Testing milk and its products; a manual for dairy students, creamery and cheese factory operators, food chemists, and dairy farmers

Front matter

TESTING MILK

AMD ITS PRODUCTS

FARAINGTON AND WOLL

Preface To First Edition

The present volume is intended for the use of dairy students, factory operators, dairymen, food chemists, and others interested in the testing or analysis of milk and its products. The subject has been largely treated in a popular manner; accuracy and clearness of statement, and systematic arrangement of the subject matter have, however, been constantly kept in mind. The aim has been to make the presentation intelligible to students with no further training than a common-school education, but their work will naturally be greatly lightened by the aid of an able teacher.

Complete directions for making tests of milk and other dairy products are givetn; difliculties which the beginner may meet with are considered in detail, and suggestions offered for avoiding them. It is expected that a factory operator or practical dairyman, by exercising common sense and ordinary care, can obtain suflScient knowledge of the subject through a study of the various chapters of this book to make tests of milk, cream, etc., even if he has had no previous experience in this line.

For the benefit of advanced dairy students who are somewhat familiar with chemistry and chemical operations, Chapter XIV has been added giving detailed instructions for the complete chemical analysis of milk and other dairy products. The detection of preservatives and of artificial butter or filled cheese has also been treated in this connection.

As the subject of milk testing is intimately connected with the payment for the milk delivered at butter- and cheese fac tories, and with factory dividends, a chapter has been devoted to a discussion of the various systems of factory book-keeping, and tables greatly facilitating the work of the factory secretary or bookkeeper have beeli prepared and are included in the Appendix.

Madison, Wis., October 1, 1897.

Preface To Twentieth Edition

Each year that passes brings some valuable contributions to our knowledge of the subjects treated in this book and a frequent revision of it is therefore desirable. The present edition contains descriptions of methods and apparatus that have stood the test of actual use during the past few years; the new information published since the last revision of the book has been carefully sifted, and what was deemed of suflScient importance has been incorporated in such detail as the scope of the book permitted; many changes and additions suggested by the experience of the authors have also been introduced. The book has, in brief, been subjected to a renewed critical examination and careful revision.

The general adoption of the book as a text or reference book in American dairy schools, as well as the favorable reception which it has been accorded by the dairy public in general, will, it is hoped, be further justified by the present revision. Acknowledgment is due to the following parties for loan of electrotypes, viz.: Creamery Pkg. Mfg. Co., Chicago, 111.; Vermont Farm Machine Co., Bellows Falls, Vt.; D. H. Burrell & Co., Little Falls, N. Y.; Henry Tromner, Philadelphia, Pa. ; Torsion Balance Valve Co., New York City; Marschall Dairy Laboratory, Madison, Wis., and International Instrument Co., Cambridge, Mass.

Madison, Wis., Dec. 15, 1910.

Chap. XIV. Chemical analysis of milk and its products 217 Appendix 259

Table I. Composition of milk and its products.

Table II. State and city standards for dairy products.

Table III. Quevenne lactometer degrees corresponding to the scale of the N. Y. Board of Health lactometers.

Table IV. Value of for specific gravities from

Table V. Correction table for specific gravity of milk.

Table VI. Per cent, of solids not fat, corresponding to to 6 per cent, of fat and lactometer readings of 26 to 36.

Directions for the use of Tables Am, VIII, IX and XI.

Table VII. j. ounds of fat in 1 to 1,000 pounds of milk testing 3 to 5.35 per cent.

Table VIII. Pounds of fat in 1 to 1,000 lbs. of cream testing 12.0 to 50.0 per cent. fat.

vi Testing Milk and Its Products.

Table IX. Amount due for butter fat, in dollars and cents, at 12 to 2o cents per pound.

Table X. Relative-value tables.

Table XI. Butter chart, showing calculated yield of butter, in pounds, from 1 to 1,000 pounds of milk testing 3.0 to 5.3 per cent, of fat.

Table XII. Overrun table, showing pounds of butter from 100 pounds of milk.

Table XIII. Yield of cheese, corresponding to 2.5 to 6 per cent, of fat, with lactometer readings of 26 to 36.

Table XIV. Comparisons of Fahrenheit and Centigrade (Celcius) thermometer scales.

Table XV. Comparison of metric and customary weights and measures.

Suggestions regarding the organization of co-operative creameries and cheese factories.

Constitution and by-laws for co-operative factory associations.

Index 291

Introduction

The need of a rapid, accurate and inexpensive method of determining the amount of butter fat in milk and other dairy products became more and more apparent, in this country and abroad, with the progress of the dairy industry, and especially with the growth of the factory system of butter and cheese making during the last few decades. So long as each farmer made his own butter and sold it to private customers or at the village grocery, it was not a matter of much importance to others whether the milk produced by his cows was rich or poor. But as creameries and cheese factories multiplied, and farmers in the dairy sections of our country became to a large extent patrons of one or the other of these, a system of equitable payment for the milk or cream delivered became a vital question.

I. Nearly all the creameries in existence in this country up to about 1890 were conducted on the cream-gathering plan: -the different patrons creamed their milk by the gravity process, and the cream was hauled to the creamery, usually twice or three times a week, where it was then ripened and churned. The patrons were paid per inch of cream furnished. This quantity was supposed to make a pound of butter, but cream from different sources, or even from the same sources at different times, varies greatly in butter-producing capacity, as will be shown under the subject of cream testing (203^). The system of paying for the number of creamery inches delivered could not therefore long give satisfaction.

The proposition to take out a small portion, a pint or half a pint, of the cream furnished by each patron, and determine the amount of butter which these samples would make on being churned in so-called test chums, found but a very limited acceptance, on account of the labor involved and the difficulty of producing a first-class article from all the small batches of butter thus obtained.

2. The introduction of the so-called oil test churn in creameries, which followed the creamery-inch system, marked a decided step in advance, and it soon came into general use in gathered-cream factories (202). In this test, glass tubes of about % inch internal diameter and nine inches long, are filled with cream to a depth of five inches, and the cream is churned ; the tubes are then placed in hot water, and the column of melted butter formed at the top is read off by means of a scale showing the number of pounds of butter per creamery inch corresponding to different depths of melted butter. While the oil test is capable of showing the difference between good and poor cream, it is not sufficiently accurate to make satisfactory distinctions between different grades of good and poor cream.^ As a result,

^ Refers to paragraph numbers.

2 Wis. Expt. Station, bulletin 12 (see also under 203).

Inti'oduction. 3

full justice cannot be done to different patrons of creameries where payments for cream delivered are made on the basis of this test.

3. In cheese factories, and since the introduction of the centrifugal cream separator, in separator creameries, the problem of just payment for the milk furnished by different patrons was no less perplexing than in the case of gathered-cream factories. By the pooling system generally adopted, each patron received payment in proportion to the number of pounds of milk delivered, irrespective of its quality. Patrons delivering rich milk naturally will not be satisfied with this system when they find that their milk is richer than that of their neighbors. The temptation to fraudulently increase the amount of milk delivered by watering, or to lower its quality by skimming, will furthermore prove too strong for some patrons; the fact that it was difficult to prove any fraud committed, from lack of a reliable and practical method of milk analysis, rendered this pooling system still more objectionable.

4. Formerly private dairymen and breeders of dairy cattle who desired to ascertain the butter-producing capacities of the individual cows in their herds were obliged to do this by the cumfbersome method of trial churnings : by saving the milk of each cow to be tested, for a day or a week, and churning separately the cream obtained. This requires a large amount of work when a number of cows are to be tested, and can not therefore be done except with cows of great excellence or by farmers having plenty of hired help. Here again the need of a practical milk test was strongly felt.

5. Introduction of milk tests. The first method which fulfilled all reasonable demands of a practical and reliable milk and cream test was the Babcock test, invented by Dr. S. M. Babcock, of the Wisconsin agricultural experiment station. A description of the test was first published in July, 1890, as bulletin No. 24 of that Station, entitled: A new method for the estimation of fat in milk, especially adapted to creameries and cheese factories. This test, which is now known in all parts of the world where dairying is an important industry, was not, however, the first method proposed for this purpose which could be successfully operated outside of chemical laboratories. It was preceded by a number of different methods, the first one published in this country being Short's method, invented by the late F. G. Short, and described in bulletin No. 16 of Wisconsin experiment station (July 1888).

6. Short*s test. In this ingenious method, a certain quantity of milk (20 cc.*) was boiled with an alkali solution and afterwards with a mixture of sulfuric and acetic acids; a layer of insoluble fatty acids separated on top of the liquid and was brought into the graduated neck of the test bottles by addition of hot water; the reading gave the per cent, of fat in the sample of milk tested.

Short's method did not find very wide application, both because it was rather lengthy and its manipulations somewhat difficult for non-chemists, and because several other methods were published shortly after it had been given to the public.

7. Other milk tests. Of these may be mentioned, besides the Babcock test already spoken of, the Failyer and Willard method,^ Parsons' method,' Cochran's test,* the Patrick or Iowa

1 See 48, footnote.

2 Kansas experiment station report, 1888, p. 149. ^N. H. experiment station report, 1888, p. 69.

* Journal of Anal. Chem., Ill (1889), p. 381.

Introduction. 5

station test/ and the Beimling (Leffmann and Beam) test.^ Of foreign methods published at about the same time, or previously, the Lactocrite/ Liebermann's method,* the Schmid,' Thorner,* Nahm/ Eose-Gottlieb,^ sin-acid method,' and the Gerber sal-method^" may be noted.

8. All these tests were similar in principle, the solids not fat of the milk being in all cases dissolved by the action of one or more chemicals, and the fat either measured as such in a narrow graduated tube, or brought into solution with ether, gasoline, etc., and a portion thereof weighed on evaporation of the solvent. While this principle is an old one, having been employed in chemical laboratories for generations, its adaptation to practical conditions, and the details as to apparatus and chemicals used were, of course, new and different in each case. The American tests given were adopted to a limited extent within the states in which they originated and even outside of them, as in the case of the Short, Patrick and Beimling methods. The Babcock test, however, soon replaced the different methods mentioned, and during the past twenty years it has now been in almost exclusive use in creameries and cheese factories in this country, where payments are made on the basis of the quality of the milk deliv-

1 la. exp. sta., bull. No. 8, Feb. 1890 ; Iowa Homestead, June 14, 1889.

2 Vermont exp. sta., bull. No. 21, September, 1890. For description of these and other volumetric methods of milk analysis, see Wiley, Agricultural Analysis, Vol. Ill, p. 490 et seq ; Wing, Milk and its Products, p. 33 et seq., and Snyder, Chemistry of Dairying, pp. 112-113.

* Fresenius' Zeitschr., 22, 383.

oChem. Centralbl., 1892, 429.

8 Milch-Zeitung, 1904, No. 27 .

"Milch-Zeitung, 1906, No. 8.

ered, as well as in the routine work in experiment station laboratories, and among milk inspectors and private dairymen.

9. The Babcock test. The main cause why the Babcock test has replaced all competitors is doubtless to be sought in its simplicity and its cheapness. It has but few manipulations, is easily learned, and is cheap, both in first cost and as regards running expenses.

The test is furthermore speedy, accurate,^ and easily applied under practical conditions, and may therefore safely be considered the best milk test available at the present time.

The method is applicable not only to whole milk, but to cream, skim milk, butter milk, whey, condensed milk, and (if a small scale for weighing out the sample is available) to cheese and butter. ^

With all its advantages, the Babcock milk test is not in every respect an ideal test. The handling of the very corrosive sulfuric acid requires constant care and attention; the speed of the tester, the strength of the acid, the temperature of the milk to be tested, and other points, require constant watching, lest the results obtained be too low or otherwise unsatisfactory. In the hands of careful operators the test can, however, always be relied upon to give most satisfactory results.

10. Foreign methods. In European countries five practical milk and cream tests, besides the Babcock test,

1 For a summary of comparative analyses made by the Babcock test and gravimetric analysis up to 1892, see Hoard's Dairyman, Oct. 7, 1892, p. 2560 ; also Schrott-Fiechtl, Milchzeitung, 1896, p. 183 et seq.

2 The Babcock test, like the ether-extraction method gives, however, somewhat too low results in the case of skim milk (97).

Introduction.

are in use at the present time, viz. : Gerher's acidhut yrometer, the sin-acid (or no-acid) test, the lactocrite, De LavaVs hutyrometer, and Fjord's centrifugal cream test."-

Of these tests the last one has never, to our knowledge, been introduced into this country, and the first four only to a limited extent.

Fig. 1. ThelGerber acid-butyrometer.

11. The Gerber method'

(fig. 1) is essentially the old Beimling method (7), worked out independently by the Swiss chemist, Dr. N. Gerber.

In this test sulfuric acid of the same strength as in the Babcock test is used, and a small quantity of amyl alcohol is added. The amyl alcohol facilitates the separation of the fat, but may introduce a source of error on account of impurities contained therein, when the results obtained with a new lot of alcohol can not be checked against gravimetric analysis or against tests made with amyl alcohol known to give correct results. This method is, however, extensively used in European countries, having there practically replaced the Babcock test or been adopted in preference to it.

11a. The sin-acid test was invented by the German chemist A. Sichler and published in 1904.^ In place of the sulfuric acid used in the Babcock and Gerber tests, Sichler employs a solution of

1 The Lister-Iiahcock milk test advertised in English papers and known as such in England, is the regular Babcock test, to which the English manufacturers have affixed their name ; the same applies to the Ahlhorn-BabcncJc and the Krugmann-Bahcock methods.

2 Gerber, Die praktische Milchpriifung, 7th edition, 1900.

^ Milchztg., 1904, p. 417. The word sin (sine) is Latin and means without; hence, when introduced into this country in 1909 the method was called the .M-o-acid test.

Testing Milk and Its Products.

Roehelle salts, sodium sulfate and sodium hydroxid/ 150 cc. of this mixture of salts are dissolved in 1 liter of water. In testing milk, 11 cc. of this solution and 0.6 cc. of "sinol" (isobutylalcohol) are added to 10 cc. of milk. After thorough mixing of the milk and solution the test bottles are placed in water of 113° F. for 3-5 minutes, when they are shaken till all the curd dissolves. They are then revi)lved in a centrifuge for 3 minutes and the results read off. By heating the bottles for 1 hour in boiling hot water correct results may be obtained without the use of a centrifuge. The main advantages of the method appear to lie in this fact and in that the use of a corrosive acid is avoided.

12. The Lactocrite was one of the earliest practical milk tests introduced. It was invented by De Laval in 1886. The acids used in this test are lactic acid (originally, acetic acid) with a mixture of hydrochloric and sulfuric acids. This test is now but rarely met with.

Fig. 2. De LavaFs butyrometer. 13. In the De Laval butyrometer (fig. 2) the same acid is used as in the Babcock test, but the tubes employed and the manipulations of the method differ materially from this test; a smaller sample of milk is taken (only 2 cc.) and a correspondingly small quantity of acid used. Where a large number of

1 Barthel-Goodwin, Methods used in Examination of Milk and Dairy Products, p. 77.

Introduction.

milk samples are tested every day, as, for instance, in milk control stations, the butyrometer may be preferable to the Babcoek test; but it requires more skill of the operator and does not work satisfactorily in case of sour, loppered, or partially churned milk.

14. Fjord's centrifugal cream tester' (fig. 3) is extensively used m Denmark and is mentioned in this connection as it furnishes, as a rule, a reliable method for comparing the quality of different lots of milk. The method was published in 1878, by the late N. J. Fjord, director of the state experiment station in Copenhagen, through whose exertions and on whose authority it was introduced into Danish creameries in the middle of the eighties. No chemicals are added in this test, the milk being simply placed in glass tubes, seven inches long and about two-thirds of an inch in diameter, and whirled for twenty minutes at a rate of 2000 revolutions per minute at 55°G (ISl^T.). The reading of the cream layer thus obtained gives the per cent, of cream, and not of butter fat, in the sample tested. One hundred and ninety-two samples of milk can be tested simultaneously. Within the limits of normal Danish herd milk, the results obtained correspond to the per cents of fat present in the samples, one per cent, of cream being equal to about 0.7 per cent, of fat; outside of these limits the test is, however, unreliable, especially in case of very rich milk and strippers' milk. Only sweet milk can be tested by this method. Milk tests proper, like the Gerber, Babcoek and De Laval tests, have during recent years been introduced into Denmark and are used in some creameries.^

^ State Danish experiment station, Copenhagen, sixth and ninth reports, 1885-7.

2 Among foreign milk tests in use abroad should also be mentioned the Lindstrom butyrometer and the Wollny re f Tactometer , both of which, in the hands of trained chemists, may prove better adapted for use where a very large number of samples are to be tested at a time, than any other available milk test.

Fig. 3.

Fjord's centrifugal cream tester.

Chapter I. Composition Of Milk And Its Products

Before taking up the discussion of the Babecck milK test, a brief description of the chemistry of milk and ith products is given, so that the student may understand what are the components of dairy products, and the relation of these to each other. Only such points as have a direct bearing upon the subject of milk testing and the use of milk tests in butter and cheese factories or private dairies will be treated in this chapter, and the reader is referred to standard works on dairying for more detailed information in regard to the composition of dairy prodr/cts.

15. Composition of milk. Milk is composed of the following substances: water, fat, casein, albumen, mUk sugar, and ash. A few other substances are present in small quantities, but they are hardly of any practical importance and will not be considered here. The components of the milk less the water are known collectively as milk solids or total solids, and the total solids less the fat, i. e., casein, albumen, milk sugar, and ash, are often spoken of as solids not fat or the non-fatty milk solids. The milk serum includes all components uf the milk less the fat ; the serum solids are therefore another name for the solids not fat; when given, they are, however, generally calculated to per cent, of milk serum, not of milk. If, e. g., a sample of milk contains nine per cent, of solids not fat, and three per cent, of fat, the milk serum will make up 97 per cent, of the

milk, and the serum solids, ^^^^ = 9.28 per cent, of the milk serum.

i6. Water. The amount of water contained in cow's milk ranges from 82 to 90 per cent. Normal cow's milk will not, as a rule, contain more than 88 per cent, of water, nor less than 84 per cent. In states where there are laws regulating the sale of milk, as is the case in eighteen states of the Union (see Appendix, Table II), the maximum limit for' water in milk in all instances but one (South Carolina) is 88 per cent. ; the state mentioned allows 88.5 per cent, of water in milk offered for sale within her borders. The effect of fraudulently increasing the water content of milk by watering is considered under Adulteration of Milk (121).

17. Fat. The fat in milk is not in solution, but suspended as very minute globules, which form an emulsion with the milk serum; the globules are present in immense numbers, viz., on the average about one hundred millions in a single drop of milk ; a quart of milk will contain about two thousand billions of fat globules, a number written with thirteen figures. The sizes of the globules in the milk from the same cows vary according to the stage of the period of lactation, the globules being largest at the beginning of the lactation period, and gradually decreasing in size with its progress. Different breeds of cows have fat globules of different average sizes; the Channel Island cows are thus noted for the relatively large fat globules of their milk, while the lowland breeds, the Ayrshire, and other breeds have "uniformly smaller globules. The diameter of average sized fat globules in fresh milkers is about .004 millimeter, or one six-thousandth of an inch; that is, it takes about six thousand such globules placed side by side to cover one inch in length. The globules of any sample of milk vary greatly in size; the largest globules are recovered in the cream when the milk is set or run through a cream separator, and the smallest ones remain in the skim milk; thoroughly skimmed separator skim milk contains only a small number of very minute fat globules.

Milk fat is composed of so-called glycerides of the fatty acids, i. e., compounds of the latter with glycerin; some of the fatty acids are insoluble in water, viz., palmitic, stearic, and oleic acids, while others are soluble and volatile, the chief ones among the latter being butyric, caprylic, and capronic acids. The glycerides of the insoluble fatty acids make up about 92 per cent, of the pure milk fat; about 8 per cent, of the glycerides of volatile fatty acids are therefore found in natural milk- (and butter-) fat. The distinction between natural and artificial butter lies mainly in this point, since artificial butter (butterine, oleomargarine) as well as other solid animal fats contain only a very small quantity of volatile fatty acids. The glycerides of the volatile fatty acids are unstable compounds, and are easily decomposed through the action of bacteria or light; the volatile fatty acids thus set free, principally butyric acid, are the cause of the unpleasant odor met with in rancid butter.

Cow*s milk generally contains between three and six per cent, of fat; in American milk we find, on the average, toward four per cent, of fat. The milk from single cows in perfect health will occasionally go below or above the limits given, but mixed herd milk rarely falls outside of these limits. The standard adopted by the U. S. government for fat in milk is 3.25 per ct. The legal standard for fat in milk in most states of the Union is 3 per cent. ; Rhode Island allows milk containing 2.5 per cent, of fat to be sold as pure, while Georgia and Minnesota require it to contain 3.5 per cent., and Massachusetts 3.7 per cent, (in the months of May and June; see Appendix, Table II).

i8. Casein and albumen. These belong to the so-called nitrogenous substances, distinguished from the other components of the milk by the fact that they contain the element nitrogen. Another name is albuminoids or protein compounds. Casein is precipitated by rennet in the presence of soluble calcium salts, and by dilute acids and certain chemicals ; albumen is not acted upon by these agents, but is coagulated by heat, a temperature of 170° P. being sufficient to effect a perfect coagulation. The casein, fat, and water, are the main components of nearly all kinds of cheese. In the manufacture of Cheddar and most other solid cheeses, the casein is coagulated by rennet, and the curd thus formed holds fat and whey mechanically, the latter containing in solution small quantities of non-fatty milk solids. The albumen goes into the whey and is lost for cheese making ; in some countries it is also made into cheese by evaporating the whey under constant stirring ; whole milk of cows or goats is often added and incorporated into such cheese {primost, gjetost).

Casein is present in milk partly in solution, in the same way as milk sugar, soluble ash-materials and albumen, and partly in suspension, in an extremely fine colloidal condition, mixed or combined with insoluble calcium phosphates. The casein and calcium phosphates in suspension in milk may be retained on a filter made of porous clay (so-called Chamberland filters).

About 80 per cent, of the nitrogenous compounds of normal cow's milk are made up of casein; the rest is largely albumen. If the amount of casein in milk be determined by precipitation Avith rennet or dilute acids, and the albumen by boiling the filtrate from the casein precipitate, it will be found that the sum of these two compounds do not make up the total quantity of nitrogenous constituents in the milk. The small remaining portion (about five per cent, of the total nitrogenous constituents) has been called by various authors, globulin, albumose, hemi-albumose, nuclein, nucleon, proteose, etc. The nitrogenous constituents of milk are very unstable compounds, and their study presents many and great difficulties ; as a result we find that no two scientists who have made a special study of these compounds agree as to their properties, aside from those of casein and albumen, or their relation to the nitrogenous substances found elsewhere in the animal body. For our purpose we may, however, consider the nitrogen compounds of milk as made up of casein and albumen, and the term casein and albumen, as used in this book, is meant to include the total nitrogenous constituents of milk, obtained by multiplying the total nitrogen content of the milk by 6.25.^

The quantity of casein in normal cow's milk will vary from 2 to 4 per cent., and^^ albumen, from .5 to .8 per cent. The total content of casein and albumen ranges between 2.5 and 4.6 per cent, the average being about 3.2 per cent. Milk with a low fat content will contain more casein and albumen than fat, while the reverse is generally true in case of milk containing more than 3.5 per cent, of fat.

19. Milk sugar or lactose belongs to the group of organic compounds known as carbohydrates. It is a commercial product manufactured from whey and is obtained in this process as pale white crystals, of less sweet taste and less soluble in water than ordinary sugar (cane sugar, sucrose). About 70 per cent, of the solids in the whey, and 33 per cent, of the milk solids, are composed of milk sugar.

When milk 'is left standing for some time, viz., from one to several days, according to the temperature of the surrounding medium it will, as a rule, turn sour and soon becomes thick and loppered. This change in the composition and appearance of the milk is brought about through the action of acid-forming bacteria on the milk sugar. These are present in ordinary milk in immense numbers, and under favorable conditions of temperature multiply rapidly, feeding on the milk sugar

1 The factor 6.25 is generally used for obtaining the casein and albumen from the total nitrogen in the milk, on the theory that protein compounds contain 16% N. ; the factor 6.37 would, however, be more correct, since casein and albumen, according to our best authorities, contain on the average 15.7 per cent of nitrogen {'^^^=Q.Z7)

as they grow, and decomposing it into lactic acid. When this change alone occurs, there is not necessarily a loss in the nutritive value of the milk, since milk sugar breaks up directly into lactic acid. This is shown by the following chemical formula: