of the patrons have been written on the labels, and these have * Iowa experiment station, bulletins 9, 11, 82. 1 44 Testing Milk and Its Products. been put on the jars. Gummed labels, 1x2^ in., answer this purpose weir. In keeping the milk from day to day, care should be taken that the cream forming on the milk does not stick to the sides of the jars in patches above the level of the milk. Unless the daily handling of the jars, and the addition of fresh portions of milk, be done with sufficient care, the cream will become lumpy, and will dry on the sides of the jars. In some cases it is nearly impossible to evenly distribute this dried cream through the entire sample so as to make the composite sample a true representative of the different lots of milk from which it has been taken. 177. Every time a new portion of milk is added to the jar it should be given a gentle horizontal rotary motion, thereby mixing the cream already formed in the jar with the milk, and rinsing off the cream sticking to its side. This manipulation also prevents the surface of the milk from be- coming covered with a layer of partially dried leathery cream. Composite samples having patches of dried cream on the inside of the jar are the result of carelessness or ignorance on the part of the operator. If a little careful attention is given to the daily handling of the composite samples, the cream which is formed in the jars can be evenly mixed again with the milk without difficulty, 178. Fallacy of averaging percentages. A compos- ite sample of milk should represent the average quality of the various lots of milk of which it is made up. This will invariably be true if a definite aliquot portion or fraction of the different lots of milk is taken. If the weights of, say ten different lots of milk are added together, and the sum divided by ten, the quotient will represent the average Composite Samples of Milk. H5 weight per lot of milk, but the average obtained in this way of the tests of the different lots, may not be the correct aver- age test of the entire quantity of milk. The accuracy of such an average figure would depend on the uniformity in the composition and weights of the ten lots of milk. When there is no uniformity, the weights of the different lots of milk as well as their tests, must be considered. An average of a number of weights can be calculated directly, but not average percentages. The following example illustrates the difference between the arithmetical average of a number of single tests, and the true average test of the various lots. Methods of calculating average percentages. I. Milk varying in ueighh and tests. II. Milk of uniform weights and tesls. Lot. Weight of milk. Test of milk. Weight of fat. Lot. Weight of milk. Test of milk. Weight of fat. I lbs. 120 ."STO H60 55 82 per ct. 8.5 5,0 5.2 3.0 4.0 lbs. 4.2 28.5 18.7 1.6 3.2 I lbs. 250 225 240 238 234 per ct. 4.2 4.0 4.3 4.1 4.4 lbs. 10.5 II . II 9.0 Ill Ill 10.3 IV IV 9.7 v. .. V 10.3 Total Total 1187 2.37 56.2 11.24 1187 237 49.8 Average True av'ge test 4.14 4.73* Average True av'ge test 4.20 4.22t 10.0 56.2 X mo 1187 4.73. f 49.8 X 100 1187 4.22. 179. The figures given in the table show that when the different lots of milk vary in test and weight, as in the first case, the correct average test of the 1187 lbs. of milk is not found by dividing the sum of these tests by five, which would give 4.14 percent; but the percentage which 56.2 (the total amount of fat in the mixed milk, in lbs.) is of 1187 10 146 Testing Milk and Its Products. (the total amount of milk, in lbs.) is 4.73, and this is the correct average test of the mixed milk made up of the five different lots. In the second case, the variations in both the weights of the different lots of milk, and their tests, are comparatively small, and both methods of calculation give therefore practically the same average test; but also in this case, the correct average test is found by dividing the total amount of fat by the total quantity of milk, making 4.22 percent, instead of 4.20 percent, which is the arithmetical mean of the five tests, the quantities of milk in the various lots do not enter into the calculation of the latter.* 180. The second example represents more nearly than the first one the actual conditions met with at creameries and cheese factories. As a rule the mixed milk from a herd of cows does not vary more in total weight or tests, within a short period of time like one to two weeks, than the fig- ures given in this example. On account of this fact, sam- ples taken, for instance, with a small dipper may give per- fectly satisfactory results to all parties. If the different lots of milk varied in weight and test from day to day, as shown in the first case, it would be necessary to use a " milk thief " or a Scovell sampling tube for taking the composite samples, as the size of the samples taken would then repre- sent an exact aliquot portion of each lot of milk (166). 181. A patron's dilemma. The following incident which occurred at the Wisconsin Dairy School creamery during the past winter, will further explain the difficulties met with in calculating the average tests of various lots of milk. * In the experiment given on p. 122, the arithmetical mean of the tests given is 5.15 percent, while the true average fat content of the milk is 4.89 perceiit. Composite Samples of Milk. 147 The weekly composite sample of the milk supplied by a patron of the creamery from his herd of 21 cows tested 4.0 percent. One day the farmer brought to the creamery a sample of the morning's milk from each of his cows, and had them tested; after adding the single tests together, and dividing the sum by 21, he obtained an average figure of 5.1 percent of fat. From this he concluded that the aver- age test of the milk from his cows ought to be 5.1, instead of 4.0, and naturally asked for an explanation. 182. The first thing done was to show him that while 5.1 was the correct average of the figures representing the tests of his twenty-one cows, it was not a correct average test of the mixed milk of all his cows, as he had not con- sidered in calculating this average, the quantities of milk yielded by each cow; the following illustration was used: Cow No. 1. yield 25 lbs. of milk, test 3.6 percent. = 0.9 lbs. of butter fat. Cow No. 2, yield 6 Ibg. of milk, test 5.0 percent, = 0.3 lbs. of butter fat. Total 31 lbs. 2)8^ 1.2 lbs. 4.3 percent. The two cows gave 31 lbs. of milk containing 1.2 lbs. of fat; the test of the mixed milk would therefore not be 4.3 percent (^^^|^), but ^^^ = 3.87 percent. If the fat in the mixed milk was calculated by the average figure 4.3 percent, 1.33 lbs. of fat would be obtained, i. e., .13 lbs. more than the cows produced. In order to further demonstrate the actual composition of the mixed milk of the twenty-one cows, the milk of each cow was weighed and tested at each of the two milkings of one day. The weights and tests showed that the cows pro- duced the following total number of pounds of milk and fat: 148 Testing Milk and Its Products. Morning milking, 113.3 lbs. of milk, containing 5.17 lbs. of fat. Night milking, 130.9 lbs. of milk, containing 4.98 lbs. of fat. The morning milk contained ' ^ — =4.56 percent of fat, and the night milk ^'^^^^^^ =3.80 percent of fat. The average of 21 tests of morning milk was 4 8 percent, and of 21 tests of night milk, 3.8 percent. The sum of the morning and night milkings gave: milk 244.2 lbs. fat 10.15 lbs. The mixed morning and night milk, therefore, con- tained 10-^5X10 0^^;^ percent of fat. This is the true aver- 244.2 age test of the morning and night milkings of these twenty- one cows, as found by weighing and testing separately the milk of each cow at both milkings. 183. The total milk was strained into a large can at the farm, both in the morning and in the evening. A sample of the mixed milk was taken in each case with a long- handled dipper as soon as the milkings were finished. When the cans of milk were delivered at the creamery, a sample of each was taken with a Scovell sampling tube. The tests of these four samples are given below, together with the re- sults from the individul tests. Morning milk. Night milk. Sample taken at the farm, with dipper.. 4.4 percent. 3.8 percent. Sample taken at creamery, with Scovell tube 4.5 percent. 3.7 percent. Calculated from weights and tests of milk from each cow 4.5 percent. 3.8 percent. The figures given show that practically uniform tests were obtained by the different methods of sampling. The sum of the weights of the milk from the diflerent cows was as follows: Composite Samples of Milk. 149 Morning milk. Nitjhl milk. Daily milk . Total milk produced 118.3 lbs. 130.9 lbs. 244.2 lbs. Milk in samples 12.3 lbs. 8.9 lbs. 21.2 lbs. Milk for family use 2.5 lbs 2.5 lbs. Milk taken to creamery 98.5 lbs. 122.0 lbs. 220.5 lbs. It has already been shown from the weights and tests of each cow's milk that the herd milk contained 4.1 percent of fat. Multiplying the total milk delivered at the creamery, 220.5 lbs., by 4.1 gives 9.04 lbs. of fat. The morning and night milkings, which were weighed and tested separately, contained the following quantities of butter fat. Morning milk 98.5 lbs. X test 4.5 = 4.43 lbs. of butter fat. Night milk 122.0 lbs. X test 3.8 = 4.63 lbs. of butter fat. Total 220.5 lbs. 9.06 lbs. By weighing, sampling and testing separately the morn- ing and night milkings of twenty-one cows, deducting the weight of milk in the samples, and what was taken out for family use, it was found that 9.04 lbs. of butter fat was sent to the creamery. The weights and tests of this same milk when delivered at the creamer}^, gave 9.06 lbs. of butter fat. 184. This example furnishes an excellent illustration of the accuracy of the Babcock test, and of the closeness of results which may be obtained at creameries when proper care is taken in weighing, sampling and testing the milk. Similar demonstrations may be made by any factory operator, and with equally satisfactory results, provided that the work is carefully done. 150 Testing Milk mid Its Products, CHAPTER XL CREAM TESTING AT CREAM=GATHER1NQ CREAflERIES. 185. The cream delivered at gathered cream factories is now in many localities tested by the Babcock test, and this has been adopted as a basis of paying for the cream in the same manner as milk is paid for at separator creameries. It has been found to be more satisfactory to both cream buyer and seller, than either the oil test churn or the space (or gauge) systems which have been used for this purpose in the past. The details of the application of the Babcock test to the practical work at cream-gathering creameries have been carefully investigated by Winton and Ogden in Connecticut,* Bartlett in Maine,t and Lindsey in Massachusetts; X and we also owe to the labors of these chemists much information concerning the present workings of other systems of paying for the cream delivered at creameries. 186. The space system. Numerous tests have shown that one si^ace or gauge of cream does no't contain a definite, uniform amount of fat. In over 100 comparisons made by Winton it was found that one space of cream I contained *Conn. experiment station, (New Haven) bull. No. 108 and 119; report 1894, pp. 214-244. t Maine experiment station, bull. 3 and 4 (S. S). J Hatch experiment station, report 1894, pp. 92-103; 1895, pp. 67-70. §The space is the volume of a cylinder, SJ^ inches in diameter and gf of an inch high. The number of spaces in each can of milk is read off before skimming by means of a scale marked on a strip of glass in the side of the can. (Conn. exp. sta., bull. No. 119). Cream Testing' at Creameries. 151 from .072 to .170 lbs. of butter fat, or on the average, .13 lbs., and the number of spaces required to make one pound of butter varied from 5.01 to 11.72. It is also claimed that in the winter season when the cream is gathered at long inter- vals, as once a week, it is necessary for the buyer to accept the seller's statement of the record of the number of cream spaces which he furnishes, since the cream cannot be left in Fig. 43. The oil test churn. the creaming cans for so long a time. These objections to the space system apply only to the method of paying for the cream, and not the manner in which the cream is obtained. 187. The oil=test churn. As stated in the introduction, the oil-test churn (fig. 43), has been used quite extensively among gathered cream factories; this system is based on the 152 Testing Milk and Its Products. number of creamery inches of cream which the various patrons deliver to the factory; one inch of cream contains 113 cubic inches (i. e., a layer of cream one inch deep in a 12-inch pail; two inches in an 8-inch pail contains 100.531 cubic inches, two inches in an 8f-inch pail 110.18 cubic in- ches, and two inches in an 8|-inch pail 113.490 cubic inches). The driver pours the patron's cream into his 12- inch gathering pail, and with his rule measures and records the depth of the cream in the can in inches and tenths of an inch. The cream is then stirred thoroughly with a ladle or a stout dipper, and a sample is taken by filling a test tube from the oil test churn outfit, to the graduation mark by means of a small conical dipper provided with a lip. A driver's case contains either two or three " cards," holding fifteen test tubes each. The tubes as filled are placed in the case, and the corresponding number is in each instance re- corded in front of the patron's name together with the number of inches of cream furnished by him. On the arrival at the creamery the tin cards holding the tubes are placed in a vessel filled with water of the temper- ature wanted for churning (say, 60° in summer, and 65 to 70° in winter). When ready for churning they are placed in the oil-test churn, the cover of the churn put on, and the sam- ples churned into butter. On the completion of the churn- ing, the cards are transferred to water of 175-190° Fahr., where they are left for at least ten minutes to melt the butter and " cook the butter milk into a curd." The oil will now be seen mixed all through the mass. The test tubes are then re-tempered to churning temperature and churned again, by which process the curd is broken into fine par- ticles, which, when the butter is re-melted, will settle to the bottom. The butter is melted after the second churning Cream Testing at Creameries. 153 by placing the tubes in water at 150-175° F., allowing them to remain therein for at least twenty minutes. If a clear separation of oil does not take place, the sample must some- times be churned three or four times until a good separation is obtained. A clear separation of the fat in oil test churn tubes is often helped by adding a little sulfuric acid to them. The length of the column of liquid butter fat thus separ- ated is determined by means of a special rule for measuring the butter oil; this shows the number of pounds and tenths of a pound of butter which an inch of cream will make; the first tenth of a pound on the rule is divided into five equal parts, so that measurements may be made to two- hundredths of a pound. The melted fat is measured with the rule, by raising the tin card holding the bottles, to about the height of the eye; the reading is recorded on the driver's tablet under Test per inch, opposite the number of the particular patron. The test per inch multiplied by the inches and tenths of an inch of cream supplied will give the butter yield in pounds, with which the patron will be credited on the books of the creamery. 188. The objection to this system of ascertaining the quality of cream delivered by different patrons lies in the fact that it determines the churnable fat, and not the total fat of the cream; the amount of the former obtained is de- pendent on many conditions beyond the control of the pat- ron, viz: the acidity and the temperature of the cream, the size of the churn or churning vessel, etc. The same reasons which caused the churn to be replaced by methods of deter- mining the total fat of the milk, in the testing of cows among dairymen and breeders, have gradually brought about the abandonment of the oil test in creameries, and the adoption of the Babcock test in its place. 154 Test in o Milk and Its Products. 189. The Babcock test for cream. Both the space system and the oil test churn used for estimating the quality of cream at creameries have now largely been replaced by the Babcock test in the more progressive creameries in this country, and composite samples of cream are collected and tested in a similar manner as is done with milk at separator creameries and cheese factories. A very satisfactory method of arrangements for working the Babcock test, which is in use in many Eastern cream- eries, is described by Winton and Ogden in the Connecticut report, previously referred to. The cream gatherer who collects the cream in large cream cans is supplied with a spring balance (1, see fig. 44), pail for sampling and weigh- ing the cream (2), sampling tube (3), and sample bottles (5). At each patron's farm he takes from his wagon the sampling pail and tube, the scales, and one small collecting bottle. He should find in the dairy of the patron the cans of perfectly sweet cream, kept at a temperature of 40° to 50° F., and protected from dirt and bad odors. Either sour or frozen cream must be rejected. The patron's number should be painted in some conspicuous place near the cream cans in his dairy house. The gatherer hangs the scale on a hook near the cream to be collected; the scale should be so made that the hand of the dial will stand at zero when the empty pail is hung on it. The cream is then poured at least twice from one can to another in order to mix it thoroughly.* * The necessity of care in mixing the cream is shown by the following illustra- tion given hy the authors referred to : Percent of fat in cream which stood for 24 hours. j^'ot mixed Surface. 28.0 Bottom. 5.0 22.0 Sample drawn with sampling tube 19.25 23.75 22.50 Poured twice 22.25 Cream Testino- at Crecnuer 1CS. 155 11 8 7 6 5 Fig. 44. Outfit for cream testing by the Babcock test at gathered cream factories. 190. When properly mixed, the cream is poured into the weighing pail, and is weighed and sampled. The authors give the following description of the cream samp- ling tube used, and directions for sampling and weighing the cream. 156 Testing Milk and Its Products. ''Sampling T«6e.— This tube, devised by Mr. Ogden, is of stout brass, about ^^ of an inch thick, and a few inches longer than the weighing pail which is used with it. On the upper end, a small brass stop-cock of the same bore is fastened. It should be nickel plated inside and out, to keep the metal smooth and free from corrosion. These tubes maybe obtained from less than f^ to over 14 inch bore. The greater the diameter of the weighing pail, the wider should be the bore of the tube. For use with pails 8 inches in diameter, a y% i^^'b bore sampling tube will serve the purpose, but when the pail has a diameter of 9 or more inches, a tube with a bore of ^A inch or more should be used. It must be borne in mind that doubling the diameter of the pail, or of the sampling tube, increases the capacity fourfold. The tube when not in use should be kept in an upright position to permit draining. ''Sampling and Weighing. — Lower the sampling tube with the cock open, to the bottom of the weighing pail which holds the mixed cream. When it is filled, raise it out of the liquid and allow it to drain for a few seconds. By this. means the tube is rinsed with the cream to be sampled and any traces of cream adhering to the tube from previous use are removed. With the cock still open, slowly lower the sampling tube to the bottom of the cream pail. After allowing a moment for the cream to rise in the tube to the same height as in the pail, close the cock and raise the sampler carefully out of the cream. As long as the cock is closed, the cream in the tube will not flow out, unless the tube is strongly jarred. Allow the cream adhering to the outside of the tube to drain off for a few seconds, then put the lower end into the 1 to IV2OZ. wide-mouth glass collecting bottle which bears the patron's number on its cork, and open the cock. The cream will then flow out of the sampler into the bottle, which is afterwards securely corked and put into the cream-gatherer's case. Immediately weigh the cream in the cream pail to the quarter or half pound, as may be judged expedient, and record the weight. " If the patron has more than one pailful, repeat with each pail- ful the operation of sampling and weighing, putting all the sam- Ci'cam Testing- at Creameries, 157 pies in one and the same bottle. Weigh, all cream collected, in one and the same sampling pail and draw a sample from each separate portion v\reighed.^' 191. After sampling and weighing each patron's cream it is poured into the driver's large can, and the sample bot- tles are carried in a case to the creamery where the con- tents of each are poured into the composite sample jar of the particular patron. The accompanying illustration (fig. 45) shows an arrangement for keeping the composite sample jars in the testing room of the creamer}^ 192, The samples of cream in the small bottles, besides furnishing the means of testing the richness of the cream, Fig. 45. Case for holding composite sample bottles at creamery. 158 Testing Milk and Its Products. give the creamery owner or manager an opportunity to in- spect the flavor of each lot of cream, and the condition in which it has been kept by the various patrons. Potas- sium bi-chromate is placed in the composite sample jars, and these are cared for and tested in the same manner as composite samples of milk (176). 193, The collecting bottles should be cleaned with cold, and afterwards with hot water, as soon as they are emptied, and before a film of cream dries on them. When washed and dried, these bottles are placed in the cases, ready for the next collecting trip. There can be no confusion of bot- tles since the corks and not the bottles are marked with the numbers of the respective patrons. 194:. When cream is bought by this system of testing composite samples, the patrons are paid for the number of butter fat contained in their cream, in exactly the same way as milk is paid for at separator creameries. It makes no difference how thick or how thin the cream may be, or how much skim milk is left in the cream when brought to the factory. Eighty pounds of cream containing 1 5 percent of fat will bring no more or less than 48 pounds of cream testing 25 percent; in either case 12 pounds of pure butter fat is delivered, which will make the same amount of butter in both cases, viz: toward 14 lbs., and both patrons should therefore receive the same amount of money. There is a small difference in the value of the two lots of cream to the creamery owner or the butter maker, in favor of the richer cream, both because its smaller bulk makes the transportation and handling expenses lighter, and because slightly less butter fat will be lost in the butter milk, a smaller quantity of this being obtained from the richer cream. But it is doubtful if the differences thus occurring Cream Testing at Creameries, 15^ are of sufficient importance to be noticed under ordinary creamery conditions; the example selected presents an ex- treme case of variation in the fat content of cream. A trial of this system at five Connecticut creameries, supplied mostly with Cooley cream, by over 175 patrons, showed that the average composition of the cream from the different patrons varied only from 16.9 to 19.8 percent fat. The cream of some patrons on certain days contained only 9.5 percent of fat, and other patrons at times had as high a test as 30 per- cent., but these great differences were largely evened up when the average quality of the cream delivered during a period of time, like a month or more, was considered. 195. Smaller differences in the composition of cream will, however, always occur, even if the same system of set- ting the milk, like the cold deep-setting process, is used, and the water is kept at the same temperature at all times. This is due to differences in the composition of the milk and its creaming qualit}^; whether largel}' from fresh cows or from late milkers; whether kept standing for a time before being set or submerged in the creamer immediately after milking and straining; diameter of creaming cans, etc. Bartlett states^ that the percentage of fat in the cream from the same cows may be increased ten percent or more by keeping the water at 70° instead of at 40° F. The higher temper- ature will give the richer cream, but the separation will not be so complete, since a richer skim milk is obtained from the milk set at the higher temperature. Separator cream is not materially influenced by the conditions mentioned, as the separator can be regulated to deliver cream of nearly uniform richness from all kinds of sweet milk. * Maine experiment station, bulletin No. 3 (S. S). i6o Testing Milk and Its Products. CHAPTER XII. CALCULATION OF BUTTER AND CHEESE YIELD. A. — Calculation of Yield of Butter. 196. Butter fat test and yield of butter. The Bab- cock test shows the amount of pure butter fat contained in a sample of milk or other dairy products. The butter ob- tained by churning cream or milk contains, in addition to pure butter fat, a certain amount of water, salt and curd. While an accurate milk test gives the total quantity of but- ter fat which there is in the sample of milk or cream tested, the churn cannot be depended upon either to leave the same amount of butter fat in the butter milk or to include the same amount of water, salt and curd in the butter at each churning. If a quantity of milk, say 3000 lbs., be thoroughly mixed in a vat, and then divided into half a dozen equal portions, a Babcock test of the different lots will show the same per- centage of butter fat in each portion. If, on the other hand, each of these lots be skimmed, and the cream ripened in different vats and churned separately, the same weight of butter from each lot of 500 lbs. of milk will not be obtained, even by the most expert butter maker, or if all the opera- tions of skimming, cream ripening, churning, sailing and butter-working were made as nearly uniform as possible. Careful operators can handle the milk and cream so that very nearly the same proportion of the fat contained in the Calculation of Butter and Cheese Yield. i6i milk is recovered in the butter in different churnings, but since the water in butter is held mechanicall}-, and is not chemically combined with it, the amount retained by the butter is quite variable in different churnings, especially since the laws governing the retention of water in butter are but imperfectly understood. 197. Variations in the composition of butter. As an illustration of the variability of butter in its composition, the analyses of the butter made in the breed tests at the World's Fair in 1893, may here be cited; the butter was in all cases made by as nearly identical methods and under as uniform conditions as could possibly be obtained by the skilled operators having this work in charge; the average composition of 350 samples of this butter, with upper and lower limits, was as shown in the following table. Composition of samples of butter, World's Fair, 1893. Water. Fat. Curd. Salt and ash. Sum of water, curd, salt and ash. Average of 350 analy- ses Percent. 11.57 8.63-15.00 Percent. 84.70 76.53-88.26 Percent. .95 .50-2.14 Percent. 2.78 1.01-8.58 Percent. 15 30 Lower and upper lim- its Analyses of fifty samples of creamery butter taken from the tubs ready for market at as many Wisconsin creameries, in 1896, indicated that no two of them were exactly alike in composition, but varied within the limits given below.* * Wisconsin experiment station, bull. 56. 11 l62 Testing Milk and Its Products. Summary of analyses of Wisconsin creamery butter. Water. Butter fat. Curd. Salt and ash. Sum of water, salt and curd. Highest. Lowest . Average Percent. 17.03 9.18 12.77 Percent 87.50 77.07 83.08 Percent. 2.45 0.36 1.28 Percent. 4.73 1.30 2.87 Percent. 22. 95 12.50 16.92 The preceding analyses show the composition of butter made at one place where every possible effort was taken to produce a uniform product, and of butter made at fifty dif- ferent creameries, where there was more or less variation in the different operations of manufacture, and in the appli- ances and machinery used. The majority of the samples of butter analj^zed were, in either case, naturally of, or very near, the average composition given, but since there is such wide variations in the composition of the butter made by the uniform methods adopted in the World's Fair breed tests, butter of a more uniform composition cannot be expected from the thousands of different creameries and private dairies which supply the general market with butter. The analyses of the fifty samples of creamery butter, given above, show that the content of the butter fat varied from 77 to over 87 percent, and according to the average of the analyses, 83 pounds of butler fat was contained in, or made, 100 lbs. of butter. There was, therefore, in this case produced 20.5 percent more butter than there was butter fat, since 83:100 : :100:x; therefore 100X100 83 120.5. Calculation of Butter and Cheese Yield. 163 198. "Overrun " of churn over test. The yield of butter is not, however, as a rule compared with the amount of butter fat contained in the butter, but with the total but- ter fat of the whole milk from which it was made. This "increase of the churn over the test" is what is generally called the overrun in creameries. The overrun obtained in different creameries, or even in the same creameries at different times, will be found to vary considerably. When the milk is accurately tested, and the butter well worked, this overrun will vary from 10 to 16 per- cent; that is, if a quantity of milk contains exactly 100 lbs. of butter fat, as found by the Babcock test or any other ac- curate method of milk testing, from 110 to 116 lbs. of butter ready for market may be made from it. Variations will occur in the speed of the separator, in the conduct of the ripening and the churning processes, and in the condition of the butter when the churn is stopped, even under the very best of care and attention to details; and abso- lutely uniform losses of fat in skim milk and butter milk, or the same water content of the butter cannot, therefore, be expected. 199. Factors influencing the overrun. The overrun is influenced by two factors: the losses of butter fat sus- tained in separating the milk and churning the cream, and the gain due to the admixture of water, salt, etc., in the manufacture of butter. Considering first the losses of fat in skim milk and butter milk, the separator usually, when run at normal capacity, will leave the same percent of fat in skim milk, whether rich or poor milk is skimmed; an excep- tion to this may be found in separating rich milk having large fat globules, or milk from fresh milkers, in either of 164 Testing Milk and Its Products. which cases the large size of the fat globules occasions a more complete separation of fat by the centrifugal force. But generally speaking, the statement holds good that the total loss of fat in separator skim milk is a factor of the quantity of milk run through the separator, rather than of its quality. It follows from this, however, that the relative losses of fat in skim milk will vary to some extent according to the quality of the milk separated. Selecting two extremes in the quality of milk, 2.5 and 6.0 percent of fat, there will be found, say .2 percent of fat in the skim milk from either lot, provided the separator is not unduly crowded, and the separation is con- ducted under normal conditions in either case. But .2 per- cent fat makes 8 percent of the total fat in the poor milk /— ^ 00 = 8), and only 3 percent of that in the rich milk. It takes 4000 lbs. of the 2.5 percent milk to furnish 100 lbs. of fat, and only 1666 lbs. of the 6 percent milk; in skim- ming the poor milk, a loss of .2 percent of fat is sustained in the skim milk from 4000 lbs. of milk, while in the rich milk a similar loss is sustained in the skim milk from only 1666 lbs. of milk. The example gives an extreme case, and one not likely to be met with in practice. The range in the richness of the milk delivered by different patrons at the factory is usually within one-half a percent of fat. In such cases the propor- tion of fat lost in skimming does not vary much, e. g., in case of milks containing 3 5 and 4 percent of fat, and varia- tions in the overrun occurring when the proper care in skim- ming, ripening and churning is taken, are due, therefore, primarily to differences in the water content of the butter made (197). Calculation of Butter and Cheese Yield. 165 200. The losses from very poor, very rich and average milk, as received at creameries and cheese factories, can be traced from the following statement; this gives the quantities of fat lost in handling milk of four grades, viz: 2.5, 3.5, 4.0 and 6.0 percent, in case of each grade calculated to a stand- ard of 100 lbs. of fat in the milk. To supply 100 lbs. of fat would require the following amounts of the different grades ot milk: 4000 lbs. of milk testing 2.5 percent will contain 100 lbs. of fat. 2857 " " " 3.5 " " '* 100 " " 2500 •' " " 4.0 " •* " 100 " " 1666 " " " 6.0 ♦' " " 100 " " Assuming that the skim milk contains .2 percent of fat and makes up 80 percent of the whole milk, and that the butter milk tests .3 percent, and forms 20 percent of the whole milk, the butter fat record of the quantities of differ- ent grades of milk containing 100 lbs. of fat will appear as follows: Fat available for butter in different grades of milk. Grade of milk. Whole milk. Skim milk. Butter milk. Total loss. Fat available for butter. 2 5 percent 4000 ftx 2.5 perct. 320C ft). .2 per ct. 800 ft). .3 perct. ft. 8.8 6.3 6.5 3.6 Percent. Fat 100 ft). 2857 ft). 3. 5 perct. 6.4 ft). 2285 ft). . 2 per ct. 2.4 ft). 572 ft). . 3 per ct. 91.2 3 5 percent Fat 100 ft). 2500 ft). 4 perct. 4.0 ft). 2000 ft). .2per ct. 1.7 ft). 500 ft). .3 perct. 93.7 Fat 100 ft). 1666K ft). 6 perct. 4.0 ft). 1.S33 ft). .2 perct. 1.5 ft. 333 ft. .3 perct. 94.5 6 percent ......... Fat 100 ft). 2.6 ft). 1.0 ft. 96.4 1 66 Testing Milk and Its Products. The table shows that with 2.5 percent milk, there is a loss of 6.4 lbs. of fat in the skim milk and 2.4 lbs. of fat in the butter milk for every 100 lbs. of fat in the whole milk, or a total loss of 8.8 lbs. from these sources. In case of 6 per- cent milk these losses are 2.6 lbs. and 1.0 lbs. for skim milk and butter milk, respectively; a total loss of 3.6 lbs., or 5.2 lbs. less than the losses with the very poor milk. This difference in the losses between 3.5 percent and 4.0 percent milk shrink to only .8 pound of fat when a quantity of milk containing 100 lbs. of fat is handled in both cases. The overrun from each of the four grades of milk can be calculated to butter containing a certain percent of fat. The fat content of butter varies greatly, as has been shown (197); assuming it to be 83 percent, on the average, the quantity of butter obtained from the 100 lbs. of fat, or rather from the portion thereof which is available for butter, in each case would be as follows : Butter cont. Available fat. SSprct.fat. 100 lbs. of fat from 4000 lbs. of 2.5 pr ct. milk, 91.2 lbs.=109.61bs. 100 " " " 2857 " 3.5 '' " 93.7 " =112.9 lbs. 100 " " '' 2500 " 4.0 " " 94.5 " =113.8 lbs. 100 " " " 1666 " 6.0 " " 96.4 " =116.1 lbs. The overrun in each case would be: For 2.5 percent milk = 109.6 — 100= 9.6 percent. 3.5 " " =112.9 — 100 = 12.9 4.0 " " =113.8 — 100 = 13.8 6.0 " " =116.1 = 100 = 16.1 All butter makers should obtain more butter from a cer- tain quantity of milk than the Babcock test shows it to con- tain butter fat, but it is impossible to know exactly how much butter fat is lost in the skim milk and the butter milk, unless these products are tested, and how much water, salt and curd the butter will contain. Calculation of Butter and Cheese Yield. 167 201. Calculation of overrun. The overrun is calcu- lated by subtracting the amount of butter fat contained in a certain quantity of milk, from the amount of butter made from it, and finding the percent which this difference is of the amount of butter fat in the milk. Example. 8000 lbs. of milk is received at the creamery on a certain day; the average test of the milk is 3.8 percent. 340 lbs. of butter was made from this milk, as shown by the weights of the packed tubs. By a simple multiplication we find that the milk contained 8000X. 038=304 lbs. of butter fat. The diflference be- tween the weight of butter and butter fat, is, therefore, 36 lbs.; 36 is — — — — =11.8 percent of the quantity of butter fat in the o04? milk; that is, the overrun for the day considered was 11.8 percent. The formula for the overrun is as follows: (b — f) 100 X— ^ h and / designating the quantities of butter and butter fat, respectively, made from or contained in a certain quantity of milk. In the preceding example, the calculation would be as -p^ii^ « (340 — 304) 100 -,. o . lollows : ^^ -—— =11.8 percent. 202. Conversion factor for butter fat. A committee of the Association of American Agricultural Colleges and Experiment Stations reported at the ninth annual conven- tion of the Association that " in the ninety-day Columbian Dairy Test, 96.67 percent of the fat in the whole milk was recovered in the butter. This butter on the average contained 82.37 percent butter fat; in other words, 117.3 pounds of but- ter were made from each 100 pounds of butter fat in the whole milk.* The exact conversion factor would be 1.173. * When 82.37 lbs. of butter fat will make 100 lbs. of butter, how much butter will 96.67 lbs. of butter fat make? 82.37:96.67 :: 100: x. x = 117.3. 1 68 Testing Milk and Its Products. As this is an awkward number to use, and as li is so nearly the same ... it has seemed best to recommend that the latter be used as the conversion factor." A resolution was adopted by this association recommend- ing that the approximate equivalent of butter be computed by multiplying the amount of butter fat by \\. These figures represent more than ordinary care in testing, skimming and churning, and probably the minimum loss of fat in the manufacturing processes. The increase of churn over test represented by one-sixth, or 16 percent, therefore may be taken as a maximum " overrun." Butter makers who report overruns of 16-20 percent do not show their expert- ness in butter making by such high figures, but their lack of accuracy in testing, or carelessness in working the butter; a large overrun may be obtained both by reading the test too low, and by leaving an excess of water in the butter through insuflScient working or other causes. 203. Butter yield from milk of different richness. a. Use of hutter chart. The approximate yield of butter from milk of different richness is shown in table IX in the Appendix. This table is founded on ordinary creamery ex- perience and will be found to come near to actual every-day conditions of creameries where modern methods are followed in the handling of the milk and its products. The table has been prepared in the following manner: It is assumed that the average loss of fat in the skim milk is .20 percent, and that 85 lbs. of skim milk is obtained from each 100 lbs. of whole milk; to this loss of fat is added that taking place in the butter milk; about 10 lbs. of butter milk is obtained per 100 lbs. of whole milk, testing on the average .30 percent. Caladation of B titter and Cheese Yield. 169 If / designate the fat in 100 lbs. of milk, then the fat recovered in the butter from 100 lbs. of milk will be f-(foX.20+^,X.3o)=f-.20 There is, on the other hand, an increase in weight in the butter made, owing the admixture of non-fatty components therein, principally water and salt. Butter packed and ready for the market will contain in the neighborhood of 84 percent of fat (197), so that the fat recovered in the but- ter must be increased by Vf = 1.19. If B therefore desig- nate the yield of butter from 100 lbs. of milk, the following formula will express the relation between yield and fat con- tent, provided there are no other factors entering into the problem, viz: B = (f — .20) 1.19 Certain mechanical losses are, however, unavoidable in the creamery, as in all other factory operations, viz: milk and cream remaining in vats and separators, butter sticking to the walls of the churn, etc. These losses have been found to average about 3 percent of the total fat in the milk handled, under normal conditions and under good manage- ment (202) ; we therefore deduct this percent from the pre- ceding value for B, and have: B=(f_.20)1.16 204. Table IX in the Appendix, founded on this formula, may be used to determine the number of pounds which the milk delivered by the various patrons within the limits of 3 and 5.3 percent will be likely to make. It presupposes good and careful work at the separator, churn and butter worker, and will generally under such conditions show yields of butter varying but little from those actually obtained. It may also be conveniently used by the butter maker or 170 Testing- Milk and Its Products, the manager to check the work in the creamery; the aver- age test of the milk received during a certain period is found by dividing the total butter fat received by the total milk, multiplying by 100; the amount of butter which the total milk of this average fat content will make according to the table, is then compared with the actual churn yield. Example. A creamery receives 200,000 lbs. of milk during a month; the milk of each patron is tested and the fat contained therein calculated. The sum of these amounts of fat may be 7583 lbs.; the average test of the milk is then 3.79 percent. Ac- cording to table IX, 10,000 lbs. of milk, testing 3.8, will make 421.2 lbs. of butter, and 200,000 lbs., therefore, 8424 lbs. of but- ter. The total quantity of butter made during the month should not var3' appreciably from this figure. 205. b. Use of overrun table. The table referred to above gives a definite calculated butter yield for each grade of milk, according to the average creamery conditions. As it may be found that this table will either give uniformly too low or too high results, table X in Appendix is included, by means of which the butter yield corresponding to overruns from 10-20 percent may be ascertained in a similar way as above described. The total yield of butter is divided by the total number of pounds of fat delivered; the quotient will give the amount of butter made from one pound of fat, and this figure mul- tiplied by the fat delivered by each patron shows the pounds of butter to be credited to each patron. To use the table, find in the upper horizontal line the number corresponding most nearly to the number of pounds of butter from one pound of fat. The vertical column in which this falls gives the pounds of butter from 100 lbs. of milk containing the percents of fat given in the outside columns (Babcock *). *Woll, Handbook for Farmers and Dairymen, p. 273. Calculation of Butter and Cheese Yield. 171 B. — Calculation of Yield of Cheese. 206. a. From fat. The appropriate yield of green Ched- dar cheese from 100 lbs. of milk may be found by multiply- ing the percent of fat in the milk by 2.7: if / designate the percent of fat in the milk, the formula will, therefore, be: Yield of cheese = 2. 7 f (I) The factor 2.7 will only hold good as the average of a large number of cases. In extensive investigations during three consecutive years. Van Slyke * found that the number of pounds of green cheese obtained for each pound of fat in the milk varied from 2.51 to 3.06, the average figures for the three years 1892- '94, incl., being 2.73, 2.71, and 2.72 lbs., respectivel}'. The richer kinds of milk will produce cheese richer in fat, and will yield a relatively a larger quantity of 7 7.82 8.07 8.32 8.57 8.88 9.08 9.33 9.58 9.84 10.09 5.3 5.4 7.59 7.84 8.09 8.34 8.60 8.85 9.10 9.36 9.61 9.86 10.11 5.4 5.5 7.61 7.86 8.11 8.36 8.62 8.87 9.12 9.389.68 9.88 10.13 5.5 5.6 7.68 7.88 3.13 8.39 8.64 8.89 9.15 9.40 9.65 9.90 10.15 5.6 5.7 7.65 7.90 8.15 8.41 8.66 8.91 9.17 9.42,9.67 9.92 10.17 5.7 5.8 7.67 7.92 8.17 8.43 8.68 8.94 9.19 9.44 9.69 9.94 10.19 5.8 5.9 7.69 7.94 8.20 8.45 8.70 8.96 9.21 9.469.71 9.96 10.22 5.9 6.0 7.71 7.96 8.22 8.47 8.72 8.98 9.23 9.48 9.73 1 9.98 10.24 6.0 Aj>fendix. 211 Directions for Use of Tables VI, VH, and IX. TABLE VI. Find the test of the milk in the first or last hori- zontal row of figures; the amounts of fat in ten thousand, thou- sands, hundreds, tens, and units of pounds of milk are then given in this vertical column. By adding the corresponding figures in any given quantity of milk, the total quantity of butter fat con- tained therein is obtained. Example. How many pounds of fat is contained in 8925 lbs. of milk testing 3.65 percent? On p. 21.3, second column the test 3.66 is found, and by going downward in this column we have: 8000 lbs 292, lbs. 900 " 32.9 " 20 " 7 " 5 " 2 " 8925 lbs. of milk. 325 8 lbs. of fat. 8925 lbs. of milk testing 3.65 percent, therefore, contains 325.8 lbs. of butter fat. TABLE VII. The price per pound is given in the outside ver- tical columns.'and the freight of butter fat in the upper and lower horizontal row of figures. The corresponding tens of pounds are found by moving the decimal point one place to the left; the units, by moving it two, and the tenths of a pound, by moving it three places to the left. The use of the table is, otherwise, as explained above. Example.^ How much money is due for 325.8 lbs. of butter fat, at Xhy^ cents per pound? In the horizontal row of figures beginning with 15^, we find: .300 lbs $46.50 20 " 3.10 5 " 77 .8 " 12 325. 8 lbs. $50. 49 325.8 lbs. of butter fat at 15 3^ cents per pound, therefore, is worth ^50.49. TABLE IX, Find the test of milk in the upper or lower hori- zontal row of figures. The amount of butter likely to be made from ten thousand, thousands, hundreds, tens, and units of pounds of milk are then given in this vertical column. The use of the table is, otherwise, as explained above in case of table VI. Example. How much butter will 5845 lbs. of milk testing 3.8 percent be apt to make under good creamery conditions? In the column headed 3.8, we find: 5000 lbs 2C9.0 lbs. 800 " 33.4 " 40 " 1.7 " 5 " 2 " 5845 lbs. 244. 3 lbs. 5845 lbs. of milk testing 3.8 percent of fat will make about 244.3 lbs. of butter, under conditions similar to those explained on pp. 168-69 of the present work. 212 Testing- Milk and Its Products, Table VI. Lbs. of fat in i to 10,000 lbs. of milk, testing 3.0 to 5.35 percent. (See p. 211). 1 3.00 3.05 3.10 3.15 3.20 3.25 3.30 3.35 3.40 3.45 3.50 3.55 t Milk Milk lbs. lbs. 10, COO 300 305 310 315 320 325 330 335 340 345 350 355 10, 000 9,000 270 275 279 284 289 293 297 302 306 311 315 320 9,000 8,000 240 244 248 252 256 260 264 268 272 276 280 284 8,000 7,000 210 214 217 221 224 228 231 235 238 242 245 249 7,000 6,000 180 183 186 189 192 195 198 201 204 207 210 213 6,000 5,000 150 168 155 158 160 163 165 168 170 173 175 178 5,000 4,000 120 122 124 126 128 130 132 134 136 138 140 142 4,000 3,000 90.0 91.5 93.0 94.5 96.0 97.5 99.0 101 102 104 105 107 3, 000 2, 000 60.0 61.0 62.0 63.0 64.0 65.0 66.0 67.0 68.0 69.0 70.0 71.0 2,000 1,000 30.0 30.5 31.0 31.5 32.0 32.5 33.0 33.5 34.0 34.5 35.0 35.5 1,000 900 27.0 27.5 27.9 28.4 28.8 29.3 29.7 30.2 30.6 31.1 31.5 32.0 900 800 24.0 24.4 24.8 25.2 25.7 26.0 26.4 26.8 27.2 27.6 28.0 28.4 800 700 21.0 21.4 21.7 22.1 22.4 22.8 23.1 23.5 23.8 24.2 24.5 24.9 700 600 18.0 18.3 18.6 18.9 19.2 19.5 19.8 20.1 20.4 20.7 21.0 21.3 600 500 15.0 15.3 15. 5 15.8 16.0 16.3 16.5 16.8 17.0 17.3 17.5 17.8 500 400 12.0 12.2 12.4 12.6 12.8 13.0 13.2 13.4 13.6 13.8 14.0 14.2 400 300 9.0 9.2 9.3 9.5 9.6 9.8 9.9 10.1 10.2 10.4 10.5 10.7 300 200 6.0 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 io.9 7.0 7.1 200 100 3.0 3.1 3.1 3.2 3.2 3.3 3.3 3.4 3.4 3.5 3.5 3.6 100 90 2.7 2.8 2.8 2.8 2.9 2.9 3.0 3.0 3.1 3.1 3.2 3.2 90 80 2.4 2.4 2.5 2.5 2.6 2.6 2.6 2.7 2.7 2.8 2.8 2.8 80 70 2.1 2.1 2.2 2.2 2.2 2.3 2.3 2.3 2 4 2.4 2.5 2.5 70 60 1.8 1.8 1.9 1.9 1.9 2.0 2.0 2.0 2.0 2.1 2.1 2.1 60 50 1.5 1.5 1.6 1.6 1.6 1.6 1.7 1.7 1.7 1.7 1.8 1.8 50 40 1.2 1.2 1.2 1.3 1.3 1.3 1.3 1.3 1.4 1.4 1.4 1.4 40 30 .9 .9 .9 .9 1.0 1.0 1.0 1 1.0 1.0 1.1 1.1 30 20 .6 .6 .6 .6 .6 .7 .7 .7 7 .7 .7 .7 20 10 .3 .3 .3 .3 .3 .3 .S .3 .3 .3 .4 .4 10 9 .3 .3 .0 .3 .3 .3 .3 .3 .3 .3 .3 .3 9 8 .2 .2 2 .3 .3 .3 .3 .3 .3 .3 .3 .3 8 7 .2 .2 .2 .2 .2 2 2 .2 .2 •2| .2 .2 / 6 .2 2 .2 .2 .2 .2 .2 .2 .2 .2I .2 .2 6 5 2 .2 .2 .2 .2 9 .2 .2 .2 2 .2 2 5 4 .1 .1 .1 .1 .1 .1 .1 .1 .1 .1 .1 .1 4 3 .1 .1 .1 .1 .1 .1 .1 .1 .1 .1 .1 .1 3 2 1 .1 .1 ...".! .1 .1 .1 .1 .1 .1 .1 .1 .1 2 1 ■« 3.00 3.05 3.10 3.15 3.20 3.25 3.30 3.35 3.40 3.45 3.50 3.55 ^ H 1 Appendix. 213 Table TI. Lbs. of fat in 1 to 10,000 lbs. of milk {Continued) 3.60 3.65 3.70 3.75 3.80 3.85 3.90 3.95 4.00 4.05 4.10 4.15 I Milk Mlik lbs. lbs. 10, 000 360 365 370 375 380 385 390 395 400 405 410 415 10, 000 9,000 324 329 333 338 342 347 351 356 360 365 369 374 9,000 8,000 288 292 296 300 304 308 312 316 320 324 328 332 8,000 7,000 252 256 269 263 266 270 273 277 280 284 287 291 7,000 6,000 216 219 222 225 228 231 234 237 240 243 246 249 6,000 5, 000 180 183 185 188 190 193 195 198 200 203 205 208 5,000 4,000 144 146 148 150 152 154 156 158 160 162 164 166 4,000 3,000 108 110 111 113 114 116 117 119 120 122 123 125 3,000 2,000 72.0 73.0 74.0 75.0 76.0 77.0 78.0 79.0 80.0 81.0 82.0 83.0 2,000 1,000 36.0 36.5 37.0 37.5 38.0 38.5 39.0 39.5 40.0 40.5 41.0 41.5: 1,000 900 32.4 32.9 33.3 33.8 34.2 34.7 35.1 35.6 36.0 36.5 36.9 37.4 900 800 28.8 29.2 29.6 30.0 30.4 30.8 31.2 31.6 32.0 32.4 32.8 33.2 800 700 25.2 25.6 25.9 26.3 26.6 27.0 27.3 27.7 28.0 28.4 28.7 29.1 700 600 21.6 21.9 22.2 22.5 22.8 23.1 23.4 23.7 24.0 24.3 24.6 24.9 600 500 18.0 18.3 18.5 18.8 19.0 19.3 19.5 19.8 20.0 20.3 20.5 20.8 500 400 14.4 14.6 14.8 15.0 15.2 15.4 15.6 15.8 16. C 16.2 16.4 16 6 400 300 10.8 11.0 11.3 11.3 11.4 11.6 11.7 11.9 12.0 12.2 12.3 12.5 300 200 7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9 8.0 8.] 8.2 8.3 200 100 3.6 3.7 3.7 3.8 3.8 3.9 3.9 4.0 4.0 4.1 4.1 4.2 100 90 3.2 3.3 3.3 3.4 3.4 3.5 3.5 3.6 3.6 3.7 3.7 3.7i 90 80 2.9 2.9 3.0 3.0 3.0 3.1 3.1 3.2 3.2 3.2 3.3 3.3 80 70 2.5 2.6 2.6 2.6 2.7 2.7 2.7 2.8 2.8 2.8 2.9 2.9 70 60 2.2 2.2 2.2 2.3 2.3 2.3 2.3 2.4 2.4 2.4 2.5 2.5 60 50 1.8 1.8 1.9 1.9 1.9 1.9 2.0 2.0 2.0 2.0 2.1 2.1 50 40 1.4 1.5 1.5 1.5 1.5 1.5 1.6 1.6 1.6 1.6 1.6 1.7 40 30 1.1 1.1 1.1 1.1 1.1 1.2 1.2 1.2 1.2 1.2 1.2 1.2 30 20 .7 .7 .7 .8 .8 .& .8 .8 .8 .8 .8 .8 20 10 .4 .4 A .4 .4 .4 .4 .4 .4 .4 . .4 .4 10 9 .3 .3 .3 .3 .3 .3 .4 .4 .4 .4 4 .4 9 8 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 8 7 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 7 6 2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 6 5 2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 5 4 !l .1 .1 .2 .2 2 .2 .2 .2 .2 .2 .2 4 3 .1 .1 .1 .1 .1 '.1 .1 .1 .1 .1 .1 .1 3 2 1 .1 .1 .1 .1 .1 .1 .1 .1 .1 .1 .1 .1 2 1 3.60 3.65 3.70 3.76 3.80 3.85 3.90 3.95 4.00 4.05 4.10 4.15 i ^ I 214 Testing Milk and Its Products. Table YI. Lbs. of fat in 1 to 10,000 lbs. of milk {Continued) 4.20 4.25 4.30 1 4.35 1 4.40 1 4.45| 4.50 4.56 4.60 4.65 4.70 1 4.75 t Milk Milk lbs. lbs. 10,0001= 420 425 430 435 440 445 450 455| 460 465 470 475 10, 000 9,000! 378 383 387 392 396 40l| 405 410 4141 419 423| 428 9,000 8,000 336 340 344 348 352 356 360 364 368 372 376 380 8,000 7,000 294 298 301 1 305 308 312[: 315 319 322 326 329 333 7,000 6,000 252 255 258' 261 264 267;' 270 273 276 279 282 285 6,000 5, 000 210 213 215 218 220 223|' 225 228 230 2331 235 238 5,000 4,000 168 170 172 174 176 178li 180l 182 184 186! 188 190 4,000 3,000 126 128 129 131 132 13411 135 137 138 140| 141 14311 3,000 2,000 84.0 85.0 86.0 87.0 88.0 89.0ll90.0 91.0 92.0 93.094.0 95.0 2,000 1,000 42.0 42.5 43.0 43.5 44.0 44.5115.0 45.5 46.0 46.5 47.0 47.5 1,000 900':'37.8 38.3 38.7 39.2 39.6 40.1140.5 41.0 41.4 41.9142.3 42.8 900 800 33.6 34.0 34.4 34.835.2 35.6:136.0 36.4 36.8 37.2 37.6 38.0 800 700 29.4 29.8 30.1 30.5 30.8 31.2131.5 31.9 32.2 32.6 32.9 33.3 700 600 25.2 25.5 25.8 26.1 26.4 26.7j27.0 27.3 27.6 27.9 28.2 28.5 600 500 21.0 21.3 21.5 21.8 22.0 22.3I22.5 22.8 23.0 23.3 23.5 23.8 500 400 16.8 17,0 17.2 17.4 17.6 17.8 18.0 18.2 18.4 18.6 18.8 19.0 400 300:12.6 12.8 12.9 13.1 13.2 13.4 13.5 13.7 13.8 14.0 14.1 14.3 300 200 8.4 8.5 8.6 8.7 8.8 8.9 9.0 9.1 9.2 9.3 9.4 9.51 200 100 4.2 4.3 4.3 4.4 4.4 4.5;: 4.5 4.6 4.6 4.7 4.7 4.8:, 100 90 3 8 3 8 3.9 3.9 4.0 4.01 4.1 4.1 4.1 4.2 4.2 4.3:1 90 80 3 4 3,4 3.4 3.5 3.5 3.6! 3.6 3.6 3.7 3.7 3.8 3.8 80 70 1 2 9 3.0 3.0 3.0 3.1 3.1' 3.2 3.2 3.2 3.3 3.3 3.3 70 60 2,5 2.6 2.6 2.6 2.6 2.7 2.7 2.7 2.8 2.8 2.8 2.9 60 50 2.1 2.1 2.2 2.2 2.2 2.2 2.3 2.3 2.3 2.3 2.4 2.4 50 40 1 7 1.7 1.7 1.7 1.8 1.8 1.8 1.8 1.8 1.9 1.9 1.9 40 80 1.3 1.3 1.3 1.3 1.3 1.3 1.4 1.4 1.4 1.4 1.4 1.4 ' 30 20 8 9 .9 .9 .9 .9 .9 .9 .9 .9 9 1.0 i 20 10 .4 .4 .4 .4 .4 .4 .5 .5 .5 .5 .5 .5 10 9 4 4 ,4 .4 .4 .4 .4 .4 .4 .4 .4 .4 i 9 8 3 ,3 ,3 .3 .4 .4 .4 .4 .4 .4 .4 .4 8 7 3 ,3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 7 6 ,3 ,3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 6 5 ? 2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 5 4 V 2 ,2 .2 .2 .2 .2 .2 .2 .2 .2 .2 4 3 1 1 1 .1 .1 .1 .1 .1 .1 .1 .1 .1 3 2 1 .1 .1 .1 .1 .1 .1 i. ...'.. .1 .1 .1 .1 .1 2 1 1 4.2C 4.25 4.30 4.35 4.4C 4.45 4.50 4.55 4.6C 4.65 4.70 4.75 t Appendix, 21 Table TI. Lbs. of fat in 1 to 10,000 lbs. of milk ( C on tinned) 1 4.80 4.85 4.90 4.95 5.00 5.05 5.10 5.15 5.20 5.25 5.30 5.35 Milk 1 Milk lbs. 1 lbs. 10, 000 9,000 480 485 490 495 500 505 510 515 520 52o 530 535 10,000 432 437 441 446 450 455 459 464 468! 4731 477! 482 i 9, 000 8,000 384 388; 392 396 400 404 408 412 416 420 422! ^28 8,000 7, 000 i 336 340; 343 347 350 354 357 361 364 368 371 375 : 7, 000 6,000 288 291! 294 297 300 303 306 309 312 315 318 321 6,000 5,000 240 243i 245 248 250 243 255 258 260 263 265 268 1 5,000 4,000 192 194' 196 198 200 202 204 206 208 210 212 214 4,000 3,000 144 146 147 149 150 152 153 155 156 158 159 161 3,000 2, 000 96.0 97.098.0 99.0 100 101 102 103 104 105 106 107 2,000 1,000 48.0 48.5 49.0 49.5 50.0 50.5 51.0 51.5 52.0 52.5 53.0 53.5 1,000 900 43.2 43.7 44.1 44.6 45.0 45.5 45.7 46.4 46.8 47.3 47.7 48.2 900 800 38.4 38.839.2 39.6 40.0 40.4 40.8 41.2 41.642.042.442.8 800 700 33.6 34.034.3 34.7 35.0 35.4 35.7 36.1 36 4 36.8 37.1137.5 700 600 28.8 29.129.4 29.7 30.0 30.3 30.6 30.9 31.231.531.8^32.1 600 500 24.0 24.324.5 24.8 25 25.3 25.5 25.8 26.0 26.3 26.626.8 500 400 19.2 19.419.6 19.8 20.0 20.2 20.4 20.6 20.821.021.221.4 400 300 14.4 14.614.7 14.9 15.0 15.2 15.3 15.5 15.6 15.8 15.9 16.1 300 200 9 6' 9.71 9.8 9.9 10.0 10.1 10.2 10.3 10.410.510.610.7 200 100 4.8 4.9 4.9 5.0 5.0 5.1 5.1 5.2 5.2 5.3 5.3 5.4 100 90 4.3 4.4 4.4 4.5 4.5 4.5 4.6 4.6 4.7 4.7 4.8 4.8 90 80 3.8 3.9 3.9 4.0 4.0 4.0 4.1 4.1 4.2| 4.2 4.2' 4.3 80 70 3.4 3.4 3.4 3.5 3.5 3.5 3.6 3.6 3.6 3.7 3.71 3.7 1 '0 60 2.9 2.9 2.9 3.0 3.0 3.0 3.1 3.1 3.1 3.2' 3.2 3.2 i 60 50 2.4 2.4 2.5 2.5 2.5 2.5 2.6 2.6 2.6 2.6| 2.7 2.7 50 40 1.9 1.9 2.0 2.0 2.0 2.0 2.01 2.1 2.1| 2.1 2.l| 2.1 40 30 1.4 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.6j 1.6 l.el 1.6 30 20 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.1 1.1} 1 1 20 10 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5j .5 .5 10 9 .4 .4 .4 .4 .5 .5 .5 .5 .5 .5i .5' .5 9 8 .4 .4 .4 .4 .4 .4 .4 .4 .4 .4 .4 .4 8 7 .3 .3 .3 .3 .4 .4 .4 .4 .4 .4 .4 .4 7 6 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 6 5 .2 .2 .2 .2 .3 .3 .3 .3 .3 .3 .3 .3 5 4 .2 .2 .2 2 .2 .2 .2 2 .2 .2 .2 .2 4 3 .1 .1 .1 A .2 .2 .2 2 .21 .2! .2 .2 3 2 .1 .1 .1 .1 .1 .1 .1 .1 .li .1 .1 .1 2 1 .1 .1 .1 .1 .1 .1 .1 .1 1 4.80 4.85 4.90 4.95 5.00 5.05 5.10 5.15 5.20 5.25 5.30 5.35 S 2l6 Testing Milk and Its Products, Table VII. Amount due for butter fat, in dollars and cents, at 12 to 25 cents per pound. 11 Pounds of butter fat. 11 "1 1,000 900 800 700 600 500 400 300 200 100 If % $ % $ $ $ % $ $ $ 12 120.00 108.00 96.00 84.00 72.00 60.00 48.00 36.00 24.00 12.00 12 12i 122.50 110.25 98.00 85.75 73.50 61.25 49.00 36.75 24.50 12.25 Vl\ 12^ 125.00 112.50 100.00 87.50 75.00 62.50 50.00 37.50 25.00 12.60 12^ 121 127.50 114.75 102.00 89.25 76.50 63.75 51.00 38.25 25.50 12.75 121 13 130.00 117.00 104.00 91.00 78.00 65.00 52.00 39.00 26.00 13.00 13 13i 132.50 119.25 106.00 92.75 79.50 66.25 53.00 39.75 26.50 13.25 \Z\ 13^ 135.00 121.50 108.00 94.50 81.00 67.50 54.00 40.50 27.00 13.50 13* 13f 137.50 123.75 110.00 96.25 82.50 68.75 55.00 41.25 27.50 13.75 131 14 140.00 126.00 112.00 98.00 84.00 70.00 56.00 42.00 28.00 14.00 14 14i 142.50 128.25 114.00 99.75 85.50 71.25 67.00 42.75 28.50 14.25 14J 14i 145.00 130.50 116.00 101.50 87.00 72.50 58.00 43.50 29.00 14 50 14^ 14| 147.50 132.75 118.00 103.25 88.50 73.75 59.00 44.25 29.50 14.75 141 15 150.00 135.00 120.00 105.00 90.00 75.00 60.00 45.00 30.00 15.00 15 15i 152.50 137.25 122.00 106.75 91.50 76.25 61.00 45.76 30.50 15.25 15i 15^ 155.00 139.50 124.00 108.50 93.00 77.50 62.00 46.50 31.00 15.50 15^ 15f 157.50 141.76 126.00 110.25 94.50 78.75 63.00 47.25 31.50 15 75 151 16 160.00 144.00 128.00 112.00 96.00 80.00 64.00 48.00 32.00 16.00 16 16i 162.50 146.25 130.00 113. 7^ 97.50 81.25 65.00 48.75 32.50 16.25 16J 16^ 165.00 148. 5C 132.00 115.50 99.00 82.50 66.00 49.50 33.00 16.50 16^ 161 167.50 150.75 134.00 117.25 100.50 83.75 67.00 60.25 33.50 16.75 16| 17 170.00 153.00 136.00 119.00 102.00 85.00 68.00 51.00 34.00 17.00 17 n\ 172.50 155.25 138.00 120.75 103.50 86.25 69.00 51.75 34.50 17.25 17i 17^ 175.00 157.50 140.00 122.50 105.00 87.50 70.00 52.50 35.00 17.50 17i 17| 177.50 159.75 142.00 124.25 106.50 87.75 71.00 53.25 35.50 17.75 171 18 180.00 162.00 144.00 126.00 108 CO 90.00 72.00 54.00 36.00 18.00 18 I84 182.50 164.25 146.00 127.75 109.50 91.25 73.00 54.75 36.50 18.25 181- 18i 185.00 166.50 148.00 129.50 111.00 92.50 74.00 55.50 37.00 18.50 18^ 18- 187.50 168.75 150.00 131.25 112.50 93.75 75.00 56.25 37.50 18.75 18| 1,000 900 800 700 600 500 400 300 200 100 Appendix. 217 Table YII. Amount due for butter fat {Continued.) (1 s Pounds of butter fat. <-' Q s.§ S.S 8-« 8-2 H 1,000 900 800 700 600 500 400 300 200 100 •r a ^1 $ 1 $ $ $ $ $ $ $ $ 19 190.00 171.00 152.00 133.00 114.00 95.00 76.00 57.00 38.00 19.00 19 19,1 192.50 173.25 154.00 134.75 115.50 96.25 77.00 57.75 38.50 19.25 191- 19^ 195.00 175.50 156.00 136.50 117.00 97.50 78.00 58.50 39.00 19.50 191 191 197.50 177.75 158.00 138.25 118.50 98.75 79.00 59.25 39.50 19.75 19f 20 200.00 180.00 160.00 140.00 120.00 100.00 80.00 60.00 40.00 20.00 20 20| 202.50 182.25 162.00 141.75 121.50 101.25 81.00 60.75 40.50 20.25 20i 20i 205.00 184.50 164.00 143.50 123.00 102.50 82.00 61.50 41.00 20.50 20* 20f 207.50 186.75 166.00 145.25 124.50 103.75 83.00 62.25 41.50 20.75 20f 21 210.00 189.00 168.00 147.00 126.00 105.00 84.00 63.00 42.00 21.00 21 2U 212.50 191.25 170.00 148.75 127.50 106.25 85.00 63.75 42.50 21.25 2U 21i 215.00 193.50 172.00 150.50 129.00 107.50 86.00 64.50 43.00 21.50 2U 21| 217.50 195.75 174.00 152.25 130.50 108.75 87.00 65.25 43.50 21.75 21| 22 220.00 198.00 176.00 154.00 132.00 110.00 88.00 66.00 44.00 22.00 22 22| 222.50 200.25 178.00 155.75 133.50 111.25 89.00 66.75 44.50 22.25 22i 22^ 225.00 202.50 180.00 157.50 135.00 112.50 90.00 67.50 45.00 22.50 22J 221 227.50 204.75 182.00 159.25 136.50 113.75 91.00 68.25 45.50 22.75 22| 23 230.00 207.00 184.00 161.00 138.00 115.00 92.00 69.00 46.00 23.00 23 234I 232.50 209.25 186.00 162.75 139.50 116.25 93.00 69.75 46.50 23.25 2Sk 23i 235.00 211.50 188.00 164.50 141.00 117.50 94.00 70.50 47.00 23.50 23h 23| 237.50 213.75 190.00 166.25 142.50 118.75 95.00 71.25 47.50 23.75 23| 24 240.00 216.00 192.00 168.00 144.00 120.00 96.00 72.00 48.00 24.00 24 241 242.50 218.25 194.00 169.75 145.50 121.25 97.00 72.75 48.50 24.50 24i 24^ 245.00 220.50 196.00 171.50 147.00 122.50 98 00 73.50 49.00 24. 5( 24* 241 247.50 222.75 198.00 173.25 148.50 123.75 99.00 74.25 49.50 24.75 24f 25 250.00 225.00 200.00 175.00 150.0U 125.00 100.00 75. 0( 50.00 25. 0( 25 1,000 900 800 700 600 500 400 300 200 100 (For directions for use, see page 211) 2l8 Testing Milk and Its Products. Table VIII. Relative value tables. (For directions for use, see pp. 180-81). Pric E OF MILK PER 100 POUNDS, IN DOLLARS AND CENTS. 3.0 .30 .31 .33 .3i .36 .37 .39 .40 .42 .43 .45 3.1 .31 .33 .34 .36 .37 .39 .40 .42 .43 .45 .46 3.2 .32 .34 .35 .37 .38 .40 .42 .43 .45 .46 .48 3.3 .33 .85 .36 .38 .40 .41 .43 .45 .46 .48 .49 3.4 .34 .36 .37 .39 .41 .42 .44 .46 .48 .49 .51 3.5 .35 .37 .38 .40 .42 .44 .45 .47 .49 .51 .52 3.6 .36 .38 .40 .41 .43 .45 .47 .49 .50 .52 .54 3.7 .37 .39 .41 .43 .44 .46 .48 .50 .52 .54 .55 3.8 .38 .40 .42 .44 .46 .47 .49 .51 .53 .55 .57 3.9 .39 .41 .43 .45 .47 .49 .51 .53 .55 .57 .58 4.0 .40 .42 .44 .46 .48 .50 .52 .54 .56 .58 .60 4.1 .41 .43 .45 .47 .49 .51 .53 .55 .57 .59 .61 4.2 .42 .44 .46 .48 .50 .52 .55 .57 .59 .61 .63 4.3 .43 .45 .47 .49 .52 .54 .56 .58 .60 .62 .64 4.4 .44 .46 .48 .51 .53 .55 .57 .59 .62 .64 M 4.5 .45 .47 .49 .52 .54 .56 ..■^8 .61 .63 .6) .67 4.6 .46 .48 .51 .63 .55 .57 .60 .62 .64 .67 .69 4.7 .47 .99 .52 .54 .56 .59 .61 .63 .66 .68 .70 4.8 .48 .50 .53 .55 .58 .60 .62 .65 .67 .70 .72 4.9 .49 .51 .54 .56 .59 .t)l .64 .66 .69 .71 .73 5.0 .50 .52 .55 .57 .60 .62 .65 .67 .70 .72 .75 5.1 .51 .54 .56 .59 .61 .64 .66 .69 .71 .74 .76 5.2 .52 .55 .57 .60 .62 .65 .68 .70 .73 .75 .78 5.3 .53 .56 .58 .61 .64 M .69 .72 .74 .77 .79 5.4 .54 .57 .59 .62 .65 .67 .70 .73 .76 .78 .81 5.5 .55 .58 .60 .63 .66 .69 .71 .74 .77 .80 .82 5.6 .56 .59 .62 .64 .67 .70 .73 .76 .78 .81 .04 5.7 .57 .60 .63 .66 .68 .71 .74 .77 .80 .83 .85 5.8 .68 .61 .64 .67 .70 .72 .75 .78 .81 .84 .87 5.9 .59 .62 .65 .68 .71 .74 .77 .80 .83 .86 .88 6.0 .60 .63 .66 .69 .72 .75 .78 .81 .84 .87 .90 Affendix. 219 Table VIII. Relative value tables (Continued). . Price of milk PER 100 POUNDS , IN DOLLARS AND CENTS. 3.0 .46 .48 .49 .51 .52 .54 .55 .57 .58 .60 3.1 .48 .50 .51 .53 .54 .56 .57 .59 .60 .62 3.2 .50 .51 .53 .54 .56 .58 .59 .61 .62 .64 3.3 .51 .58 .54 .56 .58 .59 .61 .63 .64 .66 3.4 .53 .54 .56 .58 .59 .61 .63 .65 .66 .68 3.5 .54 .56 .58 .59 .61 .63 .65 M .68 .70 3.6 .56 .58 .59 .61 .63 .65 .67 .68 .70 .72 3.7 .57 .59 .61 .63 .65 .67 .68 .70 .72 .74 3.8 .59 .61 .63 .65 .66 .68 .70 .72 .74 .76 3.9 .60 .62 .64 .66 .68 .70 .72 .74 .76 .78 4.0 .62 .64 .66 .68 .70 .72 .74 .76 .78 .80 4.1 .64 .66 .68 .70 .72 .74 .76 .78 .80 .82 4.2 .65 .67 .69 .71 .73 .76 .78 .80 .82 .84 4.3 .67 .69 .71 .73 .75 .77 .80 .82 .84 .86 4.4 .68 .70 .73 .75 .77 .79 .81 .84 .86 .88 4.5 .70 .72 .74 .76 .79 .81 .83 .85 .88 .90 4.6 .71 .74 .76 .78 .80 .83 .85 .87 .90 .92 4.7 .73 .75 .78 .80 .82 .85 .87 .89 .92 .94 4.8 .74 .77 .79 .82 .84 .86 .89 .91 .94 .96 4.9 .76 .78 .81 .83 .86 .88 .91 .93 .96 .98 5 .77 .80 .82 .85 .87 .90 .92 .95 .97 1.00 5.1 .79 .82 .84 .87 .89 .92 .94 .97 .99 1.02 5.2 .81 .83 .86 .88 .91 .94 .96 .99 1.01 1.04 5.3 .83 .85 .87 .90 .93 .95 .98 1. 01 1.03 1.06 5.4 .84 .86 .89 .92 .91 .97 1.00 1.03 1.05 1.08 5.5 .85 .88 .91 .93 .96 .99 1.02 1.04 1.07 1.10 5.6 .87 .90 .92 .95 .98 1.01 1.04 1.06 1.09 1.12 5.7 .88 .91 .94 .97 1.00 1.03 1.05 1.08 1.11 1.14 5.8 .90 .93 .96 .99 1.01 1.04 1.07 1.10 1.13 1.16 5.9 .91 .94 .97 1.00 1.03 1.06 1.09 1.12 1.15 1.18 6.0 .93 .96 .99 1.02 1.05 1.08 1.11 1.14 1.17 1.20 220 Testing Milk and Its Products. TaWe VIII. Relative value tables [Continued). 0. . Price OF MILK PER 100 POUNDS, IN DOLLARS AND CENTS. 3.0 .61 .63 .64 .66 .67 .69 .70 .72 .73 .75 3.1 .64 .65 .67 .68 .70 .71 .73 .74 .76 .78 3.2 .66 .67 .69 .70 .72 .74 .75 .77 .78 .80 3.3 .68 .69 .71 .73 .74 .76 .78 .79 .81 .83 3.4 .70 .71 .73 .75 .76 .78 .80 .82 .83 .85 3.5 .72 .73 .75 .77 .79 .80 .82 .84 .86 .88 3.6 .74 .76 . / / .79 .81 .83 85 .86 .88 .90 3.7 .76 .78 .80 .81 .83 .85 .87 .89 .91 .93 3.8 .78 .80 .82 .84 .85 .87 .89 .91 .93 .95 3.9 .80 .82 .84 .86 .88 .90 .92 .94 .96 .98 4.0 .82 .84 .86 .88 .90 .92 .94 .96 .98 1.00 4.1 .84 .86 .88 .90 .92 .94 .96 .98 1.00 1.03 4.2 .86 .88 .90 .92 .94 .97 .99 1.01 1.03 1.05 4.3 .88 .90 .92 .95 .97 .99 1.01 1.03 1.05 1.08 4.4 .90 .92 .95 .97 .99 1.01 1.03 1.06 1.08 1.10 4.5 .92 .94 .97 .99 1.01 1.03 1.06 1.08 1.10 1.13 4.6 .94 .97 .99 1.01 1.03 1.06 1.08 1.10 1.13 1.15 4.7 .96 .99 1.01 1.03 1 06 1.08 1.10 1.13 1.15 1.18 4.8 .98 1.01 1.03 1.06 1.08 1.10 1.13 1.15 1.18 1.20 4.9 1.00 1.03 1.05 1.08 1.10 1.13 1.15 1.18 1.20 1.23 5.0 1.02 1.05 1 07 1.10 1.12 1.15 1.18 1.20 1.23 1.25 5.1 1.05 1.07 1.10 1.12 1.15 1.17 1.20 1.22 1.25 1.27 5.2 1.07 1.09 1.12 1.14 1.17 1.20 1.22 1.25 1.27 1.30 5.3 1.09 1.11 1.14 1.17 1.19 1.22 1.25 1.27 1.30 1.32 5.4 1.11 1.13 1.16 1.19 1.21 1.24 1.27 1.30 1.32 1.35 5.5 1.13 1.15 1.18 1.21 1.24 1.26 1.29 1.32 1.35 1.38 5.6 1.15 1.18 1.20 1.23 1.26 1.29 1.32 1.34 1.37 1.40 5.7 1.17 1.20 1.23 1.25 1.28 1.31 1.34 1.37 1.39 1.43 5.8 1.19 1.22 1.25 1.28 1.30 1.33 1.36 1.39 1.42 1.45 5.9 1.21 1.24 1.27 1.30 1.33 1.36 1.39 1.42 1.45 1.48 6.0 1.23 1.26 1.29 1.32 1.35 1.38 1.41 1.44 1.47 1.50 Appendix. Table YIII. Relatiye value tables {Continued) 221 Price of milk per ]00 pounds, in dollars and CENTS. 3.0 .76 .78 .79 .81 .82 .84 .85 .87 .88 .90 3.1 .79 .81 .82 84 .85 .87 .88 .90 .91 .93 3.2 .82 .83 .85 .86 .88 .90 .91 .93 .94 .96 3.3 .84 .86 .87 .89 .91 .92 .94 .96 .97 .99 3.4 .87 .88 .90 .92 .93 .95 .97 .99 1.00 1.02 3.5 .89 .91 .93 .94 .96 .98 1.00 1.01 1.03 1.05 3.6 .92 .94 .95 .97 .99 1.00 1.03 1.04 1.06 1.08 3.7 .94 .96 .98 1.00 1.02 1.03 1.05 1.07 1.09 1.11 3.8 .97 .99 1.01 1.03 1.04 1.06 1.08 1.10 1.12 1.14 3.9 .99 1.01 1.03 1.05 1.07 1.09 1.11 1.13 1.15 1.17 4.0 1.02 1.04 1.06 1.08 1.10 1.12 1.14 1.16 1.18 1.20 4.1 1.05 1.07 1.09 1.11 1.13 1.15 1.17 1.19 1.21 1.23 4.2 1.07 1.09 1.11 1.13 1.15 1.18 1.20 1.22 1.24 1.26 4.3 1.10 1.12 1.14 1.16 1.18 1.20 1.23 1.25 1.27 1.29 4.4 1.12 1.14 1.17 1.19 1.21 1.23 1.25 1.28 1.30 1.32 4.5 1.15 1.17 1.19 1.21 1.24 1.26 1.28 1.30 1.33 1.35 4.6 1.17 1.20 1.22 1.24 1.26 1.29 1.31 1.33 1.36 1.38 4.7 1.20 1.22 1.25 1.27 1.29 1.32 1.34 1.36 1.39 1.41 4.8 1.22 1.25 1.27 1.30 1.32 1.34 1.37 1.39 1.42 1.44 4.9 1.25 1.27 1.30 1.32 1.35 1.37 1.40 1.42 1.45 1.47 5.0 1.27 1.30 1.32 1.35 1.37 1.40 1.42 1.45 1.47 1.50 5.1 1.30 1.33 1.35 1.38 1.40 1.43 1.45 1.48 1.50 1.53 5.2 1.33 1.35 1.37 1.40 1.43 1.46 1.48 1.51 1.53 1.56 5.3 1.35 1.38 1.40 1.43 1.46 1.48 1.51 1.54 1.56 1.59 5.4 1.38 1.40 1.43 1.46 1.48 1.51 1.54 1.57 1.59 1.62 5.5 1.40 1.43 1.46 1.48 1.51 1.54 1.57 1.60 1.62 1.65 5.6 1.43 1.46 1.48 1.51 1.54 1.57 1.60 1.62 1.65 1.68 5.7 1.45 1.48 1.51 1.54 1.57 1.60 1.62 1.65 1.68 1.71 5.8 1.48 1.51 1.54 1.57 1.59 1.62 1.65 1.68 1.71 1.74 5.9 1.50 1.53 1.56 1.59 1.62 1.65 1.68 1.71 1.74 1.77 6.0 1.53 1.56 1.59 1.62 1.65 1.68 1.71 1.74 1.77 1.80 222 Testing Milk and Its Products. Table IX. Butter chart, showing calculated yield of but= ter (in lbs.) from i to 10,000 lbs. of milk, testing 3.0 to 5.3 per cent. (See directions for use, p. 211). H 3.00 3.10 3.20 3.30 3.40 3.50 3.60 3.70 3.80 3.90 4.00 4. It I Milk Mlik lbs. lbs. io,ono 325 336 318 360 371 383 394 406 418 429 441 452 10, 000 9,000 293 302 313 324 334 345 ! 355 365 376 386 397 407 9,000 8,000 260 269 278 288 297 306 316 325 334 343 353 362 8,000 7,000 228 235 244 252 260 268 276 284 293 300 309 316 7,000 6,000 195 202 209 216 22:^ 230 236 244 251 257 265 271 6,000 5,000 163 168 174 180 186 192 197 203 209 215 221 226 5,000 4,000 130 134 139 144 148 163 158 162 167 172 176 181 4,000 3,0C0 97.5 101 104 108 111 115 118 122 125 129 132 136 3,000 2,000 65.0 07.2 69.6 72.0 74.2 76.6 78.8 81.2 83.6 85.8 88.2 90.4 2,000 1,000 32.5 33.6 34.8 36.0 37.1 38.3 39.4 40.6 41.8 43.9 44.1 45.2 1,000 900 29.3 30.2 31.3 32.4 33.4 34.5 35.5 36.5 37.6 38.6 39.7 40.7 900 800 26.0 26.9 27.8 28.8 29.7 30.6 31.5 32.5 33.4 34.3 35.3 36.2 800 700 22.8 23.5 24.4 25.2 26.0 26.8 27.6 28.4 29.3 30.0 30.9 31.6 700 600 19.5 20.2 20.9 21.6 22.3 23.0 23.6 24.4 25.1 25.7 26.5 27.1 600 500 16.3 16.8 17.4 18.0 18.6 19.2 19.7 20.3 20.9 21.5 22.1 22.6 500 400 13.0 13.4 13.9 14.4 14.8 15.3 15.8 16.2 16.7 17.2 17.6 18.1 400 300 9.7 10.1 10.4 10.8 11.1 11.5 11.8 12.2 12.5 12.9 13.2 13.6 300 200 6.5 6.7 6.9 7.2 7.4 7.6 7.9 8.1 8.3 8.6 8.8 9.0 200 100 3.2 3.4 3.5 3.6 3.7 3.8 3.9 4.1 4.2 4.3 4.4 4.5 100 90 2.9 3.0 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 4.1 90 80 2.6 2.7 2.8 2.9 3.0 3.1 3.2 3.3 3.4 3.4 3.5 3.6 80 70 2.3 2.3 2.4 2.5 2.6 2.7 2.8 2.8 2.9 3.0 3.1 3.2 70 60 1.9 2.0 2.1 2.2 2.2 2.3 2.4 2.4 2.5 2.6 2.7 2.7 60 50 1.6 1.7 1.7 1.8 1.9 1.9 2.0 2.0 2.1 2.2 2.2 2.3 50 40 1.3 1.3 1.4 1.4 1.5 1.5 1.6 1.6 1.7 1.7 1.8 1.8 40 30 1.0 1.0 1.0 1.1 1.1 1.2 1.2 1.2 1.3 1.3 1.3 1.4 30 20 .6 .7 .7 .7 .7 .8 .8 .8 .8 .9 .9 .9 20 10 .3 .3 .4 .4 .4 .4 .4 .4 4 .4 .4 .5 10 9 .3 .3 .3 .3 .3 .3 .4 .4 .4 .4 .4 .4 9 8 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .4 .4 8 7 .2 .2 .2 .3 .3 .3 .3 .3 .3 .3 .3 .3 7 6 .2 .2 .2 .2 .2 2 .2 2 .3 .3 .3 .3 6 5 .2 .V .2 2 .2 .2 .2 2 2 2 .2 5 4 .1 .1 1 .2 .2 .2 .2 .2 .2 ,'■1 .2 .2 4 3 .1 .1 .1 .1 .1 .1 .1 .1 .1 .1 .1 .1 3 2 1 .1 .1 .1 .1 .1 .1 .1 .1 .1 .1 .1 .1 i 2 1 1 1 3.00 3.10 3.20 3.30 3.40 3.50 3.60 3.70 3.80 3.90 4.00 4.10 J_ Appendix. Table IX. Butter chart (Continued). 223 1 4.20 4.30 4.40 4.50 4.60 4.70 4.80 4.90 5.00 5.10 5.20 5.30 t Milk Milk lb3. . lbs. 10, 000 ! 464 476 487 499 510 522, 534 545 557 568 580 592 10, 000 9, 000 i 418[ 42a 438 ' 449 459 4701 481 491 501 511 522 533 1 9,000 8,000 371 381 325 333 390 399 408 418; 427 436 446 454 464 ' 474 ' 8, 000 7,000 1341 349 357 ' 365 374 382 390 398 406 414 7, 000 6,000 : 278i 286 • 292 299 306 ! 313' 320 327 ' 334 341 348 , 355 6,000 5,000 i 232 238 244 250 255 261 267 273 279 284 290 296 5,000 4,000 ! 186| 190 195 200 ! 204 209, 214 218 223 227 232 237 i 4, 000 3,000 1391 143 146 150 153 157 160 164 167 170 174 178 ; 3, 000 2,000 192.8 95.2 97.4 99.8 102 104 107 109 111 114 116 118 1 2, 000 1,000 46.447.6 48.7 49.9 51.0 52.2 53.4 54.5 55.7 56.8 58.0 59.2 1,000 900 41.842.8 43.8 44.9 45.9 47.0 48.1 49.1 50.1 51.1 52.2 53.3 900 800 37-.l;38.1 39.0 39.9 40.8 41.8 42.7 43.6 44.6 45.4 46.4 47.4 800 700 32.5:33.3 34.1 34.9 35.7 36.5 37.4 38.2 39.0 39.8 40.6 41.4 i 700 600 127.8|28.6 29.2 29.9 30.6 31.3 32.0 32.7 33.4 34.1 34.8 35.5 600 500 23.2 23.8 24.4 25.0 25.5 26.1 26.7 27.3 27.9 28.4 29.0 29.6 500 400 II8.6 19.0 19.5 20.0 20.4 20.9 21.4 21.8 22.3 22.7 23.2 23.7 400 300 13.9 14.3 14.6 15.0 15.3 15.7 16.0 16.4 16.7 17.0 17.4 17.8 300 200 9.3 9.5 9.7 10.0 10.2 10.4 10.7 10.9 11.1 11.4 11.6 11.8 200 100 4.6 4.8 4.9 5.0 5.1 5.2| 5.3 5.5 5.6 5.7 5.8 5.9 100 90 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.0 5.1 5.2 5.3 90 80 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.5 4.6 4.7 80 70 3.3 3.3 3.4 3.5 3.6 3.7 3.7 3.8 3.9 4.0 4.1 4.1 70 60 2.8 2.9 2.9 3.0 3.1 3.1! 3.2 3.3 3.3 3.4 3.5 3.6 60 50 2.3 2.4 2.4 2.5 2.6 2.6: 2.7 2.7 2.8 2.8 2.9 3.0 50 40 1.9 1.9 2.0 2.0 2.C 2.1 2.1 2.2 2.2 2.3 2.3 2.4 40 30 1.4 1.4 1.5 1.5 1.5 1.6 1.6 1.6 1.7 1.7 1.7 1.8J 30 20 .9 1.0 1.0 1.0 1.0 1.0 1.1 1.1 1.1 1.1 1.2 1.2 20 10 .5 .5 .5 .5 .5 .5 j .5 .6 .6 .6 .6 .6 10 9 .4 .4 .4 .5 .5 M .5 .5 .5 .5 .5 .5! 9 8 .4 .4 .4 .4 .4 A\\ .4 .4 .5 .5 .5 . c 8 71 .3 .s .3 .4 .4 .4,1 .4 .4 .4 .4 .4 .4 7 61 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .4 .4 6 5[ .2 .2 .2 .3 .3 M .3 .3 .3 .3 .3 .3 5 4 .2 .2 .2! .2 .2 .2[! .2 .2 .2 2 .2 .2 4 3 .1 .1 2 .2 .2 .2 .2 .2 .2 .2 .2 .2 3 2 .1 .1 A .1 .1 .1 .1 .1 .1 .1 .1 .1 2 1 .1 .1 .1 .1 .1 .1 .1 .1 .1 .1 .1 .1 1 1 i 4.20 1 4.30 4.40 4.50 4.60 4.70 4.80 4.90 5.00 1 5.IO1 5.20 5.30 t 224 Testing Milk and Its Products, Table X. Overrun table, showing pounds of but= ter from one hundred lbs. of milk. See direc- tions for use, p. 170. Per cent. fat. 1. 10 1.11 1.12 1.13 1.14 1.15 1.16 1.17 1.18 1.19 1.20 Per cent, fat. 3.0 3.30 3.33 3.36 3.39 3.42 3.45 3.48 3.51 3.54 3.57 3.60 3.0 3.1 3.41 3.44 3.47 3.50 3.53 3.57 3.60 3.63 3.66 3.68 3.72 3.1 3.2 3.52 3.55 3.58 3.62 3.65 3.68 3.71 3.74 3.78 3.81 3.84 3.2 3.3 3.63 3.66 3.70 3.73 3.76 3.80 3.83 3.86 3.89 3.93 3.96 3.3 3.4 3.74 3.77 3.81 3.84 3.88 3.91 3.94 3.98 4.01 4.05 4.08 3.4 3.5 3.85 3.89 3.92 3.96 3.99 4.03 4.06 4.10 4.13 4.17 4.20 3.5 3.6 3.96 4.00 4.03 4.07 4.10 4.14 4.18 4.21 4.25 4.28 4.32 3.6 3.7 4.07 1.11 4.14 4.18 4.22 4.26 4.29 4.33 4.37 4.40 4.44 3.7 3.8 4.18 4.22 4.26 4.29 4.33 4.37 4.41 4.45 4.48 4.52 4.56 3.8 3.9 4.29 4.33 4.37 4.41 4.45 4.49 4.52 4.56 4.60 4.64 4.68 3.9 4.0 4.40 4.44 4.48 4.52 4.56 4.60 4.64 3.68 4.72 4.76 4.80 4.0 4.1 4.51 4.55 4.59 4.63 4.67 4.72 4.76 4.80 4.84 4.88 4.92 4.1 4.2 4.62 4.66 4.70 4.75 4.79 4.83 4.87 4.91 4.96 5.00 5.04 4.2 4.3 ,4.73 4.77 4.82 4.86 4.90 4.95 4.99 5.03 5.07 5.12 5.16 4.3 4.4 4.84 4.88 4.93 4.97 5.02 5.06 5.10 5.15 5.19 5.24 5.28 4.4 4.5 '4.95 5.00 5.04 5.09 5.13 5.18 5.22 5.27 5.31 5.36 5.40 4.5 4.6 5.06 5.11 5.15 5.20 5.24 5.29 5.34 5.38 5.43 5.47 5.52 4.6 4.7 5.17 5.22 5.26 5.31 0.36 5.41 5.45 5.49 5.55 5.59 5.64 4.7 4.8 5.28 5.33 5.38 5.42 5.47 5.52 5.57 5.62 5.66 5.71 5.76 4.8 4.9 5.39 5.44 5.49 5.54 5.59 5.64 5.68 5.73 5.78 5.83 5.88 4.9 5.0 5.50 5.55 5.60 5.65 5.70 5.75 5.80 5.85 5.90 5.95 6.00 5.0 5.1 5.61 5.66 5.71 5.76 5.81 5.87 5.92 5.97 6.02 6.07 6.12 5.1 5.2 5.72 5.77 5.82 5.88 5.93 5.98 6.03 6.08 6.14 6.19 6.24 5.2 5.3 5.83 5.88 5.94 5.99 6.04 6.10 6.15 6.20 6.25 6.31 6.36 5.3 5.4 5.94 5.99 6.05 6.10 6.16 6.21 6.26 6.32 6.37 6.43 6.48 5.4 5.5 6.05 6.11 6.16 6.22 6.27 6.33 6.38 6.44 6.49 6.55 6.60 5.5 5.6 6.16 6.22 6.27 6.33 6.38 6.44 6.50 6.55 6.61 6.66 6.72 5.6 5.7 6.27 6.33 6.38 6.44 6.50 6.56 6.61 6.67 6.73 6.78 6.84 5.7 5.8 6.38 6.44 6.50 6.55 6.61 6.67 6.73 6.79 6.84 6.90 6.96 5.8 5.9 6.49 6.55 6.61 6.67 6.73 6.79 6.84 6.90 6.96 7.02 7.08 5.9 6.0 6.60 6.66 6.72 6.78 6.84 6 . 90 6.96 7.02 7.08 7.14 7.20 6.0 Affendi: 225 Table XI. Yield of cheese, corresponding to 2.5 to 6 percent of fat with lactometer readings 26 to 36. (See p. 172). Lactometrr degrees. 26 27 28 1 29 30 31 32 33 34 35 36 2.5 7.28 7.41 7.54 7.67' 7.81 7.94 8.07 8.20 8. S3 8.47 8.60 2.5 2.6 7.44 7.57 7.70 7.83 7.96 8.09 8 . 22 8.35 8.49 8.62 8.76 2.6 2.7 7.59 7.72 7.85 7.99 8.12 8.25 8.38 8.51 8.64 8.77 8.91 2.7 2.8 7.74 7.87 8.00 8.14 8.2; 8.40 8.53 8.67 8.80 8.91 9.07 2.8 2.9 7.90 8.03 8.16 8.30 8.34 8.56 8.69 8.82 8.95 9.09 9.22 2.9 3.0 8.05 8.18 8.31 8.45 8.58 8.71 8.81 8.97 9.11 9.24 9.37 3.0 3.1 8.21 8.34 8.47 8.60 8.74 8.87 9.00 9.13 9.26 9.39 9.53 3.1 3.2 8.36 8.49 8.62 8.75 8.89 9.02 9.15 9.28 9.42 9.55 9.68 3.2 3.3 8.52 8.65 8.78 8.91 9.05 9.18 9.31 9.44 9.57 9.70 9.84 3.3 3.4 8.67 8.80 8.93 9.06 9.20 9.33 9.46 9.59 9.73 9.86 9.99 3.4 3.5 8.82 8.96 9.09 9.22 9.3o 9.48 9.62 9.75 9.88 10.01 10.15 3.5 3.6 8.98 9.11 6.24 9^37 9.50 9.63 9.77 9.90 10.03 10.17 10.30 3.6 3.7 9.13 9.26 9.39 9.52 9.65 9.78 9.92 10.05 10.19 10.32 10.46 3.7 3.8 9.29 9.42 9.55 9.68 9.81 9.94 10.08 10.21 10.34 10.48 10.61 3.8 3.9 9.44 9.57 9.70 8.84 9.97 10.10 10.23 10.36 10.50 10.64 10.77 3.9 4.0 9.60 9.73 9.86 10.00 10.13 10.26 10.39 10.53 10.66 10.79 10.93 4.0 4.1 9.75 9.88 10.02 10.15 10.28 10.39 10.54 10.68 10.81 10.94 11.08 4.1 4.2 9.90 10.03 10.17 10.30 10.43 10.57 10.70 10.84 10.97 11.10 11.24 4.2 4.3 10.06 10.19 10.32 10.45 10.58 10.72 10.85 10.99 11.12 11.25 11.39 4.3 4.4 10.21 10.34 10.48 10.61 10.74 10.87 11.00 11.14 11.27 11.41 11.55 4.4 4.5 10.36 10.49 10.63 10.76 10.79 11.03 11.16 11.29 11.42 11.56 11.70 4.5 4.6 10.52 10.65 10.78 10.92 11.05 11.18 11.31 11.45 11.58 11.71 11.85 4.6 4.7 10.67 10.81 10.94 11.07 11.20 LI 34 11.47 11.60 11.73 11.87 12.01 4.7 4.8 10.83 10.96 11.09 11.22 11.36 11.49 11.62 11.76 11.89 12.02 12.16 4.8 4.9 10.9- 11.11 11.25 11.38 11.51 11.60 11.78 11.91 12.04 12.18 12.32 4.9 5.0 11.14 11.27 11.40 11.54 11.67 11.80 11.93 12.07 12.20 12.34 12.48 5.0 5.1 11.29 11.42 11.55 11.69 11.82 11.96 12.09 12 23 12.36 12.49 12.63 5.1 5.2 11.45 11.58 11.71 11.85 11.98 12.11 12.24 12.38 12.52 12.66 12.80 5.2 5.3 11.60 11.73 11.86 11.99 12.13 12.27 12.40 12.53 12.67 12.71 12.85 5.3 5.4 11.76 11.89 12.02 12.16 12.29 12.42 12.55 12.69 12.83 12.97 13.01 5.4 5.5 11.91 12.04 12.17 12.31 12.44 12.5812.71 12.85 12.99 13.12 13.25 5.5 5.6 12.07 12.20 12.33 12.47 12.60 12.73 12.87 13.00 13.14 13.28 13.41 5.6 5.7 12.22 12.35 12.48 12.52 12.75 12.89 13.02 13.16 13.30 13.44 13.57 5.7 5.8 12.38 12.51 12.64 12.77 12.91 13.05 13.18 13.31 13.45 13.59 13.72 5.8 5.9 12.53 12.66 11.79 12.93 13.06 13.19 13.33 13.47ll3.60'i3.74 13.87 5.9 6.0 12.69 12.82 12.95 12.09 13.22 13.35 13.49 13.62 13.75 13.89 14.02 6.0 226 Testing Milk and Its Prodiccts. Table XII. Comparisons of Farenheit and centigrade (Celsius) thermometer scales. Fahren- Centi- Fahren- Centi- Fahren- Centi- heit. grade. heit. grade. heit. grade. + 212 + 100 + 176 + 80 + 140 + 60 211 99.44 175 79.44 139 59.44 210 98.89 174 78.89 138 58.89 209 98.33 173 78.33 137 58.33 208 97.78 172 77.78 136 57.78 207 97.22 171 77.22 135 57.22 206 96.67 170 76.67 154 56.67 205 96.11 169 76.11 153 56.11 204 95.55 168 75.55 152 55.55 203 95 167 75 131 55 202 94.44 166 74.44 130 54.44 201 93.89 165 73.89 129 53.89 200 93.33 164 72.33 128 53.33 199 92.78 163 72.78 127 52.78 198 92.22 162 71.22 126 52.22 197 91.67 161 71.67 125 51.67 196 91.11 16'» 71.11 124 51.11 195 90.55 159 70.55 123 50.55 194 90 158 70 122 50 193 89.44 157 69.44 121 49.44 192 88.89 156 68.89 120 48.89 191 88.33 155 68.33 119 48.33 190 87.78 154 67.78 118 47.78 189 87.22 153 67.22 117 47.22 188 86.67 152 66.67 116 46.67 187 86.11 151 66.11 115 46.11 186 85.55 150 65.55 114 45.55 185 85 149 65 113 45 184 84.44 148 64.44 112 44.44 183 83.89 147 63.89 111 43.89 182 83.33 146 63.33 110 43.33 181 82.78 145 62.78 ' 109 42.78 180 82.22 144 62.22 i 108 42.22 179 81.67 143 61.67 107 41.67 178 81.11 142 61.11 106 41.11 177 80.55 141 60.55 ! 105 40.55 Appendix. 227 Table XII. Comparisons of tliermometer scales {Continued) Fahren- Centi- Fahren- Centi- Fahren- Centi- heit. grade. heit. grade. heit. grade. + 104 + 40 + 68 + 20 + 32 + 103 39.44 67 19.44 31 —0.55 102 38.89 66 18.89 30 1.11 101 38.33 65 18.33 29 1.67 100 37.78 64 17.78 28 2.22 99 37.22 63 17.22 27 2.78 98 36.67 62 16.67 26 3.33 97 36.11 61 16.11 25 3.89 96 35.55 60 15.55 24 4.44 95 35 59 15 23 5 94 34.44 58 14.44 22 5.55 93 33.89 57 13.89 21 6.11 92 33.33 56 13.33 20 6.67 91 32.78 55 12.78 19 7.22 90 32.22 54 12.22 18 7.78 89 31.67 53 11.67 17 8.33 88 31.11 52 11.11 16 8.89 87 30.55 51 10.55 15 9.44 86 30 50 10 14 10 85 29.44 49 9.44 13 10.55 84 28.89 48 8.89 12 11.11 83 28.33 47 8.33 11 11.67 82 27.78 46 7.78 10 12.22 81 27.22 45 7 22 9 12.78 80 26.67 44 6^67 8 13.33 79 26.11 43 6.11 t 13.89 78 25.55 42 5.55 6 14.44 77 25 41 5 5 15.00 76 24.44 40 4.44 4 15.55 75 23.89 39 3.89 3 16.11 74 23.33 38 3.33 2 16.67 73 22.78 37 2.78 1 17.22 72 22.22 36 2.22 17.78 71 21.67 35 1.67 —1 18.33 70 21.11 34 1.11 2 18.89 69 20.55 33 0.55 3 19.44 To convert deg. Fahrenheit to corresponding deg. Centigrade : Subtract 32, multiply difference by 5, and divide by 9. Example: Which degree Centigrade corresponds to 110° F.? 110 — 32 — 78; 78 X 5 = 390 ; 390 -^ 9 = 43.33. To convert deg. Centigrade to corresponding deg. Fahrenheit : Multiply by 9, divide product by 5, and add 32 to quotient. Example : Which degree Fahrenheit corresponds to 95.5° C? 95.5 X 9 — 859.5; 859.5 ^ 5 = 171 9 ; 171.9 + 32 = 203.6. Table XIII. Comparison of metric and customary weights and measures. Customary- weights and measures. 1 inch 1 foot 1 mile 1 square inch. 1 square foot. 1 square yard 1 acre 1 cubic inch.... 1 cubic foot..,. 1 cubic yard... 1 bushel 1 fluid ounce... 1 quart 1 gallon 1 grain.. 1 ounce (av.).. 1 pound (av.) Equivalents in metric system 2.54 centimeters. .3048 meter. 1.6094 kilometers. 6.452 sq. centimeters 9.29 sq. decimeters. .836 sq. meter. .4047 hectare. 16.387 c. c. .0283 cub. meter. .765 cub. meter, .3525 hectoliter. 29.57 c. c. .9464 liter. 3.7854 liters. 64.8 milligrams. 28.35 grams. .4536 kilogram. Metric weights and measures. 1 meter 1 meter 1 kilometer 1 sq. centimeter. 1 square meter.. 1 square meter.. 1 hectare 1 c. c 1 cub. decimeter. 1 cub. meter 1 hectoliter 1 c. c 1 liter 1 decaliter 1 gram 1 gram 1 kilogram Equivalents in customary system. 39.37 inches. 1.0936 yards. .6214 mile. .155 sq. inch. 10.764 sq. feet. 1.196 sq. yards. 2.471 acres. .061 cubic inch. 61.023 cubic inches. 35.314 cub. feet. 2.8377 bushels. .0338 fluid ounce. 1.0567 quarts. 2.6417 quarts. 15.43 grains. .035274 ounce. 2.2046 pounds (av.) (228) Appendix, 229 JUQQE5TI0NS regarding the organization of co= operative creameries and cheese factories. When the farmers of a neighborhood are considering the estab- lishment of a creamery or cheese factory, they should first of all make an accurate canvas of the locality to ascertain the number of cows that can be depended on to supply the factory with milk. The area which may be drawn from, will vary according to the kind of factory which it is desired to operate. A successful sepa- rator creamery will need at least 400 cows within a radius of four to five miles from the proposed factory.* Small cheese fac- tories can be operated with less milk, and gathered-cream and batter factories, generally cover a much larger territory than that mentioned. In all cases, however, the question of the number of cows con- tributing to the enterprise must be fully settled before further steps are taken, since this is the vital point, and one upon which success will largely depend. Methods of organization. The farmers should form their own organization, and not accept articles of agreement proposed by traveling agents. An agreement to supply milk from a stated number of cows should be signed by all who expect to join the association. When a sufficient number of cows has been pledged to insure the successful operation of a factory, the farmers agree- ing to supply milk should meet and form an organization. This may be done according to either of the following plans which have been known to give good satisfaction. Raising money for building and equipment. First.— Each member will sign an agreement to pay on or be- fore a given date for a certain number of shares or stock in the company at dollars per share; or, Second.— Xn elected board of directors may be authorized to borrow a sum of money not exceedieg thousand dollars on their individual responsibility, and the sum of cents, (usually five cents) per hundred pounds of milk received at the factory shall be reserved for the payment of this borrowed money. Bull. 56, Wisconsin experiment station. 230 Testing Milk and Its Products. Constitution and by-laws of co-operative associations are drawn up and signed by the prospective members of the associa- tion as soon as possible after it has been determined to form such an association. As it is impossible to include in an illustration all the articles and rules that ma\' be found useful in each partic- ular instance, the following suggestions in regard to some of the points to be included in the documents are given as a guide only. It may be found advisable to modifj' them in various ways to meet the needs of the organization to be formed. After the constitution and by-laws have been drawn up and made plain to all the members of the association, they should be printed and copies distributed to all parties interested. CONSTITUTION OR Articles of Agreement of the Association.* 1. The undersigned, residents within the counties of. , state of. , hereby agree to become members of the co-operative association, which is formed for the purpose of man- ufacturing butter or cheese from w^hole milk. 2. The regular meetings of the association shall be held annu- ally on the day of the month of Special meetings may be called by the president, or on written request of one-third of the members of the association, provided three days' notice of such meeting is sent to all members. Meetings of the board of directors may be called in the same way, either by the president, or by any two members of the board of directors. 3. The members of the association, or three of the board of directors, shall constitute a quorum for the transaction of busi- ness. 4. The officers of the association shall include president, secre- tary, treasurer, one of whom is also elected manager, and these officers together with three other members of the association * The following publications have been freely used in preparing this constitu- tion and by-laws: Woll, Handbook f. Farmers and Dairymen; Minn, experiment station, bull. No. 35; Ontario Agricultural College, special bulletin, May 1897. Appendix. 231 shall constitute the board of directors. Each of these six officers shall be elected at the annual meeting and hold office for one year, or until their successors have been elected and qualified. \ny va- cancies in the board of directors may be filled by the directors un- til the next annual meeting of the association. 5. The duties of the president shall be to preside at all meet- ings of the association, and perform the usual duties of such pre- siding officers. He shall sign all drafts and documents of any kind relating to the business of the association, and pay all money which comes into his possession by virtue of his office, to the treasurer, taking his receipt therefor. He shall call special meetings of the association whenever it is deemed necessar^^ In the absence of the president, one of the board of directors shall temporarily fill the position. 6. The secretary shall attend all business meetings of the asso- ciation and of the board of directors, and shall keep a careful rec- ord of the minutes of the meetings. He shall also give notices of all meetings and all appointments on committees, etc. He shall sign all papers issued, conduct the correspondence and general business of the association, and keep a correct financial account between the association and its members. He shall have charge of all property of the association, not otherwise disposed of, give bonds for the faithful performance of his duties, and receive such compensation for his services as the board of directors may deter- mine. 7. The treasurer shall receive and give receipt for all money belonging to the association, and pa^^ out the same upon orders signed by the president and the secretary. He shall give such bonds as the board of directors may require. 8. The board of directors shall audit the accounts of the asso- ciation, invests its funds, appoint agents, and determine all com- pensations. They shall prescribe and enforce the rules and regula- tions of the factory. They shall cause to be kept a record of the weights and tests of the milk or cream received from each patron, the products sold, the running expenses, etc., and shall divide among the patrons the money due them each month. They shall 232 Testing Milk and Its Products. also make some provision for the withdrawal of any member from the association, aod make a report in detail to the association at the annual meeting. Such report shall include the gross amount of milk handled during the year, the receipts from pro- ducts sold, and all other receipts, the amount paid for milk, also for running expenses, and a complete statement of all other methods pertaining to the business association. 9. Among the rules and regulations to be enforced by the board of directors may be included some or all of the following : a. Patrons shall furnish all the milk from all the cows promised at the organization of the association. b. Only sweet and pure milk will be accepted at the factory, and any tainted or sour milk milk shall be refused. c. The milk of each patron shall be tested at least three times a month. d. Any patron proved to be guilty of watering, skimming or otherwise adulterating the milk sent to the factory, or by taking more than 80 pounds of skim milk or whey for ever^^ 100 pounds of whole milk delivered to the factory, shall be fined as agreed by the association. e. A patron's premises may be inspected at any time by the board of directors, or their authorized agent, for the purpose of suggesting improvements in the methods of caring for the milk or the cows, in drainage and general cleanliness; or to secure sam- ples of the milk of his cows for examination when it is deemed necessary. 10. Any changes or amendments to the by-laws or constitu- tion of the association must be made in writing b^^ the parties proposing the same, and posted prominently in a conspicuous place at the creamery at least two weeks previous to their being acted upon. Such changes to be in force must be adopted by a two-thirds vote of the stockholders. 11. In voting at any annual or special meeting of the associa- tion, the members shall b^ entitled to one vote for each cow sup- plying milk to the factorj^ or for each share of the stock owned by them, as agreed upon. INDEX The numbers refer to pages in the book. Acid measures, 42. Acidity of cream, 104; estimation of, 106. Acidity of milk, cause of, 94; determina- tion of, 94, 195; methods of testing, 95. Adulteration of milk, 88; calculation of, 90. Adulterated butter, 200; cheese, 203. Albumen 13, determination of, in milk, 191, 193. Albaminoids, 13. Albumose, 14. Alkaline tablet test, 99, Alkaline tabs, 101). Amphoteric reaction of milk, 94. Appendix, 205. Artificial butter, detection of, 200. Ash, determination of, in butter, 198; in cheese, 2)3; in milk, 17, 194. Automatic milk scale, 121. Babcock test, the, 6, 2-5; Bartlett's modi- fication of, 63; directions for, 2t; dis- cussion of details, 34; for butter milk, 74, 77; for cheese, 77; for condensed milk, 79; for cream, 64, 151; for skim milk, 74; for whey, 74, 77; glassware used in, 34; modificitions of, 62; scales for weighing cream, cheese, etc., 71; water to be used in, 60. Bartlett's modification of Babcock test, 63. Beimling test, 5. Bi-carbonate of soda, detection of. in milk, 196. Bi-chromate of pjtash, 141. Board of health degrees, 83. Boracic acid, 195. Borax, 195. B. & W. bottle, 76. Butter, artificial, 13; detection of, 200. Butter chart, 222; use of, 168. Butter, chemical analysis of, 197; com- plete analysis in same sample, 199; composition of, 205; determination of ash, 198; casein, 198; fat, 198; water, 197; sampling for analysis, 197; varia- tions in composition, 161; yield, cal- culation of, 160. Butter fat, conversion factor for, 167; determination of specific gravity, 200; volatile fatty acids, 201; expansion co- efficient, 33; price per pound, 174; table showing amounts due for, at 12 to 25 cents per pound, 216; test and yield of butter, 160. Butter milk, Babcock test for, 74, 77; chemical analysis of, 197: composition of, 205. Calculation of adulteration, 90; of milk solids, 85; of yield of butter, 160, 167, 169; of cheese, 171; of dividends, at creameries, 174; at cheese factories, 182. Calibration of glassware, 43. Carbohydrates, 15. Caseia, 13; determination of, in butter, 198; in cheese, 203; in milk, 191, 192. Centrifugal machines, 47. Chamberland filters, 14. Cheese, 77; calculating yield of, from casein and fat, 173; from fat, 171; fiom Folids not fat and fat, 171; composition, 205; chemical analysis of, 202; deter- mination of ash, 263; casein, 203; fat, 202; water, 202; "filled", detection of, 203; sampling, 77; yield, calculation of, 160, 171; yield of, and quality of milk, relation between, 171. Cheese factories, calculating dividends at, 182; co-operative, 185; proprietary, 184. Chemical analysis of butter, 197, 199; butter milk, 197; cheese, 202; milk, 186; skim milk, 197; whey, 197. Cholesterin in milk, 19. Citric acid in milk, 19. Cleaning solutions for test of bottles, 39. Cleaning test bottles, 36; apparatus for, 38. Cochran's test, 4. Coloring matter, foreign, in milk, de- tection of, 92. Colostrum milk, 19; composition of, 205. Composite samples, 134; care of, 143; 234 Testing Milk and Its Products. case for holding, 157; methods of tak- ing, 134; preservatives for, 140. Composite sampling, by use of drip sample, 136; one-third sample pipette, 138; Scovell sampling tube, 136; tin dipper, 134. Composition of butter, 205; butter milk, 205; cheese, 205; colostrum milk, 205; cream, 205; milk, 205; skim milk, 205; whey, 205. Condensed milk, composition of, 205; testing of, 79. Conversion factor for butter fat, 167. Conversion tables for thermometer scales, 226; for weights and measures, 228. Cow, a, when to test, 123. Coivs, number of tests required in test- ing, 121. Cows' milk, composition of, 205. Cream, acidity of, 104; avoiding errors of measuring in testing, 67; Babcock test for, 154; bottles, the bulb-necked, 68; the VVinton, 09; care in sampling, necessity of, 154; determination of acidity in, 100, 108; errors of measur- ing in testing, 65; separation of, in- fluence of temperature, 159; spaces, 150; specific gravity, 66; testing, 64 testing outfit, 155; testing at cream eries, 150; use of 5 c. c. pipette in, 71 use of milk test bottles in, 69; test bottles 67; weighing in cream testing 71; weight delivered by a 17.6 ec. pi pette, 66. Creameries, calculating dividends at 174, 176; co-operative, 175; cream test ing at, 150; proprietary, 175. Creamery inch, 1. Curd test, the Wisconsin, 111. DeLaval's butyrometer, 8. Devarda's acidimeter, 99. Diameter of tester and speed required, relation between, 50. Dividends, calculating, at cheese factories, 182; at creameries, 174. Draining rack for test bottles, 39. Expansion coeflBcient for butter fat, 33. Failyer and Willard's test, 4. Farrington's alkaline tablet test, 99. Fat, 12; color of, an index to strength of acid used, 58; content, causes of variation in, 120; determination of, in butter, 198; in cheese, 202; in milk, 190; globules, 12; influence of tem- perature on separation of, 59; measur- ing of, in cream testing, 73; in milk testing, 32; pounds in 1-10,000 lbs. of milk, testing 3 to 5.35 per cent., 212; speed required for complete separa- tion of, 48. Fermentation test, the, 113. Filled cheese, detection of, 203. Fjord's centrifugal cream test, 9. Fluorids, detection of, in milk, 195. Food, influence of, on quality of milk, 131. Fool pipettes, 40. Formaline, detection of, in milk, 197. Frozen milk, sampling of, 24. Gauges of cream, 150. Gerber's acid-butyrometer, 7; fermenta- tion test, 113. Glassware used in the Babcock test, 34; calibration of, 43. Globulin, 14. Glycerides of fatty acids, 12. Goat cheese, 14. Grain feeding, heavy, influence of, on quality of milk, 130. Hand testers, 52. Hemi-albumose, 14. Herd milk, variations in, 128; ranges in variation, of 129. Hypoxanthin, 19. Introduction, 1. Iowa station test, 5. Lactic acid in milk, 16. Lactocrite, 5. Lactose, 15. Lactoc hrome, 10. Lactometer, the, and its application, 80; degrees, 81; N. Y. board of health, 83; Quevenne, 80; reading the, 84; time of taking readings, 85. Index. 235 Lecithin iu milk, 19. LetFmann and Beam test, 5. Le2:al standards for milk, 89, 206. Liebermann's method, 5. Manns' test, 96. Measuring fat column in testing cream, 73; in testing milk, 32. Meicury, calibration with, 43; cleaning, 44. Metric and customary systems of weights and measures, comjiarison of, 228. Milk, acidity of, 94; adulteration of, 88; amphoteric reaction of, 94; ash, com- position of, 18; chemical analysis of, 186; cholesterin in, 19; citric acid in, 19; colostrum, 19; composition of, 11; table showing composition of, 205; composite sampling of, 134; condensed, 79, 205; correction table for specific gravity of, 208; detection of preserva- tives in, 109; determination of acidity, 106, 195; of ash, 194; of casein and albumen, 191, 192, 193; of fat, 190; of milk sugar, 193; of specific gravity, 186; of water, 188, 190; fat available for butter in different grades of, 165; from cows in heat, 89; from sick cows, 89; from single cows, sampling of, 126; variations in, 116; frozen, sampling of, 24; gases, 19; hypoxanthin, 19; lacto- chrome, 19; lecithin, 19; mineral com- ponents, 17; partially churned, sampl. Ing of, 21; quality of, influence of food, 131; of heavy grain feeding, 130; of pasture, 181; method of improving, 132; samjiling, 20; scales, 124; serum, 11; skimming, 90; solids, 11; calcula- tion of, 85; sour, sampling of, 23, 26; standards, 89, 206; sugar, 15; testing purity of. 111; urea, 19; watering of, 91; watering and skimming, 91. Milk test, a practical, need of, 1; re- quirements of, 6; bottle, use of, in testing cream, 69; Russian, 62. Milk tests, Beimling (Leffmann and Beamj 5; Cochran, 4; DeLaval butyrometer, 8; Failyer and Wil- lard, 4; Fjord, 9; foreign, 7; Gerber acid-butyrometer, 8; introduction of, 3; lactocrite, 5; Liebermann, 6; Par- son, 4; Patrick (Iowa station test,) 5; Roese-Gottlieb, 5; Schmied, 5; Short, 4; Thoerner, 5. Milk testing, 26; on the farm, 116. N. Y. board of health lactometer, 83; degrees corresponding to Quevenne lactometer degrees, 207. Non-fatty milk solids, 11, Normal solutions, 96. "No-tin" test, 52. Nucleiu, 14. Oil test churn, 2, 151. Ohlsson test bottle, 76. Oleomargarine, detection of, 200; cheese, detection of, 203, One-third sampling pipette, use of, 138. Organization of co-operative creameries and cheese factories, suggestions con- cerning, 229. Overrun, 163; calculation of, 167; factors influencing, 163; t?ble, 224; use of, 170. Parson's test, 4. Pasture, influence of, on quality of milk, 131. Patrick's test. 5. Patron's dilemma, a, 146. Percentages, average, methods of cal- culation, 145; fallacy of averaging, 144. Phenolphtalein, 97. Pipettes, 39; proper construction of points, 40. Potassium bi-cbromate, 141. Power testers, 53. Preservaline, 109; detection of, in milk, 109. Preservatives, for composite samples, 140; in milk, detection of, 196. Primost, 14. Proteose, 14. Quevenne lactometer, the, 80; degrees corresponding to scale ofN. Y. board of health lactometer, 207. Record of tests of cows, 126. Reichert number, 202. Reichert-Wollny method, 201. 236 Testing Milk and Its Products. Relative value tables. 178, 180, 218. Resei'voir for water in Babcock test, 61. Roese-Gottlieb method, 5. Russian milk test, the, 62. Salicylic acid, in milk, detection of, 196. Sampling cheese, 77; milk. 20, 26; milk from single cows, 126. Schmied method, the, 5. Scovell sampling tube, 136. Serum solids, IJ. Short's test, 4. Skimming of milk, detection of, 90. Skimmilk, Babcock test for, 74; chemi- cal analysis of, 197; composition of, 205; test bottles, 76, 77. Solids not fat, 11; formula for calculat- ing, 86; table showing, corresponding to 0-6 per cent, fat and 26-36 lacto- meter degrees, 209. Sour milk, sampling of, 23. Spice system, the, 150. Specific gravity, 80; cylinders, 84; in- fluence of temperature, 81; of butter fat. determination of, 200: of milk, determination of, 186; temperature correction table, 208. Speed required for complete separation of fat, 48 Spillman's cylinder, 195. Steam turbine testers, 53. Sulfuric acid, 54; table showing strength of, 56; testing strength of, 54. Swedish acid bottles, 42. Swedish acid tester, 57. Tank for cleaning test bottles, 40. Test bottle?, 34; apparatus for cleaning, 38; cleaning, 36; double-necked, 76; draining-rack for, 39; marking, 35; for cream testing, 67; for skim milk testing, 76, 77; rack for use in cream- eries and cheese factories, 139; tank for cleaning, 40. Testers, hand, 52; p .wer, 53. Testing cows, number of tests required during a period of lactation, 121. Testing milk and its products, 1; on the farm, 116. Test sample, siie of, 128. Tests of cows, record of, 126. Thermometer scales, comparison of, 226. Thoerner's method, 5. Total solids in milk, 11; determination of, 189. Volatile acids, 201. Waste acid jar, 37. Water, calibration with, 45; determina- tion of, in butter, 197; in cheese, 202; in milk, 183, 190; reservoir for, 61; to be used in the Babcock test, 60. Watering of milk, detection of, 91, Watering and skimming of milk, de- tection of, 91. Weights and measures, comparison o f metric and customary, 228. Whey, Babcock test for, 74, 77; chemical analysis of, 197; composition of, 205. Winton cream bottle, the, 69. Wisconsin creamery butter, summary of analyses, 162. Wisconsin curd test, the. 111. World's Fair breed tests, composition of butter from, 161. Yield of butter, calculation of, 160; and butter fat test, 160; from milk of dif- ferent richness, le**; table showing, from 1 to 10,000 lbs. of milk, testing 3 to 5.35 per cent., 222 Yield of cheese, calculation of, 160, 171; relation between, and quality of milk» 171; table showing, corresponding to 2.5 to 6 per cent, of fat, with lacto- meter readings of 26 to 36, 225. n \K UK i '^%I, ^i^tsTM LIBRARY OF CONGRESS ODDOflTSHblt.