'^J ..j;..'i- :.■-•■ . •:-tia.i;li-'-ili.'!.i!r'i.,:iJ:'k.'.^r:fS:'l.}.Ar ': ' ' MODERN METHODS 3^ TESTING MILK & MILK PRODUCTS '^;'f-T ^ri,*, j'ig, K>4-iM-:''.!»!viCi/ll3i;ri-« m i.~.-ip'r;-i .-,1 - ; ;;;;hr 4t-..i-^ri .- ir.' -••-;: :: ■■■ . ■ .. -• -t.-t.-,!,! '.' ri , .-1 ifr!:; ■:; ; , •,.• :■• ,■:■,• ;■ •^. ■.■ ;r;.-i.';':-.,:- •: ,..:'x r '.»" t4;''':S» i^,-r'/4l''^-W)B«Jtt.-';^iii-iM'nr.n4-r.M-:x\".>.\: 'J .■:-.y.-'l.i':.i.jy,lUi i'l -i - - .- J ^vvh'->H . .ri • if;';1"r'vri ,^.i,:.;.;ji;.i LUCIUS L.VAN SLYKE Class 3£i-^:3 Book _ Copyright}!^. COPYRIGHT DEPOSIT. THE ORIGINAL BABCOCK TESTER Modern Methods qf Testing Milk and Products 5» A HANDBOOK PREPARED FOR THE USE OF DAIRY STUDENTS, BUTTER MAKERS, CHEESE MAKERS, PRO- DUCERS OF MILK, OPERATORS IN CONDENSERIES, MANAGERS OF MILK-SHIPPING STATIONS, MILK- INSPECTORS, PHYSICIANS, ETC. ::: ::: ::: ::: ::: LUCIUS L: VAN SLYKE Chemist of the New York Agricultural Experiment Station ILLUSTRATED REVISED EDITION NEW YORK ORANGE JUDD COMPANY LONDON Kegan Paul, Trench, Trubner & Co., Limited 1907 UBRAHY of COMGRESS Two Oootes Recaivad AUG 20 190^ , Ooovrifht Entry GLASS A XXC, iW. OOP^ -/^z-ji^ Copyright, 1906, 1907, Bt ORANGE JUDD COMPANY All Rights Reserved [entered at stationers' hall, LONDON, ENGLAND] PREFACE To attain the highest degree of success in the pro- duction of milk and in the manufacture of its pro- ducts, it has become essential to acquire some knowl- edge of the methods of testing milk and milk products. The application of these methods to dairying has re- sulted in lifting the dairy industry to a higher plane of intelligence, and in effecting changes of great economic importance, among which may be briefly mentioned: (i) Greater justice rendered milk pro- ducers in paying for milk according to its quality. (2) Prevention of large losses, once very common, in the manufacture of butter and cheese. (3) Improvement of methods of manufacture through better control of details. (4) Increase of yield of products made from a given amount of milk. (5) Improvement in the uni- formity and quality of manufactured dairy products. This little book has been prepared for the use of dairy students, cheese-makers, butter-makers, produc- ers of milk, operators in condenseries, managers of shipping-stations, milk-inspectors, and others inter- ested. Physicians who are specialists in infant-feed- ing will find the book useful in testing human milk as well as cows' milk that is modified or to be modified. No previous chemical training is required for oper- ating successfully the methods described. Any intel- ligent person who can labor with painstaking patience and appreciate the value of attention to little details should be able to master these methods with a rea- VI PREFACE sonable amount of work. The assistance of a trained teacher will, of course, make the task easier. No one, whatever his educational preparation, can hope to use these or any similar methods successfully who can not or will not follow instructions accurately and ex- ercise patience in mastering every minute detail. In the preparation of this work, the writer has tried to keep in mind the following points: (i) Accuracy, simplicity and clearness of statement. (2) Making prominent, as far as practicable, the reasons for each step in ^ach process. (3) Emphasis of common diffi- culties and instructions for overcoming them. (4) Im- pressing students with the necessity of precision and care in performing every detail given. (5) Selection of the methods approved by experience. (6) Avoidance of such technical methods as require unusual skill or equipment. (7) Omission of unnecessary details. (8) Embodiment of the results of the most recent investi- gations. (9) The special needs of those for whose use the work is designed. The scope of this work is far from exhaustive, but the methods selected are given with necessary com- pleteness. Chemical methods, requiring elaborate equipment and extended special training, are purposely omitted. Any one desiring a full description of such methods can obtain it by addressing a request to the U. S. Department of Agriculture, Bureau of Chem- istry, Washington, D. C, asking for a copy of "Meth- ods of Analysis adopted by the Association of Official Agricultural Chemists." The methods that have been compiled here are in large measure the direct result of the work of our PREFACE Vil agricultural experiment stations, and afford some in- dication of the direction and value of the work done by these institutions. In the preparation of Chapter XIII, valuable assist- ance has been kindly rendered by Mr. George A. Smith, Dairy Expert of this station. L. L. Van Slyke. Nezv York Agricultural Experiment Station, 1906. PREFACE TO REVISED EDITION Two editions of this book have been issued within the first year after its pubHcation. Before printing another edition it seems advisable to make some ad- ditions that have been suggested by friendly criticism. A section on the "Sampling and Testing of Gathered Cream" is added to Chapter V. The author has also rewritten Chapter XIV. Gray's rapid test for water in butter, which was included as an appendix in the second edition, is transferred to the body of the book under Chapter VL For the marked favor with which this little book has been received, appreciation can not be expressed in any way more substantial than by making an effort to keep the subject matter up to date. L. L. Van Sl\;ke. Geneva, N. Y., September, ig>oy. CONTENTS I Chemistry of Cow's Milk and Milk Products ... 1 II Methods of Sampling Milk 20 III The Babcock Test — Description of Apparatus and Materials 32 IV Method of Operating the Babcock Test .... 53 V Method of Testing Cream by the Babcock Test . . 69 VI Methods of Testing Skim-milk, Whey, Butter, Cheese, etc. 81 VII Methods of Testing Acidity of Milk and Milk Prod- ucts 96 VIII Methods of Testing the Bacterial Condition of Milk 113 IX Methods of Testing Milk by Rennet-Extract and Pepsin 121 X Methods of Testing Specific Gravity and Solids of Milk by the Lactometer . . . . . . . .127 ix X CONTENTS XI Methods of Testing Milk and Milk Products for Adul- terations . . • • 141 XII The Babcock Test applied to Farm Conditions . . 151 XIII Methods of Commercial Testing and Scoring of But- ter and Cheese io8 XIV Methods of Commercial Testing and Scoring of Milk and Cream 182 XV Arithmetic of Milk and Milk Products 193 Index 215 ILLUSTRATIONS PAGE The Original Babcock Tester Frontispiece Composite-Sample Jars 25 Rack for Composite Samples 26 Sampling-Dipper 27 Scovell Milk-Sampler 27 Equity Milk-Sampler 27 Milk-Testing Bottle 34 Milk-Measuring Pipette 35 Greiner's Automatic Pipette 36 Wagner's Pipette 36 Acid-Measure 36 Acid-Burette and Stand 37 Automatic Burette 37 Steam-Turbine Tester 38 Hand-Tester 38 Small Hand-Tester 39 Electric Centrifuge 40 Hydrometer for Testing Strength of Sulphuric Acid 43 Milk-Bottle Tester 46 Testing Accuracy of Milk-Bottle 46 Burette and Support 48 Waste-Jar for Emptying Test-Bottles 50 Test-Bottle Rinser 50 Test-Bottle Draining Rack 51 Farrington's Bottle-Cleaner — Bottle-Holder Empty . 51 xi xii ILLUSTRATIONS Farrington's Bottle-Cleaner — Bottle-Holder Immersed 52 Farrington's Bottle-Cleaner — Bottle-Holder Draining 52 Correct Way of Holding Pipette and Bottle ... 55 Wrong Way of Holding Pipette and Bottle .... 56 Measuring Fat-Column ^. . 63 Automatic Russian Pipette 67 Russian Test-Bottle . 67 Bulb-Necked Cream-Bottle 70 Straight-Necked Cream-Bottle 71 Cream-Testing Scales 72 Cream-Sampling Sieve 74 Bottles for Testing Skim-Milk 82 Bottle for Testing Butter 86 Wagner's Improved Form of Gray's Water-Test . 91 Mann's Acid Test 101 Farrington's Alkaline-Tablet Test ...... 104 Spillman's Acid-Cylinder 105 Purdue Alkali-Test 107 Hand-Centrifuge for Sedimentation Work . . . . 118 Tube for Sedimentation Work 118 Bausch & Lomb Electric Centrifuge .118 International Instrument Co.'s Electric Centrifuge . 119 Glass for Collecting Sediment in Milk 120 Monrad Rennet-Test . 121 Marschall Rennet-Test 123 Quevenne Lactometer 130 Cylinder for Lactometer . 133 Comparison of Different Specific Gravity Scales . . 134 Richmond's SliJe-Rule ...... -. . . . 138 Butter-Trier . 159 Modern Methods of Testing Milk and Milk Products CHAPTER I Chemistry of Cows' Milk and Milk Products The normal milk of cows contains the following compounds and classes of compounds: (i) Water. (4) Milk-sugar. (2) Fat. (5) Salts or ash. (3) Nitrogen compounds or proteids. (6) Gases. WATER The water present in milk, however much its pres- ence may be disguised, is the compound of hydrogen and oxygen with which we are everywhere familiar. The water in milk serves the purpose of holding in solution the soluble constituents of the milk, and it also acts as a diluent, better fitting the mixture f5r animal nutrition. Variation. — The amount of water normally con- tained in milk varies, depending upon such conditions as individuality, breed, stage of lactation, age, char- acter of food, amount of water drunk, state of health, etc. In the case of single milkings of individual cows, the water may vary from 82 to 90 per cent, or more. In the case of milk from herds of cows, the water varies less, usually ranging from 86 to 88 per cent. 2 MODERN METHODS OF TESTING MILK The influence of breed. — The following figures, ' from the records of the N. Y. Agricultural Experi- ment Station at Geneva, illustrate the influence of breed upon the water content of milk : Per cent, of NAME OF BREED Water ill milk Holstein Friesian 88.20 American Holderness 87.35 Ayrshire 87.25 Short Horn 85.70 Devon 85.50 Guernsey 85.ID Jersey 84.60 The influence of lactation. — The variation of water in milk, as affected by advance of the lactation period, is illustrated by the following figures, which cover a period of ten months from the time of calving: Per cent, of MONTH OF LACTATION Water in milk I 86.00 2 86.50 3 86.53 4 86.36 - 5 86.25 6 86.00 7 85.82 8 85.67 9 85.54 10 85.17 There is noticeable a general tendency for the amount of water in milk to increase for the first three months of lactation, after which there is a con- tinuous decrease to the end of the lactation period. CHEMISTRY OF COWS MILK 3 Total solids. — Under the general term of tbfal solids or milk-solids, we indicate the constituents of the milk other than water (and gases). The per cent, of water in milk subtracted from lOO gives the percentage of milk-solids, which include fat, proteids, milk-sugar and salts or ash. The amount of solids in milk varies with the same conditions that affect the percentage of water in milk, but, of course, in just the reverse manner. Most states orescribe a legal standard for milk-solids, usually 12 per cent., and milk containing less than the legal amount is regarded as adulterated. MILK-FAT The composition of milk-fat. — Milk-fat, also called hutter-fat, is not a single chemical compound, but is a somewhat variable mixture of several different com- pounds called glycerides. Each glyceride is formed by the chemical union of glycerin as a base with some acid or acids of a particular kind. These glycerin-acid compounds, or glycerides, of milk-fat contain about ten different acids, some being present in small propor- tions. The four following acids enter most largely into the composition of milk-fat, in the form of their combinations with glycerin: Palmitic acid, oleic acid, myristic acid and butyric acid. The compounds, or glycerides, formed by the combination of glycerin and the acids, have special names derived from the acids ; thus, we have palmitin (glycerin combined with palm- itic acid), butyrin (glycerin combined with butyric acid), olein, etc. Milk- fat contains, on an average, about 40 per cent, of palmitin, 34 per cent, of olein, 10 per cent, of myristin, 6 per cent, of butyrin, and 4 MODERN METHODS OF TESTING MILK from less than i to nearly 3 per cent, of each of the glycerides of other acids. Milk-fat contains about 12.5 per C2nt. of glycerin in combination with the acids. The proportions of these constituents of milk- fat vary somewhat, and this variation influences the character of the milk-fat. Thus, palmitin and myris- tin tend to make milk-fat harder, while olein and buty- rin have the opposite tendency. The acids contained in milk-fat or butter-fat may be divided into two groups : ( i ) The acids in one group (palmitic, oleic, myristic, stearic, lauric) are insoluble in water and non-volatile, while (2) the other acids (butyric, caproic, etc.,) are more or less completely soluble in water and are volatile. These differences afford a practical basis for distinguishing pure butter from artificial butter. Of the fat-acids contained in butter- fat, about 87.5 per cent, consists of the insoluble fat-acids, while in other forms of animal fat (beef- fat, lard, etc.,) the amount of these insoluble fat-acids is considerably greater. The amount of volatile fat-acids in milk-fat or butter-fat is much greater than in other forms of animal fat. Fat-globules in milk. — Milk-fat is present in milk, not in solution, but suspended in the form of very small, transparent globules. Globules varying in size between one twenty-five hundredth and one fifteen- thousandth of an inch in diameter are the ones most commonly present. The average size of fat-globules in milk is somewhat more than one ten-thousandth of an inch in diameter. The smaller globules are more numerous than the larger ones. In one drop of aver- age milk there are more than one hundred million fat- CHEMISTRY OF COWS MILK 5 globules. Skim-milk contains fewer and smaller glo- bules than whole milk, while the reverse is true of cream. The large globules do not differ in composi- tion from the small ones. The size and number of fat-globules in milk are influenced by such conditions as advance of lactation, breed of cow, food, age, health, different milkings, different parts of the same milking, etc. It was formerly believed generally, and is still by some, that the fat-globules of milk are 'surrounded by a membranous covering, or else by a semi-liquid, al- buminous layer. We may, however, accept it as es- tablished beyond reasonable doubt that the fat-globules of milk have no special covering of any kind, but are simply minute particles of fat floating free in milk in the form of an emulsion. Fat-globules quite generally retain their individuality even in butter and cheese. Amount of fat in milk. — Normal milk varies greatly in its fat content, containing from below 2 to over 10 per cent., if we consider single milkings of individual cows. The milk from herds of cows varies in fat more commonly between the limits of 3 and 5 per cent. The average amount of milk- fat in milk pro- duced in this country, taking the true average for the entire year, lies somewhere near 4 per cent., perhaps a little under. Many of the conditions that affect the percentage of fat in milk are fairly well known, while others are little understood. We will briefly consider some of tjie well-recognized conditions that influence the fat content of milk. (i) Influence of individuality of cozv on fat con- tent of milk. — It is uncommon to find in a herd of MODERN METHODS OF TESTING MILK COWS two individuals whose milk contains the same per cent, of fat, whether we consider single milkings or the average of many milkings. (2) Influence of breed of cozv on fat content of milk. — It is well known that the per cent, of fat in milk varies in a somewhat characteristic way with the kind of breed of cow. While there is marked variation in individuals of the same breed, there is found to be a fairly uniform difference, more or less marked, if we consider the averages of several indi- viduals. It is largely owing to this influence that we find the milk of one country differing from that of another, or the milk of one section of a country differing from that of another section. For example, the average amount of fat in milk in Germany and Holland is fully one-half per cent, lower than in this country, because the prevailing breeds of cows there are those producing milk comparatively low in fat. The following figures, taken from the records of the New York (Geneva) Agricultural Experiment Station, represent averages of many individuals for several periods of lactation: NAME OF BREED Holstein Friesiaii . Ayrshire .... American Holderness Short Horn . Devon Guernsey .... Jersey (3) Infliience of age of cozv on fat content of milk. — So far as published data throw light upon this Pel • cent, of fat in milk Average Lowest Highest . 3-36 2.88 3.85 . 3.60 3-20 4.24 . 373 3-49 392 . 4-44 4.28 4-56 . 4.60 430 5-23 • 5-30- 4.51 6.13 . 560 4.96' 6.09 CHEMISTRY OF COWS' MILK 7 point, there appears to be a tendency for milk to be- come less rich in fat with each succeeding period of lactation, especially after the second, though individ- ual exceptions are not infrequent. More data are needed to settle the question definitely. (4) Influence of advance of lactation on the fat content of milk. — In general, it is found that the per cent, of fat in milk increases as the stage of lactation advances after the third month, as illustrated by the following data from the records of the New York (Geneva) Station, covering lo months from the time of calving: NUMBER OF Per cent, of MONTH OF LACTATION fat itt milk I 4-54 2 4-33 3 4-28 4 . 4-39 5 4.38 6 4-53 7 4.56 8 4.66 9 4.79 10 5.00 (5) Variation of time between milkings in rela- tion to the fat content of milk. — As a rule, the longer the time between two successive milkings, the smaller is the per cent, of fat in the milk; and the shorter the time between milkings, the greater the per cent, of fat. When the time between milkings is uniformly equal, the variation of fat in milk is small, provided the gen- eral environment of the animal is the same. How- ever, as there are not commonly such entirely uniform 8 MODERN METHODS OF TESTING MILK conditions of surroundings during the day and night, there appears to be a common tendency for the pres- ence of a httle more fat in the morning's milk, even when milkings are apart the same length of time. (6) Variation of fat content in different portions of milk drawn from the udder. — The following figures, taken from the writer's records, illustrate the general rule that the first milk drawn contains least fat, the milk last drawn (strippings) being the richest in fat: Per cent, of fat in milk cow I cow 2 cow 3 First portion drawn .... 0.90 1.60 1.60 Second portion drawn . . . 2.60 3.20 3.25 Third portion drawn .... 5.35 4.10 5.00 Fourth portion drawn (strip'gs) 9.80 8.10 8.30 It is also known that the per cent, of fat in milk varies in different quarters of the udder of a cow, and also varies more or less in each quarter with the order in which the teats are milked. THE NITROGEN COMPOUNDS OF MILK Some confusion prevails in respect to the names of the nitrogen compounds of milk. They have been spoken of as albuminoids, proteids, etc. Frequently the word c'asein is erroneously used to include all the nitrogen compounds of milk. How many nitrogen or proteid compounds are pres- ent in normal milk? What are they? Different work- ers have reported from one to seven or more. The chemical evidence at hand justifies us in the belief that fresh, normal milk contains not more than three or, perhaps, four nitrogen-containing or proteid bodies. CHEMISTRY OF COWS MILK , 9 viz., casein, albumin, globulin and galactase. Globulin and galactase are present in so small quantities that we can properly regard casein and albumin, quantita- tively, as being essentially the nitrogen compounds of milk. Milk-Casein is the most important nitrogen com- pound in milk, because, (ist) it is the one present in largest quantity; (2d) its presence makes it possible to convert milk into cheese; and (-3d) it has a high value as food. Milk-casein is most familiar to us in the form of the solid, white substance called curd, which forms in milk when it sours (though, strictly speaking, this well-known, white substance is not milk- casein, but casein lactate). (i) Composition of milk-casein. — Casein is a very complex chemical compound, containing the elements carbon, oxygen, hydrogen, nitrogen, sulphur, and phos- phorus. In milk the proteid molecule of casein is com- bined with calcium, or some calcium compound, and hence the proper chemical name of milk-casein is calcium casein. It exists in milk, not in solution, but in the form of extremely minute, solid, gelatinous par- ticles in suspension. The slime found in the bowl of centrifugal separators consists, to a considerable ex- tent, of milk-casein. (2) Action of acids upon milk-casein. — When milk sours in the ordinary way, the lactic acid formed acts upon the calcium casein, two definite changes taking place. First, the lactic acid combines with the cal- cium of the calcium casein, forming calcium-free ca- sein, or simply casein set free from its combination with calcium. When more lactic acid forms, the sec- lO .MODERN METHODS OF TESTING MILK end change takes place, the free casein taking up the acid without definite combination and forming a sub- stance which is famihar as the curd of sour milk. Sim- ilar changes occur when milk is treated with other acids, such as hydrochloric, acetic, sulphuric, etc. Free casein is insoluble in water and also in very dilute acids at ordinary temperatures. The action of acids on cal- cium casein and on free casein is hastened by increase of temperature. Casein dissolves easily in an excess of acid, forming soluble casein salts. (3) Action of alkalis on milk-casein. — Dilute solu- tions of alkalis (caustic soda, ammonia, etc.) act upon casein and its salts with acids, forming compounds that dissolve easily in water. These alkali compounds of casein are not affected by rennet. Some of these compounds are found in commerce as food and me- dicinal preparations under such names as Plasmon, Nutrose, Santogene, Eucasein, Galactogene, etc. (4) Action of heat on milk-casein. — Heat alone un- der ordinary conditions, even at the boiling point of water, does not coagulate the casein in milk. Casein may be coagulated by heating under pressure at a tem- perature of about 270° F. The browning of milk heated under pressure is more or less due to changes in the casein. The formation of a peculiar skin on the surface of milk heated above 140° F. is largely due to the calcium casein of the milk and not to albu- min as was formerly supposed. The skin itself con- tains practically all of the constituents of the milk and may be regarded as a kind of evaporated milk. (5) Action of rennet on milk-casein. — One of the most characteristic properties of the calcium casein CHEMISTRY OF COWS MILK 11 of milk is its coagulation by the enzym or chemical ferment contained in rennet, which is an extract of the mucous membrane of a calf's stomach. This property makes possible the manufacture of cheese from milk. The curd formed by the action of rennet is called para- casein or, more properly, calcium paracasein. There appears to be little or no chemical difference between calcium casein and calcium paracasein. The coagula- tion of calcium casein produced by rennet is quite dif- ferent from that produced by acids. Calcium paraca- sein behaves towards acids and alkalis much like cal- cium casein. (6) Other changes caused in milk-casein. — Under the action of chemical reagents, of enzyms and of va- rious organisms, calcium casein and paracasein may be changed into a large number of other substances. Among the compounds and classes of compounds thus formed are paranuclein, albumoses, peptones, amides (crystallizable bodies) and ammonia. These products are never found in normal milk as it leaves the cow, but may be present in milk that has stood some time. Milk- Albumin. — AI ilk-albumin differs from milk- casein in composition and behavior. Thus, milk-albu- min (i) is not acted upon by rennet; (2) is not coag- ulated by acids at ordinary temperatures; (3) is co- agulated by heat alone, though not completely, above 160° F. ; and (4) is in solution in milk. Milk-Globulin. — This compound is present only in small quantities in normal milk and is of no special importance, so far as known. Galactase. — This substance is an unorganized fer- ment, or a mixture of such ferments, present in normal milk. It somewhat resembles pepsin in its action, be- 12 MODERN METHODS OF TESTING MILK ing able to coagulate milk-casein and then digest it or make it soluble. It is present in very small amounts in milk and its action is very slow. It has never been isolated from milk in pure form. It is probably a nitro- gen-containing substance. Our knowledge of galac- tase is very far from complete. Amounts of casein and albumin in milk. — In single milkings of individual cows, the casein and albumin, taken together, vary from 2.5 to 6 per cent, and average about 3.2 per cent. Milk-casein varies in amount from 2 to 4 per cent, and averages about 2.5 per cent. Albumin varies from 0.5 to 0.9 per cent, and averages about 0.7 per cent. The amount of ca- sein in relation to albumin varies greatly. On an average, milk contains about 3.6 parts of casein for one of albumin, or, stated another way, casein consti- tutes about 80 per cent, of the nitrogen compounds of milk. The amount of casein and albumin in milk is influ- enced by many conditions, such as influence the gen- eral composition of the milk, among wdiich are individ- uality, breed, advance of lactation, etc. As the lacta- tion period advances, there is a general tendency on the part of casein and albumin in milk to increase. Relation of fat and nitrogen compounds in milk. — In normal milk containing over 3 per cent, of fat, the amount of casein and albumin is rarely greater than the amount of fat, especially in the milk of herds of cows. When the per cent, of fat is less than that of the nitrogen compounds, the milk may generally be regarded as skimmed, especially in the case of milk from herds. CHEMISTRY OF COWS MILK I3 MILK-SUGAR Milk-sugar, also called lactose, is present in cows' milk in solution. In general composition, it resembles ordinary sugar, but it is less sweet and less soluble in water. The amount of sugar in milk varies from below 4 to over 6 per cent, and averages about 5 per cent. Its importance in dairy work, especially in con- nection with the manufacture of butter and cheese, comes from the ease with which it is converted into lactic acid by certain forms of bacteria. In the ordi- nary souring of milk, the amount of milk-sugar de- creases somewhat more than one-fourth and there is formed as a maximum about 0.9 per cent, of lactic acid. More acid may be formed after some time. Hence, sour milk, when two or three days old, con- tains only 3.5 to 4 per cent, of milk-sugar. The sugar of milk passes largely into the whey in cheese-making and forms over 70 per cent, of the solids in whey. The milk-sugar of commerce is usually prepared by evaporating whey and purifying the impure product first obtained. THE SALTS OF MILK The salts of milk, commonly included under the term "ash," are present in only small amounts, 0.7 per cent, on the average ; but they have important relations to milk and its products. Our knowledge of these compounds is very incomplete. The salts of milk are commonly spoken of as the ash or mineral constitu- ents. This conception is somewhat misleading, be- cause the materials appearing in the ash of milk are, to some considerable extent, combined in organic com- pounds, instead of existing in the milk as separate 14 MODERN METHODS OF TESTING MILK inorganic bodies. The ash represents in amount, therefore, more than the so-called mineral constitu- ents of milk and less than the salts of milk. While the ash in milk amounts to about 0.7 per cent., the amount of salts probably approximates 0.9 per cent. A portion of the salts of milk is in solution, including such compounds as calcium citrate, sodium chloride, potassium acid phosphate, etc., while a portion (tri- calcium phosphate) appears to be in suspension in the form of very finely divided particles. THE GASES OF MILK Milk contains more or less oxygen and nitrogen, these gases being carried into it mechanically from the air in the process of milking. It contains also, when freshly drawn, carbon dioxide, already present in the udder milk, there being probably between 3 and 4 per cent, by volume, a portion of which escapes at once while being drawn from the udder under usual conditions. GENERAL SUMMARY Milk contains water, fat, casein, albumin, sugar, salts, carbon dioxide and some other constituents in small quantities. The fat and casein and some of the salts are in suspension and not in solution, while al- bumin, sugar and the larger portion of the salts are held in solution by the water. As a matter of convenience, the compovmds of milk are divided into certain arbitrary groups. By one system of division, the compounds of milk are ar- ranged in two classes: — (i) Water, and (2) milk- solids (or total solids), this second class including CHEMISTRY OF COWS MILK 15 fat, casein, albumin, sugar, salts (ash), etc. Another division is made on the basis of the milk- fat into (i) fat and (2) milk-serum, which includes all the milk constituents except the fat. Separator skim-milk is nearly pure milk-serum. Then we have the milk- solids subdivided into (i) fat and (2) solids-not-fat (casein, albumin, sugar, salts (ash), etc.) The following arrangement shows the general rela- tion of the compounds contained in milk, the figures indicating the percentages present in average milk: Water 87.1 fFat 3.9 'Nitrogen compounds < 3.2 MILK. \ 12.9 Solids ■{ Solids-not-fat 1 Casein, 2.5 Albumin, etc. 0.7 9.0 Milk-sugar 5.1 Ash (salts) 0.7 {Carbon dioxide Nitrogen Oxygen AVERAGE ANALYSIS OF COWS MILK WATER TOTAL SOLIDS FAT CASEIN ALBU- MIN SUGAR ASH Average of 5,552 American an- alyses c o m- piled by the author Per ct. 87.1 87.4 Per ct. 12.9 12.6 Per ct. 3.9 3.75 Per ct. 2.5 2.45 Per ct. 0.7 0.7 Per ct. 5,1 5.0 Per ct. 07 Average cheese- factory milk for the season (May to Nov.) inN. Y. State... 0.7 i6 MODERN METHODS OF TESTING MILK REPRESENTATIVE ANALYSES OF PRODUCTS AND BY-PRODUCTS OF MILK •TOTAL ALBU- WATER SOLIDS FAT CASEIN MIN SUGAR ASH Per ct. Per Ct. Per ct. Per ct. Per ct. Per ct. Per ct. Butter 13.0 87.0 83.5 1.0 .... 2.51 Cheddar Cheese (green) 36.8 63.2 33.75 23.75 2 .... 5.73 Skim-milk (separator) 90.3 9.7 0.10 2.75 O.SO 5.25 0.80 Whey 93.4 6.6 0.35 0.10 0.75 4.80 0.60 Buttermilk 90.6 9.4 0.10 2.80 0.80 4.40* 0.70 1 Salt. 2 Paracasein. ^ Salt and Ash. * .60 per cent, lactic acid in addition. DEFINITIONS AND STANDARDS OF MILK AND MILK PRODUCTS The United States Department of Agriculture has estabhshed official standards for purity of dairy and other food products, defining also what is meant by the terms used in designating different materials. These definitions and standards have been most carefully worked out by members of the Association of Official Agricultural Chemists, several years having been de- voted to the collection of data. The official definitions and standards relating to milk and milk products are as follows: MILK Definitions. — i. Milk (zvholc milk) is the lacteal secretion obtained by the complete milking of one or more healthy cows, properly fed and kept, exclud- ing that obtained within fifteen days before and five days after calving. 2. Blended milk is milk modified in its composition CHEMISTRY OF COWS' MILK 1/ SO as to have a definite and stated percentage of one or more of its constituents. 3. Skim-milk is milk from which a part or all of the cream has been removed. 4. Buttermilk is the product that remains when but- ter is removed from milk or cream in the process of churning. 5. Pasteurized milk is milk that has been heated below boiling but sufficiently to kill most of the act- ive organisms present and immediately cooled to fifty degrees (50°) Fahr. or lower to retard the develop- ment of their spores. 6. Sterilised milk is milk that has been heated at the temperature of boiling water or higher for a length of time sufficient to kill all organisms present. 7. Condensed milk is milk from which a considera- ble portion of water has been evaporated. 8. Szveetened condensed milk is milk from which a considerable portion of water has been evaporated and to which sugar (sucrose) has been added. 9. Condensed skim-milk is skim-milk from which a considerable portion of water has been evaporated. Standards. — i. Standard milk contains not less than twelve (12) per cent, of total solids, not less than eight and one-half (8.5) per cent, of solids-not-fat, and not less than three and one-quarter (3.25) per cent, of milk- fat. 2. Standard skim-milk contains not less than nine and one-quarter (9.25) per cent, of milk-solids. 3. Standard condensed milk contains not less than twenty-eight (28) per cent, of milk-solids, of which not less than one-fourth is milk-fat. l8 MODERN METHODS OF TESTING MILK 4. Standard szveetcncd condensed milk contains not less than twenty-eight (28) per cent, of milk-soUds of which not less than one-fourth is milk-fat. MILK-FAT OR BUTTER-FAT Definition. — Milk-fat or butter- fat is the fat of milk. Standard. — Standard milk-fat or butter-fat has a Reichert-Meissl number not less than twenty- four (24) and a specific gravity not less than 0.905 (^^^) CREAM Definitions. — i. Cream is that portion of milk, rich in butter-fat, wdiich rises to the surface of milk on standing, or is separated from it by centrifugal force. 2. Evaporated cream is cream from which a consid- erable portion of water has been evaporated. Standard. — Standard cream contains not less than eighteen (18) per cent, of milk- fat. BUTTER Definitions. — i. Butter is the product made by gathering in any manner the fat of fresh or ripened milk or cream into a mass, which also contains a small portion of the other milk constituents, with or with- out salt. 2. Renovated or process butter is the product made by melting butter and reworking, without the addition or use of chemicals or any substances except milk, cream or salt. Standards. — i. Standard butter contains noi less than eighty-two and five-tenths (82.5) per cent, of CHEMISTRY OF COWS MILK IQ butter-fat. By acts of Congress approved August 2, 1886, and May 9, 1902, butter may also contain additional coloring matter. 2. Standard renovated or process butter contains not more than sixteen (16) per cent, of water and at least eighty-two and five-tenths (82.5) per cent, of but- ter-fat. CHEESE Definitions. — i. Cheese is the solid and ripened pro- duct made by coagulating the casein of milk by means of rennet or acids, with or without the addition of ripening ferments and seasoning. By act of Congress, approved June 6, 1896, cheese may also contain addi- tional coloring matter. 2. Whole-milk or full-cream cheese is cheese made from milk from which no portion of the fat has been removed. 3. Skim-milk cheese is cheese made from milk from which any portion of the fat has been removed. 4. Cream-cheese is cheese made from milk and cream, or milk containing not less than six (6) per cent, of fat. Standard. — Standard zvhole-milk or full-cream cheese contains, in the water- free substance, not less than fifty (50) per cent, butter-fat. MISCELLANEOUS MILK PRODUCTS Definitions. — 1. Whey is the product remaining af- ter the removal of fat and casein from milk in the pro- cess of cheese-making. 2. Kumiss is the product made by the alcoholic fer- mentation of mares' or cows' milk, with or without the addition of sugar (sucrose). CHAPTER II Methods of Sampling Milk Too much emphasis can not be placed upon the im- portance of taking for analysis a sample of milk that truly represents the whole body of milk from which the sample is taken. This statement applies equally to any product or by-product of milk that is to be tested. Before a sample for testing is taken, the body of milk from which the sample is to be drawn should be uniform throughout in composition. Several con- ditions may disturb the desired uniformity of compo- sition of a mass of milk, among which are the fol- lowing : (i) Separation of fat. (2) Partial churning of fat. (3) Freezing of milk. (4) Souring of milk. SAMPLING MILK WHEN FAT HAS SEPARATED The rapidity with which fat-globules rise to the surface of milk in the form of cream is well known. Therefore, milk standing at rest soon loses its uni- formity of composition, the upper layers containing more fat than the lower ones. On this account it is always necessary, just before taking a sample of milk for testing, to make sure that the body of milk to be tested has an even composition throughout. Milk in which fat separation is slight. — In milk 20 METHODS OF SAMPLING MILK 21 in which there is no visible separation of cream, even- ness of mixing may be best effected b\' pouring the milk from one vessel to another several times immedi- ately before each sample is drawn for testing. Stir- ring milk, as with a dipper, is less effective than pouring. Milk in which fat separation is marked. — In milk in which the cream has separated in a visible layer, the pouring needs to be done a greater number of times than in cases where the separation of cream is not noticeable ; and, in order to prevent possible churn- ing of particles of cream, the agitation should be as gentle as may be consistent with thorough mixing. Milk containing dried cream. — In cases where the cream is somewhat dried or hardened, the milk should be warmed to 105° or 110° F. for 5 or 10 minutes to allow the cream to melt. The milk is then vigor- ously agitated and immediately sampled. SAMPLING MILK WHEN FAT IS PARTIALLY CHURNED Milk- fat may separate from milk in the form of small butter-granules, as (i) when the mixing or shaking of the sample to be tested is done too vio- lently; (2) when milk in cans is excessively agitated in transportation; and (3) when bottles, partly full of milk, are sent by mail or express. In such partially churned milk it is difficult to get a representative sample, and the results of testing are, at best, only approximate, unless special measures are resorted to in sampling. Distributing fat by warming. — In the case of par- 22 MODERN METHODS OF TESTING MILK tially churned milk, the fat may be redistributed in the milk by warming it to 105° or 110° F. long enough to melt the butter-granules, after which the sample is vigorously shaken, until the fat is evenly distributed through the milk, and then the sample is drawn at once for analysis. Dissolving fat in ether. — Another method of treat- ing partially churned milk, previous to sampling, is to shake the milk with 5 per cent, of its volume of ether until the fat-granules are redissolved and then, after further vigorous shaking, to take the sample at once. In this case it is necessary to make a correc- tion by adding to the results 5 per cent, or one-twenti- eth of the result obtained. For example, a milk, treated with 5 per cent, of ether, and giving, on test- ing, 3 per cent, of fat, should have added .15 (5 per cent, of 3), making the corrected result 3.15 per cent. When ether is used, extra care must be observed in mixing the acid and milk, (see p. 58) as the heat de- veloped may cause the ether to boil up out of the neck of the test-bottle. Measures for preventing the churning of fat in milk. — It is better to prevent the churning of fat in milk than to be put to the extra trouble required to get a good sample from milk that has in it fat-granules. Cans of milk, when necessarily exposed to much mo- tion in transportation, should be made as nearly full as possible. In the case of bottles of milk sent by mail or express for analysis, the churning of fat may be prevented in the following manner: Fill the bot- tle full of milk to overflowing. Then push in tightly a stopper of cork or rubber in which has been made METHODS OF SAMPLING MILK 23 from top to bottom a hole one-eighth inch in diameter or less. Finally, push a close-fitting plug of wood or a glass rod into the hole in the stopper. SAMPLING FROZEN MILK Frozen milk is of very uneven composition in dif- ferent portions of its mass. The crystals of ice con- tained in it consist largely of water, while the liquid portion contains most of the milk-solids. In such cases it is necessary to melt the frozen portion by warming, and then to mix well by gentle pouring from one vessel to another, after which the sample is at once taken for testing. SAMPLING MILK COAGULATED BY SOURING A sample of thickened, sour milk can not, without special treatment, be taken so as to give reliable results in fat determination. In ordinary curdled milk the percentage of fat remains unchanged in amount, but it is not evenly distributed through the milk. In or- der to overcome this difficulty, the curdled casein lactate must be dissolved before sampling. This is done by adding to the milk a strong solution of caus- tic soda or potash (lye), or strong ammonia water, to the extent of 5 or 10 per cent, of the volume of the milk used for sampling. The alkali is shaken with the milk until the mixture becomes completely liquid, after which the sample is at once drawn for testing. It is necessary to make a correction by adding to the results 5 or 10 per cent, of the amount of fat found, according to the amount of alkali solution used. In place of using a solution of alkali, one can add, in 24 MODERN METHODS OF TESTING MILK small portions at a time, finely powdered caustic soda or potash, allowing the milk to stand some time after each addition of powdered alkali and shaking vigor- ously, the additions of alkali and the agitation being continued only until the milk becomes completely liquid. In using the alkali in solid form, no correction of re- sults needs to be made. The alkali solution or tab- lets described on page 96 may be used. A darkening of the milk by alkali may occur without affecting the results of the test. In testing, caution must be ob- served when adding sulphuric acid (see p. 56) to milk in which an alkali has been used, since an unusual de- gree of heat is produced and the contents of the test- bottle may spurt out. The acid must be added slowly and mixed with the milk much more deliberately than usual. COMPOSITE SAMPLING OF MILK Composite samples of milk. — A mixture of daily samples of milk, taken from day to day for several days in succession, is known as a composite sample. In commercial work at creameries, cheese-factories, milk-shipping stations, etc., where the number of pa- trons is large, a daily test of the milk for its fat con- tent is impracticable. To obviate the great amount of work involved in making daily tests, a jar is provided for the milk of each patron and in this jar is placed a sample of each day's milk, when it is delivered, these daily samples being mixed and allowed to accumulate for a period of one or two weeks. A determination of fat in such a composite sample gives the average percentage of fat in the milk for the period covered METHODS OF SAMPLING MILK 25 by the mixture of daily samples. This method has been proved to be as accurate as that of testing each sample daily by itself, but there are several precau- tions to be observed carefully in applying this method in commercial practice. The conditions that are necessary for success in using the method of composite sampling may be con- sidered under the following heads: (i) Systematic FIG. I COMPOSITE-SAMPLE JAR FIG. 2 COMPOSITE-SAMPLE JAR preparation, (2) methods of taking daily samples, (3) use of preservatives, (4) care of composite samples, (5) ^S^ of composite samples, and (6) preparation of composite samples for sampling and testing. Systematic preparation for taking composite sam- ples. — A round glass jar or bottle, holding a pint or quart, should be provided for each patron. The forms given in Figs, i and 2 are suitable, or ordinary Mason 26 MODERN METHODS OF TESTING MILK fruit- jars may be used. Whatever form of composite- sample jar or bottle be used, the stopper or cover should fit perfectly tight, so as to prevent any possi- ble evaporation of water from the sample of milk, and care should be taken to keep the covers or stoppers tight. Each sam- ple-bottle or jar should be labeled with a name or number easily iden- tifying the patron furnishing the milk. The jars should be arranged in definite order on a rack (Fig- 3). placed conveniently near the point where the milk is deliv- ered. As explained later, some preservative is used in each jar. Taking daily samples for composite samples. — Each day when milk is delivered, the sample should be taken immediately after the milk has been poured into the weighing can before weighing, and should then be placed at once in the composite jar or bottle prepared for it. Two methods of sampling are in common use, (i) by means of a small dipper, and (2) by means of a sampling-tube. ( I ) Taking sample zvifh dipper. — A half-ounce dip- per (Fig. 4) is used for taking the sample from the weigh-can, as soon as the milk is poured in. The FIG. 3 — RACK FOR COMPOSITE SAMPLES METHODS OF SAMPLING MILK 27 sample is at once placed in its proper jar or bottle. Providing the milk is thoroughly mixed in the weigh- can and the quantity of milk delivered by a patron FIG. 4 SAMPLING-DIPPER FIG. 5 SCOViiLL SAMPLER FIG. 6 EQUITY SAMPLER from day to day does not vary much, this method of sampling gives correct results. (2) Taking sample with sampling-tube. — There are different types of sampling-tubes (Figs. 5 and 6), of which the Scovell sampler is one of the best. In this 28 MODERN METHODS OF TESTING MILK instrument the main tube is open at both ends, the lower end closely fitting into a cap furnished with three elliptical openings. When the sampler, open at the bottom, is let down into a can of milk, the liquid pours into the openings and fills the tube to the height of the milk in the can. When the cap comes in con- tact with the bottom of the can, the tube slides down and closes the openings, after which the tube can be withdrawn and its contents emptied into the com- posite jar. The tube method of sampling possesses two marked advantages over the dipper method : ( i ) It always takes an aliquot portion, or uniform proportion, of the milk, representing a small column of the milk from top to bottom; and (2) it provides a strictly rep- resentative sample of the milk, even when sampling is delayed, because it takes a uniform amount from each layer of milk, going from top to bottom. THE USE OF PRESERVATIVES IN COMPOSITE SAMPLES The successful use of composite samples is made possible only by the presence of some substance which will keep the milk from curdling. Three preserva- tives have been found especially useful for this pur- pose: (i) Corrosive sublimate, (2) formalin, and (3) bichromate of potash. Corrosive sublimate, known chemically as mercuric chloride, has the advantage of being a more powerful antiseptic than the other substances, much smaller quantities being effective in keeping milk longer, but it has the disadvantage of being a violent poison. METHODS OF SAMPLING MILK 29 When this is used as a milk preservative, it is a wise precaution to add a Httle coloring matter to the milk in order to warn every one of its abnormal character. Corrosive sublimate, mixed with coloring matter, is put up in convenient tablet form and has found exten- sive use in preserving composite samples. All things considered, it is probably the most satisfactory of the preservatives commonly employed. Formalin is a liquid containing about 40 per cent, of the chemical compound known as formaldehyde. It is an effective antiseptic and has the advantage of being in liquid form. One cubic centimeter of forma- lin should keep a pint or quart sample of milk two weeks or more. Formalin possesses the disadvantage of so hardening the milk-casein that it is not as readily dissolved by sulphuric acid (see p. 58) as is the casein of untreated milk. An excessive use of corrosive sub- limate may produce a similar hardening of casein. Bichromate of potash, also called potassium bichro- mate, is extensively used in preserving samples of milk for testing. It is best to use it in powdered form. It has the following advantages : ( i ) It is compara- tively inexpensive. (2) It colors milk yellow and thus shows its presence. (3) It is not a very violent poison, though not entirely harmless. (4) It is efficient in keeping milk for one or two weeks. However, it has some disadvantages as a preservative of composite samples of milk: (i) If too much bichromate is used, the solution of the casein in sulphuric acid is some- what difficult and the final results of testing may not be clear. (2) In hot weather, it is often difficult to keep samples without using an excessive amount of 30 MODERN METHODS OF TESTING MILK bichromate. (3) Lactic acid in milk considerably re- duces the efficiency of bichromate in preserving milk. (4) Samples of milk preserved with bichromate are apt, when exposed to light, to form a tough skin on the surface, which interferes with proper sampling. The amount of potassium bichromate to be used in composite samples is about 8 or 10 grains for half a pint to a pint of milk. The bichromate is put up for sale in tablets of convenient size, ready for use in pre- serving milk samples. Bichromate can be satisfactor- ily used even in hot weather, if the samples are kept in a dark, cool place most of the time. CARE OF COMPOSITE SAMPLES In caring for composite samples of milk or cream, some special precautions must be observed, (i) Com- posite sample jars must be kept covered tight to pre- vent evaporation of water, which would result in giv- ing a test for fat higher than the correct amount. (2) They should be kept in a cool place, so that the small- est possible amount of preservative will need to be used. (3) They should be kept in the dark most of the time, since direct sunlight may cause the formation of a tough cream, rendering difficult the taking of a good sample for testing. (4) When the daily sample of milk is added to the composite sample, the contents of the jar should be mixed by giving the jar a gentle, rotary motion. Unless this is done regularly each day, the cream that rises be- comes tough, especially where it is in contact with the sides of the jar, and this condition makes it difficult to get a proper sample for testing. This daily mixing METHODS OF SAMPLING MILK 3 1 also insures the complete solution and distribution of the preservative through the milk, which is an es- sential condition of success in keeping samples. (5) If a composite sample shows any dried or churned cream, the sample should be warmed to 105° or 110° F. for some minutes and then agitated vigorously be- fore drawing the sample for testing. AGE OF COMPOSITE SAMPLES WHEN TESTED It is advisable to make the fat-test in composite samples, when they have been accumulating for a week or ten days. In any case the limit should be placed at two weeks. The custom practiced by some of testing composite samples only once a month should be severely condemned. When samples are kept longer than two weeks, it is more difficult to get a perfectly reliable test for fat. PREPARATION OF COMPOSITE SAMPLES FOR SAMPLING AND TESTING When a composite sample is to be tested, it is treated like any other sample previous to taking the sample for testing, as has already been described in the first part of this chapter on pp. 20-24. CHAPTER III The Babcock Test — Description of Apparatus and Material The Babcock test is a method for ascertaining the amount of fat in milk and milk products. It was de- vised by S. M. Babcock, Ph.D., chief chemist of the Wisconsin Agricultural Experiment Station, and was first made public in 1890. There are in use, espe- cially in Europe, other tests, which are more or less imitations or modifications of the Babcock test, such as the Gerber test or acid-butyrometer and DeLaval's butyrometer. The Babcock test solved the problem of a rapid, ac- curate, inexpensive and simple method of testing milk and milk products for fat, and it has found extensive application in many lines of dairying, as may be shown by mention of the following important results coming from its use : ( i ) The payment for milk according to its fat content has been made practicable. (2) Mak- ers of butter and cheese have been able to detect and prevent abnormal losses of fat in the process of man- ufacture. (3) It has enabled milk producers to detect unprofitable cows, thus furnishing an intelligent guide in improving their herds. (4) It has done more than any other means to stop the watering and skimming of milk in connection with creameries and cheese-factor- ies. (5) It has been of great service in scientific dairy investigations and has, in general, been a source of educational inspiration. 32 THE BABCOCK TEST 33 PRINCIPLES AT BASIS OF BABCOCK TEST This method is based on the action of two agents : (i) the action of strong sulphuric acid upon the con- stituents of milk-serum, and (2) the action of cen- trifugal force. Action of sulphuric acid in Babcock test. — The sul- phuric acid used in the Babcock test performs, at least, three functions, which we will consider briefly. (i) Action on serum-solids of milk. — Strong sul- phuric acid acts chemically and physically upon the milk-serum solids (casein, sugar, albumin and salts) in such a way as to destroy that strong mechanical, ad- hesive influence exerted by the milk-serum solids, which tends to prevent the fat-globules separating from the form of an emulsion. When this influence is overcome, the fat-globules are more free to collect in a mass. (2) Heat furnished by action of sulphuric acid. — The action of sulphuric acid upon the water of milk- serum and also upon the organic solids of the serum generates so much heat that the fat-globules easily lose their individuality and run together, a condition favor- ing rapid separation of fat from serum. (3) Specific gravity of serum increased by sulphuric acid. — The sulphuric acid, being nearly twice as heavy as milk, increases the difference in specific gravity be- tween the milk-fat and the liquid surrounding it. The milk-fat, being much lighter, more readily rises to the surface of the heavy liquid. Action of centrifugal force in Babcock test. — The action of the sulphuric acid having released the milk- fat largely from the form of an emulsion in the milk- 34 MODERN METHODS OF TESTING MILK serum, the completion of the separa- tion of fat is effected by centrifugal force. When the bottles containing the mixture of milk and acid are whirled, the centrifugal force acts more strongly upon the heavier por- tion, that is, the mixture of acid and milk-serum. Hence this heavy mix- ture is forced to the outside, which is the bottom of the bottle, while the much lighter fat is forced to the top. A small amount of fat (.i to .2 per cent.) remains unseparated under usual conditions. The following apparatus and ma- terial are used in making the test : ( i ) Test-bottles, (2) pipette for measur- ing milk, (3) acid-measure, (4) tes- ter or centrifugal machine, and (5) sulphuric acid. TEST-BOTTLES The usual form of bottle used in testing milk is shown in Fig. 7. The neck of the bottle is marked with a scale so graduated that each small di- vision represents .2 per cent, and five of these divisions, making one large division, represent i per cent., when we use 17.5CC.* or 18 grams of milk. The marks extend from o to 10 J FIG. 7 MILK-TESTING BOTTLE *cc. is the abbreviation for cubic centimeters (see p. 205). THE BABCOCK TEST 35 / \ 17.6 C.C. per cent. Why do these divisions represent exact percentages by weight of fat in milk, when no weighing is done in testing milk? We use, in testing, 17.5 cc. of milk, which is known to weigh almost exactly 18 grams. The graduated portion of the neck of the test-bottle is made to hold exactly 2 cc. be- tween the o and 10 marks. Since i cc. of pure milk-fat is known to weigh .9 gram, 2 cc. of milk- fat, the amount required to fill the neck between the'o and 10 marks, weighs 1.8 (.9x2) grams, which amount is just 10 per cent, of the 18 grams of milk sample used in testing. The divisions on the neck of the test-bot- tle should be accurate and uniform ; the lines should run straight across the neck and not obliquely. When the marks and numbers become indistinct from use, they can be rendered clear by rubbing the scale over with the lead of a pencil or with a cloth having on it a little printer's ink or black paint. When in use, each bottle should be numbered or labeled in a distinctive way. MILK-MEASURING PIPETTE The form of pipette in common use is shown in Fig. 8. Other forms are shown in Figs. 9 and 10. The pipette should hold 17.6 cc. when filled to the mark. Since about .1 cc. of milk will adhere to the in- 36 MODERN METHODS OF TESTING MILK side, such a pipette will furnish a sample amounting to 17.5 CO. of milk, which weighs about 18 grams, i cc. of milk weighing about 1.03 grams on an average. The accuracy of the test, so far as regards the amount of sample taken, depends upon the exactness of the pi- pette in holding 17.6 cc. The mark on the stem should, for convenience, be two inches or more from the up- per end of the pipette. MEASURE FOR ACID A cylinder of glass, like that shown in Fig. II, with a lip to pour from and a single mark at 17.5 cc, is the form commonly used. Other forms are shown in Figs. 12 and 13. These latter forms, made so as to hold enough acid for 20 FIG. 9 AUTOMATIC PIPETTE FIG. 10 Wagner's pipette 'N I7.4CC. A FIG. II ACID- MEASURE THE BABCOCK TEST 37 or more tests, are probably the most convenient where many samples are to be tested at the same time. IdkN^SC^ FIG. 12 ACID-BURETTE AND STAND FIG. 13 AUTOMATIC BURETTE THE CENTRIFUGAL MACHINE, OR TESTER The centrifugal machine used in the Babcock test is commonly called the Babcock tester. Various forms have been devised, varying in size from those adapted for a single duplicate test up to the needs of large fac- tories. The designs of recent years are much superior to the early forms. Some of the different types are H CO H H I— I m .« H I H 38 THE BABCOCK TEST 39- represented in Figs. 14, 15, 16 and 17. In general they all consist of a revolving disc placed in a hori- zontal position, and provided with swinging pockets, FIG. 16 — SMALL HAND-TESTER in which the test-bottles are placed. When at rest, the pockets hang down, permitting the bottles to stand upright. When the disc is in motion the pock- ets swing out, carrying the bottles to a horizontal posi- tion, the necks of the bottles being directed in toward the center. The testers should be made to carry an even number of bottles. The steam-turbine tester is the best form of centrifugal for factory work. It has the advantage of maintaining a uniform rate of speed and, in addition, the contents of the bottles are kept hot, and hot water is supplied. In some forms, in which the exhaust steam is not carried away and in 40 MODERN METHODS OF TESTING MILK which no dampers are provided in the cover, the steam testers may heat the fat too high. For use on farms, hand-testers are available. It is always necessary that FIG. 1 7 — ELECTRIC CENTRIFUGAL MACHINE OF LATEST DESIGN FOR USE IN BABCOCK TEST Capable of unusually high speed — safe, accurate and convenient. Made by International Instrument Co. Cambridge, Mass. the tester should be securely fastened to a firm founda- tion and so set that the revolving disc is level. The centrifugal should run smoothly, without jar or trem- ble, when going at full speed. Estimating speed of centrifugal tester. — In order to cause separation of the most fat possible, the cen- trifugal disc must move at a sufficient speed. The re- quired number of revolutions depends upon the diame- ter of the disc, to the edge of which the test-bottles are THE BABCOCK TEST 4I attached. The smaller the wheel, the greater must be the number of revolutions a minute. Farrington and Woll have prepared the following table, showing the necessary number of revolutions for different sizes of testers : DiAAiETER OF No. of I'evoliitions WHEEL IN INCHES of disc per minute 10 1074 12 980 14 • • 909 16 848 18 800 20 759 22 724 24 693 In the case of steam-turbine testers, they are, or should be, made to run at the desired speed under a definite head of steam. These testers should always be provided with a pressure-gage, and a speed-indi- cator is also desirable. In the case of hand-testers, the speed can be ascer- tained in the following manner : Give the handle one full turn and count the number of times a given point on the disc goes round. Suppose, for example, that the diameter of the disc is i6 inches and that it revolves 14 times for one turn of the handle. Such a disc ought to revolve 848 times per minute according to the pre- ceding table. The handle must be turned around as many times a minute as 14 is contained in 848 in or- der to attain the desired speed, which is found to be about 60 times, or once a second. Then, with watch in hand, regulate the turning of the handle until it 42 . MODERN METHODS OF TESTING MILK is made to turn 60 times a minute. The proper speed once attained should be kept up during the testing of a sample. The efficiency of whirling can be further tested by treating different samples of the same milk at different rates of speed, the highest per cent, of fat beyond which there is no increase, showing the right speed. KIND OF ACID USED IN BABCOCK TEST The acid used in the Babcock test is commercial sul- phuric acid, commonly known as oil of vitriol. It should not be quite as strong as the strongest com- mercial acid. While the strong acid has a specific gravity of about 1.84, the acid used in the test should be between 1.82 and 1.83 at 60° F. Effect of weak acid. — If the acid is weaker than that indicated by specific gravity 1.82, there is danger that some of the coagulated casein may not be com- pletely redissolved and this, mixing with the fat, makes the fat-column in the test-bottle more or less pale and cloudy, when it should be clear and usually golden yellow in color. In addition, there is apt to be a collection of cloudy matter at the foot of the fat- column, obscuring the line of division and making sharp reading difficult. The use of more than 17.5 cc. of acid not too weak may give good results. Effect of too strong acid. — When the acid is too much above specific gravity 1.83, the fat-column is dark in color. There is a layer of black material be- low it, and the amount of fat is difficult to read with accuracy. When the acid is too strong, it is possible to secure accurate results by using less than 17.5 cc. THE BABCOCK TEST . 43 4^ r S of acid, the exact quantity being determined by trying different amounts of acid, until the fat-column obtained is clear and yellow. Strong acid, if allowed to stand open to the air, will in time absorb enough moisture to reduce it to proper strength. By far the best plan is to purchase the acid of guaranteed spe- cific gravity 1.82 to 1.83, since all dairy-supply houses now furnish such acid, and then take pains to keep the acid in tightly stoppered bottles when not in use. Testing strength of acid. — The strength of sulphuric acid may be con- veniently tested by a specially designed hydrometer (Fig 18). This instru- ment or acidometer is simply allowed to float in the sulphuric acid, which must be at 60° F., and the specific gravity is read from the scale where it coincides with the upper surface of the liquid, which should be between the scale-marks 1.82 and 1.83. No acidometer should be used whose ac- curacy is not reliably guaranteed. acid Reducing the strength of strong acid. — With the aid of an acidometer, it is possible to purchase strong sulphuric acid and dilute it to proper strength. This is not advised for the average worker. When this is done, extreme caution must be used in diluting the acid. Never pour water into strong sulphuric acid, but always add the acid to the water. The amount of FIG. 18 HYDROMETER FOR TESTING STRENGTH OF SULPHURIC 44 MODERN METHODS OF TESTING MILK dilution depends upon the strength of the acid used. One should start with a small dilution and increase gradually until the specific gravity of the acid becomes 1.82 to 1.83. After diluting the acid with water, the mixture becomes hot, and it is necessary to allow it to cool to 60° F. before testing with the acidometer. Useful indications regarding strength of acid. — After one has acquired some skill in making the Bab- cock test, one can readily tell whether the acid is too strong or too weak from its action when mixed with milk in the test-bottle. One bases his judgment on the rapidity with which the milk-casein is coagulated and redissolved, and also upon the quickness with which, and the degree to which, the mixture of acid and milk turns dark. Keeping acid from air. — The acid should be kept in tightly stoppered bottles, because, if exposed to air, it absorbs moisture and becomes too weak. The stop- per should be glass, since a common cork stopper is soon destroyed by the acid, and even rubber is not long satisfactory. Care in handling sulphuric acid. — Strong sulphuric acid is extremely corrosive and is dangerous to han- dle except with care. In contact with articles like clothing or leather, it quickly ruins them, while on the skin it causes serious burns in a short time. If sulphuric acid gets upon one's skin, it should be imme- diately and thoroughly washed with an abundance of water, and this may be followed by washing with di- lute ammonia or sodium carbonate. In case acid gets on the clothing, treat it first with abundance of water and then with ammonia. Red discoloration on cloth- THE BABCOCK TEST 45 ing caused by acid may be remedied by treatment with ammonia, if not too long delayed. Acid on tables, floors, etc., may be neutralized by treatment with washing soda or other alkali. METHODS OF TESTING ACCURACY OF APPARATUS The correctness of the graduation of the glassware used in the Babcock test is a fundamental condition of accuracy in the results obtained. In some states all graduated glassware used in the Babcock test must be tested by the state and found correct before its use is permitted in commercial operations. Reliable deal- ers guarantee the accuracy of their glassware, and it is found to be much more reliable than formerly. How- ever, it is a safe precaution always to test new appa- ratus before using it. Testing graduated glassware is known technically as calibration. Testing or calibrating milk-bottles. — Test-bottles which show a variation, above or below, of more than one division, or .2 per cent., in the lo per cent, scale, should not be used. The different methods of test- ing will now be considered. (i) Testing with special bottle-tester. — The quick- est method of testing the accuracy of the scale of a test-bottle is to use a special device, which is essentially a simple brass plunger (Fig. 19). This instrument is divided into two equal portions, each part being made of such a size as to displace exactly one cubic centi- meter of liquid. This bottle-tester is used as follovv's : The test-bottle is filled to the zero mark with milk, or one may use water or, better, wood alcohol, impartiiig 4b MODERN METHODS OF TESTING MILK FIG. 19 color to the water or alcohol by adding some black aniline or carmine ink. Fill the bottle nearly to the zero mark and then finish with a pipette or dropper, adding a drop at a time just to the mark. Any drops of liquid adhering to the inside ^\ 2 * walls of the neck must be re- ^^ * "^ moved, using conveniently a strip of blotting or filter pa- per. The tester is then slowly lowered into the neck of the test-bottle until the liquid rises half way between the two sections of the instrument, when the upper surface of the liquid should be at the 5 per cent, mark (Fig 20), if the scale is correct to this point. If the surface of the liquid is above or below the 5 per cent, mark, then the scale is incorrect to that ex- tent. After the accuracy of the 5 per cent, mark is tested, the instrument is then lowered into the bottle until the liquid rises about one-eisrhth of an .Si FIG. 20 TESTING MILK-BOTTLE ACCURACY OF i^-^ch aboVC thc tOp of thc Up- TESTER MILK-BOTTLE . r i xr per section of the tester. If the upper surface of the liquid is level with the 10 per cent, mark, the graduation is correct at that point. The graduation of the scale is regarded as correct, if THE BABCOCK TEST 47 the tester shows the 5 and 10 per cent, marks to be correct. In explanation of the use of this form of bottle-tes- ter, it is to be remembered that the neck of the milk- bottle is so graduated as to hold 2 cc. between the o and 10 marks ; hence, the volume between the o and 5 marks should be i cc, and that between the 5 and 10 marks should be also i cc. The brass plunger is so made that each section displaces, or forces up into the neck, i cc. of liquid, the whole instrument displac- ing 2 cc. This tester therefore gives two tests of the scale, one at the 5 per cent, mark and the other at the 10 mark. Some. of these instruments are made to test the 4 and 8 per cent, points, so that with two testers, one can, if desired, test the accuracy of the scale at the 4, 5, 8 and 10 points. There are also testers of the same form made for cream-bottles. In using this bottle-tester, the following precautions are to be observed : (i) Have the upper surface of the liquid exactly on a level with the zero mark in the neck of the test- bottle before putting the tester in. (2) Clean the inside walls of the neck of the bot- tle from adhering liquid before testing. (3) No air-bubbles should be allowed to adhere to the tester when it is below the liquid. (4) The tester should be dry each time before using. (2) Testing zvith mercury. — From an accurately graduated burette (Fig. 21), measure 2 cc. of clean mercury into the bottle to be tested. Then push down into the neck of the bottle as far as the top line of 48 MODERN METHODS OF TESTING MILK i graduation a close-fitting cork or plug, cut off square at the lower end. Turn the bottle upside down, caus- ing the mercury to run into the neck. The mercury just fills the space in the neck between the o and lo mark, if the graduation is accurate. The same mercury can be used in the same way in test- ing one bottle after another by transferring all the mercury from one bottle to another, which may be conveniently done by slipping a piece of elastic rub- ber tubing over the ends of the necks of the two bottles. In using the same mercury for test- ing one bottle after another, no mercury must be lost in trans- ferring, and none must be left in the bottle last tested. The inside walls of the test-bottle must be dry and clean in order to prevent any mercury adhering. (3) Testing with ivatcr. — The bottle to be tested is filled with water, slightly colored with carmine ink, or otherwise, up to the zero mark. The coloring of the water makes easier the reading of the height of the liquid. Any water adhering to the inside of the neck is removed by a strip of blotting or filter paper. Then one runs into the test-bottle 2 cc. of colored water from a burette or pipette, graduated to twentieths of a cubic centimeter. The upper surface of the liquid should be on a level with the 10 per cent, mark if the scale is correct. Any part of the scale can be similarly FIG. 21 BURETTE AND SUPPORT THE BABCOCK TEST 49 tested, remembering that each per cent, on the scale should contain just .2 cc. of liquid. Testing accuracy of pipette. — When many pipettes are to be tested, one runs into one pipette from an accurately graduated burette (Fig. 21), 17.6 cc. of mercury, closing the lower end of the pipette. The mercury should fill the pipette just to the 17.6 cc. mark, if the mark is correct. The same mercury can be transferred to other pipettes in succession. Care must be taken to have the pipettes clean and dry in- side and that all the mercury is transferred without loss. When only one or a few pipettes need testing, wa- ter can be used, running from a burette into each pi- pette 17.6 cc. of water, which should just fill the pipette to the mark, if accurate. Testing accuracy of acid measure. — Ordinarily the acid measure does not need testing, since a little varia- tion does not affect the results. When desired, it can be tested by running in water or milk from a 17.6 cc. pipette, known to be accurate. KEEPING GLASSWARE CLEAN It is very important that the test-bottles and the pipettes used in the Babcock test should be kept as clean as possible from fat adhering to the inside sur- face. Unless a special effort is made, the bottles quickly become covered inside with a film of fat, which may be sufficient to increase appreciably the results obtained when the bottles are used in testing. The bottles should be kept entirely free from any fat- 50 MODERN METHODS OF TESTING MILK film and the wall should be clear and bright. This can be accomplished without serious trouble. As soon as a test is completed and the amount of fat read, the test-bottle, while still warm, should be emptied. This may easily be done by having a large earthenware jar or crock, covered with a board (Fig, 22), in which are several holes large enough to admit easily the necks of test-bottles. The bottle is in- verted, the neck run down through one of these holes, and at the same time the bottle is shaken up and down in order to remove the white calcium sulphate de- posited on the bottom of the bottle during the test. Then, when one is ready to clean up all the bottles that have been used, each one is rinsed with 8 or 10 FIG. 22 — WASTE-JAR FOR EMPTYING TEST-BOTTLES FIG. 2^ — TEST-COTTl.E UJNSER THE BABCOCK TEST 51 FIG, 24 — TEST-BOTTLE DRAINING RACK cc. of a solution, consisting of one ounce of potassium bichromate dissolved in one pint of sulphuric acid. Then a test-bottle brush is run once up and down the neck of each bottle, and finally each is well rinsed with hot water. There are available several devices which may be found convenient and time-saving where many bottles are used daily. Among these de- vices may be mentioned a bottle-rinser (Fig. 23), a drain-rack (Fig. 24), and a bottle- washer (Figs. 25a, b, and c), described by Farrington (Bulletin 129, Wis. Agr. Exp. Station, pp. 22- FIG. 25a — BOTTLE-HOLDER^ EMPTY ^4)' 52 CHAPTER IV Method of Operating^ the Babcock Test In describing the method of operating the Babcock test, when determining the amount of fat in milk, special attention will be called at each step to such difficulties as may occur, and emphasis will be placed upon such precautions as experience has shown to be necessary in order to obtain accurate results. In brief outline, the different steps may be stated as follows: I Mix thoroughly sample of milk, which is at 60° to 2. Quickly fill pipette to mark with milk. 3. Run milk into test-bottle. 4. Fill acid-measure to mark with acid and pour into test-bottle. 5. (i) Mix milk and acid thoroughly by rotary motion; (2) let stand 2 to 5 minutes; and (3) mix again. 6. Put test-bottles in tester (centrifuge) and whirl 4 or 5 minutes at proper speed. 7. (i) Add fairly hot water up to neck of bottles; (2) whirl one minute; (3) add hot water to 8 or 9 per cent, mark; and (4) whirl one minute. 8. Read results at temperature of about 130° F, PREPARING SAMPLES OF MILK FOR TESTING The milk, which should be at a temperature of 60° to 70° F., is thoroughly mixed by pouring from one vessel to another two or three times, at least, imme- 53 54 MODERN METHODS OF TESTING MILK diately before taking the sample for testing. The spe- cial methods of preparing milk for sampling under various conditions are fully considered in Chap. II, p. 20. The fat must be evenly distributed through the milk just before sampling. Every sample of milk should always be tested in duplicate, that is, two tests should be made at the same time. This insures greater accuracy. If the results of the duplicate test do not agree, there is an error somewhere and the work must be repeated. Also, in case one test is lost and another sample can not be obtained, the remaining test can be used, and the whole work will not be lost. TAKING SAMPLES OF MILK WITH PIPETTE The measuring pipette (Fig. 8, p. 35), is filled at once after the thorough mixing of the milk. This is done by placing the lower end of the pipette well down in the milk and sucking up the milk until it reaches a point in the pipette somewhat above the mark around its upper stem. Then the forefinger, zvJiich must he dry, is quickly placed over the upper end of the pi- pette before the milk runs down below the mark. By lightening the pressure of the finger on the end of the pipette, the milk is allowed to flow out slowly un- til its upper surface just reaches the mark on the stem. Some practice is necessary before one can eas- ily and rapidly manipulate the pipette with accuracy. The pipette must be kept very clean. When sam- ples of several dififerent milks are to be drawn in suc- cession, the pipette may be satisfactorily rinsed by THE BABCOCK TEST 55 drawing it full of the milk next to be sampled, this por- tion being thrown away. TRANSFERRING SAMPLE OF MILK FROM PIPETTE TO TEST-BOTTLE Having filled the pipette just to the 17.6 CO. mark, one holds the pipette obliquely to the bottle, placing the point of its lower end within the neck and against the side of the neck of the test-bottle. The right way of holding the pipette is shown in Fig. 26. By loosening the finger at the upper end of the pipette, one allows the milk to flow slowly down the inside of the V neck. The small portion of milk adhering to the inside of the pipette is nearly all carried into the bottle by blowing through the pipette several times before removing it from the neck of the bottle. Not a drop of the milk should be fig. 26— correct way of allowed to spill outside the bottle in transferring from the pipette. It is not intended to remove every trace of milk from the pipette into the bottle, since allowance for HOLDING PIPETTE AND BOTTLE 56 MODERN METHODS OF TESTING MILK what remains is made in the construction of the pipette. Special experiments having shown that .i cc. of milk will remain adhering to the inside, the pipette is made to hold 17.6 cc. to the mark, but is expected to deliver into the bottle only 17.5 cc, the exact amount required for the test. In delivering the milk, the pipette must never be held perpendicularly in a line with the neck of the test-bottle, running the milk straight down as shown in Fig. 26a, since the narrow neck may easily choke up with milk and run over the top. MEASURING AND ADDING ACID When the samples of milk are in the test-bottles, the acid- measure (Fig. II, p. 36), is filled to the 17.5 cc. mark and the acid (see p. 42) is poured FIG. 26a— WRONG WAY OF iuto the tcst-bottlc. The acid HOLDING PIPETTE AND should bc at 3. tcmpcraturc of BOTTLE 60° to 70° F. Much carc must be exercised in pouring the acid into the test-bottle containing the milk. The test-bottle is held in an inclined position, so that the THE BABCOCK TEST 57 acid will follow the inside wall down to the bottom, and the pouring should be slow and steady. Thus handled, the acid, being much heavier than the milk, forms a layer by itself at the bottom of the bottle, while the milk forms a separate layer by itself on top of the acid. While pouring in the acid, it is well to turn the test-bottle around slowly so the acid may in turn come in contact with different portions of the inside walls of the neck and wash down any adhering milk. Unless this is done, some milk may remain on the wall of the neck, in which case it will not be prop- erly acted on by the acid, and the fat-column will con- tain particles of undissolved casein. If one attempts to pour the acid straight down the neck of the bottle, two difficulties are liable to occur: (i) The neck may easily choke up and the acid over- flow on the operator's hands. (2) The acid may drop into and partially mix with the milk, in which case black particles may appear on the upper surface of the acid layer and later, mixing in the fat-column, may interfere with accurate reading of the results. Temperature of milk and acid. — It is directed to have the milk and acid at a temperature of 60° to 70° F. before they are placed in the test-bottle. There are good reasons for this precaution. If the milk or acid is decidedly cold, as may easily happen in cold weather, the action of the acid may not be vigorous enough to redissolve completely the coagulated casein, thus producing white specks or a cloudy appearance in and below the fat-column at the end of the test. On the other hand, if the milk or acid is at too high a temperature, as may easily happen in hot weather, 58 MODERN METHODS OF TESTING MILK the action of the acid is much the same as if it were too strong, producing dark-colored specks or a dark- ened appearance in and below the fat-column. Ex- perience shows that when the milk and acid are at a temperature between 60° and 70° F., there is no dan- ger of too slight or too strong action of acid. More acid can be used at lower temperatures and less at higher temperatures with satisfactory results, but this involves experimenting; the best way will be to use the regular amount of acid and regulate the tempera- ture of the milk and acid. MIXING MILK AND ACID IN TEST-BOTTLE When the measured amount of acid has been placed in the test-bottle, the acid and milk should be thor- oughly mixed. This is best done by giving the bot- tle a rotary motion, with gentle shaking, until the whole mass becomes liquid and free from solid parti- cles of casein. Much motion up and down should be avoided, since milk might be thrown up into the neck of the bottle beyond reach of the acid, in which case coagulated casein would contaminate the fat-column and impair the results. When the acid and milk first mix, the casein is coagulated in a somewhat solid mass, which gradu- ally redissolves as the mixing becomes complete. The mixing, once begun, should continue until the casein appears to be redissolved. If the operation of mixing milk and acid is incomplete or is interrupted, black particles may appear in the fat-column at the end of the test. It is a wise precaution to allow the bottle to stand THE BABCOCK TEST 59 2 to 5 minutes after the mixing appears complete and then to agitate a second time with rotary motion just before placing in the tester. The action of the sulphuric acid upon the water and organic solids of the milk produces a marked degree of heat, as soon as the acid and milk begin to mix. The color of the solution becomes yellow at first and then passes through varying darker shades of yellow to violet, brown and finally dark-brown, if the acid is of the right strength. (See p. 42.) The coloration is due to the action of the acid upon the milk-sugar and milk- casein. Too strong acid produces a dense black color. In samples of milk containing too much bichromate of potash, the color becomes greenish black. Samples of milk that have been preserved for some time with bichromate or formalin, especially when the preservative is used in larger than usual amounts, re- quire more time and agitation to redissolve the coag- ulated casein than do ordinary samples, since these preservatives harden the coagulated casein. ( See p. 29) . WHIRLING THE TEST-BOTTLES The test-bottles containing the mixture of milk and acid, after being agitated a second time as stated above, are placed in the centrifugal tester (p. 37), and whirled. This is better done soon after the milk and acid are mixed, but it may be delayed without harm for 24 hours, in which case, however, the bottles should be placed in water at 160° to 180° F. for 15 or 20 minutes before whirling. An even number of bottles should be whirled at the same time and they should be placed about the disc 6o MODERN METHODS OF TESTING MILK in pairs opposite to each other, so that the equihbrium of the tester will not be disturbed. When all the sam- ples to be tested are placed in the tester, the cover is placed on the jacket and the machine turned for 4 or 5 minutes at proper speed, 600 to 1,200 revolutions per minute, according to the diameter of the centri- fugal disc. (p. 40.) The whirling brings the fat to the top of the mix- ture in the test-bottle. The whirling of the bottles should never be done without having the cover on the jacket, for two reasons : ( i ) The cover prevents the cooling of the fat in the test-bottles during whirling, and (2) the operator is protected from injury in case a bottle should break and scatter its contents while being whirled. In the case of hand-testers, it may be necessary to put hot water in the jacket in cold weather in order to keep the bottles warm enough. ADDING HOT WATER TO THE TEST-BOTTLES When the bottles have been whirled 4 or 5 minutes, moderately hot water is added to each bottle until the contents come to the lower end of the neck. The water may be added with a pipette or by means of any convenient arrangement. The cover of the ma- chine is replaced and the bottles are whirled at full speed for one minute. Hot water .is again added to the bottles until the fat, which is lighter than the rest of the liquid, rises in the neck to the 8 or 9 per cent mark. One must be careful never to run the water above the 10 per cent. mark. The whirling is then re- peated for one minute at full speed. THE BABCOCK TEST 6l Three points deserve attention in this connection: (i) The temperature of the water added, (2) the kind of water used and (3) the number of times water is added. (i) The temperature of the water added should be above 120° F. The aim in general should be to have the temperature of the fat at the close of whirling at 130° or 140° F., and the temperature of the water added should have reference to this fact. However, any effect of too hot or too cold water can be remedied after the final whirling by adjusting the temperature as needed. (2) Clean, pure, distilled water is the best form to use and, next, soft rain water. Hard water may seri- ously affect the results. Objections to hard water may in most cases be overcome by thorough boiling or by previous treatment with a few drops of sul- phuric acid. (3) Some operators add the hot water only once, filling the bottle to near the top of the neck immedi- ately after the first whirling. The advantage of ad- ding the water in two portions is that the fat is washed free from adhering impurities, since the fat-column is often mixed with various particles which render the reading uncertain and frequently too high. READING RESULTS IN PERCENTAGE OF FAT After the last whirling is completed, the test-bottles are removed from the tester, one at a time, in order to read the results of the test. To ascertain the amount of fat, hold the test-bottle upright, having the graduated scale of the neck of the bottle on a level 62 MODERN METHODS OF TESTING MILK with the eye. Notice the divisions marking the high- est and lowest Hmits of the fat-cohimn. The differ- ence between them gives directly the per cent, of fat in the milk tested. The readings can be made accu- rately to one-half of a division, that is, to one-tenth of one per cent. Some test-bottles are provided with a regulator which moves the bottom of the fat-column to a level with the nearest numbered mark. In connection with the measuring of the fat-column, the following points deserve attention : ( i ) Using divi- ders to assist in reading, (2) the temperature of the fat-column, (3) the upper and lower limits of the fat-column, (4) the correct appearance of the fat- column, (5) defects in appearance of the fat-column. (i) If one uses test-bottles not provided with a reg- ulator for adjusting the level of the fat-column, the reading of the percentage of fat on the scale may be made with less liability of error by measuring the length of the fat-column with a pair of dividers, one point of which is placed at the bottom and the other at the upper limit of the fat-column. The dividers are then removed and one point is placed on the zero mark of the scale on the bottle used, when the other point v/ill be at the exact per cent, of fat in the milk tested. (2) The temperature of the fat, when it is read or measured, should be above 120° F. and not above 140° F., preferably about 130° F. This will insure sharply defined upper and lower limits of the fat-col- umn. In case the contents of the bottles are below 120° F., the bottles should be placed for 15 or 20 min- utes in water that has a temperature of 130° to 140° F., THE BABCOCK TEST 63 before the reading is made. This usually needs to be done in cold weather, when hand-testers are used, especially if no hot water has been placed in the jacket during the whirling. If the fat is above 150° F., it should be allowed to cool to 140° F. or below before reading the results. Too high temperatures give too high results, because the fat-column expands. (3) The line of division between the fat-column and the liquid beneath is nearly a straight line when the testing is properly done, and one need have no doubt about the reading of the scale at this point. But the upper surface of the fat-column is concave instead of straight, which may cause some uncertainty as to the exact point at which the reading should be made on the scale. The correct reading is taken at the line where the upper surface of the fat- column meets the sides of the neck, the very highest point at which the fat-column is seen. The reading should not be made from the dark line or meniscus lower down, which is caused by the refraction of the curved surface. The points at which the readings should be made are shown in Fig. 2'], indicated as A and B. Results read this way agree with those obtained by gravimetric analysis. The objection may be raised that we get too high results by reading from the extreme top points of the fat- column, just as if the upper surface were straight at B 6 S ^-4 — ..J A 2 ^— i ^^^^^—0 FIG. 27 MEASURING FAT-COLUMN 64 MODERN METHODS OF TESTING MILK these points instead of concave. While there is such an apparent error, the excessive reading thus caused is only enough to make up for the loss of fat which can not be separated from the rest of the liquid by centrifugal force and brought into the fat-column. The amount of fat thus left in the mixture of milk-serum and acid is ordinarily about .2 per cent, and this is about the amount of excess obtained by the approved method of reading the upper limit of the fat-column. (4) The fat appearing in the neck of the test-bot- tle at the end of a successful test is of a clear, yellow color, and the line of division between its lower limit and the acid solution beneath it is sharply distinct. However, the fat is apt to be light-colored in the case of milk from cows far along in lactation. (5) The fat-column may show certain defects, if the conditions of the test have not been properly car- ried out, among which are (a) black particles below or above or in the fat-column, or a darkened appear- ance of the whole column of fat; (b) white particles below or above or in the fat-column, or a cloudy ap- pearance of the whole column; and (c) bubbles on the surface. (a) Black particles in the neck of the test-bottle at the end of the test, or a darkened appearance of the fat itself, are due to one or more of the following causes: (i) Too strong acid (above 1.83 specific grav- ity), (2) too much acid (more than 18 cc), (3) too high temperature of the milk or acid (over 75° F.), (4) allowing milk and acid to stand in test-bottle too long before mixing, (5) allowing the acid to drop THE BABCOCK TEST 65 through the milk when poured into the test-bottle, (6) interrupting the mixing of the milk and acid af- ter beginning and before completion. (b) White particles of undissolved casein below or above or in the fat-column, or a cloudy appearance of the fat, are due to one or more of the following causes : (i) Too weak acid (below 1.82 specific gravity), (2) insufficient amount of acid (less than 17 cc), (3) too low temperature of milk or acid (below 60° F.), (4) incomplete mixing of milk and acid, (5) insufficient speed of tester. Sometimes when the fat is not clear, good results may be obtained by allowing the bottles to cool enough for the fat to harden some, and then warming in wa- ter at 140° F. before reading. (c) Bubbles of gas, appearing as foam on the top of the fat-column, are generally due to the use in the test-bottle of hard water containing carbonates. This condition may be prevented by adding to the water, previous to use, a few drops of sulphuric acid. When the foam appears and interferes with the reading, a few drops of alcohol are put on the top of the fat-col- umn and the reading is at once made. The alcohol causes the bubbles to disappear and produces a sharp line of division between the fat and alcohol. If the alcohol is allowed to be in contact with the fat for some time before the reading is made, the alcohol and fat mix and increase the height of the fat-column, thus producing misleading results. 66 MODERN METHODS OF TESTING MILK OUTLINE STATEMENT OF SOME SPECIAL PRECAUTIONS 1. Always make tests in duplicate. 2. Make sure that the sample is a representative one. 3. Have the temperature of the milk and acid at 60° to 70° F. before putting in test-bottle. 4. Use only acid of right strength. 5. Mix milk and acid thoroughly as soon as acid is added. 6. Mix a second time after a short interval. 7. Make sure that the tester runs at right speed and does not jar. 8. Use only clean, soft water in filling bottles. 9. Read bottles before they cool and at about 130° F. 10. To insure accuracy, read each test twice. SOME MODIFICATIONS OF THE BABCOCK TEST One frequently sees references in dairy literature to other forms of tests for milk. As a matter of in- formation, we will notice a few of the modifications of the Babcock test together with other forms that are in use, giving good results. The Russian Test. — This is a modification of the Babcock test, differing mainly in respect to some of the mechanical details. A special automatic pipette is used (Fig. 28), a special form of test-bottle (Fig. 2Sa), the neck being separate from the rest of the bottle, and also a special form of acid-measure. The pipette and acid-measure are one-half the usual size. The milk and acid are run into the bottle very easily, and the bot- tles are filled with hot water automatically while the machine is in motion, the tester also being of a special form. THE BABCOCK TEST 67 The Gerber Butyrometer. — Special forms of tes- ter, test-bottle, etc., are used. The test uses 11 cc. of milk, 10 cc. of sulphuric acid (1.825 specific gravity), FIG, 28 AUTOMATIC "RUSSIAN' =-J P PIPETTE FI-G. 28a 'RUSSIAN'' TEST-BOTTLE and I cc. of amyl alcohol. The operations are carried out about the same as in the Babcock test. The "Sinacid" (no acid) Test. — The distinctive fea- ture of this test is that, in place of sulphuric acid, it uses a patented mixture, consisting of sodium phosphate and citrate, which, unlike sulphuric acid, is entirely free from any dangerous properties ; it uses also a colored 68 MODERN METHODS OF TESTING MILK alcoholic solution. After mixing the "sinacid" liquid with the milk, the mixture is heated to 200° F. for 5 minutes before being whirled in the tester. The re- sults do not always appear to agree closely with those given by other tests, according to the reports of some operators. There is, moreover, some natural preju- dice against using a process, any part of which is patented. Gerber's "Sal" Test. — Gerber has just published a method in which no acid is used. The full details are not given, since the method is to be patented. Good results are reported. CHAPTER V • Method of Testing* Cream by the Babcock Test The Babcock test can be used in ascertaining the amount of fat in cream, but certain precautions and modifications are necessary to insure correct results. A special form of test-bottle is generally used. Spe- cial pains must be taken in sampling cream. For strictly accurate work, the cream sample must be weighed for testing. USE OF MILK TEST-BOTTLES IN TESTING CREAM Test-bottles used in testing milk can be used in testing cream only under special conditions. Cream containing over lo per cent, of fat will fill the neck of the test-bottle too full for measurement, when we take 1 8 grams (about 17.5 cc.) to test. This difficulty may be overcome in two ways : ( i ) By using a sample of cream less than 18 grams, and (2) by dividing an 18-gram sample in roughly equal parts between two or more bottles, according to its richness in fat. In the former case the per cent, of fat read is increased by a correction to be considered later. In the second case, the tests are made as in case of milk and the per- centages found in the different bottles are added, the sum being the per cent, of fat in the cream tested. The volume of cream in each test-bottle is always made up 69 70 MODERN METHODS OF TESTING MILK to about 17.5 cc. by adding water to the cream and mixing before adding acid. SPECIAL CREAM-TESTING BOTTLES To test in one bottle an 18-gram sample of cream containing over 10 per cent, of fat, the neck must be made to hold more than the neck of a milk-bottle, that is, more than 2 cc. This additional space must be ob- tained (i) by using a neck of larger diameter, keeping the length the same as in the milk-bottle or (2) by mak- ing the neck longer, keeping the diam- eter the same. There is this objection to making the necks of test-bottles too large in diameter, that the divisions on the scale come nearer together and the reading of the results is less accurate. If the neck is made long enough to allow as fine graduation as in the milk-testing bottles, then the bottles are too long to use in an ordinary tester. Bulb-necked cream-bottle. — Both of the difficulties mentioned above are overcome by having a bulb made in the neck of the bottle (Fig. 29). In this cream-testing bottle the gradua- tion is as fine as in the milk-testing bottles, the smallest division repre- senting .2 per cent., and the scale ex- FiG. 29 BULB-NECKED CREAM -BOTTLE TESTING CREAM BY THE BABCOCK TEST 71 £30 po |15 go P 1^0 tends from o to 25 per cent. Hence, in these bottles we can use 18 grams of cream, provided the cream does not contain over 25 per cent, of fat, while, with cream testing over 25 and up to 50 per cent., a 9- gram sample can be used for the test. In using this form of bottle, care must be observed when water is added near the close of the test. Neither the lower nor the upper surface of the fat-column must be allowed to come in the bulb, since, obviously, it is impossible then to read the results. This style of bottle is also made with a scale measuring 50 per cent., the smallest divisions being .25 per cent. Straight-necked cream-bottles. — The other varieties of cream-testing bottles are straight-necked, varying in respect to (i) diameter, (2) length of neck, and (3) fineness of gradua- tion. These vary in length from the ordinary size to 9 and 10 inches, the longer sizes requiring special testers. In capacity, the necks vary from 6 to 20 cc, which is equivalent to 30 to 100 per cent., when an 18-gram sample of cream is used. In fineness of graduation, the smallest divisions vary in different bot- tles from .2 to I per cent. One type of straight-necked bottle is shown in Fig. 30. For strictly accurate work, bottles graduated to .2 per cent, should be used; and the use of bottles whose finest division is more than .5 per cent, should be condemned, except for rough work. FIG. 30 STRAIGHT-NECKED CREAM-BOTTLE 72 MODERN METHODS OF TESTING MILK ^ CREAM-TESTING SCALES For weighing samples of cream, a set of accurate scales is required. Differ- ent forms are illustrated in Figs. 31 and 32. Scales with agate bearings are much preferable to other forms, since the agate bear- ings do not rust. Torsion balances and those with steel bearings are liable to rust when kept in damp FIG. 31 CREAM-TESTING SCALES FIG. 32 — CREAM-TESTING SCALES places, and this in time makes them unreliable for accurate work. The scales should be kept in perfect condition and tested for accuracy from time to time. METHOD OF SAMPLING CREAM Cream, from which the test sample is to be taken, whether for part of a composite sample or for direct testing, must be made as uniform as possible through- out its mass. The best method of sampling any but TESTING CREAM BY THE BABCOCK TEST 73 small amounts of cream is by means of a sampling- tube (Fig. 5, p. 27), which must be cleaned for each sample before using. Frozen cream must be thawed and mixed before sampling. Large lots of dried cream or partially churned cream can not be sam- pled with complete accuracy. Cream adhering to the outside of the sampler should not be allowed to go into the test sample. METHOD OF KEEPING CREAM SAMPLES In keeping composite samples of cream, the same precautions are to be observed as in the case of milk (p. 30). With thick cream special effort may be needed to cause the preservative, especially potassium bichromate, to dissolve and be distributed through the sample. This may be done by warming the jars and mixing the cream carefully by a gentle rotary mo- tion. PREPARATION OF SAMPLES FOR TEST Before taking from the cream sample the amount to be used in the test, the cream must have its fat evenly distributed. To accomplish this, the cream is heated to 105° or 110° F. until it is quite fluid. As recommended by Hills (Bulletin 100, Vermont Exp. Sta., p. 5), the cream is passed through a small sieve (Fig. 33), such as is commonly used in kitchens. Any remaining lumps found in the sieve are rubbed through the meshes by the finger, after which the cream is thoroughly mixed by pouring from one cup to another. The sample is then quickly taken for testing. 74 MODERN METHODS OF TESTING MILX WHY USE OF PIPETTE IS INACCURATE FOR CREAM The use of a pipette in measuring cream samples is inaccurate for the following reasons : ( i ) More fat adheres to the inside of a pipette than in case of milk, the error increasing with the thickness of cream. (2) The weight of cream decreases as the per cent, of fat in cream increases, since milk-fat is lighter than the other constituents of cream. The scale of the test-bottle is based on the use of 18 grams of material, but the amount of cream that oc- FiG. 33 cupies the volume filled by CREAM-SAMPLING SIEVE jg granis of milk ( 17.5 cc.) is found to be more or less below 18 grams according to the increased percentage of fat in the cream, run- ning even below 17 grams in very rich cream. (3) Separator cream is more or less filled with bubbles of air, and ripened cream contains gases produced by fermentation. These decrease the weight of a given volume of cream. For the preceding reasons, the result of trying to measure by pipette a sample of cream to be used for testing its fat content is that less cream will be used than should be, and therefore the results will be too low. Any system of volumetric measurement pro- posed is open to some uncertainty and inaccuracy. The use of a pipette in testing cream is justifiable only for work that is not expected to be strictly accurate. TESTING CREAM BY THE BABCOCK TEST 75 If one uses a pipette in measuring cream for test- ing, somewhat more accurate results are generally ob- tained by measuring i8 cc. of cream and also rinsing the pipette into the test-bottle with a small amount of water. Pipettes are obtainable which have an i8 cc. mark as well as a 17.6 cc. WEIGHING SAMPLE OF CREAM The operation of weighing cream is simple. One places the empty test-bottle on one pan of the scales and balances it by a slide-weight or some form of counterpoise. One then places an 18-gram weight on the other pan, after which the pipette is filled with cream somewhat above the 17.6 cc. mark, and this is run into the bottle, the last portion being run in more slowly, until the two scale-pans just counterbalance each other. A little practice enables one to weigh the exact amount rapidly. In case the amount of cream taken in the pipette is not enough, agitate the sample, draw a little more into the pipette and run this slowly into the bottle until it counter-bal- ances the weight. In case too much cream is run into the bottle, the surplus can easily be withdrawn by the pipette. No cream must be allowed to get on the outside of the bottle or on the scale-pan while the weighing is done. Using less than 18 grams. — In case of very rich cream it is preferable to use 9 grams for a sample, in which case the reading of the fat-column is multiplied by 2 ; or the 18-gram sample may be divided between two cream-bottles, in which case water should be ad- ded to each bottle so as to bring the liquid to about 18 cc. in volume. The results of the test in the two 76 MODERN METHODS OF TESTING MILK bottles are added. It is more convenient after a little practice to weigh exactly 9 or 18 grams than to run in an approximate amount of cream and weigh that accurately. In case one uses any amount less than 18 grams for a sample, it is necessary to correct as follows the per cent, of fat read : Divide 18 by the number of grams of cream used and multiply the result by the per cent, of fat read in the test. For example, one uses 13.5 grams of cream and the result reads 15.6 per cent, of fat. Divide 18 by 13.5, which gives 1.33, and multiply this by 15.6, which equals 20.8 per cent., the true per- centage of fat in the sample. SPECIAL POINTS ABOUT TESTING CREAM When one uses 18 grams of cream in one test-bot- tle, the operation is completed as in testing milk. When one uses less than 18 grams in one test-bottle, enough water is added to make the volume about 18 cc. Thus, when an 18-gram sample is divided be- tween two bottles, or a 9-gram sample in one bottle is used, we add 9 cc. of water to the cream, mix them thoroughly and then proceed as usual. The water and cream should always be well mixed before adding acid. When it is difficult to get a clear column of fat, it may be remedied by the use of a little less acid, the exact amount to be found by trial. In general, the richer the cream, the less acid is required, 15 cc. often being enough. The full amount of acid is used when- ever the cream is diluted with water before adding acid, as in the case of taking a 9-gram sample, or di- viding an 18-gram sample between two cream-bottles. TESTING CREAM BY THE BABCOCK TEST ^J Sometimes it is well to let the bottles stand a little while after being mixed with the acid, until the mix- ture turns dark, before whirling. When the liquid just below the fat-column is milky and the fat looks cloudy and light, exact reading is difficult. In such cases the bottles may be placed in water at 130° to 140° F. for 15 or 20 minutes before whirling, or, if this fails, the fat may be solidified by placing the bottles in cold water after the last whirl- ing and then heated to 130° or 140° F. before reading. READING THE PER CENT. OF FAT IN CREAM- TESTING BOTTLES In reading the results of a cream test, more care is needed than in testing milk, especially if wide- necked bottles are used in which the finest divisions represent .5 per cent, or more. For accurate work, one should use narrow-necked cream-bottles in which the results can be read to .2 per cent. In no case, should there be used in the same cream- ery at the same time different varieties of cream-bot- tles, having necks of widely-varying diameter. It has been shown by Webster (Bulletin No. 58, Bureau of Animal Industry, Dairy Division, U. S. Dept. of Agr.), that the results obtained with cream-bottles having necks of varying diameters are wide apart, when read in the ordinary manner. This is due to the fact that the depth of the meniscus increases with the diameter of the neck of the bottle. The wider the neck the greater the error, this being in the direction of too high results. yS MODERN METHODS OF TESTING MILK SAMPLING AND TESTING GATHERED CREAM In many dairying communities, hand separators are used by dairymen, and cream instead of milk is sent to the creamery. The cream of several patrons is gathered at intervals by some one employed for the purpose. The ideal plan would be to keep each pa- tron's cream separate until delivered and sampled at the creamery, but usually the different lots of cream, being comparatively small in amount, are not kept separate, being put together indiscriminately into large cans on the gathering wagons, as a matter of conven- ience. Under such circumstances, it becomes neces- sary to obtain at each farm a sample of the patron's cream before it is mixed with other lots. Outlining the general procedure, the person collecting cream car- ries a sample bottle for each patron, puts into this a sample of cream, takes this to the creamery and either tests at once or makes up a composite sample for each patron from the samples gathered during a period of one or two weeks. These composite samples are care- fully prepared and kept in the manner described on pp. 24-31 and p. y^)' The test for fat is made accord- ing to the directions given on pp. 69-77. Taking up the details, the method of sampling gath- ered cream on farms is carried out in the following manner : The person who collects the cream is equipped with (i) large cans into which different lots of cream are poured after weighing and sampling, (2) spring balance, (3) two pails for weighing cream and mixing before sampling, (4) sampling-tube, and (5) a case containing sample-bottles. At each dairy the TESTING CREAM BY THE BABCOCK TEST 79 collector takes the weighing and sampling equipment from the wagon, including one sample-bottle marked so as to identify the individual patron whose cream is put into it. The cream is poured from one pail to another twice or more until it is completely mixed, and then it is quickly weighed, after which a sample is at once taken and placed in the sample-bottle. The sam- ples should be taken by means of a sampling-tube (p. 2."^^, which, after using, should be rinsed each time with the cream next to be sampled. This is done simply by lowering the tube in the usual manner to the bottcr.i of the pail that holds the cream and then drawing the tube out and allowing all the cream to run back into the pail. After this rinsing, the sample is drawn an J, after allowing the cream adhering to the outside of the tube to drain off for a few seconds, the contents of the tube are transferred to the sample-bottle. In case a patron has more than one pailful of cream, a sample is taken from each weighing, the same Vv'eighing-pail being used each time ; in this case the different samples are put into one sample-bottle. After each patron's cream is weighed and sampled, it is poured into the collector's large can. The sam- ple-bottles are carried to the creamery and the con- tents of each transferred to the corresponding com- posite-sample jar. The collecting bottles must be thoroughly cleaned and dried before using again. If any difficulty is experienced in the management of the composite samples, it is advisable to test each sam- ple of cream after it reaches the creamery and not pre- pare composite samples. Cream that is frozen or sour can not be handled sat- 86 MODERN METHODS OF TESTING MILK isfactorily for sampling or testing and should not be taken. For judging and scoring the quality of cream according to the method given in Chapter XIV, one may make use of the collected sample before it is put into the composite sample. CHAPTER VI Method of Testing^ Skim-Milk, Whey, Butter, Cheese, Etc. Dairy products, such as cheese, condensed milk, and milk powders can be tested for fat by the Babcock test, and also by-products such as skim-milk, butter- milk and whey. In general, the operation is carried out as in testing milk, but some special modifications are necessary. METHOD OF TESTING SKIM-MILK, WHEY, ETC. In testing materials containing only .2 or .3 per cent, of fat, two difficulties are experienced : ( i ) In the ordinary test-bottle, the reading of so small amounts of fat can not be easily done with accuracy. (2) Some fat is necessarily left in the mixture of acid and milk- serum, which may constitute an important factor when the total fat content is small. Special forms of test-bottles used in testing whey, etc. — To enable one to make readings of small amounts of fat with increased accuracy, special forms of double- necked test-bottles have been devised, which are so graduated as to enable readings to be made as low as .01 per cent. (Figs. 34 and 35.) In using these bottles, the milk and acid are delivered into the larger neck, the fat being driven up into the small neck by the hot water added toward the end of the test. 81 82 MODERN METHODS OF TESTING l>ilLK Enough water is added to bring the fat-cokimn into the middle of the small neck. In mixing milk and acid and in running in hot water, care must be taken to prevent any liquid but fat going into the small neck or fine measuring-tube. The stoppers in the bottles FIG. 34 FIG. 35 BOTTLES FOR TESTING SKIM-MILK are used to adjust the fat-column for reading. These double-necked bottles should be placed in the tester in such a way that the filling-tube is toward the center, thus avoiding the danger of having any fat caught between this tube and the side of the bottle when re- suming the upright position after whirling. Separating fat from mixture in bottle. — Attention has previously (p. 64) been called to the difficulty ex- perienced in separating all the fat from the mixture TESTING SKIM-MILK, WHEY, CHEESE, ETC. 83 of acid and milk-serum. Under ordinary conditions of working, materials low in fat, like skim-milk, may fail to give up to the fat-coluran .05 or even .1 per cent, of fat. Some double the reading of fat when it is below .1 per cent, in order to make allowance for the unseparated fat. The fat may be separated from the rest of the liquid more completely by proceeding as follows : Use 20 cc. of sulphuric acid, whirl the bot- tles at full speed one or two minutes longer than usual and read the fat when at a temperature of 130° to 140° F. Steam-turbine testers, which keep the bot- tles hot, give best results. Any test of these by-pro- ducts showing less than .05 per cent, of fat is open to the suspicion of being defective. Skim-milk and buttermilk are treated alike. In working with whey, it is noticeable that after adding acid the mixture turns dark very slowly, due to the presence of less sugar and to the absence of casein. Less than the usual amount of acid is sufficient for whey, 8 or 10 cc. frequently being sufficient. METHOD OF TESTING CHEESE In applying the Babcock test to cheese, it is neces- sary to prepare the sample in a special way and to weigh the amount used. A 9-gram sample is a con- venient amount to use, in which case the reading of fat is multiplied by 2; or an 18-gram sample may be divided between two cream-bottles, the final readings being added. Cream-testing bottles graduated to .2 per cent, divisions should be used. 84 MODERN METHODS OF TESTING MILK "^ Sampling cheese for fat-test. — Since different por- tions of the same cheese vary in composition, special means must be used to get a representative sample. The sample for testing is prepared as follows : When a cheese can be cut, a narrow, wedge-shaped seg- ment is taken, reaching from the outer edge to the center. This is cut into strips and passed through a meat-grinding machine two or three times. This mass is carefully mixed, and from this a 9-gram sample is weighed into a cream-testing bottle, or a 4.5-gram sample into an ordinary milk-testing bottle. When cheese can not be cut, samples are obtained by a cheese-trier. If possible, three plugs should be drawn, one at the center, one about an inch from the outer edge, and one at a point half way between the other two. If only one plug can be drawn, this should be taken at a point about half way between the mar- gin and center. The plugs should be taken perpen- dicular to the end surface of the cheese and should reach either entirely through the cheese or just half way. The plugs should be made fine by grinding or cutting and carefully mixed before weighing samples. In preparing samples, they should not be exposed to the air longer than necessary, since loss of moisture should be prevented as much as possible before weigh- ing. Testing sample of cheese. — The sample is weighed into the bottle. Then one adds about 15 cc. of hot water in the test-bottle and agitates until the water disintegrates the cheese ; this may be hastened by add- ing a few cubic centimeters of acid, and keeping the bottle in slightly warm water. When no more lumps TESTING SKIM-MILK, WHEY, CHEESE, ETC. L5 are seen in the liquid, 17.5 cc. of .sulphuric acid is added. The test is then completed in the usual way. To obtain the per cent, of fat in the cheese, if less than 18 grams is used, multiply the reading of fat by 18 and divide by the number of grams of cheese used in the test. METHOD OF TESTING BUTTER The Babcock test has not been adapted to deter- mine accurately the amount of fat in butter. A mass of butter is so variable in its composition, owing to the uneven distribution of water, that it is difficult to obtain a representative sample when only a small amount is used. Since butter contains over 80 per cent, of fat, a sample of less than 9 grams must be used in a test-bottle made to measure less than 40 per cent, of fat. Approximate results can be obtained by observing certain precautions. Preparing sample of butter. — With a butter-trier draw several samples, aggregating 4 to 8 ounces. Place these samples in a fruit- jar or composite-sam- ple bottle, melt completely by placing in fairly hot water with the jar closed; then remove from hot wa- ter and shake vigorously for one or two minutes, after which moderate agitation is continued until the but- ter solidifies. The cooling may be hastened by hold- ing the jar under a stream of cold water, continuing to shake the bottle vigorously until the butter hardens. Testing sample of butter. — On a cream-scale place a bottle of the form shown. After balancing the bot- tle, one takes on the point of a knife from different 86 MODERN METHODS OF TESTING MILK parts of the sample of buLtcr, prepared as above de- scribed, small pieces of butter and places them in the funnel-shaped tube of the bottle until the sample weighs 9 grams. The bottle is then placed in hot water until the butter melts and runs into the test-bottle. The fat adhering to the P ,,€====» inside wall of the funnel is washed down into the bottle with 8.8 cc. of hot water and the same amount of sulphuric acid (8.8 cc.) is added. After the contents are thoroughly mixed, the bottle is placed in a tester and whirled for about 5 min- utes. To read the result, fill the funnel- shaped tube with hot water, which raises the fat-column in the graduated neck. By gently pressing down or gently draw- ing up the rubber cork at the top of the graduated neck, the fat-column may be moved in the graduated neck so as to bring the lower end of the fat-column level with zero mark, which is indicated by a ring below the bulb. The percent- age of fat is read directly. Before read- ing the test, let the bottle stand for ten minutes in water at 120° — I30°F. FIG. BOTTLE FOR USE IN TESTING BUTTER GRAY'S RAPID TEST FOR WATER IN BUTTER The method described below for the determination of water in butter has been recently devised by Mr. C. E. Gray, Assistant Dairyman in the Dairy Divisiro, Bureau of Animal Industry, U. S. Department of Ag"- riculture. (Circular No. 100, Dec. 31, 1906.) The TESTING SKIM-MILK, WHEY, CHEESE, ETC. 87 operation requires 15 to 20 minutes. The principle of the test depends on the separation of water from but- ter by distillation in the presence of a slightly volatile mixture of high boiling point which does not mix with water. The apparatus will be furnished by dairy-supply houses. Apparatus. — The following apparatus is required for the test: (i) Balance and zveights. — A crearn-testing bal- ance (see p. 72), sensitive to 0.025 gram; and two weights, one 5-gram and one lo-gram. (2) Graduate. — Measuring 6 cc. (3) Burner. — A gas burner or an alcohol lamp. (4) Paper. — Parchment, perfectly dry, 5 by 5 inches in size. (5) Special testing apparatus. — This includes three essential pieces, (a) a flask (A Fig. 36a) holding about 70 cc, (b) a receiving bulb with graduated tube (C Fig. 36a), and (c) a condensing jacket (E Fig. 36a). The receiving bulb is connected with flask A by means of a rubber stopper (B). G is a ground glass stopper fitting tightly into the tube above the gradu- ated scale. Each stopper and tube should be marked so that each stopper shall always be used with its cor- responding tube. ' The graduation of the tube is done after the glass stopper is fitted, so that the zero mark is the end of the stopper. Each mark of the graduated scale represents 0.2 per cent, of water when a lo-gram sample of butter is used. The glass condensing jacket (E Fig. 36a) is connected to the graduated tube by a rubber stopper (D Fig. 36a). OQ MODERN METHODS OF TESTING MILK Tifi tl" lis/ TJfi' (6) Rubber stoppers. — The rubber stopper B is so acted upon in testing by the heat and reagent, that a new one is required after about lOO tests. Reagents. — Reagents are required as follows : (i) Amyl reagent. — This consists of a mixture of 5 parts of amyl acetate and I part of amyl valerianate, and these must be entirely free from water- soluble impurities. (2) Alcohol. — This is used for burn- ing when gas is not available. Operation of the test. — The details of the method of determining water in butter are performed as follows : (i) Preparation of sample — See p. 85. (2) Weighing sample. — Place on each pan one sheet of parchment paper and balance accurately. Weigh ex- actly 10 grams of sample (see p. 75), p!?.cing in center of paper. (3) Transferring sample to Hask. — The paper, containing the lo-gram sample, is folded in such a shape that the paper and butter may be slipped into flask A without losing any of the FIG. 36a ;r.AY's APPARATUS r.ZADY FO-.USE IN SaUiplc. (4) Adding amyl reagent TKSTING WATER IN BUTTER -Fill the TESTING SKIM-MILK, WHEY, CHEESE, ETC. 89 graduate, which must be dry, with amyl reagent to the 6 cc. mark and pour into flask A. (5) Connecting the apparatus. — Connect the appa- ratus as shown in Fig. 36a, fill the condensing jacket E with cool water to within one inch of the top (F Fig. 36a) and remove the glass stopper. (6) Distilling the water from the butter. — Heat the bottom of flask A over the flame of the burner. The butter soon melts, mixes with the amyl reagent, and later the water passes as steam into the receiving bulb (C), in which it is condensed and held. The conden- sation in the graduated part of the tube must be care- fully watched and the steam not allowed to go higher that the 15 per cent, mark, in order to avoid risk of losing water. If it goes higher than this, remove the flame for a short time. If the mixture in flask A shows a tendency to foam over, remove the flame. In case of continued foaming, cool the flask and contents, add about 2 cc. more of amyl reagent and continue heating. (7) Hozu to knozv when the distillation is complete. — When the mixture in the flask becomes brown and all crackling noise in boiling ceases, the water has all been driven from the flask. This usually requires from 5 to 8 minutes. (8) Disconnecting apparatus. — After completing the distillation, flask A is disconnected from stopper B and the glass stopper G is firmly placed in tube C. The tube is then inverted, transferring the water from the receiving tube into the graduated tube, care being taken not to allow any water to be lost through the 90 MODERN METHODS OF TESTING MILK small bent tube inside the bulb. The water is then poured from the condensing jacket, after which the jacket may be removed. (9) Separating amyl reagent from water. — Some amyl reagent is distilled with the water, but does no harm as they can be easily separated. This is done as follows : Hold the receiving bulb in the palm of the hand and the stoppered end away from the body ; then raise to a horizontal position and swing at arm's length sharply downward to the side. This is repeated several' times until the line of separation between the water and amyl reagent is sharply distinct, no amyl reagent being seen with the water, or vice versa. The tube should then be held a short time with the stoppered end downward and the amyl reagent in the bulb agi- tated in order to rinse down any water adhering to the sides of the bulb. ( 10) Reading the test. — After cooling the tube and contents, the percentage of water may be read directly from the scale, when a lo-gram sample is used. The water is in the bottom of the tube. The reading is made to the lower part of the meniscus. Cleaning the apparatus. — The flask and tube should always be perfectly dry before being used. An occa- sional cleaning with washing soda in hot water or other cleansing mixture will keep the apparatus sufh- ciently clean. After making the test, tube C is emptied by holding the stoppered end downward, removing the stopper and allowing contents to flow out quickly. The tube is further cleaned and dried by swinging in the manner described above under (9). TESTING SKIM-MILK, WHEY, CHEESE, ETC. 91 Improved form of Gray's apparatus. — The Wagner Glass Works, of New York, has de- vised an improved form of ap- paratus for Gray's test (Fig. 36b. It is simple, durable and conveniently operated. The wa- ter content of the butter is col- lected in the graduated bore as rapidly as distilled from the butter, and the result may be read directly from the grad- uated bore without swinging or otherwise changing the position of the apparatus. There are no stoppers to remove, to replace, or to lose. Steps (8) and (9) (pp. 89-90) are obviated by this improved apparatus. METHODS OF TESTING CONDENSED MILK FIG. 36b Wagner's improved form OF gray's water-test For condensed milk containing no added sugar. — (i) Weigh 9 grams into a milk-testing bottle, add about 10 cc. of water, then the usual amount of acid. Complete test as for milk. Multiply fat-reading by 2. (2) Weigh 40 grams of the well-mixed sample into a 100 cc. glass cylinder, such as is used in the alkaline- tablet test (Fig. 38, p. 104). Add water to the 100 cc. mark and shake until well mixed. Then, with a 17.6 cc. pipette, take for testing a sample of this solution, representing 7 grams of the undiluted sample, and 92 MODERN METHODS OF TESTING MILK complete the test as usual. Multiply the fat-reading by i8 and divide by 7 to find the correct per cent, of fat in the sample. (3) Approximate results may be obtained by di- luting a measured amount of condensed milk with an equal volume of water, agitating the mixture until uniform. Then take sample for testing with a 17.6 cc. pipette and proceed as usual. The per cent, of fat read is multiplied by 2. For condensed milk containing added sugar. — Many brands of condensed milk contain added cane sugar, which in testing is so blackened by the acid as to make the results unreliable. This trouble can be over- come by special treatment devised by Farrington. Make a solution of 40 grams of condensed milk in TOO cc. of water, as directed above in (2). With a 17.6 cc. pipette, measure the same amount as for a milk test into a milk-testing bottle. Add about 3 cc. of the sulphuric acid used in the test and mix the acid and milk by shaking vigorously. The acid is added to coagulate the curd and enclose the fat, allowing the sugar to separate in the surrounding liquid. The curd is compacted into a lump by whirling the test- bottles in a steam-turbine tester at high speed and at a temperature of 200° F. After this whirling, the bottles are taken from the tester and the liquid portion, containing much of the sugar, is carefully poured from the neck without breaking the lump of curd. Then an addition of 10 cc. of water is made to the test-bottles, the curd is shaken up to wash out more sugar, and again 3 cc. of sulphuric acid added. The bottles are again whirled and the liquid portion de- TESTING SKIM-MILK, WHEY, CHEESE, ETC. 93 canted. Then the test is completed by adding lo cc. of water, 17.5 cc. of sulphuric acid, and proceeding as usual. Correct the fat- reading by multiplying by 18 and dividing by 7. METHOD OF TESTING FOR FAT IN INFANT FOODS Some infant foods contain so much sugar that, like sweetened condensed milk, it is impossible to use the Babcock test unmodified. Cochran recommends a method for the determination of fat in such cases, which he has found very useful (Journal of the Amer- ican Chemical Society, Vol. 27, p. 908). A double- tubed test-bottle is used, the larger tube reaching only a short distance into the bowl of the bottle. To determine fat in sweetened condensed milk, make a solution of the condensed milk in the manner given above under (2). Of this solution put 17.5 cc. in the test-bottle, add 9 cc. of 80 per cent, acetic acid and 9 cc. of strong sulphuric acid. Mix the acids and milk and set the bottle in hot water until the mixture in the bottle turns a coffee-brown color. Then remove the bottle, cool, add 15 cc. of ether and mix thoroughly with the liquid. The flask is again placed in hot wa- ter and the ether allowed to evaporate. A layer of fat will appear floating on the surface of the liquid. By pouring hot water into the side tube, the liquid fat is raised into the graduated tube, where its amount can be read. The reading is corrected by multiplying by 18 and dividing by 7. In the case of powdered infant foods, like malted milk, weigh 6 grams of the powder into the bottle, 94 MODERN METHODS OF TESTING MILK add 17.5 cc. of water, dissolve the powder as com- pletely as possible, and proceed as above. Multiply the reading by 3. METHOD OF TESTING DRIED MILK OR MILK POWDER Successful means of drying milk have recently been devised, and products are appearing on the market in the form of dried skim-milk and dried milk. These materials are in the form of fine flaky or powdery substances. Owing to the great advantages of hand- ling milk in such forms, these products are destined to find extensive use, and the desirability of testing them is obvious. The Babcock test, when applied to these materials, gives results much below the truth. The wTiter has obtained quite satisfactory results by a combination of the Gottlieb and Babcock methods of determining fat. The process used is as follows : Dissolve 10 grams of milk powder in 100 cc. of water. Take 10 cc. of this solution in a 100 cc. glass cylinder (Fig. 38, p. 104). Add I cc. of strong ammonia and shake until thor- oughly mixed with the solution. Then add the fol- lowing reagents, one after the other, shaking vigorously after each addition before adding the next material : 10 cc. of 92 per cent, alcohol, 25 cc. of washed ether, and 25 cc. of gasoline or, better, petroleum ether (boiling point below 80° C). The cylinder is closed with a tightly-fitting, moistened cork stopper. The contents of the cylinder, after thorough shaking, are poured into a 200 cc. beaker, the cylinder being rinsed with a little gasoline, and this being added to the TESTING SKIM-MILK, WHEY, CHEESE, ETC. 95 beaker. The beaker is then placed on a steam-bath or in boihng water and kept there until the ether, alco- hol and gasoline are completely evaporated. The re- maining liquid is then poured into a milk-testing bot- tle, the beaker is rinsed with a little ether, which is also added to the test-bottle. The test-bottle is then placed in boiling water a few minutes to evaporate the ether. After cooling the contents of the test-bot- tle to about 70° F., one adds 17.5 cc. of sulphuric acid and completes the operation as in the case of milk by the Babcock test. The reading is multiplied by 18. In some cases, as in dried skim-milk, it will be desira- ble to make the original solution more concentrated by having 20 or more grams in 100 cc. of solution. The following precautions must be observed in using this method : ( 1 ) The milk-powder solution must be made uni- form before sampling. (2) The shaking of the mixture must be vigorous and thorough after the addition of each of the reagents used. (3) The evaporation of the reagents added must be complete; otherwise, the final results are apt to be too high. (4) The evaporation must not be carried so far as to cause any appearance of solid particles in the liquid. When this happens, the fat-column contains dark ma- terial, which makes the results uncertain. This method is applicable to skim-milk, whey and buttermilk, which do not usually give high enough re- sults by the Babcock method. CHAPTER VII Methods of Testing the Acidity of Milk and Milk Products It is often necessary to know how much acid is present in milk, cream, whey, etc. The amount of acid in milk may be a suggestive indication of the age of milk and of its care. In butter-making, the uniformity of the product depends largely upon the ripening of cream, which can be well controlled only by knowing its acidity. In cheese-making, it is at times important to know whether milk contains too much acid and it is also quite essential to have some knowledge of the amount of acid present in the milk and whey at different stages of the operation. THE CAUSES OF ACIDITY IN MILK AND ITS PRODUCTS We may distinguish two kinds of acidity in milk and its products: (i) Apparent acidity, and (2) acid- ity due to lactic acid. The apparent acidity is due to the presence in normal milk of casein and acid phos- phates, which have the power, like free acids, of neu- tralizing alkalis. This apparent acidity in fresh milk is about .07 or .08 per cent, on the average. It varies with different conditions, increasing, for instance, with advance of lactation. Acidity due to lactic acid is formed in milk after it is drawn, and is caused by the action of certain forms 96 TESTING ACIDITY OF MILK AND MILK PRODUCTS 97 of bacteria upon milk-sugar. In general, when milk contains over .10 per cent, of acid, it may safely be assumed that it contains some lactic acid. The amount of lactic acid present in milk may be approximately found by subtracting .10 from the total amount of acid apparently present. However, in speaking of the acidity of milk, we usually mean the total acidity, and not that due to lactic acid alone. One can not com- monly detect a sour taste in milk that has a total acidity under .3 per cent. GENERAL PRINCIPLES OF TESTING ACIDITY The method of determining the amount of acid in milk and its products is based upon the well-known chemical action taking place between acids and alka- lis. Whenever we bring together in solution an acid and an alkali, they combine with each other and form a third compound, the acid and alkali disappearing as such. The acid and alkali are said to neutralize each other and the process is called neutralization. For example, if we add together some hydrochloric (muri- atic) acid and sodium hydroxide (caustic soda) in right proportions, we shall have neither hydrochloric acid nor caustic soda, but a new compound, sodium chloride (common salt), which has been formed by the action of the acid and alkali upon each other. The hydrochloric acid used in the experiment tastes sour and biting, while the caustic soda solution has a pecu- liar odor, feels soapy on the skin, and, if strong enough, destroys the skin. After these two compounds are brought tog^ether in proper proportions, there is no 98 MODERN METHODS OF TESTING MILK longer observed any sour taste of acid or soapy feel- ing or odor of alkali, because the acid and alkali have neutralized each other and have combined to form simply common salt, the presence of which is noticed by its taste. The solution is said to be neither alkaline nor acid, but neutral. USE OF INDICATORS IN TESTING ACIDITY In working with acids and alkalis, it .is necessary to have some simple means of knowing when a solu- tion is acid, alkaline or neutral (neither acid nor alka- line). This can be found by using some substance, called an indicator, which is so acted on by alkalis and acids as to undergo changes of color, being changed one color by alkalis and a different color by acids. One substance which finds wide use as an indiccitor is a chemical compound called phenolphthal- cin, a solution of which is turned pink by alkalis and colorless by acids. For use in testing acidity, one dis- solves 10 grams of the dry powder in 300 cc. of 90 per cent, alcohol. It is necessary to use only 5 or 10 drops of this solution as indicator. GENERAL APPLICATION OF PRINCIPLES OF NEUTRALIZATION What use can be made of the foregoing facts in finding the per cent, of acid in a solution? For sim- plicity, we will use the following illustrative experi- ment: In a glass or teacup we put 100 cc. of a solu- tion containing .25 per cent, of lactic acid and add 5 or 10 drops of indicator solution. Into this mixture TESTING ACIDITY OF MILK AND MILK TRODUCTS 99 we run from a graduated cylinder or burette some standard solution of caustic soda, prepared by dis- solving 4 parts by weight of pure caustic soda in i,033 parts of distilled water. This solution of caustic soda we add, a little at a time, to the solution of lactic acid, stirring or otherwise agitating the mixture thor- oughly after each addition. The pink color that ap- pears when the caustic soda solution is added disap- pears on stirring. After the alkali has been added several times, the color disappears less rapidly each time. The gradual addition of the alkali is continued until finally the pink color does not disappear readily on continued agitation but remains for some moments. The neutralization of the acid by the alkali is complete, and the addition of alkali stops at this point. The ap- pearance of the pink color throughout the body of the liquid means that enough alkali has been added to combine with the lactic acid, and a little more, one drop of the alkali solution being enough to produce the pink color with the indicator after the acid is neu- tralized. The liquid in the cup contained at the start only a solution of lactic acid. As soon as we added alkali, it combined with the lactic acid, forming the neutral compound, sodium lactate. We then had a mixture of lactic acid and sodium lactate. As we con- tinued to add alkali, the amount of sodium lactate in- creased, while the amount of lactix. acid decreased. Finally, a point is reached when tlie solution in the cup contains no free lactic acid, but only sodium lac- tate, and the addition of one more drop of alkali turns the indicator pink, producing a more lasting coloration throughout the solution and showing that the acid LOff. lOO MODERN METHODS OF TESTING MILK has been completely neutralized, that is, changed into sodium lactate. Having completed the neutralization of the acid, we examine the burette or the graduated cylinder con- taining the alkali to find exactly how much alkali so- lution has been used in neutralizing the acid. The lactic acid has required, say, 28 cc. of soda solution. Each cubic centimeter of alkali neutralized by acid corresponds to .009 gram of acid, and 28 cc. would therefore correspond to .25 gram of lactic acid. This amount of lactic acid in 100 cc. is .25 per cent. The process of making a chemical determination by means of a standard solution is known as titration. NEUTRALIZATION METHOD APPLIED TO TEST- ING ACIDITY OF MILK AND MILK PRODUCTS In practical dairy work, one is freed from the ne- cessity of preparing standard solutions, except in a simple way, and the calculations needed to figure the results are direct and easy. The caustic soda solution is prepared in such strength that a certain amount of it equals .1 per cent, of lactic acid, when a certain amount of milk or other substance is used. All tests for the acidity of milk and its products are based upon the general principles previously described and differ from one another simply in some of the details of carrying out the process. There are now available sev- eral forms of so-called acid tests, among which we shall describe the following : "Mann's acid test," "Far- rington's alkaline-tablet test," and the ''Purdue alkali test." TESTING ACIDITY OF MILK AND MILK PRODUCTS Id DESCRIPTION OF MANN'S ACID TEST Apparatus and materials. — The apparatus and ma- terials used in this test are : ( i ) A 50 cc. glass burette, with stopcock, (2) a 50 cc. pipette for measuring sam- ples, (3) a glass beaker and stirring- rod, (4) an alcoholic solution of phenolphthalein, and (5) a standardized alkali solution la- beled ''neutralizer," each cubic centimeter of which contains enough sodium hydrox- ide (.004 gram) to neutralize .009 gram of lactic acid. The outfit is shown in Fig. Operating the test. — (i) Measuring sample. — Into a clean beaker or white por- celain cup, one measures with the pipette exactly 50 cc. of the material (milk, cream, etc.) to be tested. The pipette is then rinsed by filling with water, pre- ferably distilled, and this rinsing water is added to the sample. (2) Adding alkali. — The indicator (5 or 10 drops) is then added to the sample. The burette is then filled to the zero mark with the neutralizer, and some of it is run into the sample in the beaker or cup. A pink color appears, but disappears on stirring. The addi- tion of alkali in portions is continued with care, the sample being stirred after each addition. It will be noticed sooner or later, according to the amount of acid present, that, after each addition of alkali, the FIG. 37 — MANN's acid test 102 MODERN METHODS OF TESTING MILK pink color disappears more slowly, showing- that the acid is becoming more nearly neutralized. When the color disappears quite slowly, the neutralizer should be added a drop at a time. Finally a point is reached when, after the addition of one drop of alkali, the pink color does not disappear even after stirring lo or 15 seconds. This indicates that the acid is com- pletely neutralized. Add no more alkali. The pink color will slowly disappear after standing some time, even when the solution is alkaline. Some experience enables one easily to know w^hen the pink color is suf- ficiently permanent. (3) Calculating results. — Having completed the ad- dition of alkali, the burette is examined to see how many cubic centimeters of neutralizer have been used. The percentage of acid for 50 cc. or any amount of sample taken is calculated from the following for- mula : Per cent, of acid=:^^- ""• ""l ^"^^^j "^ f ^^ xioo. No. cc. of sample taken When one uses 50 cc. of the sample, this formula becomes Per cent, of acid = No. cc, of alkali X .018. From this last formula one can prepare a table which gives the results and saves making calculations at every test: TESTING ACIDITY OF MILK x\ND MILK PRODUCTS IO3 C.C. OF ALKALI SOLUTION Per ccut. of acid (neutralizer) used in sample tested I 018 2 036 3 054 4 072 5 09 10 18 15 27 20 36 30 54 40, etc 72, etc. By using the first formula -above, similar tables can be prepared when the amount of sample taken is any amount other than 50 cc. Using Babcock Pipette with Mann's Test. — If a 17.6 cc. pipette is used in measuring the liquid to be tested, the percentage of acid may be found by divid- ing by 2, the number of cubic centimeters of Mann's alkali used, the result being expressed as tenths per cent. Thus, if the liquid tested neutralizes 13 cc. of alkali, the acidity is 13 divided by 2, which, expressed as tenths, is .65 per cent. By diluting Mann's ''neutral- izer" with an equal volume of water, the number of cubic centimeters gives directly the percentage of acid- ity, expressed as tenths. FARRINGTON'S ALKALINE-TABLET TEST Apparatus and materials. — In this method the same chemical principles are employed as in Mann's test, but the alkali, in the form of a carbonate, and in- dicator are mixed together in the form of solid tab- lets. Each tablet contains enough alkali to neutralize 104 MODERN METHODS OF TESTING MILK .034 gram of lactic acid. The apparatus (Fig. 38) consists of a 17.6 cc. pipette, a white porcelain cup, and a 100 cc. graduated glass cylinder provided with a rubber stopper. Preparing the alkali solution. — In using the alka- line tablets, one first puts 5 tablets into the graduated 1 "JJ 100 cc. cylinder, and fills ** this to the 97 cc. mark with clean, soft water, prefera- bly distilled. The cylinder is then tightly corked and laid on its side until the tablets dissolve, which re- quires several hours. The FIG. 38— farrington's cylinder must be kept ALKALINE-TABLET TEST ^-^^^^^ ^^^y^^^ ^^ ^^^^ none of the solution can be lost while the tablets are dissolving. A slight flocculent residue will not dis- solve, consisting of some inert matter used in the mak- ing of the tablets. Precautions in using alkali solution. — The solu- tion should always be shaken well before using. It should always be kept tightly stoppered when not in use. The solution, if kept perfectly stoppered, will not greatly change in a week. Solutions older than this may change their strength and should not be used. It is a wise precaution to throw away solutions previ- ously used for some days and to prepare a fresh solu- tion. The solid tablets do not change if kept dry. Operation of alkaline-tablet test. — The material (milk, cream, etc.) to be tested is thoroughly mixed, TESTING ACIDITY OF MILK AND MILK PRODUCTS IO5 and 17.6 cc. is then measured into a clean porcelain cup with a pipette. The pipette is rinsed by filling with water, which is added to the sample in the cup. The tablet solution prepared in the manner described is poured slowly into the cup in small portions at a time, and the mixture is agitated by rotating the cup after each addition of alkali. When the pink color does not disappear readily on agitating (p. 102), the number of cubic centimeters of alkali used is read from the grad- uated cylinder. Each cubic centimeter of alkaline-tab- let solution used equals .01 per cent, of lactic acid in the sample tested. For example, the use of 25 cc. of alkali solution equals .25 per cent, of acid; 40 cc. equals .40 per cent., etc. SPILLMAN'S MODIFICATION OF ALKALINE-TABLET TEST The apparatus consists of an ordi- nary teacup, a regular 17.6 cc. pipette, a quart Mason fruit- jar, and "Spill- man's acid-test cylinder" (Fig. 39). The alkali solution is prepared by dis- solving alkaline tablets in water at the rate of 5 tablets for one cylinder of water filled to the mark 8, the so- lution being made and kept in the fruit- jar. Observe the precautions given above in using tablet solutions. In making the test, put 17.6 cc. of the material to be tested in a teacup, pour into the cup the alkali solution in the manner de- scribed above, until the pink color remains. Then fcr ^=^ iP^ FIG. 39 spillman's acid- test CYLINDER Io6 MODERN METHODS OF TESTING MILK pour the contents of the teacup into the Spihman cyl- inder and read the scale at the surface of the liquid in the cylinder. The results indicate the acidity in tenths of one per cent. The cylinder reads as high as 8 tenths. TABLET TEST MODIFIED FOR RAPID ESTIMA- TION OF ACIDITY It is often desirable to ascertain quickly whether milk or cream contains more or less than .2 or .3 per cent, of acid. Farrington and Woll have devised the following method : An alkali solution is prepared by dissolving in an 8-ounce bottle 2 tablets for each ounce of water used. A No. 10 brass cartridge shell, on which a wire handle is soldered, is used for meas- uring the sample to be tested and also the alkali. A cartridgeful of milk or cream, is placed in a teacup and then a cartridgeful of the alkali solution is added. The contents of the cup are mixed by a rotary motion. If the sample tested remains white, it contains over .2 per cent, of acidity; if a pink color remains, the acidity is less than .2 per cent. The intensity of the pink color indicates the relative amount of acid pres- ent, since the color will be more intense in proportion as there is less acid. Any other measure may be used in place of the brass cartridge-shell, but in every case care must be taken to use equal amounts of milk or cream and of alkali solution. TESTING ACIDITY OF MILK AND MILK PRODUCTS IO7 THE PURDUE ALKALI TEST This test was devised by H. E. Van Norman (Bul- letin No. 104, Purdue Univ. Agr. Exp. Sta., 1905.) The following apparatus (Fig. 40) is used: (i) An A I76CC [ C.C. 6E°F. 100 — ( 90—11 80 — a 70 — 30 60 40 50 50 40 — 60 I30---70 20— « 10 90 DILUTE ALKAU SOLUTION ICC=.Ol7o ^'TH 17.6 CC PIPEHE yOF CREAM ETC.^ FIG. 40 — PURDUE ALKALI TEST ordinary 17.6 cc. pipette, (2) a 100 cc. cylinder, such as is used in the alkaline-tablet test, (3) a 2-quart bot- tle graduated at 1850 cc, (4) sealed bottles, each con- taining 37 cc. of normal solution of sodium hydroxide (caustic soda), (5) a bottle of phenolphthalein indi- I08 MODERN METHODS OF TESTING MILK cator and (6) an ordinary white teacup. The alkaU solution must be of guaranteed accuracy and should be obtained only from a reliable chemical or dairy- supply house. The test is conducted as follows : Into the 1850 cc. bottle one empties the small bottle, containing- 37 cc. of normal alkali solution, rinsing the small bottle with water once or twice and adding the rinsings to the large bottle, which is then filled to the mark with wa- ter. Use only clean, soft or, preferably, distilled wa- ter. This large bottle is kept tightly stoppered. Meas- ure into a white teacup 17.6 cc. of the material to be tested, rinse the pipette with clean water, adding the rinse water to the sample in the cup, and then add 5 or 10 drops of indicator. From the graduated cyl- inder, filled to the 100 cc. mark with the dilute alkali solution, one pours the solution in portions into the sample to be tested, agitating after each addition of alkali, and continuing the addition of alkali solution until the pink color does not disappear on agitation. Then, from the cylinder read the number of cubic cen- timeters of alkali used. Each cubic centimeter corre- sponds to .01 per cent of acid. COMPARISON OF DIFFERENT FORMS OF ACID TESTS Mann's acid test has the advantage of furnishmg the alkaline solution ready for use at any time. It has this serious disadvantage, that the alkaline solu- tion is in constant danger of becoming weakened by absorption of carbon dioxide when exposed to the air. A weakened alkali solution gives higher results of TESTING ACIDITY OF MILK AND MILK PRODUCTS IO9 acidity than the truth. It has also the disadvantage of requiring a calculation to obtain the results in the form of per cent., and any other form of statement ought not to be used. The alkaline-tablet test has the advantage of giving the results directly in percentages. The tablets do not change if kept dry, and the use of a fresh solution avoids all uncertainity of strength. The cost of trans- portation is insignificant compared with that of the neutralizer. The tablets are not always absolutely uniform in strength of alkali, but the variations are not sufficient to make serious differences in results. In the Purdue modification, the small bottles of al- kali do not lose strength so long as they are kept properly stoppered, and therefore any number can be ordered at a time. Being more concentrated than Mann's neutralizer, the cost of trc^nsportation is less. The solution is more quickly and conveniently prepared than the tablet solution. The strength of the small bottles of alkali is said to be more strictly uniform than in the case of tablets. By high dilution of the alkali solution when used, the chance of error is reduced. TESTING THE ACIDITY OF WHEY Whey may be tested by any of the methods de- scribed. Owing to the comparatively low acidity of whey in the operation of cheese-making, it is desirable either to have the alkali dilute (i cc. of alkali equal to .01 per cent, acid), or else to take twice as much whey for testing as in the case of cream, the final results being corrected by dividing by 2. The whey should be free from particles of curd, since curd has the no MODERN METHODS OF TESTING MILK power of neutralizing alkali to some extent. The per- centage of acid in milk can be used as a guide in ripening the milk before adding rennet, in the rate of heating (cooking) the curd, in the regulation of the piling of the curd, in the time of milling, etc. TESTING THE ACIDITY OF CHEESE From a sample of cheese, prepared in the manner described on page 8v, weigh nine grams and to this add water at a temperature of ioo° to iio° F. until the volume equals 90 to 100 cc. Agitate vigorously and filter. To the filtrate add alkali solution, each cubic centimeter of which equals .01 per cent, of lac- tic acid, carrying out the test as with milk, cream, etc. The number of cubic centimeters of alkali used, mul- tiplied by 2, equals the per cent, of acid in the cheese. Much higher results are obtained if one treats the cheese instead of its water extract with alkali, because the nitrogen compounds of the cheese neutralize alkali. RELATION OF FAT IN CREAM TO ACIDITY OF CREAM-RIPENING Cream rich in fat ripens, that is, becomes acid, more slowly than cream poor in fat. This is so, because the larger the percentage of fat in cream the smaller is the percentage of sugar, and the sugar is the source of lactic acid. The favorable influence of ripening upon the process of churning is believed to be due to the action of the acid upon the calcium casein of the cream, converting it into calcium lactate and so lessening its tenacious hold upon the fat-globules in emulsion. The fat itself of the cream is not changed. The amount TESTING ACIDITY OF MILK AND MILK PRODUCTS III of acid to be formed in cream-ripening is, therefore, to be governed more by the amount of calcium casein in the cream than by any other constituent. The less fat there is in normal cream, the more casein there will be, and the greater the per cent, of acidity needed. The more fat there is in cream, the less calcium casein there will be, and the less the amount of acid needed. These statements conform to practical experience. Thus, it is found that in cream containing 25 per cent, of fat, it is necessary to produce nearly .7 per cent, of acid in order to get the results sought by ripening, while, in cream containing 35 per cent, of fat, less than .6 per cent, of acid is sufficient. To ascertain how much acid should be formed in cream before churning, the following rule is suggested by Van Norman (Bulletin 104, Purdue Univ. Agr. Exp. Sta.) : From 100 substract the per cent, of fat in the cream tested and multiply the result by .9, or, expressed as a formula, (100 — per cent, of fat in cream )x.9. For example: Cream with 20 per cent, fat requires .']2 per cent, of acidity. Cream with 25 per cent, fat requires .67 per cent, of acidity. Cream with 30 per cent, fat requires .(yT, per cent, of acidity. Cream with 35 per cent, fat requires .58 per cent, of acidity. The use of .9 as a factor for multiplying may not suit all conditions and some other factor, .8 for exam- ple, may be used. Each operator may experiment and 112 MODERN METHODS OF TESTING MILK easily find what per cent, of acidity is best adapted to the production of the butter suiting his market, and then a table like the above can be made, using .9 or some other factor. OUTLINE STATEMENT OF SOME SPECIAL PRECAUTIONS IN TESTING ACIDITY 1. The material to be tested for acidity must be thoroughly mixed before sampling for a test. 2. The water used in preparing the alkali solution and in rinsing the pipette should be neither acid nor alkaline and should be soft and clean. Use distilled w^ater if possible. 3. The alkaline tablets should be kept dry in well- stoppered bottles. 4. The alkali solution, in whatever form used, must be kept from contact with the air as much as possible to prevent changing strength either through evapora- tion or absorption of carbon dioxide. 5. When alkaline tablets are used, prepare a fresh solution in order to be sure of its strength, if there is any reason for uncertainty. 6. The tests should be made in a good light so that one can easily see the appearance of the longer-lasting pink color at the end of the reaction. 7. The appearance of the pink color at the end of the test can usually be more sharply seen by diluting the material examined with three or four times its vol- ume of distilled water. CHAPTER VIII Method of Testing the Bacterial Condition of Milk A thorough bacteriological examination of milk re- quires somewhat extended special training, but there are methods for ascertaining the general bacterial con- dition of milk which are available in the hands of any careful worker, and which afford most valuable sug- gestions in regard to the cleanliness of milk. The methods of studying the bacterial condition of milk which we shall notice are ( i ) determining the acidity, (2) the fermentation method known in its most avail- able form as the ''Wisconsin Curd-Test," and (3) test- ing for dirt in suspension. THE REI^ATION OF ACIDITY TO THE BACTE- RIAL CONDITION OF MILK The details of the method of determining acidity in milk have been given in Chapter VII. We need say here only a few words in regard to the interpretation of the results as they relate to the cleanly character of milk and its fitness for use. The amount of acid in milk is generally an indication of the care given to the milk after being drawn from the cow ; to some extent, it indicates the age of the milk, especially if the tem- perature at which it has been kept, is known ; and, if the age of the milk is known, the acidity indicates, to some extent, the temperature at which the milk has 113 114 MODERN METHODS OF TESTING MILK been kept. The cleanliness of the milk-pails and other vessels and utensils with which the milk comes in con- tact is another important factor in influencing acidity. The average acidity of English market milk, sup- posed to be 12 to i8 hours old, is .18 per cent., and of German milk, .13 to .18 per cent. Market milk should not, in any case, contain over .2 per cent of total acid- ity when it reaches the consumer, and should generally be under .15 per cent. This (.2 per cent.) is also the highest limit allowable for milk that is to be used for cheese-making. The test for acidity can thus be made a very useful indication of the bacterial condition of milk so far as it relates to the acidity, and to the abundance of those forms that produce lactic acid. THE FERMENTATION OR WISCONSIN CURD-TEST Milk frequently contains objectionable forms of organisms or ferments that are not made perceptible by ordinary methods of observation. The condition arises particularly in milk used for cheese-making and may result in serious injury to the quality of the cheese. The Wisconsin Experiment Station (Wis. Exp. Sta. I2th and 15th Annual Reports, 1895 and 1898) has applied certain principles to the development of a test that enables one to identify milk containing cer- tain forms of undesirable ferments likely to do serious injury. This method is based, in general, upon the plan of making conditions favorable for the rapid de- velopment of the ferments present in milk. Apparatus. — The apparatus consists of the follow- ing parts: (i) Pint glass jars or tin cans with covers; TESTING THE BACTERIAL CONDITION OF MILK II5 (2) a well insulated tank to hold the jars, (3) rennet extract, (4) a thermometer, (5) a case-knife or simi- lar instrument for cutting curd, and (6) a small pipette for measuring rennet-extract. Operation of test. — The test is conducted as fol- lows : The jars, including covers, just previous to use, are sterilized with live steam, scalding water or dry heat (212° F.). Each jar or can is filled about two-thirds full with the milk to be tested and the ster- ilized cover put on at once. The jars are then placed in the tank which is filled with water at 100° to 102° F. up to the upper surface of the milk in the jars. The temperature of the water should be kept at 100° to 102° F. during the whole operation. To hasten the warming of the milk, the jars are taken out and shaken occasionally. The temperature of the milk is observed with a sterile thermometer, and when the milk has reached 98° F., one adds 10 drops of rennet-extract to each jar and mixes thoroughly by giving the con- tents of the jar a rotary motion. When the milk has coagulated, it is allowed to stand until it is firm, usu- a!l}' about 20 minutes. To enable the whey to sepa- rate more readily, the curd is then cut fine with a thin knife, which must be carefully rinsed with hot water after finishing each jar and before using it in another, in order to avoid carrying contamination from one milk to another and spoiling the test. The curd is allowed to settle completely. When the whey has been separating half an hour, the samples are exam- ined for flavor by smelling, after which the whey is carefully poured out of the jars and this is re- Il6 MODERN METHODS OF TESTING MILK peated at intervals of 30 to 40 minutes for 8 hours or more. Under the favorable conditions of tempera- ture, similar to those employed in cheese-making, the organisms present develop readily and reveal their presence in different characteristic ways. The jars are finally opened, any whey present is drained off, and the following tests are applied: (i) The curd is cut into two pieces. The curd will be solid and free from holes on the cut surfaces, if the milk is not tainted. If it is spongy and full of holes, it con- tains those undesirable organisms that produce gases in the curd and injure it for cheese-making, showing in the form of "floating curds" and ''huffy" cheese. The holes are usually small, their common name be- ing "pin-holes." (2) The curd is examined with ref- erence to any marked disagreeable odors that may be present. Some undesirable organisms reveal their presence by smell without making spongy curd. This may, perhaps, be best perceived by smelling of a freshly cut surface of the curd. Offensive odors are, of course, an undesirable indication. Special appara- tus for performing the test is furnished by dairy-sup- ply houses, but pint fruit- jars and other home-made appliances will answer satisfactorily. By this method one can learn what particular lot of milk among several is responsible for undesirable fermentations. Moreover, having traced the source of contamination to a single herd of cows, it is easily possible, by applying the test to single cows, to ascer- tain which individual or individuals may be the source of trouble. TESTING THE BACTERIAL CONDITION OF MILK II7 Precautions, — Two points must be carefully ob- served in carrying out this test: (i) The temperature must be kept as near 98° F. as possible, in order that the bacteria may develop as desired. This can be done by keeping the temperature of the w^ater sur- rounding the jars at 100° to 102° F. The tempera- ture must be v/atched. (2) The thermometer and the knife used should be made not only clean but sterile each time after using in one sample before placing them in another. GERBER'S FERMENTATION TEST This test consists in heating milk in tubes 6 hours at 104° to 106° F. and then observing the odor, ap- pearance, taste, etc., for abnormal qualities. The milk is heated a second time for 6 hours at 104° to 106° F. Any abnormal coagulation of the milk is noticed, such as holes due to gas. Gerber states that milk coagu- lating in less than 12 hours is abnormal, due either to the abnormal character of the milk itself when drawn or to improper care after being drawn. Milk that does not curdle within 24 to 48 hours is open to the suspicion of containing preservatives and should be examined for such substances, METHOD OF TESTING MILK FOR SUSPENDED DIRT The amount of dirt in suspension in milk can be estimated without serious difficulty. The best way is to provide a small centrifugal machine that is made to run at higher speed than the Babcock tester. A ii8 MODERN METHODS OF TESTING MILK FIG. 41 — HAND-CENTRIFUGE FOR SEDIMENTATION WORK form of hand-centrifuge is shown in Fig. 41. Special graduated tubes (Fig. 42) are made to use in this. The milk to be tested is stirred thoroughly, the tube is filled to the highest mark, placed in the pocket of the centri- fuge and whirled several minutes. The sediment col- lects at the bottom and can be easily measured by read- ing the amount on the scale. In Fig. 43 is shown a Bausch and Lomb electric centrifuge. This company also furnishes hand-centrifuges capable of 3,000 to 8,000 revolutions per FIG. 42 — TUBE FOR SEDIMENTATION WORK FIG. 43 — BAUSCH & LOMB ELECTRIC CENTRIFUGE Speed 1,000 revolutions per minute ^H f* iig I20 MODERN METHODS OF TESTING MILK minute. Their centrifuges and tubes can be used also in testing for fat in milk by the Babcock method. In Fig. 44 is shown another form of electrical cen- trifuge which is very satisfactory FIG. 45 for collecting sediments. GLASS FOR COLLECTING ^ uicthod Icss accuratc, but SEDIMENT IN MILK fairly satisfactory in the absence of any better means, is to place about 4 ounces of milk in a test-glass (Fig. 45) and let it stand for one or two hours. TJhe dirt collects in the bottom and its amount can be roughly estimated by the eye. CHAPTER IX Methods of Testing Milk by Ren net -Extract and Pepsin In cheese-making it is necessary to have some means of finding out when the rennet-extract should be ad- ded to milk in order to secure the best results in the process. This is usually known as "testing the ripe- ness of milk." Two methods are in common use for this purpose: (i) The Alonrad test and (2) the Marschall test. THE MONRAD TEST This test is based upon the amount of time re- quired for a definite quantity of milk at a given tem- FIG. 46 — MONRAD RENNET-TEST perature to become coagulated by a fixed quantity of rennet. The pieces of apparatus (Fig. 46) required are the following: (i) A tin cylinder for measuring milk, holding, when full, 160 cc, (2) a 5 cc. pipette, (3) a i2T 122 MODERN METHODS OF TESTING MILK 50 cc. glass flask, (4) a thermometer, and (5) a half- pint tin basin. In testing the ripeness of milk by means of rennet- extract, one first prepares a dilute solution of the rennet, as follows : One measures with the small pi- pette 5 cc. of rennet-extract into the 50 cc. flask. The pipette is then rinsed twice with water by sucking it full of cold, clean water to the mark, the rinsings also being run into the 50 cc. flask. The flask is then filled with water to the 50 cc. mark, and the contents are well mixed by shaking. The next step is to fill the tin cylinder with the well-mixed milk to be tested and this is emptied into the half-pint basin. The milk must be at the temperature at which one adds the rennet in cheese-making, which is generally about 85° or 86° F. To the milk at the desired temperature, one adds 5 cc. of the diluted rennet solution, mixes it through the milk quickly, using the thermometer as a stirrer. The exact time when the rennet-extract is added to the milk is noted b}^ the second-hand of a watch and then again wdien the milk has coagulated ; the number of seconds. required to coagulate the milk is recorded. The exact point of coagulation can be seen more sharply by scattering a few particles of charcoal (as the blackened end of a partly burned match) on the surface of the milk, and then with the thermometer starting the surface into motion around the dish. The black particles stop the instant the milk coagulates. By using a stop-watch, great accu- racy and delicacy can be attained. Care should be taken to keep the temperature of the milk at 85° or 86° F., testing frequently with the thermometer; and, TESTING MILK BY RENNET-EXTRACT AND PEPSIN 1 23 in case the temperature drops, it can be raised by plac- ing the basin of milk in warm water. In ordinary cheddar cheese-making, milk is ready for the addition of rennet when it coagulates in 30 to 60 seconds un- der the foregoing conditions. THE MARSCHALL TEST In this test the same general procedure is followed as in the Monrad test, but the rate of coagulation is observed in a different way. The following pieces of apparatus (Fig. 47) are used: (a) A testing cup or basin, of about a pint capacity, for holding the milk to be tested. On the inside wall of this cup there FIG. 47 MARSCHALL RENNET-TEST are graduated spaces be- ginning with zero at the top and going by half -divis- ions to 7 near the bottom of the cup, while in the bot- tom of the cup is a glass tube w4th a very small bore, (b) An ounce bottle with a mark on it to indicate 20 cc. (c) A spatula for stirring the milk, (d) A i cc. pipette. The operation of conducting this test is as follows : Measure with the pipette i cc, of the rennet-extract used and empty it into the ounce bottle, previously half filled with clean cold water. Rinse the pipette two or three times by drawing water into it from the bottle and allowing it to run back into the bottle. Mix well by shaking. Then place the milk to be tested 124 MODERN METHODS OF TESTING MILK in the test-cup, setting it in a level position and allow- ing the milk to run out at the bottom. Taking the bot- tle of diluted rennet in one hand and the spatula in the other, watch the level of the milk in the cup. The moment the upper surface of the milk drops to the zero mark, pour the diluted rennet into the milk and stir well. Then leave it alone. When the milk coagu- lates, it stops running through the glass tube. From the graduated scale, read the number of spaces un- covered on the inside of the cup, showing how many divisions of milk have run out. The more slowly the milk coagulates, the larger the amount that runs out; the more quickly the milk coagulates, the smaller the amount that runs out and the fewer spaces there are uncovered. When about 2^ spaces are vmcovered, the milk is ready for addition of rennet. The tempera- ture must be watched, being tested at the start and finish, especially in a cold room. Some objectionable features of the Marschall test should be noticed. A difference in the size of the bore of the glass tube in the bottom of the cup obviously makes a difiference in the results. It is found that the size of the bore of the glass tubing varies in different cups. Therefore, the results given by one cup can not be compared with those of another, unless they are tested on the same milk and found to agree. Special pains must be taken to keep the tube open, since a lit- tle dirt quickly stops it. The Marschall test is con- venient for ordinary work, but is not capable of as great delicacy as is the Monrad test. Results obtained by different workers can be compared by the Monrad test, but not by the Marschall test. TESTING MILK BY RENNET-EXTRACT AND PEPSIN 12 j METHOD OF TESTING RENNET-EXTRACTS Rennet-extract is prepared by soaking calves' stom- achs in dilute brine. This treatment dissolves from the mucous membrane the enzym or chemical ferment that has the property of coagulating milk-casein, a property upon which the process of cheese-making is dependent. The ferment contained in rennet-extracts appears to be the same as pepsin in regard to its ac- tion upon milk casein. Different brands of rennet-ex- tract vary somewhat in their strength, that is, the rapidity and completeness with which they coagulate milk when used in the same amount. It is therefore important to have a means of testing their strength, in order that their value may be definitely known and that cheese-makers may be able to know in advance of using how much they must use for the best results. The Monrad and Marschall tests are available for this purpose. In order to test the comparative strength of differ- ent rennet-extracts, one treats different portions of the same milk with the different extracts to be tested. In all other respects, the details of the methods pre- viously given are followed. All conditions must be kept alike in the different tests. The strength of the rennet-extracts is shown by the rapidity with which the milk is coagulated ; the stronger the rennet, the less the time of coagulation. METHOD OF TESTING PEPSIN Pepsin is beginning to be used in cheese-making as a substitute for rennet-extract. V^ivian has worked out the important details. The scale-pepsin, of strength 126 MODERN METHODS OF TESTING MILK known as 1-3,000, prepared from stomachs of sheep, is recommended. It may be used at the rate of 5 grams for 1,000 pounds of milk. In testing scale- pepsin by the rennet test, one can dissolve the scale- pepsin at the rate of 5 grams in 4 ounces of water and use this solution exactly like a rennet-extract with milk. It should be tested in comparison with a sam- ple of rennet-extract whose use in cheese-making has been tested, the test being made on different portions of the same milk. TESTING THE AGE OF MILK BY RENNET-TEST The age of milks and the care with which they have been kept can also be tested in a comparative way by the rennet-test, since with the same rennet-extract or pepsin solution different milks generally coagulate more rapidly in proportion to the amount of acid con- tained in them, especially if the amount of lactic acid is considerable. CHAPTER X Methods of Testing Specific Gravity and Solids of Milk by the Lactometer The specific gravity of milk may furnish important information, which becomes of special value when taken in connection with the amount of fat present. Thus, with the data furnished by the specific gravity and the per cent, of fat, we can easily calculate the amount of solids in milk and the amount of solids- not-fat. THE SPECIFIC GRAVITY OF MILK Definition of specific gravity. — By the specific grav- ity of milk, we mean the weight of a given bulk or volume of milk as compared with the weight of an equal volume of water at the same temperature. To illustrate, suppose we have a vat which, when just full of water, contains exactly i,ooo pounds of water at 60° F. Now, if we fill such a vat full of milk of aver- age composition at the same temperature, this amount of milk weighs 1,032 pounds. This is so because the milk contains, in addition to the water in it, several solid substances heavier than water. In this illustra- tion we express the relation or ratio of the equal vol- umes of water and milk by dividing 1,032 by 1,000; the result, 1.032, is the specific gravity of the milk. 127 128 MODERN METHODS OF TESTING MILK Variation in specific gravity of milk. — Since the specific gravity of milk largely depends upon the amount of solids in it heavier than water, the specific gravity should vary, since we know that the amount of solids in milk varies considerably. And so we find the specific gravity of some milks below 1.030 and of some others above 1.035; ^^^^^ most normal milks from herds of cows have specific gravities lying between 1.030 and 1.034. The solids of milk heavier than water are casein, albumin and milk-sugar. They constitute the solids- not-fat of milk and have a specific gravity of about 1.500. Effect of milk-fat on specific gravity of milk. — Milk-fat is lighter than water, its specific gravity be- ing about 0.900 compared with that of water as i.ooo. Therefore, an increase of fat in milk, relative to the other solids, lowers the specific gravity of milk. Thus, by adding cream to normal milk, we can I'^ike its specific gravity lower than that of normal milk. On the other hand, by removing fat from milk, we in- crease the specific gravity, because we remove what is liehter, and leave what is heavier, than water. Effect of adding water and other substances to milk. — Water being lighter than milk, the specific gravity of milk is lowered by addition of water. There- fore, it is easily possible by removing cream from nor- mal milk to increase the specific gravity and then, by adding water, to decrease the specific gravity again to that of normal milk. The addition of sugar, salt or any similar substance to milk increases the specific gravity. Since water has been the most common adul- TESTING BY THE LACTOMETER I29 terant of milk, it was formerly thought that such adulteration could readily be detected by ascertaining the specific gravity; but the results of using the spe- cific gravity may be very misleading, when consid- ered without reference to any other factor. INFLUENCE OF TEMPERATURE ON SPECIFIC GRAVITY Most liquids expand when heated and contract when cooled. A vessel full of milk or water at 40° F. will overflow when heated considerably higher, that is, will hold less of the fluid, and so" the same volume weighs less at higher than at lower temperatures. From this it is readily seen that the specific gravity of a liquid like water or milk grows less when its tem- perature increases. On the other hand, a vessel full of water at 200° F. is not full when cooled to 40° F. The same weight of water occupies less volume and its specific gravit}^ is higher. Decrease of tempera- ture increases the specific gravity of liquids. It is therefore necessary in measuring the specific gravities of different liquids to have the measurements made at the same temperature, if they are to be comparable. The temperature commonly used is 60° F. METHOD OF TESTING THE SPECIFIC GRAVITY OF LIQUIDS The specific gravity of liquids is readily measured by an instrument known as a hydrometer. The use of such an instrument is based on the fact that, when a solid body floats in a liquid, it displaces an amount of liquid equal in weight to the weight of the floating I30 MODERN METHODS OF TESTING MILK body. Thus it sinks deeper in a light hquid than in a heavy one, because it takes a larger volume of the light liquid to equal the weight of the floating body. Such an instrument is graduated as the re- sult of extensive experiments, so that the specific gravity of the liquid in which the hydrometer is placed can be read at the point where the scale is even with the upper surface of the liquid. A hydrometer is cor- rect only for the temperature used in stand- ardizing it. When a hydrometer has a scale specially adapted to the limits of the specific gravity of milk, it is called a lactom- eter. Of the various lactometers made, only two are sufficiently used to deserve atten- tion: (i) The Quevenne, and (2) the New York Board of Health, lactometers. THE QUEVENNE LACTOMETER Description. — This instrument (Fig. 48) is a hydrometer the scale of which is di- vided into 25 equal parts, going from 15 to 40. Each division is called a degree, and FIG. 48 every fifth degree is numbered on the scale. LACTOMETER ^^^ poiut marked 1 5 corresponds to the point marked specific gravity 1.015 on an ordinary hydrometer, and is the point to which it will sink when placed in liquids whose specific gravity is 1. 01 5. The 40 degree mark on the Quevenne lactom- eter corresponds to the specific gravity 1.040 mark on a hydrometer. The relation between specific grav- TESTING BY THE LACTOMETER I3I ity and the scale of the Quevenne lactometer is shown as follows : SPECIFIC Reading of GRAVITY Quevenne lactometer 1.015 15 1.020 .20 1.025 25 1.030 30 1.035, etc 35, etc Corrections for temperature. — The Quevenne lac- tometer is graduated to give correct results at 60° F. When it is used in milk, the milk should be at 60° F., or, if at some temperature above or below 60° F., a correction of the lactometer reading must be made. This correction can be closely made by adding to the lactometer reading .1 for each degree above 60° F., or by subtracting .1 for each degree below 60° F. For. more exact corrections, consult table on the fol- lowing page. The Quevenne lactometer should carry a thermom- eter, the scale of which is placed for convenience above the lactometer scale. Process of using Quevenne lactometer. — The sam- ple of milk to be tested for specific gravity is brought to a temperature between 50° and 70° F. For con- venience the milk is placed in a cylinder (Fig. 49), which is nearly filled. The lactometer is carefully lowered into the milk until it floats and is allowed to remain half a minute or more. Then one reads and records (i) the point at which the lactometer scale comes in contact with the upper surface of the milk; and (2) the temperature. The lactometer reading is 132 &; ,_ c» (M (M CO eo rf TP in iO w o CO 1- on C5 O 1-1 ©» eo •>!}< in CO ? "^^ ot a* at O^ eo eo eo eo CO eo CO fe "2 bi <3^ o ,_ T-, ,-, ©I a ©» eo eo eo '^ o «c t- 00 a o .,_ ©» eo •«i" in CO § O^ at Oi Oi Ol eo eo eo eo CO eo eo fa 13^ OS o o o ■^ rH o ©J ©^ ©I ©» o iO i- 00 o: o ,_ ©I 00 -t in CO 00 ^^ Oi Oi w 5< eo eo eo CO eo eo CO fa <3^ i- 00 00 00 OS OS OS o c o _ o iO CO t- 00 OS o ,_ eo ■^ in CO 1- ^^ J 0* l?t ©J ©* CO eo eo eo eo eo fa' tn CD CO CO CD I> J> t- 00 00 00 o 40 o t^ 00 OS o ,-, ©» CO Tr in <1^ c-t Qi a ©» ©* eo eo eo 00 fa Tt- in in in in in in CD CD CO CO lO CO i- 00 OS o ■^ ©« eo i< in «5 _ -t 1-1 1-1 ©J ©I ©J ©J ©» m CO i- QO OS o .,>_, ©8 eo •rV lO u K ■S ^^ Oi c< Ol a eo eo CO 00 CO 07 fa 1'^ »n CO J- (T) OS o ^ ©T eo f in < (M Gi OJ ©< ©I 00 eo to eo eo eo § S w fa OS OS OS OS OS OS OS OS OS OS OS o Tf m CO I- 00 OS c ,_l St Oi ■^ 9? ^^ 0* ot Ol ©! ©J ct eo eo eo eo 90 fa u bo 00 00 00 00 00 00 00 I- i- t- I- o ■* in CO i^ on «?s o ■^ o* on •<* <1^ Qt ©t ©< ©J ©« ©1 CO eo eo CO eo fa' u bo l- I- t- t- l- I- CO CD CD CO CO o TT in o t- 00 OS o ,_ ©» eo Tf fa° O^ (N eo s K^^ bo W TJ* rt< TT ■* eo eo 00 eo ©J ©? o ■>*< in CO 1- 00 o o .^ ©» eo T»< * eo eo eo eo ©i ©J ©J tH rH o o rr m CO t- 00 o o .,_ ©I ■CO ■Tt" lO ^^ (H (M Oi ©t ©* ©* CO eo 00 eo eo fa' "vibis 00 5« Qi ©J ©J 1-1 ,_, ^ z> o o o ^ in CO «- 00 OS o _ Si 00 e^ Qi (M ot ©< ©J ©t eo 00 CO CO eo fa' «^ ei OJ ■ to- «0 '0- JO- 100 z »5" 70- - • t.030- ^03^■- no- 1?0- V s N 3>- 40' L Comparison of the two lactometer scales. — In compar- ing the scales of the Quevenne and Board of Health lactom- eters, the following points will make clear their relations : ( i ) The zero point on both instruments in- dicates the specific gravity of water, that is, I. GOO. (2) On the B. of H. lac- tometer, the 100 di- visions or degrees from o to 100 corre- spond to 29 divisions on the Quevenne. Therefore one de- gree on the B. of H. lactometer c o r r e- sponds to .29 degree on the Quevenne. To convert the B. of H. lactometer reading into that of the Quevenne, multiply the reading of the former by .29. The relation of the specific gravity scale of a hydrometer FIG. 50 — COMPARISON OF DIFFERENT SPECIFIC GRAVITY SCALES .Sis specific gravity scale; A^'is New York Board of Health lactonaeter; Q is Quevenne lactometer. TESTING BY THE LACTOMETER 135 to the scales of the Ouevenne and B. of H. lactometer is shown in Fig. 50. TABLE II. — DEGREES ON QUEVENNE LACTOMETER CORRESPONDING TO DEGREES ON NEW YORK BOARD OF HEALTH LACTOMETER BOARD OF BOARD OF BOARD OF HEALTH QUEVENNE HEALTH QUEVENNE HEALTH QUEVENNE Degrees Degrees Degrees Degrees Degrees Degrees 60 17.4 81 23.5 101 29.3 61 17.7 82 23.8 102 29.6 62 18.0 83 24.1 103 29.9 63 18.3 84 24.4 104 30.2 64 18.6 85 24.6 105 30.5 65 18.8 86 24.9 106 30.7 66 18.1 87 25.2 107 31.0 67 19.4 88 25.5 108 31.3 68 19.7 89 25.8 109 31.6 69 20.0 90 26.1 110 31.9 70 20.3 91 26.4 111 32.2 71 20.6 92 26.7 112 32.5 72 20.9 93 27.0 113 32.8 73 21.2 94 27.3 114 33.1 74 21.5 95 27.6 115 33.4 75 21.7 96 27.8 116 33.6 76 22.0 97 28.1 117 33.9 77 22.3 98 28.4 118 34.2 78 22.6 99 28.7 119 34.5 79 22.9 100 29.0 120 34.8 80 23.2 .... .... .... 136 MODERN METHODS OF TESTING MILK Corrections for temperature. — In using the B. of H. lactometer, correction is made for temperatures above or below 60° F. For each degree of tempera- ture of milk above 60° F., add .3 to the lactometer reading, and for each degree below 60° F. subtract .3 from the reading. PRECAUTIONS IN TESTING SPECIFIC GRAVITY OF MILK 1. Milk should, for best results, not be examined until I to 2 hours or more after milking, since the specific gravity of milk is lower for a while after be- ing drawn than it is later, due chiefly to the presence of gases. 2. The sample of milk must be completely mixed. 3. The lactometer must be kept clean. 4. In milk which has been preserved by potassium bichromate, the specific gravity is about one degree higher than in the normal milk, in case the usual amount of bichromate has been added. (See p. 30). VALUE OF LACTOMETER IN DETECTING ADUL- TERATED MILK The value of the lactometer in detecting adulterated, especially watered, milk was formerly overestimated. Taken alone, the results given by the lactometer may be thoroughly unreliable and misleading. It has come to be quite generally recognized that the proper use of the lactometer in milk inspection is largely to indicate whether a sample is suspicious and to furnish a guide as to whether it is necessary to take a sam- ple for further detailed investigation by chemical anal- TESTING BY THE LACTOMETER 1 37 ysis. As already stated, a milk which is both skimmed and watered may appear to be entirely normal by the lactometer. METHOD OF TESTING MILK FOR SOLIDS BY LACTOMETER As the result of extended studies of the relations ex- isting between the specific gravity of milk, milk-fat and milk-solids, rules have been formulated by means of which it is possible to calculate with a close degree of approximation the total solids of milk, when one knows the percentage of fat and the (Quevenne) lac- tometer reading. Babcock's formulas for solids and solids-not-fato — The following formulas were devised by Dr„ Bab- cock: (i) Formula for determining solids-not-fat, — Sol- ids-not-fat=}4L+.2f, in which L is the reading of the Quevenne lactometer and f is the per cent, of fat in the milk. (2) Formida for determining solids in milk. — Total solids=:^L+i.2f. These formulas can be expressed in the form of rules as follows : Rule I. — To find the per cent, of solids-not-fat in milk, divide the reading of the Quevenne lactometer by 4, and to the result add the per cent, of fat in the milk multiplied by .2. Rule 2. — To find the per cent, of solids in milk, di- vide the Quevenne lactometer reading by 4, and to the result add the per cent, of fat multiplied by 1.2. Examples: — A milk containing 4 per cent, of fat 138 MODERN METHODS OF TESTING MILK SlFl^'iT >'1 = FIG. 51 Richmond's slide-rule shows a lactometer reading of 32, What is the per cent, (a) of soHds-not-fat, (b) of total solids? (a) The lactometer reading (32), di- vided by 4, equals 8. The per cent, of fat (4), multiplied by .2, equals .8. Add- ing 8 and .8, we obtain 8.8 as the per cent, of solids-not-fat. b) The per cent, of total solids in the milk is 12.8 per cent. ;''for the lactometer reading, divided by 4, equals 8, the per cent, of fat (4) multiplied by 1.2 equals 4.8, and 8 plus 4.8 equals 12.8. Richmond's slide-rule for calculating solids. — Instead of going through the details of calculation to estimate solids in milk,.Richmond uses a slide-rule which is a clever mechanical calculating device. (Fig. 51). The results obtained in this manner agree closely with those given by Babcock's formulas. The method of using the slide-rule is as follows: De- termine (i) the Quevenne lactometer reading, (2) the temperature of the milk, and (3) the per cent, of fat in the milk. Then set the central slide of the rule so that the observed lactometer read- ing is opposite the 60 degree (tempera- ture) mark. The true lactometer read- ing is found opposite the line indicating the observed temperature of the milk. Having thus corrected the lactometer reading for temperatures TESTING BY THE LACTOMETER I39 Other than 60° F., next set the arrow on the sliding portion of the rule opposite the per cent, of fat found in the milk and read the total solids contained in the milk corresponding with the corrected lactometer read- ing or specific gravity. To illustrate, suppose the lactometer reading of a sample of milk at 70° F. is 30 and the per cent, of fat is 4. To correct for temperature and find what the lactometer reading would be at 60° F., the lactometer reading (30) is placed opposite the little arrow at 60 on the temperature scale. Then, looking at the point of temperature 70, we find opposite this point 31.3, which is the corrected or true reading. Next, we place the arrow opposite the 4 per cent, mark, as the milk contains 4 per cent, of fat, and then notice where the point 31.3 (specific gravity), comes in contact with the solids scale. It corresponds closely to 12.8, which is the per cent, of total solids in the sample of milk examined. Some practice with this slide-rule enables one to work rapidly. Specific gravity of milk-solids. — The following rule has been proposed by Fleischmann for calculating the specific gravity of milk-solids : Multiply the specific gravity of the milk by 100, from the result subtract 100 and divide this result by the specific gravity of the milk. Subtract the last result from the per cent, of total solids in the milk and then divide by this re- sult the per cent of total solids of the milk. This may also be expressed by the following formula: Sp. gr. milk-solids= milk-solids milk-solids — (looXsp. gr.) — loo sp. gr. 140 MODERN METHODS OF TESTING MILK Example: A sample of milk contains 12.5 per cent, of solids and has a specific gravity of 1.031 ; — what is the specific gravity of the milk-solids? 100 X 1. 031 — 100 „ ^^. , 12.5 , ^^ C: !5 = 3-Oo6; 12.5—3.006 = 9.494; — •?- = 1.32 1. 031 ^ -^ ^^"^ 9.494 This calculation may assist in determining whether a sample of suspected milk has been adulterated. The variations of the specific gravity of milk-solids is slight, ranging between 1.25 and 1.34. Milks richer in fat have solids of lower specific gravity. The specific gravity of milk-solids is not changed by watering milk, but is increased by removing fat or by addition of skimmed milk. Hence, milk whose solids have a specific gravity above 1.34 is suspected of being skimmed, while a specific gravity above 1.40 is re- garded as clear evidence of skimming. CHAPTER XI Methods of Testing Milk and Milk Products for Adulterations Milk is commonly adulterated in one of the follow- ing ways: (i) By addition of water, (2) by removal of fat (skimming) or addition of skim-milk, (3) by addition of substances not normally found in milk; such as preservatives and coloring matter. All these forms of adulteration may occur in the same milk. DETECTION OF ADDED WATER IN MILK Since water in milk is the same chemical compound as the water found everywhere else, it is impossible to identify added water in milk by any direct test for special properties. The presence of added water in milk can be learned with certainty only by indirect means and even then not with certainty in all sus- pected cases. An examination of milk direct from the cow or herd, when this is possible, may settle the question of watering. The lactometer, while unrelia- ble as a sure means of detecting added water in milk, may give a helpful suggestion, used as a preliminary test. Thus, if a milk shows a specific gravity under 1.028, it is open to the suspicion of being watered, and should then be carefully examined in other ways. Most states fix legal standards for the per cent, of water, solids, fat, and solids-not-fat in milk, and any 141 142 MODERN METHODS OF TESTING MILK milk falling below the fixed limit in composition is regarded as adulterated. Thus, a standard common to several states is 12 per cent, of solids and 3 per cent, of fat. This means also that such legal-standard milk must not contain more than 88 per cent, of water or less than 9 per cent, of solids-not-fat. The relations of the different constituents of milk have been studied and formulas have been devised which enable one in an approximate way to tell how much water has been added to a sample of milk be- yond the amount allowed by the standards. These formulas are based on the assumption that the limits fixed by the legal standard represent the lowest amounts of solids and fat found in normal milk, and they are correct only when the original milk contains the lowest percentages given in the legal standard. In calculating the amount of added water in milk, the amount of solids-not-fat (total solids minus fat) is used as a basis. The procedure is as follows : (i) Determine the per cent, of fat in the suspected sample. (2) Take the lactometer (Quevenne) test. (3) Determine the amount of solids-not-fat ac- cording to the formula, ^L+.2f. (p. 137). (4) Apply the following rule: Multiply the per cent, of solids-not-fat by 100 and divide the result by the legal standard for solids-not-fat. Subtract the last result from 100 and the result is the per cent, of ad- ded water in the sample of suspected milk. This rule is expressed in the form of the following formula: _ , , , , per cent, of solids-not-fat X loo. Per cent, of added water=ioo- legal standard for solids-not-fat. • TESTING MILK FOR ADULTERATIONS I43 This formula gives only the amount of water added beyond the limit fixed by the legal standard and is correct only if the original milk contained the amount of solids-not-fat prescribed by the standard (usually 9 per cent.). Hence, in cases of watered milk, the calculated amount of water added is gener- ally less than the real amount added. Example: A milk is found to contain 3 per cent, of fat and to show a lactometer reading of 2^. Applying the formula for finding the amount of solids-not-fat, the per cent, is 7.35. If the legal standard for solids- not-fat is 9, then the formula becomes 7.35 X 100 „ 100— ^-^^1^^^— =18.3, the per cent, of added water that is contained in the milk, assuming that it contained 9 per cent, of solids- not-fat before being watered. The following rule can also be used: Add together the lactometer reading and the per cent, of fat present in the milk, divide the sum by 36, multiply the result by 100 and subtract the last result from 100. Ex- pressed as a formula, this becomes Per cent, of | • .,. ,^ lactometer reading + per cent, of f at \/ ^ ^^ added water r"" ^^1^=^°° ^6 XlOO. An examination of the serum of milk by means of a refractometer gives, probably, the most reliable means of detecting added water in milk, but this method is available only for special workers. For its details see "Food Inspection and Analysis," by Leach, (p. 139)- 144 MODERN METHODS OF TESTING MILK DETECTION OF SKIMMED MILK The percentage of fat in milk in relation to the other milk-solids is reduced either (i) by direct re- moval of fat through some process of skimming or (2) by the addition of separator skim-milk to nor- mal milk. Milk containing less than 3 per cent, of fat is generally skimmed. Watering milk does not disturb the relations of the constituents of milk to one another, since it reduces the percentages of all uniformly, but the removal of fat does very seriously affect the amounts of the constituents in respect to their relative percentages. In skimming milk, the solid constituent most largely removed is fat, com- paratively little casein, sugar, etc., being taken with the fat. The removal of fat therefore leaves the milk containing less fat but with most of its casein, sugar, etc., still remaining. In normal herd milk, containing over 3 per cent, of fat, the percentage of fat is rarely as low as the percentage of casein and albumin. In 5,500 analyses of samples of American milks, compiled by the author, with a fat content lying between 3 and 5 per cent., the fat averages 3.92 per cent., and the ca- sein and albumin together, 3.20 per cent. ; that is, for i part of casein and albumin there is an average of 1.225 pa-rts of fat. In skimming such milk, the fat may be decreased to i per cent, or .1 per cent., but the re- maining milk still contains about 3.20 per cent, of casein and albumin. Milk is open to the suspicion of being skimmed, when the percentage of fat falls be- low that of the casein and albumin. The percentage of fat removed, based on the legal TESTING MILK FOR ADULTER.\TIONS 1 45 standard, may be calculated by the following rule: Alultiply the per cent, of fat in the milk by loo, di- vide the result by the legal standard for fat and sub- tract this from lOo; or expressed as a formula: The per cent, of fat removed=ioo — ^ ^ ^°° This formula is true only for milks originally containing 3 per cent, of fat and so its results are generally much below the truth. For example, in a milk containing originally 5 per cent, of fat, which has been skimmed to 2.50 per cent., thus removing 50 per cent, of the fat in the milk, the above formula would indicate that only 16.6 per cent, of the fat had been removed. In most cases results nearer the actual truth are given by substituting 3.75 for 3 in the formula. GENERAL METHOD FOR JUDGING WATERED AND SKIMMED MILK Having found in a sample of milk ( i ) the per cent, of fat, (2) the specific gravity of the milk and (3) of the milk-solids, (4) the per cent, of solids, and (5) of solids-not-fat, one may arrive at fairly safe conclu- sions in regard to the watering and skimming by making comparison with the percentages of constitu- ents present in average normal milk. In forming such conclusions, the following facts should be kept in mind: 1. Water has a lower specific gravity than milk. 2. Watering milk decreases (a) the lactometer read- ing, (b) the fat, (c) total solids, and (d) solids-not-fat. 3. Water has a higher specific gravity than milk-fat. 146 MODERN METHODS OF TESTING MILK 4. Skimming milk (a) increases the lactometer read- ing, (b) decreases the fat and total solids, (c) slightly increases the solids-not-fat, and (d) increases the spe- cific gravity of the milk-solids. 5. Skimming and watering decrease all constituents, but lower the fat more in proportion than the solids and solids-not-fat. 6. Skimming and watering may produce the same specific gravity as in normal milk. 7. The amount of fat in milk is more variable than the amount of solids-not-fat. 8. Herd milk which shows a lactometer reading above 33.5, along with a low percentage of fat, and a specific gravity of solids above 1.40, can be regarded as skimmed. 9. Herd milk showing a lactometer reading below 28 may be regarded as watered, especially with low fat, solids and solids-not-fat. Milk is watered when (i) the specific gravity of the milk is low, (2) the percentage of fat and solids- not-fat is low and (3) the specific gravity of the milk- solids is between 1.25 and 1.35. Milk is skimmed when (i) the specific gravity of the milk and of the milk-solids is high; when (2) the per cent, of solids-not-fat is high, and when (3) the per cent, of fat and solids is low. Milk is watered and skimmed when (i) the spe- cific gravity of the milk is normal or otherwise, (2) the specific gravi|:y of the milk-solids is normal or high, and (3) the per cent, of fat and solids-not-fat is low, TESTING MILK FOR ADULTERATIONS 1 47 DETECTION OF FOREIGN SUBSTANCES IN MILK The foreign substances most frequently found in milk are preservatives and coloring matters. The pre- servatives in common use are formalin, boric acid, borax and sodium bicarbonate. The coloring matters generally used are annatto and coal-tar dyes (azo- colors), which are added to milk to make it look rich, and, especially in case of skimmed and watered milk, to cover up the signs of such adulterations. Test for annatto. — To lo cc. of milk in a test-tube add 10 cc. of ether, shake vigorously and let stand un- til the ether separates on top of the milk. If annatto is present, the layer of ether will be yellow, the depth of color depending on the amount of annatto present. Test for Coal-Tar Dyes. — The azo-colors, which are the ones most commonly used in coloring milk, may be detected by adding lo cc. of milk to lo cc. of strong hydrochloric acid and mixing, when a pink coloration appears. Tests for Formalin. — Formalin, which is a 40 per cent, solution of formaldehyde, is commonly di- luted and sold under such names as ''Freezine," "Ice- line," etc., which contain from 2 to 6 per cent, of for- maldehyde. In making the Babcock test in milk, the presence of formalin may be shown when a marked violet layer forms at the junction of the acid and milk just after pouring the acid into the test-bottle. The test may also be performed by taking 10 cc. of milk in a test-tube or Babcock test-bottle, and adding about 5 cc. of sulphuric acid, such as is used in the Babcock test, pouring the acid down the 'side of the tube so that it does not mix with the milk. 148 MODERN METHODS OF TESTING MILK Leach's test, which is more deHcate, is performed as follows: Make a solution of hydrochloric acid (spe- cific gravity 1.2) containing 2 cc. of 10 per cent, ferric chloride per liter. Add 10 cc. of this solution to 10 cc. of milk in a white teacup and heat slowly over a flame to boiling, giving the cup a rotary motion. If formalin is present, a violet coloration appears, vary- ing in depth with the amount present. Test for borax and boric acid. — To 25 cc. of milk add lime water, until the milk is alkaline, evaporate to dryness and burn to an ash in a small porcelain or platinum dish. To the ash add a few drops of dilute hydrochloric acid, not too much ; then add a few drops of water and place in this water solution a strip of turmeric-paper (obtainable at drug-stores). Then dry the paper, when a cherry-red color will appear on the paper if either borax or boric acid is present. This test is made still more certain by moistening the reddened paper with a drop of an alkali solution, when the pa- per turns to a dark-olive color in the presence of borax or boric acid. Test for sodium carbonate. — To 10 cc. of milk add 10 cc. of alcohol and a few drops of a i per cent, so- lution of rosolic acid. Carbonates are present if a rose-red color appears, while pure milk shows a brown- ish-yellow color. ADULTERATIONS OF CREAM The adulterants of cream are the same as those for milk and are detected in the same manner. Gelatine and sucrate of lime are used to some extent to give cream a greater consistency. TESTING MILK FOR ADULTERATIONS I49 ADULTERATIONS OF BUTTER The most common adulteration of butter is substi- tution, in part or in whole, of fat other than butter- fat, such as products from beef-fat and lard. Occa- sionally preservatives are found, such as occur in milk. ''Renovated" or "process" butter is made from refuse butter that can not be disposed of otherwise on the market. Excessive water or casein should be regarded as an adulteration. Harmless coloring matter has been universally allowed. The absolute identification of such adulterants as oleomargarin requires somewhat elaborate chemical methods. Only simple tests can be given here. Foam-test for oleomargarin and "renovated" but- ter. — Melt in an ordinary tablespoon a piece of the suspected butter about the size of a small chestnut, holding it over a small flame, — a candle flame will do. Stir the fat, while melting, with a match or similar stirrer. Then lower the spoon into the flame and let the fat boil vigorously, stirring thoroughly several times during the boiling and not neglecting the outer edges. Oleomargarin and "renovated" butter boil with marked noise, sputtering more or less and pro- ducing little or no foam. Genuine butter generally boils with much less noise and foams up vigorously. Milk-Test for oleomargarin. — In a tin measuring- cup take about one gill of sweet milk or water, heat to about 140° F. and then add a slightly rounded tea- spoonful of the suspected sample. Stir with a small piece of wood, about the size of a match or smaller, until the fat is melted. Then immerse the cup to about one-third of its height in a pan of water in which there 150 MODERN METHODS OF TESTING MILK are several large pieces of ice. Stir the liquid contin- uously, alternating a circular and crosswise motion, until the fat hardens, when it can be easily collected into one lump by means of the wooden stirrer, if it is oleomargarin ; but, if butter, the fat will form little granules and can not be collected in one lump. When milk is used in the test, it should contain as little fat as possible. In this test "renovated" butter behaves like genuine butter. ADULTERATIONS OF CHEESE Only two kinds of adulteration are common in American cheddar cheese: (i) The removal of fat in varying degrees producing so-called skim-cheese, and (2) the use of fat other than milk- fat, producing the so-called filled cheese. Harmless coloring matter is allowed. Cheese containing less than 32 per cent, of fat can be regarded as having been made from milk containing less than its normal amount of fat. The per cent, of fat in filled cheese is generally lower than in cheese made from normal milk. CHAPTER XII The Babcock Test Applied to Farm Conditions The Babcock test finds application on the farm of every dairyman in one or more of the following ways : i) In testing the quality of milk in respect to fat produced by individual cows and by the herd. (2) In testing cream. (3) In testing skim-milk. (4) In testing buttermilk. (5) In testing milk and cream as a means of self- protection. TESTING COWS The most effective test of the value of a dairy cow is the production of milk and of milk-fat. Evidence has been carefully collected showing that many cows in this country are kept at an actual loss. The owners of such cows may be conscious of the fact that they are not prospering, but without having any idea of the cause. The amount of fat in milk required for various purposes differs somewhat. For ordi- nary market purposes, where consumers take as a matter of course any kind of milk delivered to them, the most profitable cow is the one producing a large yield of milk, which generally means a low percentage of fat, frequently just enough to keep above the legal standard. The statement applies to milk sold by bulk 151 152 MODERN METHODS OF TESTING MILK or by weight alone, whether sold for direct consump- tion or taken to a cheese- factory or creamery. But whenever milk is paid for according to its percentage of fat, as in certain forms of market milk, at cream- eries, at condenseries, and at progressive cheese-fac- tories, the cow producing the largest amount of milk- fat will nearly always be found the most profitable. As a rule, a pound of milk-fat can be produced at less cost in rich milk than in poor milk. The only method of ascertaining accurately the value of a cow or of a herd for the production of milk-fat is by testing the milk. The real object of a test is to find the total num- ber of pounds of fat produced in the milk for a defin- ite period of time, the most satisfactory imit being one period of lactation, that is, from the time of calving to the time of becoming dry. In testing the value of a cow for the production of milk- fat, two factors must be considered: (i) The amount of milk produced and (2) the per cent, of fat in the milk. The first amount is obtained by weigh- ing the milk, and the second by testing the milk by the Babcock test. From these data the amount of milk-fat produced is easily found. In applying the Babcock test on the farm to indi- vidual cows, certain details need to be considered, such as (i) the duration of the testing, (2) the fre- quency of testing, (3) the method of sampling, (4) weighing the milk, (5) keeping records, and (6) cal- culating results. In carrying out the work of the milk-test, all necessary details are given in Chapter IV, p. 53- Duration of testing. — For best results, the tests FARM CONDITIONS 1 53 should be made at intervals for a whole period of lac- tation. Frequency of testing. — It is not practicable to test the milk of every milking for fat and it is not neces- sary. On the other hand, the testing of a single milk- ing or of a day's milk or even of a week's milk is in- sufficient, since, for many reasons, the percentage of fat may vary greatly from one time to another. The following plan combines a high degree of accuracy with the least amount of labor : Make the first fat- test in about two weeks after the cow calves and then repeat it regularly once in two weeks during the period of lactation. Even a monthly testing will, however, give fairly accurate results. Method of sampling. — When a single cow's milk is to be tested, the following precautions should be observed in taking the sample : ( I ) The cow must be milked dry at the milking pre- vious to the one to be tested. (2) On the day of milk- ing for the test, the cow is milked as completely as possible each time. (3) After the morning's milk- ing, the milk is well mixed by pouring from one pail to another or by stirring with a dipper, and about a gill is at once dipped out and poured into a pint fruit- jar, which has been thoroughly cleaned and scalded. The sample is kept in a cool place. Repeat the samp- ling with the evening's milk or with each milking, if the cow is milked more than twice a day, adding a sam- ple of each to the jar containing the morning's milk. (4) Make a test before the milk can sour, mixing well before taking samples for the test by pouring back and forth a few times from one vessel to another. If 154 MODERN METHODS OF TESTING MILK it is impossible to make the test promptly, add bi- chromate of potash to preserve the sample, as directed on p. 30. (5) In testing the milk of several cows at the same time, label each sample- jar with the number or name of the cow furnishing the milk. (6) If the milk is to be tested also for solids by the lactometer, take about a half-pint sample from each milking. More strictly accurate results are secured if each milking is sampled by a tube, as stated on p. 27. Weighing milk. — In testing a cow, the milk must always be weighed on the testing day immediately af- ter the milking is completed. As it is so easy to weigh milk, it is desirable to weigh the milk at every milking, or, at least, on two or three days each week. Accurate spring scales of moderate cost are available. Keeping records. — Records of each cow tested should be carefully kept, the following facts being re- corded: (i) Date, (2) name of cow, (3) pounds of milk given, (4) per cent, of fat in milk, (5) lactometer reading, if desired. Calculating results. — The following data can be de- rived by calculation from the facts recorded above : (i) Pounds of fat produced on day of test, (2) pounds of fat and milk produced each month, (3) pounds of fat and milk produced for one period of lactation. The amount of fat on the day of the test is found by multiplying the total number of pounds of milk given by the per cent, of fat found and dividing by 100. For example, if the day's yield of milk is 25 pounds and the per cent, of fat is 4, the day's milk contains i pound of milk- fat. (See p. 193). The amount of milk and fat produced each month FARM CONDITIONS 155 is found as follows, when the test is made once in two weeks : Add the daily yields of milk for the day of the test and for one week before and one week after the test, thus obtaining the milk yield for 15 days. Multiply this sum by the per cent, of fat found on the day of the test and the result is the fat yield for half a month. This added to the next half month gives the yield of fat for the month. The monthly yields of milk and fat, added together at the end of the period of lactation, give the total yields for the period. APPLICATION OF RESULTS OF TESTING INDI- VIDUAL COWS A progressive dairyman will discard from his herd any animal that can not produce, at least, 200 pounds of milk-fat in a year, especially if the milk is sold on the basis of its fat content ; and he will aim, by means of intelligent breeding, feeding and care, to increase the annual yield of milk-fat to 250 or 300 pounds for each cow. TESTING CREAM ON THE FARM There are several conditions under which it is of advantage to test cream on the farm in order to know its fat content. When a dairyman is producing cream to sell directly to consumers, it is important to know its percentage of fat, in order that it may be uniform from day to day, whatever the desired percentage may be. The work of the cream-separator may be controlled ad- 156 MODERN METHODS OF TESTING MILK vantageously only by knowing the percentage of fat in the cream prockiced. In states where a certain per- centage of fat in cream is required by law, it is im- portant for the dairyman to know that his product is up to standard before he sells it. In making butter on the farm, better results can be secured by having the cream of a uniform rich- ness in fat, and the percentage of fat in cream can be accurately known and regulated only by testing. TESTING SKIM-MILK AND BUTTERMILK ON THE FARM The completeness with which fat is removed from milk by different methods of creaming, whether by separator or by gravity processes, can be known accu- rately only by testing the skim-milk for its fat content. With the knowledge furnished by testing, one is in position to prevent further losses when they are known to exist. Similarly, the efficiency of churning may be found by testing the buttermilk for its fat content. TESTING MILK AND CREAM FOR SELF- PROTECTION When dairymen sell milk or cream to milk-dealers, creameries, cheese-factories, shipping-stations, con- denseries, etc., on the basis of the per cent, of fat in milk, it is often a matter of satisfaction to know that the tests which serve as a basis of payment are cor- rect. If a dairyman will take pains to acquire the skill necessary to perform the operations of the Babcock test, he can satisfy himself easily in regard to the ac- FARM CONDITIONS 1 57 curacy of the tests of his milk made by others. In cases where a purchaser reports the test lower than it is, his dishonesty can be detected by means of home testing. It is also important for the dairyman who sells milk directly to consumers to know that his milk is above the legal standard. Much annoyance and expense may sometimes be saved by knowing the percentage of fat and solids in the milk one sells. CHAPTER XIII Methods of Commercial Testing and Scoring of Butter and Cheese In commercial transactions in butter and cheese, certain points or qualities have been adopted as a basis or standard in judging the commercial value of these products. The terms used in expressing the different qualities vary considerably in different market cen- ters, and the same expression is used with different meanings by different persons. Frequently individ- uals use terms that are strictly local or personal. It is desirable that there should be a uniform usage and a common understanding in respect to the terms used in judging dairy products. The attempt is made here to discuss the terms in common use and to define them as well as may be, in the hope that it may serve as a beginning in bringing about a general agreement in respect to the use and understanding of the expres- sions employed in testing and scoring dairy products. The definitions here given can hardly be expected to be in full agreement with the usage of everyone, since individuals differ from one another so much in their use of these terms. SAMPLING AND TESTING BUTTER In obtaining a sample of butter from a package for examination, a butter-trier (Fig. 52) is used. This is inserted its whole length, if possible, into the but- 158 COMMERCIAL TESTING OF BUTTER AND CHEESE 159 ter, turned around once and then drawn out, bringing with it a long, round plug as a samplCo The plug, as soon as drawn, is examined for flavor by smelling and next by tasting. It is then broken across to examine the grain or texture, and then other qualities are examined in turn. TERMS USED IN DESCRIBING QUALITIES OF BUTTER The qualities that have been selected to serve as a basis or standard in the commercial testing and scoring of but- ter are as follows: (i) Flavor, (2) tex- ture, (3) body, (4) moisture, (5) color, (6) salt and (7) appearance. Flavor. — By flavor is meant the qual- ity that is perceptible to the senses of smell and taste. The sense of smell is, as a rule, capable of being developed so as to be more highly sensitive than the sense of taste in detecting variations of flavor. The flavor in normal butter is due to the formation of certain organic compounds in minute quantities during the cream-ripeninpf process. What spe- , BUTTER TRIER cinc compounds these are has not yet been learned. The odor is not that of lactic acid, since that is odorless. Testing Flavor. — The flavor is obtained by placing the plug of butter under the nose as soon as possible after the plug is drawn. A portion of the butter is also tasted. l6o MODERN METHODS OF TESTING MILK Terms describing flavors. — The following terms are selected from the great variety of names that are applied to various flavors found in butter: (i) Per- fect, (2) quick, (3) clean, (4) light, (5) buttermilk, (6) ranci-d, (7) tallowy, (8) cowy, (9) fishy, (10) tainted, (11) stable, (12) weedy, (13) cheesy. (i) Perfect flavor applies to butter which possesses the characteristic aroma and taste of high-grade but- ter in a well-marked degree. It is difficult to de- scribe this flavor adequately, but it is commonly char- acterized as nutty, clean, pleasantly aromatic, delicate and sweet. Perhaps the best description of it is to liken it to the flavor of clean, well-ripened cream. It should be entirely free from rancidity or any unusual flavor, (2) Quick flavor is so delicate and volatile that it disappears quickly; "high" is also applied to the same condition. (3) Clean flavor is free from every trace of unpleas- ant aroma or taste. (4) Light or flat flavor in butter indicates absence of marked flavor, due to lack of cream-ripening, to ex- cessive washing of granules and to other conditions. (5) Buttermilk flavor is somewhat sour in taste and like buttermilk in aroma. It is due to the presence of an excessive amount of buttermilk in the butter. (6) Rancid flavor is that of butyric acid, the pres- ence of which is due to the use of over-aged cream or milk or to age of butter, in which butyric acid fer- mentation has occurred. When the flavor is strong, it produces an impleasant, strangling or choking sen- sation in a sensitive throat. The odor is very pene- trating and lasting. COMMERCIAL TESTING OF BUTTER AND CHEESE l6l (7) Tallozvy flavor is like that of tallow. (8) Cowy flavor refers to the animal odor, particu- larly as noticed in the breath of a cow. It appears to be especially prominent in cows freshly turned into pasture. (9) Fishy odor is rather suggestive of salted cod- fish. It is usually due to a special form of fermentation appearing in the milk and cream. (10) Tainted flavor covers a variety of odors and tastes that are offensive in varying degrees. (11) Stable flavor is the one characteristic of cow manure. (12) Weedy flavor includes such abnormal flavors as may come from onions, leeks, cabbages, turnips, etc. (13) Cheesy flavor suggests the flavor of cheese and is due to fermentation changes in the proteid of but- ter; it is more common in unsalted butter. Texture. — The texture of butter refers to what is called the grain and depends upon the condition of the butter-granules. In its first formation in churning, butter appears in very small, irregular grains or gran- ules. These grains retain their individuality in large measure throughout the Vest of the process of butter- making and even in the finished product. The more distinct the individuality of the granules can be kept in making the butter into a solid mass, the better is the texture. Testing texture. — The granular texture of butter is seen when a plug or chunk of butter is broken into parts transversely, giving somewhat the fractured ap- pearance seen in broken steel and free from a smooth, greasy appearance. Another method of testing tex- l62 MODERN METHODS OF TESTING MILK ture is to pass a knife-blade or butter-trier through the butter ; when it is withdrawn, the trier is clean and free from any greasy appearance, if the texture is good. Terms describing texture. — The terms used to de- scribe texture are (i) perfect, (2) poor grain, and (3) salvy. ( 1 ) Perfect texture in butter is shown by the gran- ular formation, as described above. (2) Poor grain texture in butter is shown by less marked grain and a more or less smooth, greasy ap- pearance on the broken surfaces. (3) Salvy texture applies to butter in which the grain is more or less destroyed and the smooth, greasy appearance of the broken surface is very marked. Defective texture in butter is caused by allowing the butter-granules in the churn to become too large and by working too much or at too high a tempera- ture. The granular texture of butter is completely de- stroyed by warming butter to near its melting point. Body. — By this term is meant the quality of consis- tency, firmness or hardness, as shown by not melting or softening too easily. Testing body. — The body of a sample of butter can be ascertained by pressing a portion of the plug be- tween the thumb and fingers, and also by pressing be- tween the tongue and roof of the mouth. Terms describing body. — The terms used to de- scribe the body of butter are: (i) perfect, firm or solid, (2) hard or tallowy, (3) weak-bodied, (4) sticky. COMMERCIAL TESTING OF BUTTER AND CHEESE 163 (i) Perfect body in butter is shown by firmness or solidity under proper conditions of temperature. When pressed between the fingers or on the tongue it shows a certain amount of resistance. (2) Hard or talloivy body is shown by excessive so- lidity, being characteristic of butter made from cows far along in lactation, or in the case of cows heavily fed on cotton-seed meal. (3) Weak-bodied butter is lacking in firmness, more or less soft, melting more easily on warming than a perfect-bodied butter. Weak-bodied butters are usu- ally salvy in texture and high in moisture. Certain feeds, such as gluten meal, tend to increase the soft- ness of butter. (4) Sticky body in butter is shown by extreme soft- ness amounting to stickiness. Moisture. — The water in butter should be so thor- oughly incorporated with the fat that it does not appear in the form of free beads of water visible to the eye. Water should not run ofif the trier when a sample is drawn. The water should also be clear and trans- parent. Testing moisture. — The sample of butter is exam- ined for the appearance of moisture or brine in respect to the completeness of its incorporation and its clear- ness. Terms describing moisture. — The following terms are used to describe the condition of moisture in but- ter: (i) Perfect, (2) excessive, (3) milky or turbid. (i) Perfect moisture in butter is shown by the ab- sence of any visible moisture in the form of drops. (2) Excessive moisture is shown by the presence of 164 MODERN METHODS OF TESTING MILK water easily apparent to the eye. Butter may some- times contain so much water as to be called "mushy." (3) Milky or turbid moisture or brine appears more or less milky, being due to the presence of too much buttermilk. Relation of texture, body and moisture. — Considera- ble confusion prevails in the use of the terms texture, body and moisture. Some use the term texture to in- clude also body and moisture; others use the term body to include texture, while others use the expres- sion "body and grain" to cover all three qualities. Tex- ture and body and moisture may be influenced by the same conditions and may be, to some extent, interde- pendent, but in reality they are distinct properties and, if they were treated as such, needless confusion would be avoided. Color. — The color of butter varies in different mar- kets according to requirements, but most of the but- ter made in the United States has, as its standard, an even, bright, straw-yellow. Most butter in commerce is colored artificially, so as to maintain a uniform ap- pearance at all seasons of the year. Somewhat dif- ferent shades of color are demanded by different mar- kets. Testing color. — The quality of color is tested simply by inspection with the eye. The thumb-nail is run along the surface of the plug near the edge of the trier, and the fresh surface thus made is examined. The examiner carries in his mind the shade of what he regards as an ideal color and judges the sample under examination by its comparison with his ideal. It would lead to easier methods of comparison and COMMERCIAL TESTING OF BUTTER AND CHEESE 165 more uniform results if there could be agreed upon a certain shade of color which should serve as a na- tional standard as far as possible. Such a color stand- ard could be furnished butter-makers and examiners of butter. Along with such a standard color, there could be prepared a scale of shades which could serve as a basis for scoring color. Terms describing color. — The terms used in de- scribing the color of butter are: (i) perfect, (2) light, (3) high, (4) reddish, (5) mottled, and (6) white- specked. (i) Perfect color in butter is a straw-yellow, bright, and uniform throughout the mass. A plug of butter held between the light and the eye should be evenly translucent and not opaque or cloudy. (2) Light color is shown by insufficient color, the yellow being too pale. (3) High color is deeper yellow than called for by perfect color. (4) Reddish color is self-explanatory and is due to excessive use of coloring material. (5) Mottled color in butter is shown by the appear- ance of light-colored portions, which may be in spots or streaks or waves. The term wavy is often used to indicate a variation of color that is just perceptible. They are not seen as readily on a sample plug drawn by a trier as they can be by cutting a lump of butter across so as to show a smooth, broad surface. Slight mottling is apt to escape observation when the exam- ination is made only of a plug. Mottling is due to the action of salt upon buttermilk retained in the but- ter. The light portions owe their color to the pres- l66 MODERN METHODS OF TESTING MILK ence of the casein lactate of buttermilk. Removal of buttermilk from the butter-granules prevents mottling. (Bulletin No. 263, N. Y. Agr. Exp. Sta. 1905). (6) White-specked color in butter appears in white specks of varying size, but usually small. They are due to particles of coagulated casein lactate produced in cream by over-ripening, and also to dried cream particles, caused by lack of stirring during the process of ripening. Salt. — The amount of salt in butter varies with dif- ferent markets ; but, whatever the amount used, it should be completely dissolved and evenly distributed through the mass of butter. Testing butter for salt. — The quality of butter as affected by salt is examined by tasting, sight and feel- ing. Undissolved particles of salt, when they can not be felt on the tongue or seen, can be detected by rub- bing some of the butter between the fingers. Terms describing salt. — The terms used in de- scribing the quality of butter in relation to salt are the following: (i) Perfect, (2) too salty, (3) flat, (4) gritty, (5) uneven. (i) Perfect quality in respect to salt in butter is shown as follows : The salt must be in the proportion demanded by the market ; it must be entirely dissolved and evenly distributed. (2) Too salty butter contains more salt than the market demands. (3) Flat butter is lacking in salt for the market re- quirements. (4) Gritty butter contains undissolved salt. COMMERCIAL TESTING OF BUTTER AND CHEESE 167 (5) Uneven salt in butter is lack of uniformity, some portions of butter being more salty than others. Appearance. — Under this head are included the manner of packing, the attractive appearance of the package, cleanliness, etc. Testing appearance. — When the cover of the pack- age is removed for sampling the butter, the appearance of the surface of the butter is noticed. The outside of the package is also examined. The two general quali- ties that must be kept in mind in this connection are cleanliness and neatness. Terms describing appearance. — The quality of ap- pearance of butter may be considered under two heads, (i) finish and (2) package. (i) Finish in appearance, in connection with exam- ining butter, refers to the manner of packing. The finish is perfect when the package is lined with paraf- fin or with a good quality of parchment paper, neatly placed, and the package well filled, the surface being even and bright. The package should be just evenly full. The top should be neatly covered with cheese- cloth saturated with brine. (2) Package. — The package is regarded as perfect when of good material, well-made, clean, and neat in appearance. In the same lot of butter the packages should all be alike in size and shape. SCORING BUTTER The different qualities indicated above are used in a specific manner for judging and fixing the com- mercial value of butter. l68 MODERN METHODS OF TESTING MILK Scale of points. — To each quality is assigned a defi- nite numerical value and these numbers are called a scale of points. The following scale of points is in common use in many markets of this country, the num- bers indicating perfect quality in each case, and . the totals aggregating loo: Flavor, 45 points. Color, 15 points. Texture, (10) ) Salt, 10 points. Body, (10) -25 points. Appearance, 5 points. Moisture, (5) ) Total, 100 points. Method of scoring. — In scoring a sample of but- ter, an examination is made with reference to each of the qualities mentioned. In those qualities in which it is perfect, it is given the values or points assigned above. If the butter is defective in any quality, that is, short of perfect, then a smaller value is given than the one indicated above in the scale of points ; the more de- fective the butter is in any quality, the lower is the value or number of points given it. When all the qualities have been scored, the numbers of points as- signed to them are added and the total is the score of the butter under examination. It can readily be seen that judgment, trained by ex- perience, is required to assign to each quality its proper number of points. The sense of smell and of taste must be highly developed by training in the field of experience. The eye and touch must also be trained by special experience. Score-cards. — For convenience, score-cards are used in keeping records of the results of scoring where many samples are examined. The following form illustrates a commercial score-card: COMMERCIAL TESTING OF BUTTER AND CHEESE I69 NAME OR NUMBER IDENTIFYING SAMPLE, DATE JUDGE QUALITY SCORE-POINTS Sample 1 Sample Sample 3 Sample 4 Flavor 45 10 10 5 15 10 5 45 8 10 5 13 10 4 40 10 8 3 14 10 5 36 8 8 4 13 10 4 32 Texture 7 Bodv 7 Moisture 4 Color .0...0 Salt 12 8 Appearance 5 95 90 83 75 These scores, under the system of grading described below, would be graded as follows: Sample i, "ex- tras ;" sample 2, "firsts ;" sample 3, "seconds ;" and sample 4, "thirds." In commerical scoring, reasons for the number of points given are not stated ; but in dairy schools and competitive public exhibitions, where educational pur- poses are in view, the reason for each score should be given. The following form of score-card for such purposes is a suggestion, which may be modified to suit any special conditions : Butter-Scoring — Numerical and Descriptive Card Date Jiidge Name or number identifying butter I/O MODERN METHODS OF TESTING MILK NUMERICAL SCORE. Perfection — Flavor, Texture, Body, Moisture, Color, Salt, Appearance (45) (10) (10) (5) (15) (10) (5) Score given — — — — — — — DESCRIPTIVE SCORE (check defects below). Flavor Texture Body Moisture Color Salt Appear- ance Perfect Quick Perfect Perfect Perfect Perfect Perfect Finish Clean I^ight Buttermilk Rancid Tallowy Cowy Poor- grain Salvy Firm Hard Weak- bodied Exces- sive Milky I,ight High Reddish Too salty Flat Gritty Package Fishy Tainted Sticky Mottled Uneven Stable Weedy Cheesy Wavy Specks CLASSES AND GRADES OF BUTTER The following system for classifying and grading butter is taken from the regulations of the New York Mercantile Exchange: Classification : — 1. Creamery Butter includes butter made in a cream- ery from cream obtained by the separator system, or from gathered cream. 2. Imitation Creamery Butter includes butter churned by the dairyman, collected in its unsalted, un- worked condition, and worked, salted and packed by the dealer or shipper. 3. Dairy Butter includes such as is made, salted and packed by the dairyman and offered in its orig- inal package. 4. Factory Butter is butter collected in rolls, lumps, COMMERCIAL TESTING OF BUTTER AND CHEESE 171 or in whole packages, and reworked by the dealer or shipper. 5. Renovated Butter is that made by taking pure butter and melting the same and rechurning with fresh milk, cream or skim-milk, or other equivalent pro- cess. 6. Grease consists of all grades of butter below Fourths free from adulteration. 7. Known Marks is a term used to include such butter as is known to the trade under some particular mark or designation and must grade as Extras, if creamery, and as Firsts, if reworked butter, in the season in which it is offered, unless otherwise speci- fied. Grades: — Grades of butter must conform to all the following requirements and are not determined by score alone. 1. Extras must be of the highest grades of butter made in the season when offered under the different classifications ; 90 per cent, shall be up to the following standard and the balance must not grade below Firsts : (i)Flavor must be fine, sweet, clean and fresh, if of current make ; and fine, sweet and clean, if held. (2) Body must be firm, smooth and uniform. (3) Color should be a light straw shade, even and uniform. (4) Salt should be medium. (5) Package should be good, uniform and clean. (6) Score must average 93 points or higher. 2. Firsts is a grade just below Extras and must be fine butter for the season when made and offered, under the different classifications, and up to the fol- lowing standard: 172 MODERN METHODS OF TESTING MILK (i) Flavor must be good, sweet, clean and fresh, if of current make ; and good, sweet and clean, if held. (2) Body must be good and uniform. (3) Color must be reasonably uniform, neither too high nor too light. (4) Salt should be medium. (5) Packages should be good and uniform. (6) Score must average 87 points or higher. 3. Seconds is a grade just below Firsts and must be good for the season when offered under the differ- ent classifications, and up to the following standard : (i) Flavor must be reasonably good and sweet. (2) Body, if creamery or dairy, must be solid-boring. If factory or renovated, must be 90 per cent, solid-bor- ii^?- (3) Color must be fairly uniform. (4) Salt may be high, medium or light. (5) Package should be good and uniform. (6) Score must average 80 points or higher. 4. Thirds is a grade just below Seconds. ( I ) Flavor must be reasonably good ; may be strong on top and sides. (2) Body should be fair-boring, if creamery or dairy, and at least 50 per cent, boring a full trier, if factory or renovated. (3) Color may be irregular. (4) Salt may be high, light or irregular. (5) Packages should be fairly uniform. (6) Score must average 75 points or higher. 5. Fourths is a grade just below Thirds and may consist of promiscuous lots. (i) Flavor may be off and strong on tops and sides. (2) Body is not required to draw a full trier. (3) Color may be irregular. (4) Salt may be high, light, or irregular. (5) Package may be of any kind men- tioned at time of sale. COMMERCIAL TESTING OF BUTTER AND CHEESE I73 SAMPLING AND TESTING CHEESE Only the ordinary American cheese, usually made by the cheddar system, is here considered. For com- mercial testing, cheese is sampled by a cheese-trier in much the same manner as butter. The plug should always be drawn from the top and not from the side in order to avoid injuring the protective power of the bandage. The plug drawn is examined by smelling, feeling, appearance, etc., in reference to the various qualities mentioned below. TERMS USED IN DESCRIBING QUALITIES OF CHEESE The following qualities have been selected to serve as a basis in the commercial testing and scoring of cheese: (i) Flavor, (2) texture, (3) body, (4) color, (5) salt, and (6) appearance. Flavor. — By flavor is meant the quality that is per- ceptible to the smell and taste. The sense of smell is depended upon in testing flavor in cheese much more largely than is the sense of taste, because, in ex- amining a large number of samples of cheese in suc- cession, constant tasting soon dulls not only the sense of taste but also that of smell. Flavor in cheese is due to the formation of some unknown compound or compounds during the ripening process. Testing flavor in cheese. — The flavor is best ob- tained by direct smelling of the plug as soon as it is drawn and, in addition, by crushing and warming some of the cheese in the hand and then smelling. Terms used in describing cheese flavors. — From a great variety of names applied to various flavors 174 MODERN METHODS OF TESTING MILK found in cheese, the following terms are selected for consideration: (i) Perfect, (2) high or quick, (3) clean, (4) low or flat, (5) strong, (6) too much acid, (7) too little acid, (8) sour, (9) sweet or fruity, (10) rancid, (11) tallowy, (12) tainted, (13) stable, (14) weedy, (15) bitter, (16) cowy. (i) Perfect flavor applies to cheese when it some- what resembles that of first-class butter with an added quality of its own that is characteristic but cannot be described further than to call it cheese-like. It is sometimes described as "nutty." This flavor should be marked, but not strong. It should be free from all other flavors, particularly the more or less offen- sive products of undesirable fermentations. The taste should be mild and somewhat lasting, but should not be so sharp as to "bite" the tongue. (2) High or quick flavor is a delicate flavor that dis- appears quickly. (3) Clean flavor is free from every trace of unpleas- ant aroma or taste. (4) Loiv or flat flavor applies to slight traces, or absence, of flavor; it is insipid. (5) Strong flavor is a good flavor very pronounced but free from everything offensive ; it is a good flavor strongly developed. (6) Too much acid applies to flavor that smells somewhat sour but does not taste sour. (7) Too little acid applies to a mild flavor, lacking in character. (8) Sour flavor is characterized by a sour taste when the cheese is fresh, owing to the presence of too much whey. COMMERCIAL TESTING OF BUTTER AND CHEESE 1/5 (9) Sweet or fruity flavor is suggestive of artificial pineapple odor and is somewhat "sickish." (10) Rancid flavor is that of butyric acid, more com- mon in old cheese than in young. When very strong, it affects a delicate throat with a slight sensation of choking or strangling. (11) Tallozvy flavor is lik^ that of tallow. (12) Tainted flavor includes a variety of odors, mildly to strongly offensive. (13) Stable flavor suggests the smell of cow- ma- nure. (14) Weedy flavor applies to such abnormal flavors as come from onions, leeks, cabbages, ragweed, etc. (15) Bitter flavor is self-descriptive. It is often due to certain fermentations that develop when a cheese is undersalted. (16) Cozvy flavor is suggestive of the breath of a cow and may develop in cheese from some form of fermentation. Texture. — Texture, as applied to cheese, refers chiefly to compactness or appearance of solidity, and has a meaning quite dift'erent from what it has when used with, reference to butter. It is quite common to regard the ''body" as a part of the texture, but the two qualities are clearly distinct. Testing texture in cheese. — The texture of cheese is tested by an examination of the plug with reference to the presence of holes. The plug is broken in two and the broken ends examined for the characteristic flinty appearance. Terms describing texture. — The following terms are among those most commonly used in describing 176 MODERN METHODS OF TESTING MILK texture: (i) Perfect, (2) close, (3) loose, (4) me- chanical holes, (5) gas or pin-holes, (6) Swiss holes, (i) Perfect texture in cheese is shown when a plug or a cut surface of the inside of the cheese presents to the eye a solid, compact, continuous appearance, free from breaks, holes and chunks. When a plug is broken in two, it should show a flaky appearance, termed a ''flinty" break, resembling the surface of broken flint or steel. (2) Close texture describes the appearance of a cut surface of cheese when free from all kinds of holes. (3) Loose or porous texture is indicated by lack of solid compactness, being more or less full of holes, which vary from a few to enough to make a spongy appearance. (4) Mechanical holes in cheese are irregular, open spaces, caused by the incomplete cementing of the pieces of curd in the press. (5) Gas-holes or pin-holes are small holes, produced by gaseous products of fermentation. (6) Szviss holes are fairly large, round holes, such as are present in Emmenthaler cheese. Body. — This term, used in connection with cheese, refers to the consistency, firmness or substance of cheese. It is largely influenced by the amount of fat and moisture in cheese. Testing body. — This quality is found by pressing a piece of cheese between the thumb and fingers. Terms describing body. — The following terms are among those used in describing the body of cheese: (i) Perfect, (2) solid or firm, (3) smooth, (4) silky, (5) waxy, (6) pasty or salvy, (7) stiflf, corky, or COMMERCIAL TESTING OF BUTTER AND CHEESE I77 curdy, (8) weak-bodied, (9) mealy, (10) gritty, (11) watery, (12) over-dry. ( 1 ) Perfect body in cheese is indicated when it feels solid, firm and smooth in its consistency or substance. It does not crumble under pressure. A plug drawn from a cheese of perfect body should be smooth in ap- pearance and not "fuzzy." (2) Solid or Urm body is indicated when cheese of- fers a certain amount of resistance under pressure, somewhat like that shown by a piece of fat pork or cold butter. The term meaty is also used. (3) Smooth-hodx^d cheese, when pressed between the thumb and fingers, feels smooth and velvet-like, as distinct from harsh, gritty or mealy. (4) vS'i7^3;-bodied cheese is smooth in feeling but not over-solid in consistency. (5) ^a,rv-bodied cheese is much the same as silky but possessing more firmness or solidity. (6) Pasty or salvy cheese is very soft, usually from an excess of moisture. When pressed, it sticks to the fingers. (7) Stiff, corky or curdy cheese is hard, tough, over- firm; it does not crush down readily when pressed in the hand. (8) PF^a^-bodied cheese is very soft, lacking in firmness but not necessarily sticky like pasty cheese. (9) Mealy cheese breaks down in fine crumbs when pressed. (10) GnV^y-bodied cheese feels harsh and gritty under pressure. (11) Watery-hodi^d cheese is excessively soft, pasty and sticky. 178 MODERN METHODS OF TESTING MILK (12) In an over-dry cheese the body is very hard or mealy. Color. — The color of cheese varies considerably, whether artificially colored or not. There appears to be an increasing demand for uncolored cheese. The coloring varies from a pale yellow to a reddish yel- low, according to the demands of special markets. Testing color. — The color is tested by inspection with the eye, the examiner noticing particularly unev- enness and any extreme condition of color. Terms describing color. — Color in cheese is de- scribed by the following terms: (i) Perfect, (2) straight, (3) translucent, (4) white specks, (5) streak- ed, (6) wavy, (7) mottled, (8) acid-cut, (9) high, (10) light, (11) uncolored. (i) Perfect color in cheese is indicated by evenness of color throughout the mass. A plug held between the eye and light should appear somewhat translucent. (2) Straight color is an even, uniform color through the whole cheese. (3) Translucent applies to color in cheese which appears slightly translucent when the plug is held be- tween the eye and the light. (4) White specks is a term that describes itself. Such specks in cheese are a defect. They may appear in cheese cured at low temperature. (5) Streaked color indicates that there are light- colored portions in the form of streaks. {6)Wavy color applies to lighter portions appear- ing in the form of waves. (7) Mottled color shows in cheese in lighter-colored spots of fairly large size^ more or less irregular. COMMERCIAL TESTING OF BUTTER AND CHEESE I79 (8) Acid-cut color is shown in cheese when consid- erable portions of the cheese have been made lighter in color by the presence of too much acid (whey). (9) High color is indicated by a reddish color, caused by using too much coloring matter. How- ever, the question of color is a relative one, because the demand in different markets varies from uncolored to extremely high color. ( 10) Light color is the term usually used in describ- ing cheese that has been made uniformly dead white by the action of too much acid (whey). (11) Uncolored cheddar cheese is not white but of a light amber shade. Salt. — The amount of salt in cheese varies somewhat with different markets. There is seldom experienced difficulty of uneven salting in cheese, because the salt slowly permeates the cheese in the ripening process. Little variations usually occur in different parts of the same cheese, but are so slight as to be incapable of being noticed by ordinary methods of examination. Testing cheese for salt. — The quality of cheese as influenced by the salt is found simply by tasting. Terms used in describing salt. — In describing the relation of salt to cheese, the following terms are used : (i) Perfect, (2) too much, (3) too little. (i) Perfect applies to salt in cheese when just enough has been used to impart a sufficient taste of salt. (2) Too much salt is indicated by salty taste. Too much salt in cheese causes a dry, mealy texture, over- firm body and imperfect flavor. (3) Too little salt is shown by insipidity of taste. l8o MODERN METHODS OF TESTING MILK It is usually accompanied by bitter flavor and porous texture. Appearance. — This term refers to the general ap- pearance of the cheese to the eye in respect to uni- formity, neatness and cleanliness. It may also include the boxing. One system, as in the case of butter, de- scribes under ''finish" the appearance of the cheese, and under "packages" the boxing; and we will follow this method here. Testing appearance. — When the cover of the box is removed for sampling, in the case of boxed cheese, the appearance of the cheese is noticed and the box itself is examined. Cleanliness and neatness are the points to observe in judging appearance. Terms describing appearance. — The general terms used in describing appearance are (i) finish and (2) package. (i) Finish in appearance, in order to be perfect, must meet the following requirements : The rind must be smooth, even in color, free from cracks and fairly hard. The bandage must be without wrinkles and must be neatly rounded over the edges about an inch and a half on each end of the cheese. The sides of the cheese should be straight and of uniform height all around. The faults of appearance in finish are as follows, the terms being self-descriptive: (i) Cracks, (2) light spots, (3) roughness in rind, (4) uneven edges, (5) wrinkles in bandage, (6) lack of uniformity in ends and in height, (7) bulging out at sides or ends. (2) Package. — The packages or boxes are regarded as perfect when of good material, well made, strong. COMMERCIAL TESTING OF BUTTER AND CHEESE l8l clean, close-fitting, uniform in size and in undamaged condition. SCORING CHEESE The qualities described in the preceding pages are used for judging and fixing the commercial value of cheese. Scale of points. — The following scale of points is in use in many places, the numbers indicating perfect quality in each case and the totals aggregating too: Flavor, 50 Body, 15 Salt, 5 Texture, 15 Color, 10 Appearance, 5 In the practice of many markets, salt is omitted and appearance is given 10 points. Method of scoring. — The general procedure is es- sentially the same as that already described in connec- tion with butter (p. 160). Method of grading cheese. — The same general prin- ciples apply as in grading butter (p. 171). One class- ification is into (i) "fancy," (2) ''firsts," and (3) ''seconds." In the Canadian market, there are first, second and third grades. Score-cards for cheese can be prepared in a manner similar to those previously suggested for butter (p. 170) . As in the case of butter, the testing, scoring and grading of cheese demand good judgment trained by experience. The sense of smell and touch must be well developed. No formal classification or grading of cheese is made by the New York Mercantile Exchange, as is done in the case of butter. CHAPTER XIV Methods of Commercial Testing and Scoring^ of Milk and Cream In the past few years, there has been a growing sentiment that some method of testing the commer- cial value of milk and cream, similar to that used in judging butter and cheese, is much needed. Some helpful work has been done in this direction by the Dairy Division of the U. S. Department of Agricul- ture. The writer believes an efficient system will be developed in. the near future, and, as a basis for such a system, the present discussion is offered. For convenience of treatment, we will classify milk and cream according to the purposes for which they are used, as follows: (i) Market milk and cream, when sold for direct consumption as such. (2) Cream- ery milk and cream, when used for butter-making. (3) Cheese-factory milk, when used for cheese-ma- king. (4) Certified and standardized milk, when pro- duced under specific sanitary conditions and reaching a guaranteed standard of composition. TERMS USED IN DESCRIBING QUALITIES OF MARKET MILK The qualities that are selected as a basis in the com- mercial testing and scoring of milk are the following : (i) Composition, (2) keepmg-power, (3) flavor, and, we may add, though of minor importance, (4) color, 182 I COMMERCIAL TESTING OF MILK AND CREAM 1 83 and (5) appearance. For all practical purposes color and package can be omitted. In educational competi- tions they may be included. Composition. — By composition is meant the per cent, of fat and of solids-not-fat in milk. The com- position of milk, other things being equal, is the direct and only practicable measure of value as food, and this must constitute an important factor in judging the value of market milk. Testing composition. — The percentages of fat and of solids-not-fat are found in the manner described on p. 53 and p. 137. Terms describing composition. — Only two terms need to be used in describing the composition of milk : (i) perfect and (2) defective. ( 1 ) Perfect, as applied to testing and scoring milk, means milk containing not less than 4 per cent, of fat and not less than 9 per cent, of solids-not-fat. (2) Defective applies to milk containing less fat or solids-not-fat than required for milk of ''perfect" composition. The figures selected to indicate milk "perfect" in composition represent as nearly as possible normal milk of average composition. Injustice would obvi- ously be done by selecting 4.5 or 5 per cent, as the amount of fat to represent milk perfect in compo- sition. It would, on the other hand, be very unsatis- factory to take a so-called legal standard as repre- senting perfect composition, because it is not a stand- ard at all but simply a method of prescribing the low- est permissible amounts of fat and solids that will be legally tolerated in milk. 184 MODERN METHODS OF TESTING MILK Keeping-power is an expression used to indicate in a general way the length of time milk remains sweet and palatable for direct consumption. In esti- mating the commercial value of milk, this is an im- portant factor; since milk that is sour or otherwise unpalatable, or milk containing the products of any form of undesirable fermentation, is comparatively valueless for direct consumption, however rich it may be in fat and other solids. Keeping-power is a matter of much importance in warm weather, especially in the case of milk that is transported long distances before reaching the consumer. Testing keeping-power. — The keeping-power of milk is tested by making determinations of (i) the acidity (p. 96), (2) the dirt in suspension (p. 117), (3) the fermentation test (p. 114), and (4) the num- ber of bacteria (when practicable). Terms describing keeping-power. — The terms used in describing the keeping-power of milk are (i) per- fect, (2) acidity, (3) dirt in suspension, (4) unde- sirable fermentations, and (5) number of bacteria per cubic centimeter. (i) Perfect. — Milk is called perfect in keeping- power (a) when its acidity is not above 0.18 per cent., (b) when it contains no dirt in suspension, (c) when the fermentation test reveals nothing abnormal, and (d) when the number of bacteria does not exceed 100,000 per cubic centimeter. (2) Acidity is used to mean the amount of appar- ent total acid calculated as lactic, as shown by the amount of alkali neutralized. For discussion of the relation of acidity to the temperature and cleanliness of milk, see pp. 11 3- 114. COMMERCIAL TESTING OF MILK AND CREAM 185 (3) Dirt in suspension. The amount of dirt sus- pended in milk may usually be regarded as a rough measure of the germ content or cleanliness of milk, since the visible dirt we find in milk is generally the same in source and kind. (4) Undesirable fermentations refer to the results of the fermentation test. They may reveal themselves in causing porous, spongy curd and in producing offensive odors. (5) Number of bacteria per cubic centimeter is a self-descriptive term. Flavor, applied to milk, is used to mean the odor and taste. The abnormal odors and tastes noticeable in market milk, otherwise good, come from three sources: (i) From certain things eaten by the cow, as leeks, onions, rag- weed, cabbage, etc. (2) From the direct absorption of strong-smelling substances present in the air surrounding the milk, such as ma- nure, tobacco smoke, ensilage, etc. (3) From stable filth dropping bodily into the milk. Testing flavor. — This is done by tasting and smell- ing the milk. The presence of abnormal odors can be more readily perceived by heating the milk for a few minutes to 100° F. in a closed bottle or jar and then smelling at once on opening the vessel. Terms describing flavor. — The following terms may be used in describing the flavor of market milk : (i) Perfect, (2) stable or cow manure, (3) cowy or animal, (4) weedy, caused by leeks, rag-weed, etc., (5) vegetable, such as cabbage, turnip, ensilage, etc., (6) bitter, (7) fishy, (8) sour, (9) tainted. (i) Perfect flavor in market milk is indicated by t86 modern methods of testing milk freedom from all traces of abnormal odor and taste. There should be no marked odor and no trace of any offensive smell. The taste should be palatable, slightly saline and rich, without any unpalatable after-taste. It should not be flat or insipid. The other terms are mostly self-descriptive. Their number could be extended to cover more minute de- tails. The term fainted is used to cover miscellaneous offensive flavors not included under other terms. Color in relation to the testing and scoring of mar- ket milk explains itself. It is in itself of little prac- tical importance and may usually be omitted except in case of educational competitions. Testing color. — The milk is examined for color by direct inspection in a clear light. Terms describing color. — In describing the color of market milk, the following terms may be used : (i) Perfect, (2) white, (3) bluish, (4) high color, (5) reddish. (i) Perfect as applied to color in milk indicates a yellowish color, not too pronounced, strikingly differ- ent from the white or bluish color of skim-milk, but not as deep as the color of cream. The other terms explain themselves. High color may be caused by artificial coloring, usually producing a reddish tint. Appearance. — This term refers to the appearance of the can, bottle or other vessel containing the milk and applies to these in respect to uniformity, neatness and cleanliness. This is of relatively small practical importance and may usually be omitted in commercial work. It finds a use in educational competitions. I COMMERCIAL TESTING OF MILK AND CREAM 187 SCORING MARKET MILK The qualities described above are intended for use in the commercial judg-ing and scoring of market milk. For practical purposes it will be sufficient ordi- narily to make use of the first three qualities men- tioned, viz: (i) composition, (2) keeping-power, and (3) flavor. Scale of points. — In fixing a scale of points, we have a score of 100 to distribute among the three qualities last mentioned. How many points of the 100 shall be assigned to each? Here is a chance for wide varia- tion of opinion. In actual experience, flavor does not hold the same important relation to market milk that it does to cheese and butter. A flavor that is imper- ceptible in the ordinary consumption of milk usually becomes concentrated in the process of butter-making or cheese-making and seriously affects the quality of the final product. A flavor must be very bad to render milk useless for cooking or for purposes other than direct drinking. Hence, flavor should not receive so high a score in the case of market milk as in the case of butter or cheese. On the other hand, composition should receive a relatively high score, since, other things being equal, it alone governs the relative values of different milks. The importance of keeping-power has already been considered. For the various reasons given, the following scale of points is suggested as a desirable one for practical use : Composition 40 Keeping-power 30 Flavor 30 l88 MODERN METHODS OF TESTING MILK The numbers indicate perfect quality in each case and the totals aggregate loo. The numbers assigned could be varied considerably and still give equally satisfactory results in practise. Method of scoring. — The milk is examined in the manner previously described and defects are indicated by making deductions from the perfect score in the following manner : (i) Composition. — The perfect score of 40 points is reduced one point for each o.i per cent, below 4 per cent, of fat and one point for each o.i per cent, below 9 per cent, of solids-not-fat. (2) Keeping-pozver. — The perfect score of 30 is to be reduced (a) one point for each o.oi per cent, of acidity above 0.18; (b) a certain number of points, according to the judgment of the examiner, for dirt in suspension; (c) a certain number of points for any abnormal results shown by the fermentation test ; and (d) one point for each 100,000 bacteria above 100,000 in I cubic centimeter of milk, when this determination is made. (3) Flavor. — The perfect score of 30 is reduced by the presence of abnormal odors or tastes, the examiner using his judgment as to the amount of reduction. JUDGING MILK FOR MANUFACTURE OF BUTTER AND CHEESE For milk that is to be used for butter-making or cheese-making, a somewhat different method is sug- gested for judging and scoring. Since such milk is or should be paid for on the basis of the fat, the sys- COMMERCIAL TESTING OF MILK AND CREAM 189 tern of judging and scoring should be such as to affect the vakie or the amount of the fat for which payment is received. Such a method is given in the following statements : (i) Score the milk directly for two qualities only, (a) keeping-power and (b) flavor. For perfection, allow 50 points for keeping-power and 50 points for flavor. (2) From the total amount (pounds) of fat con- tained in the milk furnished by each patron, deduct, according to the results of scoring, a certain amount of fat, which is found by multiplying the total amount of fat in the milk by 0.25 per cent, and this result by the number of points scored below 100. (3) Use the number of pounds of fat thus found as the actual amount of fat on which to base dividends. To illustrate, a patron furnishes 1,000 pounds of milk, containing 4 per cent, of fat, which scores 90 in keep- ing-power and flavor. What deduction shall be made from the fat for the score of 90? The amount (1,000 pounds) of milk furnished contains 40 pounds of fat. This number (40) multiplied by 0.25 per cent. (Rule I) P- 193) equals o.i pound of fat, which multiplied by 10, the number of points scored below 100, equals i pound. The total amount (40 pounds) of fat, dimin- ished by the subtraction of i, leaves 39 pounds, which amount would be used as a basis in making this pa- tron's dividend. The judging of flavor and keeping-power is carried out as previously described for market milk. In cheese-making, it is important to make full use of the fermentation test (p. 114). 190 MODERN METHODS OF TESTING MILK COMMERCIAL TESTING AND SCORING OF CER- TIFIED AND STANDARDIZED MILK Certified milk usually guarantees (i) the per cent, of fat, (2) the per cent, of total solids or solids-not- fat, and (3) bacteria below a specified number. Stand- ardized milk usually guarantees only the per cent, of fat. The examination and scoring of certified or of stand- ardized milk are conducted in the same manner as in the case of market milk, except that the scoring is based upon the guarantees, so far as these are given. The guaranteed per cent, of fat and of solids-not-fat and the number of bacteria are to be taken as repre- senting the perfect score in place of the figures given above for market milk, and deductions from the per- fect score for defects are made on the basis of the guarantees. For example, if a certified milk is guar- anteed to contain 5 per cent, of fat, then, in order to score 40 in composition, the milk must contain 5 per cent, of fat and, in case of any shortage, a proportion- ate deduction should be made from the perfect score of 40. COMMERCIAL TESTING AND SCORING OF CREAM Market cream is that which is sold for direct con- sumption. The method of examining and scoring market cream is essentially the same as in the case of market milk. In composition, cream is not exam- ined or scored for solids but for fat only. Another quality might be added in case of cream, that of body, COMMERCIAL TESTING OF MILK AND CREAM I9I by which is meant the consistency of the cream, espe- cially with reference to the presence or absence of lumps, complete evenness of consistency being required for perfection of body. In composition, not less than 20 per cent, of fat should be used as a basis for perfection and from the perfect score of 40 is deducted one point for each 0.5 per cent, of fat below 20. Acidity and flavor are of special importance in market cream. The acidity should not be allowed to be above 0.2 per cent, for perfect cream. Creamery cream is that which is intended for but- ter-making. The same general procedure is employed as in the case of milk that is to be used for butter- making, which, adapted to cream, is as follows : (i) Score the cream directly for two qualities only, (a) keeping-power and (b) flavor. For perfection allow 50 points for each quality. (2) From the total amount (pounds) of fat con- tained in the cream furnished by each patron deduct, according to the results of scoring, a certain amount of fat, which is found by multiplying the total amount of fat in the cream by 0.25 per cent, and this result by the number of points scored below 100. (3) Use the number of pounds of fat thus found as a basis for making dividends. To illustrate, a patron furnishes 500 pounds of cream, containing 20 per cent, of fat, which scores 90 in keeping-power and flavor. What deduction shall be made from the fat for the score of 90 ? The amount (500 pounds) of cream furnished contains 100 pounds of fat. This number (100) multiplied by 0.25 per 192 MODERN METHODS OF TESTING MILK cent. (Rule i, p. 193) equals 0.25 pound of fat, which multiplied by 10, the number of points scored below 100, equals 2.5 pounds. The total amount (100 pounds) of fat, diminished by the subtraction of 2.5, leaves 97.5 pounds of fat, which amount would be used as a basis in making this patron's dividend. I I 1 i CHAPTER XV Arithmetic of Milk and Milk Products In connection with the testing of milk and milk products, especially in some of the practical applica- tions, various arithmetical calculations are often nec- essary. Special attention may need to be given to the methods employed in solving such problems as are presented, and a few pages are here devoted to the treatment of some of the more common problems in a systematic, comprehensive form, convenient for ready reference. In creameries, cheese-factories, etc., where much arithmetical work is involved in making divi- dends, saving of time is effected by using calculations or tables, which are published in book form. I. FINDING WEIGHT OF ANY CONSTITUENT Rule. — To find the weight of any constituent in milk or milk products, when the weight of the milk or its product and the per cent, of the constituent are known, multiply the weight by the number indicating the per cent, of the constituent and divide the result by lOO. Example : How many pounds of fat in 675 pounds of milk testing 4.6 per cent, of fat? ^25>-]-,: .: .rVitt::': ■f^jir.-^sBifnu-i'f .. .,,,.i.!.l:.:..i'ug^.^-::I^:7i-i,;^gpJJj;yi] :: •„ ...•■-•i u.:. - V«< -■'5-,;,-.";^ tpn.-:;.!!'"-.- aferh'.':?:.-.:; ■ •.-::? ;t^ -.j i:!^rr«j».»jtji;-.iif}ir;«a'r.ti::.-^ .l:.r. .■".ricl :«.-S'.-' •-•;ii:-'ii.'j-»< 5 :►' .-.•- -i.'.'t.- ■' Jii;:?.;^;;.-..--Tr..-r.:,r: t- ... .- ■ -■-'/."..:.. Urj—'-'^-'?- -■"-f«"--^-lr' '' -i ' ■ " ■ ' ' ,- .. :,-■■■;„■ :::;y.t :.iiiTIt^•^j?^.'^.«jij<^•i,'tif|■-•:^:;.,-fJ ;^ It ^i . ;i . [«^I>^ jWiiij^ j^jr " . '■'"nLTj'n:,} . .'rj'-.jt-.i^vinfl -..••._-M ;'.'».;,■.. iJ.-;>.'. trwv.ijiri .!/■ • .-r ■■'',]' Ji''-TlH .{.,l.-.X--,'-J. J, LIBRARY OF CONGRESS ■^^■^^^-■^^SM^^^^^M j:'.t,. ^ . aval^^H^^^^^^^^I I. ...,.« •• t «< • M^^^^^^^^^^^^^^^^^^^^^^^^^^H ^:: :;;:; -'.';:''m;^>*4tt$^^^^^^^^^^^^^^^^^^^^l ' ■ ;.,u.:L3:i7u.:Sv^.■>';'-;;k>rvil^n«^='=l^^^ '• '■■'^■■.■.■■'— .-u;*^|M^J :';r-. ••f- ■■ • '■■.ViaH;.':::":' .••^VJf"'!' «'•.->, ,»;»« -i.'rrftiiu.-tiJ^iis;: ...„.,... ^v/>--Pw^j:2-,;?-:;u:if,Bru«2aij|| .-..--•.•;.. 'OH ,'-:..::i.:-^--^;^-j,:£f:« ;■■ ir:rf;i.^>^^::^;;f:'!:;rJ5^.;.i;=^:^ -•r*— 'rl^^^l ■;...i;L:;;i:,- ■ ;.4H;.ii?'^™ ■■■^'!!'¥^:'r^i2^L:- ''''t'«;;;'?':^^'!r''=i?1^ -'■■■: I- --:\ V. ".^ri. -^Jfe, ;.; „ ,;., .-,;^„_^ - ■ ;..-;-.-;-.v.rv;:: vrri,i^ •'•;rii-- r-ix-;:.^