•aoamoo A)ivNi)iaxaA axvxs 'a "n aui .^ 
 
 doasn 3HJ. aod /g'ls' 
 
 3H1 
 
 'AXISU3AINn -n3NUOO 
 
The original of tiiis book is in 
 tine Cornell University Library. 
 
 There are no known copyright restrictions in 
 the United States on the use of the text. 
 
 http://www.archive.org/details/cu31924000935068 
 
Demy 8vo., pp. Zl^, price 12s. &d. net. 
 
 APPLIED BACTERIOLOGY, 
 
 ®n 3nfro^ucfore §(mb6ooft for f^e QJse of ^fubenfs, 
 (JjClebicftf <t)fficer6 of f eaft^, ®nftfe6f0, anb <&f^er6. 
 
 BY 
 
 T. H. PEAEMAIN and C. G. MOOR, M.A. 
 
 Mtmbeirs of the Society of Public Analysts, 
 Associates British Institute of Public Health, etc. 
 
 Illustrated with thirty-three Woodcdts and eighty-one 
 Coloured Figures of Preparations, Cultures, etc. 
 
 EXTRACTS FROM PRESS NOTICES. 
 The British Medical Journal, December 19, 1896, 
 
 ' This handbook appears to contain all the bacteriological information ' which a 
 medical officer of health is likely to require. . . . The authors have selected and 
 arranged their material with judgment, and have succeeded in presenting it in a 
 scientific spirit. Thus the importance of the chemical examination of water is shown 
 as clearly as the incapacity of such analysis to reveal bacterial contamination, and 
 the remarks on sand-filters account in advance for the liability to failure which has 
 since been so strikingly shown in London. ... On many subjects, as, for instance, 
 in the account of the examination of disinfectants, the authors show a just appre- 
 ciation of the limits within which inferences from experiments are justified ; and 
 those who have approached bacteriology after the study of the physical sciences will 
 know how rare this sense is in the exposition of the latter science.' 
 
 The Lancet, October 31, 1896. 
 ■ This work, though it claims to be especially suitable for medical officers of 
 health, may be accepted as appealing to a somewhat wider clientMe, as, although the 
 ground covered necessarily comes within the domain of a medical officer of health, 
 there is little in the book which should not be thoroughly understood by every 
 student of bacteriology. . . . The book is well got up, and the coloured illustrations 
 of the naked-eye growths and microscopic appearances of organisms add very dis- 
 tinctly to the value of the work, which, taken altogether, may be recommended as a 
 convenient and accurate account of the methods with which a public analyst should 
 be conversant in order that he may carry on bacteriological investigations with any 
 chance of obtaining satisfactory results, and we congratulate the authors on the 
 manner in which they have placed before their confreres the information that they 
 have been able to collect.' 
 
 The Chemical News, October 23, 1896. 
 
 'In the face of a writer who asserts in the columns of an esteemed medical 
 contemporary that "the whole modern science of bacteriology is a gigantic mistake," 
 we must venture to pronounce this work a timely and useful contribution to our 
 hygienic literature, and to offer to the authors our warmest congratulations. . . . 
 We think this book should be studied not merely by professional men, but by the 
 members of County Councils, and all persons having to concern themselves with 
 public health. ' 
 
 LONDON: BAILLlfiRE, TINDALL & COX, 
 
 20 & 21, Kins WitLiAii Stkbbt, Strand. 
 
CHINOSOL. 
 
 An absolutely non-poisonous, non-corrosive, most- 
 powerful 
 
 Antiseptic, Disinfectant, & Deodorant. 
 
 TEADE MABK. 
 
 CHINOSOL is a germicide at least forty times stronger than Carbolic 
 Acid, and considerably more effective than Corrosive Sublimate, without 
 possessing the dangerous and objectionable character of either of the two (see 
 published bacteriological reports and general literature). 
 
 CHINOSOL presents the following further favourable 
 points over all known Antiseptics : 
 
 It is readily soluble in warm and cold water in any proportion; its solu- 
 tions keep active for a long time ; it does not coagulate albumen ; it is 
 quickly resorbed through the skin ; it favours granulation ; it does not 
 stain or damage linen, nor does it injuriously affect the skin of the hands 
 even in the strongest solution ; and it is a 
 
 Most efficient DISINFECTANT, and acts very promptly as 
 a DEOBOMANT. 
 
 CHINOSOL is sold in the following forms and prepara- 
 tions : 
 
 Pure Chinosol in crystalline powder ; tablets ; antiseptic surgical dressings 
 (prepared in John Milne's Antiseptic Dressings Factory) ; medicinal, veteri- 
 nary, and toilet soaps ; dusting powder for wounds and destroying parasites ; 
 ointments, etc., etc. For Inodorous Disinfection the following preparations 
 are offered : Crude Chinosol granular ; Crude Chinosol Powder F ; Chinosol 
 Fluid and Chinosol Disinfecting Powder for sprinkling. 
 
 CAUTION. 
 
 CHINOSOL is a chemical product protected in England by Letters 
 Patent. It is sold in various packages, each bearing the above Trade -mark 
 and the address : B. KUHN, 36, ST. MAEY-AT-HILL, LONDON Sole 
 Licensee for the United Kingdom, the British Colonies, and Depend- 
 encies. Sellers and purchasers of packages of Chinosol not provided with 
 the above Trade-mark and address render themselves liable to 1 1 
 oeedings. ^ 
 
THE ANALYSIS OF FOOD AND DRUGS 
 
THE ANALYSIS OF FOOD 
 MB DRUGS. 
 
 iart I. 
 MILK AND MILK PRODUCTS. 
 
 T. H. PEARMAIN and C. G. MOOE, M.A. (caniab.), 
 
 *" Members of the Society of Public AnalystS| 
 
 authors of 
 * a manual of applied bacteriology,' * aids to the analysis of food and drugs,' 
 
 ETC, 
 
 LONDON : 
 BAILLIEEB, TINDALL AND COX, 
 
 20 & 21, King William Strkbt, Stband. 
 
 [PAEIS AND MADRID.] 
 
 1897. 
 
 All rights reserved. 
 

 
 SMI 
 
PREFACE. 
 
 The ready sale attending our 'Aids to the Analysis of 
 Food and Drugs ' encourages us to hope for a favourable 
 reception for a larger work on the same subject. 
 
 Our aim is to produce a book convenient for laboratory 
 use, which shall contain all that is required for everyday 
 routine work without in any way pretending to be an 
 exhaustive manual on the subject. 
 
 As the work may run to some 800 or 900 pages, we have 
 decided to bring it out in parts, so that each portion may 
 be up to date. 
 
 Each part will be complete in itself, and will have an 
 index, while a general index will be issued with the last 
 part. 
 
 In order to keep the book within manageable limits, we 
 are compelled to deal briefly, or sometimes merely to refer 
 to researches which would be rightly included in a larger 
 work. 
 
 The length of the various articles must not be taken as 
 a criterion of the importance of the subjects dealt with, as 
 it often happens that much work has been done on com- 
 paratively unimportant articles, while on other subjects of 
 great importance very little information has been published. 
 
 Most of the grosser forms of adulteration mentioned in 
 the older text-books have disappeared, owing to the opera- 
 
VI PREFACE 
 
 tion of the Food and Drugs Act. The chief need at the 
 present time appears to be the formulation of standards to 
 which foods should conform. There is, of course, much to 
 be said on both sides as regards the effect such standards 
 would have if adopted; but they would in our opinion 
 be generally productive of good, and would tend to the 
 ' levelling up ' rather than to the ' levelling down ' of the 
 quality of food. 
 
 The need of a proper standard for milk, one of the most 
 important foods of all, is admitted by everyone who is not 
 interested in its sophistication ; while it seems equally 
 advisable that some limit should be imposed by law as to 
 the amount of water in bread and other foods, and a 
 minimum set for the fat in butter, cheese, cream, and 
 condensed milks. We have endeavoured to indicate in 
 each article to what standards and tests it should conform. 
 
 As the law relating to the adulteration of food and drugs 
 is likely to undergo changes in the near future, we have 
 made but few references to legal matters, which are, how- 
 ever, admirably dealt with in a small work recently pub- 
 lished, ' The Law and Chemistry of Foods and Drugs,' by 
 Cecil H. Cribb and Mansfield Eobinson. 
 
 In conclusion, we beg to offer our sincere thanks to 
 all who have helped with suggestions or information, 
 particularly to Dr. J. A. Voelcker, Mr. Alfred Smetham, 
 and Mr. John Benson. 
 
 T. H. P. 
 C. G. M. 
 
 4, Danes Inn, Steand, London, W.C, 
 June, 1897. 
 
CONTENTS. 
 
 KILE. 
 
 Physical characters — Composition — Milk-fat — Proteids — Milk- 
 sugar — Ash of milk — Colostrum — Human and other milk^- 
 Influence of health, age and food^ Variations in the composition 
 of nulk— Numher of samples examined by public analysts — 
 Milk standards — Abnormal milks — Skimmed and separated 
 milk — Butter-milk — Frozen milk — Koumiss and Kephir — 
 Methbds of analysis — Becknagel's phenomenon — Total solids 
 — Estimation of the ash — Estimation of fat^ — Mechanical 
 methods — Bell's method — Macfarlane's method — Werner- 
 Schmidt method — Calculation method — Richmond's slide rule 
 — Estimation of proteids — Estimation of sugar by Payy, and 
 polarimetrie method— Detection and estimation of preservatives 
 in milk — Borax and boric acid — Formaldehyde — Sahcylic acid 
 — ^Potassiimi chromate — The bacteriology of milk — Sterihzation 
 of milk 1—65 
 
 CREAM. 
 
 Devonshire cream— Separators— Standard of fat in cream — Gela- 
 tine as an adulterant of cream — Estimation of fat in cream 66 — 68 
 
 CONDENSED MILK. 
 
 Number of samples examined by pubhc analysts — Varieties of 
 condensed mUk — Unsweetened milk — Sweetened mUks — 
 Sweetened partly skimmed mUks— Sweetened skimmed mUks 
 — Composition of condensed milk — Analysis of condensed 
 mUks — Estimation of total solids, ash, proteids, milk-sugar, 
 fat — Standard for condensed mUk — Humanized condensed 
 milk 69—84 
 
viii CONTENTS 
 
 BUTTER. 
 
 Composition of butter and butter-fat-Margarine— Number of 
 samples examined by pubUc analysts— Analysis of butter- 
 Estimation of water, salt, curd, fat— The Eeichert process— 
 The Valenta test— Determination of soluble and insoluble 
 fatty acids — Specific gravity of butter-fat — "Oleo-refracto- 
 jueter — Optical method of examination — Preservatives — 
 Bacteriology of butter - 85—105 
 
 CHEESE. 
 
 General method of manufacture — Bacteriology of cheese — Eennet 
 — Starters — Classification of cheeses — American Cheddar — 
 Camembert — Cheddar — Cheshire — Cream — Derbyshire — 
 Dutch — Gloucester — Gorgonzola — Gruyere — Leicester — 
 Parmesan — Eoquefort — Skimmed milk cheese — Stilton — 
 Wensleydale — Margarine cheese — Number of cheeses 
 examined by pubUc analysts — Standard for cheese — Analysis 
 of cheese 106—129 
 
 Index - - 130—132 
 
THE ANALYSIS OF 
 
 MILK AND MILK-PRODUCTS 
 
 MILE. 
 
 The milk of various animals has always been used and 
 valued as food, and, judging from the many references 
 which occur in the ancient writings to the milk of the cow, 
 ewe, and goat, it is plain that these formed an important 
 part of the dietary of man from the earliest times. In spite 
 of the fact of the very early acquaintance with butter, curd, 
 cheese, whey, and other milk products, very little of any- 
 thing approaching an accurate knowledge of the chemistry 
 of milk was known until the end of the eighteenth century, 
 and it is only comparatively recently that many of the 
 constituents and proportions of the various bodies found in 
 milk have been investigated and estimated. 
 
 Milk may be defined as the fluid which is secreted in 
 the lacteal glands of all female mammals for the nourish- 
 ment of their young. In the case of cow's milk, more 
 particularly, it should be understood to mean the whole 
 milk of perfectly healthy, properly-fed cows. 
 
 It is the only article occurring in Nature which combines 
 in the right proportions all the necessary elements requisite 
 to secure the proper nutrition. For children it is rightly 
 considered the ideal food ; but it is too voluminous to serve 
 
 1 
 
2 THE ANALYSIS OF MILK AND MILK -PRODUCTS 
 
 for the sole diet of adults. Milk differs as procured from 
 different animals, but its general characteristics are the 
 same in all. 
 
 Genuine milk is a dense opalescent white liquid, often 
 having a yellowish tinge, which has a specific gravity 
 ranging between 1'0285 to 1"035. It has a sweetish, bland 
 taste. The colour is due essentially to suspended fatty 
 globules, which occur under conditions which prevent 
 spontaneous coalescence. Milk consists of water, fat, 
 casein, albumin, milk-sugar, and mineral salts, consisting 
 chiefly of the chlorides of potassium and sodium, together 
 with the phosphates of potassium, calcium, and magnesium, 
 and traces of sulphates. 
 
 The fat is suspended in the aqueous portion of the milk 
 in the form of minute globules, forming an almost perfect 
 example of a natural emulsion. When examined under 
 the microscope, milk will be seen to consist of a turbid 
 fluid, containing fatty globules of different sizes, from 
 0'000053 to 0'0004 of an inch in diameter, frequently with 
 small particles of epithelium. Just before and after calving 
 the milk is yellowish, due to the presence of yellow bodies, 
 from 0"00025 to 0"001 of an inch in diameter. Such milk 
 is called colostrum, and, at least in the case of cows, is 
 unfit for human consumption. Pus cells, rather larger 
 than the red blood-corpuscles, granulated and with a core 
 of irregular contour, are found, as are blood-corpuscles, in 
 the milk of sick cows. They are prevented, as Soxhlet has 
 pointed out, from coalescing into an oily film by the con- 
 dition of the albumin, which isolates the fat corpuscles in 
 its meshes. The presence of an envelope, which used to be 
 thought probable, has been disproved. 
 
 When milk is allowed to stand for some time, a separa- 
 tion of the fat takes place. This is known as the rising 
 of the cream, the larger fat globules of the milk naturally 
 
MILK 3 
 
 rising first. Absolutely fresh milk is amphoteric in re- 
 action — that is to say, blue litmus is turned red, and red 
 litmus blue ; but usually by the time milk reaches the 
 consumers it is faintly acid, owing to some slight fer- 
 mentative changes taking place, resulting in the produc- 
 tion of lactic acid. 
 
 We are still in ignorance as to many of the physiological 
 changes which take place in the mammary glands, result- 
 ing in the formation of milk. No satisfactory theory has 
 yet been put forward to explain the physical change which 
 takes place in the conversion of the fibrin and albumin of 
 the blood into the casein and albumin of the milk, or as 
 to the method by which the milk-sugar is formed, or as to 
 the remarkable changes which take place in the fatty mole- 
 cules, resulting in the production of butyric and other 
 higher fatty acid radicles. As already mentioned, when 
 milk is allowed to stand for some time it undergoes 
 fermentative changes, due to a micro-organism, usually 
 Bacillus acidi lactiei. This decomposes the milk-sugar, 
 resulting in the production of lactic acid, which eventually 
 causes the milk to become sour. This change is largely • 
 dependent upon the time the milk has been allowed to 
 stand, but more particularly upon the temperature ; milk, 
 of course, becoming sour more quickly in the summer than 
 in the winter. When the fermentative change has resulted 
 in the production of about 0*4 per cent, of lactic acid, the 
 milk can be distinctly recognised by the taste to be sour. 
 Finally, when the acidity reaches 0"6 per cent., the milk 
 curdles, whereby it spontaneously separates into a solid 
 known as ' curd,' which consists of the fatty and proteid 
 constituents of the milk, and a clear liquid known as 
 'whey,' which consists essentially of a solution of milk- 
 sugar and mineral salts. This change is also brought 
 about artificially, as in the manufacture of cheese, by 
 
 1—2 
 
4 THE ANALYSIS OF MILK AND MILK-PRODITOTS 
 
 the addition of rennet. The further changes which take 
 place in the composition of the milk after curdling depend 
 upon the nature of the bacteria which have gained access, 
 various organisms giving rise to different fermentations. 
 
 As would be expected from its origin, milk is subject 
 to great variations in its composition, on account of the 
 age, breed, food, changes in health, and the general con- 
 ditions of the animal yielding it. The general average 
 composition of genuine cow's milk is as follows : 
 
 Water 
 
 ... 87-6 per cent. 
 
 Pat 
 
 ... 3-6 „ 
 
 MUk sugar ... 
 
 ... 4-8 „ 
 
 Proteids 
 
 ... 3-3 „ 
 
 Mineral matter 
 
 ... 0-73 „ 
 
 Milk Fat. — The fat of milk is very complex in its chemical 
 constitution. This will, however, be fully considered under 
 butter fat, with which it is of course identical. As already 
 stated, the fat occurs in milk in the form of minute globules, 
 from "0015 mm. to '005 mm. in diameter. These fatty 
 globules are stated by some observers to be covered by a 
 very fine film of albuminoid matter, but the presence of 
 this envelope is very doubtful, the spontaneous coalescence 
 of the fatty globules probably being prevented by the film 
 of non-fatty liquid in which the globules are suspended. 
 
 About 68 per cent, of the solid fatty acids are contained 
 in the form of palmitine and stearine; 30 per cent, as 
 olein. 
 
 Milk Proteids.— The proteids of milk consist largely of 
 casein, together with about 0-4 to 0-5 per cent, of albumin, 
 of the variety known as lactalbumin, with traces of globulin. 
 Much work is required to be done on these bodies before 
 their exact nature and their relation to one another can be 
 properly determined. It is these bodies that give to milk 
 its property of curdling when rennet or an acid is added. 
 The albumin is coagulated by boiling. The form in which 
 
MILK 5 
 
 casein exists in milk is known as caseinogen. It is not 
 known with any degree of certainty in what form the 
 casein exists, but it is possible that it is in combination 
 with the calcium and other phosphates. Acid destroys 
 the combination, precipitating the casein in an insoluble 
 condition ; but heat alone will not produce this change. 
 
 Very small proportions of other nitrogenous bodies are 
 present, including those to which are due the characteristic 
 flavour and smell of milk. An unorganized ferment or 
 enzyme, which converts starch, is also present in milk. 
 
 The albumen of milk consists mainly of casein, of which 
 the composition is by no means clear. Hamarsten regards 
 it as a combination of nuclein with some albumin. Done- 
 lewski considers it a mixture of two albumins, which are 
 themselves nuclein compounds, and others attribute it to 
 alkaline albuminates. According to Engling, dry calcium 
 phosphate is combined with the albumin of casein. The 
 casein is precipitated by dilute acids or by rennet or by 
 saturation with salt and magnesium sulphate at 37° to 
 40° C. Boiling precipitates it in small quantities, but if 
 continued for a day will throw it down almost completely. 
 In the presence, however, of organisms capable of decom- 
 posing milk-sugar, the action proceeds more rapidly in 
 proportion to the amount of acid present, and ultimately 
 becomes spontaneous. 
 
 Usually, according to Sebelein, lactalbumin, a body 
 similar to serum albumin, but not identical with it, and 
 thrown down by heating to 70° or 80° C, is present in 
 small quantities, and in colostrum or in many milks from 
 sick cows in abundance. Its presence enables the milk to 
 be readily curdled by heat even in the absence of an acid 
 reaction. 
 
 By boiling, the taste and smell of milk is substantially 
 altered, but the chemical changes are at present obscure. 
 
6 THE ANALYSIS OF MILK AND MILK-PEODUCTS 
 
 A thin film is formed on the surface, which consists of 
 casein with lime-salts. Prolonged boiling, especially at a 
 temperature above 100°, partly decomposes the milk-sugar, 
 and the milk becomes brown and bitter in taste. 
 
 Many vegetable growths are eaten by the cows without 
 mischief, but lead to the production of milk which is dis- 
 tinctly poisonous, especially to children. In the^ same way, 
 certain metallic salts, such as mercury, lead, arsenic, and 
 antimony, if ingested by the cow in the form of medicine, 
 may pass into the milk and create danger for children. 
 
 Milk-sugar, or Lactose (C12H22O11+H2O). — This variety of 
 sugar, which is an isomer of cane-sugar, is peculiar to the 
 milk of the mammalia. It is the most constant con- 
 stituent of milk, from 5 to 6 per cent, being present in 
 human milk, and 4 to 5 per cent, in the milk of most 
 of the herbivora. As milk-sugar is the most constant 
 constituent in milk, the variations in the specific gravity of 
 the whey or serum of genuine milk falls within much 
 narrower limits than in the case of the specific gravity 
 of milk itself. Lactose has only a faint degree of sweetness, 
 and is gritty when chewed. It is much less soluble than 
 cane-sugar, being soluble in six parts of cold and three of 
 boiling water. Milk-sugar is convertible by the action of 
 dilute. acids into 'galactose' and ' lactoglucose ' ; these may 
 be reunited to form lactose. Lactose is dextro-rotatory to 
 a ray of polarized light, and resembles dextrose in its 
 reducing action on Pehling's solution. 
 
 Milk-sugar is obtained commercially as a by-product 
 in the manufacture of butter and cheese. After the 
 separation of the curd by rennet, the lactose remaining 
 in solution is separated by evaporation, and purified by 
 recrystallization. It occurs in the form of rhombic 
 crystals and- prisms. That which comes into the market 
 has been crystallized on strings or wood splinters. 
 
MILK 
 
 Mineral Matter. — The mineral matter or ash of milk 
 obtained by the careful ignition of the solids of milk 
 consists of the chlorides of potassium and sodium, the 
 phosphates of calcium, magnesium, and. potassium, with 
 traces of sulphates. The ash of milk varies from 0"65 to 
 0'85 per cent. 
 
 According to Konig, the average composition of the ash 
 of milk is as follows : 
 
 Potash as KjO 
 Soda as NajO 
 Lime as CaO 
 Magnesia as MgO ... 
 Phosphoric acid as P2O5 
 Chlorine as CI 
 Sulphuric acid as SO3 
 Iron ... 
 
 24-0 per cent. 
 
 8-2 
 22-4 
 
 2-6 
 26-3 
 13-9 
 
 2-5 
 traces. 
 
 The following table gives the composition of the ash 
 of cow's milk, according to Dr. J. Bell : 
 
 Percentage of 
 
 Average 
 Milk of 
 29 Cows. 
 
 Average 
 Milk of 
 several 
 Cows. 
 
 Average 
 Milk of 
 
 several 
 Cows. 
 
 Milk of 
 
 Shorthorn 
 
 6 years 
 
 old. 
 
 Total ash 
 
 0-73 
 
 0-72 
 
 0-78 
 
 0-71 
 
 Potash — estimated as ' KgO 
 Soda „ NajO 
 Lime ,, CaO 
 Magnesia „ MgO 
 Phosphoric Anhydride PjOs 
 Sulphuric „ SO3 
 Chlorine CI 
 
 17-24 
 4-29 
 
 24-53 
 2-89 
 
 35-67 
 2-65 
 
 12-73 
 
 19-53 
 3-30 
 
 24-48 
 4-76 
 
 32-49 
 0-92 
 
 14-52 
 
 19-78 
 3-67 
 
 24-78 
 4-35 
 
 32-07 
 0-69 
 
 14-66 
 
 19-83 
 3-19 
 
 25-98 
 3-68 
 
 32-67 
 1-62 
 
 1303 
 
 Total 
 
 100-00 
 
 100-00 
 
 10000 
 
 100-00 
 
 It may be here mentioned, although only of theoretical 
 interest, that citric acid is a normal constituent of the milk 
 of various animals. Human milk contains about 0-05 per 
 cent., and cow's milk O'l to 0-15 per cent. There are also 
 frequently traces of carbonic acid and other gases in the 
 free state present in solution. 
 
8 THE ANALYSIS OF MILK AND MILK-PEO DUCTS 
 
 Colostrum. — The milk which is yielded immediately after 
 pregnancy is known as ' colostrum.' This is thick and 
 yellow, and differs from normal milk mainly in the pro- 
 portion of proteids present, which may sometimes reach as 
 high as 20 per cent. Leffmaiin and Beam give the follow- 
 ing as the average composition of ' colostrum ' : 
 
 Fat ... ... 3'5 per cent. 
 
 Proteids 12-7 
 
 Milk-sugar ... 4-2 „ 
 
 Ash 0-8 „ 
 
 Water 78-8 
 
 Human Milk. — The composition of human milk, as in the 
 case of that of cows, is largely dependent upon the health 
 and condition of life of the individual yielding it. The 
 specific gravity of human milk varies from 1029 to 1035. 
 Abnormal variations occur between the limits 1017 and 
 1036. The amount of mineral matter does not vary so 
 much as in the case of cow's milk. The proportion of 
 sugar is very nearly constant, but the fat and proteids 
 vary very much in quantity, being very much dependent 
 upon the stage of the nursing, as will be seen from the 
 following figures : 
 
 Va/riation in Fat in Human Milk (Rotch). 
 
 From 43 analyses by Leeds 2-1 to 6-9 per cent. 
 
 » ,. „ Konig 1-7 to 7-6 
 
 From 29 analyses by Chem. Lab. Coll. 
 
 Phys. and Surg., N.Y 1-1 to 5-0 
 
 Va/riation in Proteids m Human Milk (Eotoh). 
 
 From 43 analyses by Leeds 0-85 to 4-86 per cent. 
 
 „ „ „ Konig 0-57 to 4-25 „ 
 
 From 29 analyses by Chem. Lab. Coll. 
 
 Phys. and Surg., N.Y 0-9 to 3-62 
 
 The following table shows the average composition of 
 human milk, according to various observers : 
 
MILK 
 
 
 Pfeiffer, 
 
 Leeds, 
 
 Meigs, 
 
 Juhaunes- 
 
 Lu£f, 
 
 
 Leeds, 
 
 
 100 
 
 43 
 
 4» 
 
 sen, 25 
 
 1'^ 
 
 
 43 
 
 
 samples. 
 
 samples. 
 
 samples. 
 
 samples. 
 
 samples. 
 
 
 samples. 
 
 Water ... 
 
 88-5 
 
 86-8 
 
 87-2 
 
 _ 
 
 88-5 
 
 87-5 
 
 87-3 
 
 Total 
 
 
 
 
 
 
 
 
 solids ... 
 
 11-5 
 
 13-2 
 
 12-8 
 
 
 11-5 
 
 12-5 
 
 12-3 
 
 Proteids 
 
 1-9 
 
 20 
 
 10 
 
 1-1 
 
 2-4 
 
 1-7 
 
 1-7 
 
 Fat ... 
 
 31 
 
 4-1 
 
 4-3 
 
 8-2 
 
 2-4 
 
 3-6 
 
 3-4 
 
 Lactose 
 
 6-3 
 
 6-9 
 
 7-4 
 
 4-7 
 
 6-4 
 
 7-0 
 
 6-4 
 
 Ash ... 
 
 0-2 
 
 0-2 
 
 01 
 
 — 
 
 0-3 
 
 0-2 
 
 0'2 
 
 The following table shows the variations in the com- 
 position of different kinds of milk as compared with human 
 and cow's milk : 
 
 Kind of 
 Milk. 
 
 1 
 
 
 t 
 
 CQ 
 
 1 
 
 i 
 
 •St 
 
 II 
 
 Analyst, 
 
 Htunan... 
 
 87-3 
 
 12-3 
 
 3-4 
 
 6-4 
 
 1-7 
 
 0-2 
 
 1031-3 
 
 
 Cow ... 
 
 87-6 
 
 12-4 
 
 3-6 
 
 4-8 
 
 3-3 
 
 0-7 
 
 — 
 
 
 Goat ... 
 
 86'7 
 
 13-3 
 
 3-8 
 
 4-5 
 
 4-1 
 
 09 
 
 — 
 
 Eiohmond, 
 
 Ewe ... 
 
 75-2 
 
 24-8 
 
 11-3 
 
 3-6 
 
 8-8 
 
 1-1 
 
 1039-3 
 
 J. Bell. 
 
 Ewe ... 
 
 813 
 
 18-7 
 
 6-8 
 
 4-8 
 
 6-3 
 
 0-8 
 
 — 
 
 Vieth. 
 
 Ass 
 
 89-8 
 
 10-2 
 
 12 
 
 6-8 
 
 1-7 
 
 4-5 
 
 , — , 
 
 Biohmond. 
 
 Ass 
 
 89-6 
 
 10-4 
 
 1-6 
 
 6-1 
 
 2-2 
 
 0-5 
 
 — 
 
 Vieth. 
 
 Sow 
 
 89-6 
 
 10-4 
 
 4-8 
 
 3-4 
 
 1-3 
 
 0-9 
 
 — 
 
 
 Gamoose 
 
 84-2 
 
 15-8 
 
 5 5 
 
 5 4 
 
 3-8 
 
 10 
 
 1035-4 
 
 Eichmond. 
 
 From the above table, it will be seen that, taking cow's 
 milk as a standard for comparison, human milk shows an 
 increased proportion of sugar, but all the other constituents 
 are materially less. The milk of sheep is notable for its 
 enormous percentage of fat; it is not employed to any 
 extent in this country, but abroad it is largely used in the 
 manufacture of Eoquefort cheese. The milk of asses is 
 much lower in fat and solids-not-fat than cow's milk ; the 
 casein is not readily curdled, on which account it is some- 
 times used as a substitute for human milk. 
 
 As already stated, the health, age, food, etc., of the 
 animal yielding the milk have a great influence upon its 
 composition and quality. The following are the chief 
 circumstances to which these variations in the composition 
 
10 THE ANALYSIS OF MILK AND MILK-PRODTJCTS 
 
 of cow's milk depend, which of course apply equally to all 
 other animals yielding milk : 
 
 1. The Breed of the Cow. — Some breeds yield quantity, 
 others quality. Alderneys and Jersey cows yield the most 
 fat ; Shorthorns give the most casein and sugar. The 
 average capacity of a cow's udder is about 6 pints, and 
 the average annual yield of milk about 600 gallons. 
 
 2. The Time and Stage of Milking. — Cows are usually 
 milked twice a day, the morning milk usually being the 
 larger in quantity and the poorer in quality. The milk 
 which is first drawn, known as the 'fore milk,' contains 
 very much less fat than that last drawn, known as the 
 ' strippings.' This is due to a partial creaming taking 
 place in the udders. Dishonest dealers have often taken 
 advantage of this fact in adulteration cases, to have the 
 cows partially milked in the presence of ignorant witnesses, 
 the resulting milk consisting largely of the ' fore milk.' 
 
 3. The Age of the Cow. — Young cows give less milk, 
 while cows from four to seven years old give the richest 
 milk, and less milk is given with the first calf. Cows 
 usually become milkers in the third year. They give the 
 largest yield, according to Pleischmann, after the fifth until 
 the seventh calf. After the fourteenth calf they yield, as a 
 rule, no more milk. 
 
 4. The Time of Year. — The poorest milk is yielded 
 during the spring and early summer months, the richest 
 during the autumn and early winter. 
 
 5. The Mental and Physical Conditions under which the 
 Animal is kept. — If the cows are worried or driven about, 
 the quality and yield of the milk is reduced. If they are 
 kept warm and well fed, the quantity and quality of the 
 milk is naturally increased. 
 
 6. The Food. — The natural food, fresh grass, is the best ; 
 failing this, hay is the next preferable food. The greater 
 
MILK 1 1 
 
 the proportion of the nitrogen in the food, the greater is 
 the yield of the milk, the proportion of fat being especially 
 high. Beets, carrots, and swedes throw up the proportion 
 of milk-sugar. Feeding cows with brewers' grains depreci- 
 ates the quality by lowering the total solids of the milk. 
 This objectionable practice of feeding cows with brewers' 
 grains is illegal in the State of Wisconsin, U.S.A. 
 
 7. The Time since the Last Calving. — Milk is not fit for 
 use until the colostrum has disappeared, which does not 
 occur until at least four days after parturition, and it does 
 not become normal in composition until about a month has 
 elapsed. At the end of the milk period, the milk is liable 
 to be very poor in quality, and near calving-time the 
 supply almost ceases. The milk is the best in both quality 
 and quantity from the end of the first month to the begin- 
 ning of the fifth month after calving. Not more than thirty 
 days' milk can be expected. 
 
 8. The Health of the Animal. — As would be expected, the 
 health of the animal has a profound effect on both the 
 quantity and composition of the milk. In the case of 
 diseased udders, the milk is likely to be contaminated with 
 blood, pus-cells, and other morbid products. Milk itself is 
 liable to a number of diseases, particularly those of a 
 bacillary nature. These will be dealt with in detail later. 
 
 The Variations in the Composition of Cows' Milk. — For our 
 knowledge of the composition of milk, we are indebted 
 mainly in this country to the elaborate and prolonged 
 researches .of Adams, Bell, Hehner^ Eichmond, Stokes, 
 Vieth, and many others. As the result of many thousands 
 of analyses, it is found that genuine cow's milk does not 
 normally differ materially from the analysis given as typical 
 on p. 4. 
 
 It is only on the rarest occasions that milk contains less 
 than 3-5 per cent, of fat, while the average is 4 per cent. 
 
12 THE ANALYSIS OF MILK AND MILK -PRODUCTS 
 
 The amount of solids-not-fat may fall from such causes as 
 ill-health, old age, length of time from calving, and in- 
 judicious feeding. Foods such as brewers' grains, maize, 
 and certain roots will increase the quantity of milk given, 
 at the expense of its quality, unless the diet is supple- 
 mented in a suitable manner. 
 
 The evidence that genuine milk contains upwards of 
 3"5 per cent, of fat, and 8"5 per cent, of solids-not-fat, is 
 overwhelming. It is not surprising that the adulteration 
 of milk is so universal, seeing that it is possible to add 
 about 20 per cent, of water, or 30 per cent, of skimmed 
 milk, to milk of average quality, without the resulting 
 mixture falling below the present standard. Some analysts 
 make it a practice to adopt a higher standard — namely, 
 3 per cent, of fat — and report as adulterated all milks 
 which contain less than 3 per cent, of fat. This course is 
 highly to be commended, and it is to be hoped that a legal 
 standard will shortly be fixed. Milk forms, in many cases, 
 the entire diet of children and invalids, and, under the 
 present conditions, it varies so enormously that a doctor, 
 in prescribing so much milk per day, does not know within 
 30 per cent, how much nourishment he is giving, unless, 
 indeed, he realizes this fact sufficiently to insist on the 
 milk being obtained from one of the large dairy companies 
 that exercise a reliable supervision over the purity of their 
 milk. 
 
 The following number of samples of milk were examined 
 by the public analysts during the past six years : 
 
 
 Total Number 
 
 
 Percentage of 
 
 
 E^Eamined, 
 
 
 Adulteration. 
 
 1890 
 
 11,967 
 
 1,530 
 
 12-8 
 
 1891 
 
 12,151 
 
 1,633 
 
 13-4 
 
 1892 
 
 13,633 
 
 1,818 
 
 13-3 
 
 1893 
 
 15,543 
 
 2,310 
 
 14-9 
 
 1894 
 
 16,306 
 
 1,868 
 
 11-5 
 
 1895 
 
 18,307 
 
 2,030 
 
 11-1 
 
MILK 
 
 18 
 
 These figures are not unsatisfactory, but it should be 
 borne in mind that public analysts have had no authorized 
 standard of pure milk to work by during the periods above 
 mentioned, and the diminution in the number of samples 
 reported against cannot, therefore, be regarded as con- 
 clusive evidence that there is less adulteration in milk than 
 formerly. The diminution may to some extent represent a 
 change in the opinion of analysts as to what constitutes 
 adulteration. Moreover, having regard to the enormous 
 volume of the trade in milk, it may be doubted whether 
 the number of samples of this article submitted for analysis 
 is sufficient .to offer a safe basis upon which to form an 
 opinion of the extent of milk adulteration. 
 
 The earlier figures published showing the composition of 
 milk generally give too high a figure for the solids-not-fat ; 
 this was owing to the faulty methods of analysis used failing 
 to extract all the fat. 
 
 The following figures were published by Dr. James Bell, 
 of the Inland Eevenue Laboratory, Somerset House, in 
 1881 (Bell, "Chemistry of Foods," vol. ii.). The analyses 
 of the milk of 236 individual cows are given; all these 
 were said to be thoroughly milked in the presence of a 
 responsible person, and the genuineness of the sample was 
 in each case authenticated. The results were as follows : 
 
 Milk of 236 Individual Cows. 
 
 
 Highest. 
 
 Lowest. 
 
 Average. 
 
 Specific gravity at 15-5 G. ... 
 
 Total solids 
 
 Fat 
 
 Solids-not-fat 
 
 1036-9 
 16-3% 
 6-9 
 11-3 
 
 1026-7 
 10-3% 
 1-9 
 8-0 
 
 1032-1 
 12-8% 
 3-8 
 9-0 
 
 Ash 
 
 Volume of cream after 24 
 hours 
 
 0-87% 
 26 
 
 0-62% 
 2 
 
 0-71% 
 10 
 
 In addition to the above samples. Dr. Bell also published 
 
14 THE ANALYSIS OF MILK AND MILK-PKODUCTS 
 
 the analytical results upon twenty -four dairy samples 
 (mixed milk from herds of cows). The results were as 
 follows : 
 
 
 Highest. . 
 
 Lowest. 
 
 Average. 
 
 Total solids 
 
 Pat 
 
 Solids-not-fat 
 
 14-7% 
 5-1 
 9-9 
 
 11-8% 
 3-2 
 8-5 
 
 13-2% 
 4-1 
 9-0 
 
 Ash 
 
 0-78 
 
 0-63 
 
 0-72 
 
 At the suggestion of the Local Government Board, a 
 further systematic investigation was made in 1892 in the 
 Inland Eevenue Laboratory, Somerset House, under the 
 direction of Dr. James Bell, to inquire into the composition 
 of genuine milk. The investigation embraced the milk of 
 273 single cows, and the mixed milk of fifty-five dairies. 
 In each case the cow was thoroughly milked in the 
 presence of one of the assistant analysts of the depart- 
 ment, so that the genuineness of the sample was substan- 
 tiated. The results contained in the report show a sensible 
 improvement in the quality of the milk as regards the 
 proportion of fat since the investigation which was made 
 ten years previously. The samples included the milk of 
 country farms as well as of town dairies, and were 
 collected over large areas. The breed of the cows, as well 
 as the kind of food, was duly noted. The report, which is 
 dated August, 1893, has been very ably criticized by J. P. 
 Liverseege in a paper in the Analyst for January, 1895. 
 
 The average composition of milk, as shown by the above 
 report, is as follows : 
 
 Avei'ages. 
 
 ' Specific 
 Gravity. 
 
 Total 
 Sohds. 
 
 Fat. 
 
 Solids not- 
 Pat. 
 
 Milk of 273 single cows ... 
 Milk of 55 dairies 
 
 1082-3 
 1031-9 
 
 12-9% 
 12-9 
 
 4-0% 
 40 
 
 8-9% 
 8-9 
 
 The range in quality of the above samples is shown in 
 the following table compiled by J. F. Liverseege : 
 
MILK 
 
 15 
 
 Total iSolids. 
 
 55 Dairies. 
 
 273 Cows. 
 
 Uader 11-5 % 
 
 Milks 
 
 = % 
 
 7 Milks 
 
 = 3 % 
 
 11-5 to H-99 
 
 4 
 
 
 7 
 
 37 
 
 
 14 
 
 12-0 „ 12-49 
 
 6 
 
 
 11 
 
 69 
 
 
 25 
 
 12-5 „ 12-99 
 
 19 
 
 
 35 
 
 44 
 
 
 16 
 
 13-0 „ 13-99 
 
 23 
 
 
 42 
 
 82 
 
 
 30 
 
 14-0 „ 14-99 
 
 3 
 
 
 5 
 
 22 
 
 
 8 
 
 15-0 and over 
 
 
 
 
 
 
 12 
 
 
 4 
 
 
 
 
 
 
 
 
 
 
 55 
 
 
 100 
 
 273 
 
 
 100 
 
 Pat. 
 
 Mnks 
 
 = 0% 
 
 6 Milks 
 
 
 Under 2-75 % 
 
 = 2% 
 
 2-76 to 2-99 
 
 1 
 
 
 2 
 
 12 
 
 
 5 
 
 3-0 „ 3-19 
 
 1 
 
 
 2 
 
 21 
 
 
 8 
 
 3-2 „ 3-49 
 
 8 
 
 
 15 
 
 37 
 
 
 14 
 
 3-5 „ 3-99 
 
 18 
 
 
 32 
 
 75 
 
 
 27 
 
 4-0 „ 4-99 
 
 ' 26 
 
 
 47 
 
 95 
 
 
 34 
 
 5-0 and over 
 
 1 
 
 
 2 
 
 27 
 
 
 10 
 
 
 55 
 
 
 100 
 
 273 
 
 
 100 
 
 Solids-not-Fat. 
 
 Milks 
 
 = 0% 
 
 5 Milks 
 
 
 Under 8-0 % 
 
 = 2% 
 
 8-0 to 8-29 
 
 
 
 
 
 
 10 
 
 
 4 
 
 8-3 „ 8-49 
 
 2 
 
 
 4 
 
 31 
 
 
 11 
 
 8-5 „ 8-69 
 
 6 
 
 
 11 
 
 33 
 
 
 12 
 
 8-7 „ 8-99 
 
 21 
 
 
 38 
 
 95 
 
 
 35 
 
 9-0 „ 9-49 
 
 25 
 
 
 45 
 
 71 
 
 
 26 
 
 9-5 and over 
 
 1 
 
 
 2 
 
 28 
 
 
 10 
 
 
 55 
 
 
 100 
 
 273 
 
 
 100 
 
 Dr. Vieth, in a series of papers contribjited to the Analyst 
 between 1880 and 1891, gives the analyses of no less than 
 120,540 samples of milk. A summary of these results, 
 extending over a period of eleven years, together with 
 curves showing the monthly variations, was published in 
 the Analyst for May, 1892. As the methods of analysis 
 had changed somewhat during this period, the whole of the 
 results were recalculated, so as to make all the analyses 
 strictly comparable amongst themselves. 
 
 The average composition of the above 120,540 samples 
 was as follows : 
 
16 THE ANALYSIS OF MILK AND MILK-PKODUOTS 
 
 Fat 
 
 Solids-not-fat 
 Total solids . . 
 
 4"1 per cent. 
 88 „ 
 12-9 „ 
 
 The above valuable series of analyses has been continued 
 by H. Droop Eichmond, who has published yearly papers 
 in the Analyst from 1892, showing the yearly average com- 
 position. The following table is a synopsis of his results : 
 
 Year. 
 
 Number of 
 
 Samples 
 
 Exumined. 
 
 Yearly Averages. 
 
 Specific 
 Gravity. 
 
 Total 
 Solids. 
 
 Fat. 
 
 Solids-not- 
 Fat. 
 
 1892 
 1893 
 
 1894 
 1895 
 
 23,865 
 28,487 
 28,455 
 24,662 
 
 1032-0 
 1031-8 
 1032-2 
 1032-2 
 
 12-71% 
 1268 
 12 67 
 12-66 
 
 3-91% 
 
 3 91 
 38-6 
 38-4 
 
 8-80% 
 8-77 
 8-81 
 8-82 
 
 The following figures show the greatest variation between 
 the monthly averages of the above two series of analyses 
 from 1881 to 1895 inclusive : 
 
 Total solids 
 
 Fat 
 
 Solids-not-fat 
 
 12-3 to 13 7 per cent. 
 35 „ 4-6 
 8-6 „ 9-1 „ 
 
 It is very satisfactory to see that the recent Select 
 Committee upon Pood Products Adulteration have attached 
 due importance to the above unique series of analyses, 
 representing 226,009 samples of milk, made in the 
 laboratory of Aylesbury Dairy Company by Dr. Vieth and 
 his successor, H. Droop Eichmond. All these samples of 
 milk were taken from healthy cows of all kinds, fed in every 
 kind of manner, during all weathers and times of the year. 
 
 The absolute reliability of these figures as truly repre- 
 senting the composition of genuine cow's milk is fully 
 admitted by all analysts of experience, and the information 
 obtainable from them is amply sufficient to justify the 
 immediate formulation of a legal standard. The recent 
 Select Committee upon Pood Products Adulteration, how- 
 
MILK ;17 
 
 ever, have made no recommendation in this respect, but 
 state that the fixing of a standard would be ' one of the 
 most important duties that would engage the attention of 
 the proposed Court of Eeference.' 
 
 Milk Standards. — There is a very urgent need for a proper 
 legal standard of purity for milk in this country, and it is 
 to be hoped that, in the interest of purchasers, honest 
 tradesmen and public analysts generally, one will soon 
 be adopted. The standard of purity for milk generally 
 understood to be adopted by the chiefs of the Inland 
 Eevenue Department Laboratory, who are the referees 
 under the Sale of Food and Drugs Acts, is as follows : 
 
 Total solids 
 
 ... 11 '25 per cent. 
 
 Fat 
 
 ... 2-75 „ 
 
 Solids-not-fat 
 
 ... 8-5 
 
 This standard is evidently founded upon the poorest quality 
 of milks that have been known to be yielded by cows. 
 There is a certain amount of evidence to show that milk 
 as poor as this has been yielded by separate cows, who 
 were possibly in ill health, or otherwise abnormal, due to 
 improper feeding or other causes ; but it is grossly unfair 
 to judge the quality of milk by the standard of these 
 inferior samples. Milk from single cows is rarely if ever 
 met with in commerce ; almost all the milk which finds its 
 way into the market is the mixed product of two or more 
 animals, which should be above the meagre standard in 
 question. It is only fair for a purchaser of milk to expect 
 this article to be of at least, or not materially below, the 
 average quality. The insufficiency of this standard will be 
 readily seen on comparing it with other standards, when it 
 will be seen to be lower than any other with which we are 
 acquainted. 
 
 - It is not too much to say that the disgi^aceful state of 
 the milk trade in this country is fostered, if not actually 
 
 2 
 
18 THE ANALYSIS OF mtKAND MILK-PRODTJCTS 
 
 caused:, by this ridiculously low. Btandard, by *wiiich> of 
 course, public analysts are compelled to abide, or take the 
 risk ot being overruled by the referees, who are thought to 
 be infallible in the eyes of some magistrates. 
 
 The following table shows some of the standards adopted 
 for milk in various places : : , 
 
 MILK STANDARDS. 1 
 
 
 Percentage by Weight of SoUds. 
 
 state, Town, Authority, etc. 
 
 
 
 
 
 
 
 
 Fat. 
 
 ■ Non-fatty. 
 
 Total Solida. 
 
 Society of Public Analysts 
 
 3-0 
 
 8-5 
 
 H-5 
 
 Inland- Bevenue Department 
 
 2-75 
 
 8-5 
 
 H-25 i- 
 
 Central and Associated Chamber 
 
 
 
 
 of Commerce (proposed standard) 
 
 3-0 
 
 9-0 • 
 
 12-0 
 
 Canada (proposed standard) 
 
 •3-5 
 
 8-5 
 
 12-0 
 
 Paris 
 
 4-0 
 
 9-0 
 
 13-0 
 
 Berne ■ ... 
 
 3-5 
 
 9-0 
 
 12-5 
 
 Treasury iDepartment (U.S.A ) ... 
 
 3-5 
 
 9-5 
 
 13-0 ■ 
 
 Pennsylvania 
 
 3-0 
 
 9-5 
 
 12-5 
 
 PMilttdelpMa 
 
 3-5 
 
 8-5 
 
 12-0 
 
 NetyrYork- 
 
 3-0 
 
 9-0 
 
 120 
 
 Boston (Massachusetts) 
 
 — 
 
 — 
 
 13-0 
 
 New Jersey 
 
 3-0 
 
 9-0 
 
 12-0 : 
 
 Mumesota ... 
 
 3-5 
 
 9-5 
 
 13-0 
 
 Massachusetts 
 
 3-7 
 
 9-3 
 
 13-0 
 
 „ during May and June 
 
 — 
 
 -.- 
 
 12-0 . 
 
 Coliviibus (Ohio) 
 
 3-125 
 
 9-375 
 
 12-5 
 
 Baltimore 
 
 ' — 
 
 . — , 
 
 12-0 C : 
 
 Denver (Colorado) , 
 
 — 
 
 — 
 
 12-0 
 
 Omaha (Nebraska) 
 
 3-0 
 
 9-0 
 
 12-0 
 
 Portland (Oregon) 
 
 — , 
 
 — 
 
 12-0 ■ I 
 
 Lansing (Michigan) 
 
 3-0 
 
 9-5 
 
 12-5 
 
 Madison (Wisconsin) 
 
 3-0 
 
 — 
 
 . — 
 
 Burlington (Victoria) 
 
 3S 
 
 9-0 
 
 12-5 
 
 Des Moines (Iowa) 
 
 3-S 
 
 9-63 
 
 13-13 
 
 The limits adopted by the Society of Public Analysts ar^ 
 as follows : ; 
 
 Total solids 
 Fat ... ■ ..V 
 Solids-not-fat 
 
 11-5 per cent. 
 3-0 • „ 
 8-5 „ 
 
 This standard is very low whefl:,JeompaTed' with the average 
 composition of milk, and should be rfegarded as'a minimumj 
 
MILK .1ft 
 
 and any milk falling below these limits should certainly be 
 regarded as adulterated. It must not be lost sight of that 
 if as low a standard as this were made legal, there would 
 be a great temptation, especially amongst the larger milk 
 contractors, to bring the higher quality milk down to the- 
 level of the standard by means of the addition of water or 
 skiih milk, resulting in the milk supply being brought to a: 
 dead level of inferiority. 
 
 The only basis on which to found a proper standard 
 for niilk is what a' purchaser has a right to expect, milk 
 of average quality. Abnormal samples have nothing to do, 
 with the question at issue. The Courts should accept no 
 excuse for aiiy deficiencies, either natural or acquired,; 
 which result in the sample materially falling below the! 
 average quality of milk, and so not ' of the nature,,suh-! 
 stance, and quality demanded by the purchaser.' 
 
 Abnormal Milks. — ;The causes giving rise to abno^al 
 milks have already: been stated, but no practical object 
 would be attained in this small work by discussing the 
 various samples which analysts have drawn attention to 
 from time to time. A large number of these will be found 
 in the pages of the Analyst. Various papers on abnorma?! 
 milk, and an interesting discussion thereon, will be found 
 in the Analyst for January, 1893. 
 
 Skinmied and Separated Milk. — Bona-fide skimmed milk, 
 in which the cream had been partially skimmed by hand, 
 has ceased to be sold as such, and has now practically dis- 
 appeared from commerce. Next to actual watering, partial 
 skimming of whole milk, or what amounts to the same 
 thing, th« addition of separated milk to whole milk, con- 
 stitutes th« most frequent form, of adulteration of milk. 
 
 Every 10 per cent, of the fat renioved from whole milk 
 represents approximately the loss of half an ounce of cream 
 tiova. the quatt of^ milk, the value, of \yhich is about a 
 
 2—2 
 
20 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 halfpenny, thus proving a very profitable adulteration to 
 the dishonest vendor. 
 
 All cream is now separated from milk by the various 
 mechanical centrifugal separators, the best forms of which 
 do not leave much more than O'l per cent., or at most 0'2 
 per cent, of fat in the separated milk. The solids-not-fat 
 are of course, owing to the removal of the fat, proportion- 
 ally higher than in whole milk. Separated milk containing 
 less than 87 per cent, of solids-not-fat should certainly be 
 regarded as adulterated. Separated milk usually sells at 
 about three-halfpence per quart. 
 
 Butter-Milk. — Butter-milk is the residue after the removal 
 of the butter after churning. It approximates in composition 
 to separated milk. The following analyses are given by 
 Dr. Vieth : 
 
 
 Total Solids. 
 
 Fat. 
 
 Solidsnot-Pat. 
 
 Ash. 
 
 a) 
 
 9-03% 
 
 0-63% 
 
 8-40% 
 
 0-70% 
 
 (2) 
 
 802 
 
 0-65 
 
 7-37 
 
 1-29 
 
 (3) 
 
 10-70 
 
 0-54 
 
 10-16 
 
 0-82 
 
 Frozen Milk. — Partial freezing of milk produces a con- 
 centration of the solids in the part remaining liquid, while 
 the part frozen is deficient in them. In the winter, when 
 the milk has undergone a partial freezing, great care should 
 be taken to allow it to thaw, and then to thoroughly mix 
 the same before allowing it to be sold. 
 
 H. Droop Eichmond (Analyst, 1893, p. 53) has examined 
 a sample of partially frozen milk in which the ice amounted 
 to about 10 per cent. The composition of the frozen and 
 liquid parts were as follows : 
 
 
 Ice. 
 
 Liquid. 
 
 "Water 
 
 ... 96-23% 
 
 95-62% 
 
 Fat 
 
 ... 1-28 
 
 4-73 
 
 Sugar 
 
 ... 1-42 
 
 4-95 
 
 Proteids 
 
 ... 0-91 
 
 3-90 
 
 Ash 
 
 ... 0-21 
 
 0-80 
 
 Specific gravity ... 
 
 ... 1-009 
 
 1-0845 
 
 The proportions that the various constituents bear to each 
 
MILK 
 
 21 
 
 other are not markedly different in the ice and in the 
 portion which is unfrozen. 
 
 Koumiss. — Koumiss is a preparation of mares' or asses' 
 milk in a partly fermented condition, which is largely used 
 in Eussia. It is prepared as follows : The milk is allowed 
 to cool, and is then deprived of a part of its cream, a little 
 yeast is then added ; this sets up a slow fermentation, the 
 milk-sugar being converted into alcohol and lactic acid. 
 During the fermentation the milk is subjected to frequent 
 agitation, the object of which is to retain the casein in 
 suspension, which has a tendency to separate during the 
 fermentation. 
 
 Koumiss is prepared in this country from cows' milk. 
 The following analyses of Russian koumiss by Drs. Bell and 
 Hartier respectively will give an idea of the general com- 
 position of this preparation : 
 
 Lactic acid 
 
 Bell. 
 
 1-96% 
 
 Hartier. 
 1-15% 
 
 Casein 
 
 211 
 
 112 
 
 Sugar 
 
 0-40 
 
 2-20 
 
 Pat 
 
 110 
 
 120 
 
 Alcohol 
 
 2-12 
 
 1-65 
 
 Ash 
 Water 
 
 034) 
 
 91-97 ( 
 
 91'83 
 
 Kephir. — Kephir is a preparation of fermented cows' or 
 goats' mUk, very similar to koumiss, and is prepared by 
 means of a special ferment. It is used largely by the tribes 
 of the Caucasus. 
 
 Methods of Analysis. — An instrument termed Feser's 
 lactoscope, which depends on the opacity of milk, was Used 
 before exact methods of analysis were devised; but both 
 this instrument and the creamometer are too inaccurate to 
 be relied on, or to be used at all, now that we are ih 
 possession of methods at once accurate and expeditious. 
 
 We will now proceed to describe in detail the methods of 
 q,nalysis ordinarily applied to milk. 
 „ Specific Gravity. — Directly the samples are received in 
 
22 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 the laboratory the gravity should be taken, each sample 
 being carefully shaken first, to mix in any cream that has 
 risen, but avoiding the creation of a quantity of air-bubbles. 
 
 The specific gravity may be taken either by a delicate 
 hydrometer or (more exactly) by means of a Westphal 
 balance. 
 
 When milk is sampled, it is important to stir up the 
 whole quantity contained in the vessel thoroughly, so as to 
 ensure a complete distribution of the cream throughout the 
 body of the milk. This is frequently neglected by 
 Inspectors under the Acts, and they should be repeatedly 
 reminded of its necessity. If a sample is to be divided into 
 three parts, one pint should be purchased and poured into 
 bottles of such a size as to fill them almost to the cork. 
 If they are filled entirely it is difficult to shake them so as 
 to mix in the cream ; while if they are not filled up to the 
 neck, the shaking of the milk in transit may cause some of 
 the cream to be churned into butter, in which case there is 
 no alternative but to evaporate the entire quantity, and 
 apply Bell's method of analysis. The term 'Eecknagel's 
 phenomenon ' is applied to the increase observed in the 
 specific gravity of milk, which takes place some time after 
 milking. H. Droop Kichmond (Analyst, 1894, p. 76) gives 
 an example of this, the results of which were as follows : 
 
 Specific Gravity 1 J hours after milking . 
 
 .. 1031-0 
 
 )» J) "1" J) » 1) 
 
 .. 1032-2 
 
 It »i ■'■" ii 11 M 
 
 .. 1032-5 
 
 This rise in the specific gravity must not be confounded 
 with a similar rise when frothy milk is allowed to stand. 
 < Becknagel's phenomenon has been studied by H. Droop 
 Eiehmond (see Analyst, January, 1895), who considers it 
 possible that the explanation may be that a change occurs 
 in the casein due to the action of an enzyme. Another 
 possible explanation is that the small air-bubbles produced 
 jduring milking cause the gravity at first to be below, the 
 
Ma,R 
 
 23 
 
 truth ; or, again,, it is possible, that the vast raultiplication 
 ot bacteria, that occurs in milk with great rapidity, may be 
 the cause. If this is so, the phenomenon should be less 
 aipparent in winter than in summer, as cold retards the 
 growth of all micro-organisms. The samples should be 
 shaken gently just before taking the gravity, so as to mix 
 in cream, but care must be taken to avoid "the creation of a 
 quantity of air-bubbles. The specific gravity of genuine 
 milk generally falls between 1029-0 and 1034-0 at 15'5° C. 
 Tables may be used to correct the gravity taken at tempera- 
 tures near 15-5° C. to what it would have been if the milk 
 had been exactly at 15-5° C, but in practice it is easy to 
 bring them to the correct temperature by standing the 
 bottles in water. 
 
 Table for Correcting the Specific Gravity of Milk according 
 to Temperature (compiled by Dr. Vieth). — Find the tempera- 
 ture of the milk in the uppermost horizontal line, and the 
 specific gravity in the first vertical column ^ In the same 
 line with the latter, under the temperature, is given the 
 corrected specific gravity. Example: Supposing the tempe- 
 rature to be 64°, and the specific gravity (10)34, the 
 specific gravity corrected to 60° is (10)34-6. 
 
 ' 
 
 Degrees of Thermometer (Fahr.). 
 
 ' Observed 
 Specific. 
 
 55 
 
 56 
 
 24-6 
 
 57 
 
 58 
 
 59 
 
 60 
 
 61 
 
 62 
 
 63 
 
 6i 
 
 65 
 
 , 10250 
 
 24-6 
 
 24-7 
 
 24-8 
 
 24-9 
 
 25-0 
 
 25-1 
 
 25-2 
 
 25-3 
 
 25-4 
 
 25'5 
 
 1026-0 
 
 25-5 
 
 25-6 
 
 25-7 
 
 25-8 
 
 25-9 
 
 26-0 
 
 26-1 
 
 26-2 
 
 26-3 
 
 26-5 
 
 26-6 
 
 1027-0 
 
 26-5 
 
 26-6 
 
 26-7 
 
 26-8 
 
 26-9 
 
 27-0 
 
 27-1 
 
 27-3 
 
 27-4 
 
 27-5 
 
 27-6 
 
 1028 
 
 27-5 
 
 27-6 
 
 27-7 
 
 27-8 
 
 27-9 
 
 28-0 
 
 28-1 
 
 28-3 
 
 28-4 
 
 28-5 
 
 28-6 
 
 10290 
 
 28-5 
 
 28-6 
 
 28-7 
 
 28-8 
 
 28-9 
 
 29-0 
 
 29-1 
 
 29-3 
 
 29-4 
 
 29-5 
 
 29-6 
 
 10300 
 
 29-4 
 
 29-6 
 
 29-7 
 
 29-8 
 
 29-9 
 
 30-0 
 
 30-1 
 
 30-3 
 
 30-4 
 
 30-5 
 
 30-7 
 
 10310 
 
 30-4 
 
 30-5 
 
 30-6 
 
 30-8 
 
 30-9 
 
 31-0 
 
 31-2 
 
 31-3 
 
 31-4 
 
 31 -S 
 
 31-7 
 
 10320 
 
 31-4 
 
 31-5 
 
 31-6 
 
 31-7 
 
 31-9 
 
 32 'a 
 
 32-2 
 
 32-3 
 
 32-5 
 
 32-6 
 
 32-7 
 
 10330 
 
 32-4 
 
 32-5 
 
 32-6 
 
 32-7 
 
 32-9 
 
 33-0 
 
 33-2 
 
 33-3 
 
 33 '5 
 
 33-6 
 
 33'8. 
 
 1034-0 
 
 33-3 
 
 33-5 
 
 33-6 
 
 33-7 
 
 33-9 
 
 34-0 
 
 34-2 
 
 34-3 
 
 34-5, 
 
 34;6 
 
 34-8 
 
 10350 
 
 34-3 
 
 34-5 
 
 34-6 34-7 34:9 
 
 35-0 
 
 35-2 
 
 35-3 
 
 35-5 
 
 35-6 
 
 35-8 
 
24 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 Milk samples should be examined at the earliest oppor- 
 tunity, as curdling takes place very rapidly in summer, and 
 directly the milk is at all curdled it is impossible to 
 estimate the gravity accurately. If the curdling has not 
 proceeded too far, it is possible to obtain a uniform mixture 
 again, by adding a few drops of ammonia to the sample, 
 and shaking. This will enable us to measure out the- 
 required amounts by pipette, instead of resorting to the 
 more 'lengthy operation of weighing. The specific gravity 
 of milk is raised by the abstraction of fat, and lowered by 
 the addition of water ; hence, by partial skimming and 
 watering an adulterated sample may possess the same 
 gravity as that of genuine milk. Whether, therefore, the 
 specific gravity is normal or otherwise, it will be necessary 
 to estimate the fat and the total solids. 
 
 Total Solids. — The total solids in milk are estimated by 
 drying 5 grammes in a shallow dish upon a water-bath 
 until constant in weight. A platinum dish is best for this 
 purpose ; but whatever kind of dish is used, it will gene- 
 rally be necessary to dry for three hours on the water- 
 bath, and then for two hours inside a water-oven. Where 
 rigid accuracy is not required, 5 c.c. may be taken for 
 the total solid determination, correction afterwards being 
 made for the weight from the specific gravity of the 
 sample. _ 
 
 Care should be taken to ascertain that the temperature 
 inside the water-oven really reaches 100° C, as some water- 
 ovens give a much lower temperature than this. Some 
 operators prefer to take only 2'5 grammes of milk, and add 
 a drop of very dilute alcoholic acetic acid, which, of course, 
 is driven off by evaporation afterwards ; this addition 
 curdles the milk and prevents the formation of a skin over 
 the surface, which retards the drying. 
 
 The iftilk aolids should be weighed at once after drying 
 
MILK 25 
 
 as they absorb moisture from the air very readily, in 
 addition to the well-known change of weight that takes 
 place with all platinum articles, due to the occlusion of 
 gases on their surface. 
 
 The total solids of genuine milk rarely fall below 12"5 
 per cent. 
 
 The total solids of milk being on the average 12'5 per 
 cent., and the fat 3'5, it follows that the solids-not-fat 
 will be 9;0 per cent. ; of this, about 0*75 is mineral matter, 
 or ash ; about 3"6 per cent, albuminoids, and the remainder 
 4'75 per cent, milk-sugar. 
 
 Estimation of the Ash. — The residue after the determina- 
 tion of the total solids is heated cautiously at a temperature 
 as low as possible over a Bunsen burner until a white ash 
 is left. Over-heating will cause the result to be too low, 
 owing to loss of sodium chloride. The ash of normal milk 
 is about 0"75 per cent., and is slightly alkaline. If the ash 
 is found to be materially less than this, it would point to 
 watering. A marked degree of alkalinity and effervescence 
 with hydrochloric acid would suggest the addition of a 
 carbonate to the milk. 
 
 The ash of milk is of but little value as an indication of 
 purity, owing to the almost universal practice of adding 
 varying quantities of borax or boric acid to the milk, at 
 least during the summer months. 
 
 The most constant figure in normal milks is the propor- 
 tion of ash to solidB-not-fat, which averages 8'3 per cent, 
 of the solids-not-fat. It very rarely exceeds the limits, 
 8"0 to 8"5. In cases of abnormally low solids-not-fat the 
 ash generally bears a.higher ratio. 
 
 Estimation of the Fat. — There are several different 
 methods of determining the fat, which may be divided 
 into the following classes : 
 
 1. Estimation of the fat by reading off the volume of 
 
26 THE ANALYSIS OF MIEKj AND MILK-PRODUCTS 
 
 fat liberated after treatment by chemical means mth the 
 employment ' of " centrifugal force. On this priiiciple all 
 the mfechanical methods are based. 
 
 2. Estimation by simple extraction with a solvent of 
 the milk dried (a) without addition (BfeU's method) ; {&) on 
 blotting-paper (Adams' method) ; (c) on chrysotile fibre 
 (Macfarlane's method). 
 
 3. Estimation by extraction of the fat by. ether from 
 milk, after destruction of the casein by acid (Werner- 
 Schmidt). 
 
 Fig. 1. — Leffmann-Beam Machine. 
 
 Meclianical Methods. — There are several forms of centri- 
 fugal apparatus in use — namely, the Leffmann-Beam 
 machine, the Gerber, etc. Perhaps the one at present 
 most commonly used is the Leffmann-Beam machitie, 
 which is capable of giving results precisely identical with 
 the Adams' process in skilled hands. All these methods 
 depend on the liberation of the fat from the milk by 
 treatment by sulphuric acid. The application of centri- 
 fugal force then causes the heavier liquid to force the 
 
MILK 27 
 
 fat to rise up into the neck, which is graduated into 
 divisions corresponding to a tenth of a per cent, of fat by 
 weight. 
 
 Leffmann-Beam Process. — Leffmann-Beam bottles are con- 
 striicted to hold about 40 c.c, and the neck is graduated to 
 estimate up to 8 per cent, of fat. The procedure is as 
 follows : 16 c.c. of milk are run into the bottle, and 3 c.c. 
 of a mixture of fusel-oil and hydrochloric acid are added, 
 and the bottle carefully shaken so as to cause the liquids 
 to mix, but not so as to shake the milk up into the neck ; 
 9 c.c. of strong sulphuric acid are now added with agitation 
 — the best strength of acid to use is 95 per cent. The milk 
 will turn brown or purple, and the bottle is now full to 
 within 3 or 4 c.c. of the graduations, and fat is generally 
 seen beginning to separate. We now fill the bottle nearly 
 up to the zero mark with a hot mixture of sulphuric acid 
 and water in equal parts, and the bottle is whirled in the 
 machine. Care must be taken to balance the machine 
 properly by placing bottles in each receptacle, or it will 
 vibrate. A minute's whirling is generally sufficient to 
 cause a complete separation of the fat from the acid 
 liquid ; the line of demarcation should be sharp, and 
 both fat and liquid should be quite clear. The reading 
 is made from the extreme top to the extreme bottom of 
 the fat column. If there is a cloudy layer between the 
 fat and the acid liquid, it may be due to using too strong 
 acid, or to careless mixing of the milk and acid. Some 
 samples of fusel- oil cause high results, and some com- 
 parisons should be made whenever a fresh batch of fusel- 
 oil mixture is used, or when Leffmann-Beam bottles are 
 bought, though we have found these latter to be exceed- 
 ingly accurate. The object of using fusel-oil is to cause 
 the fat globules to collect together, which they would not 
 jdo so readily if sulphuric acid were employed alone. This 
 
28 THE ANALYSIS OF MILK AND MILK-PKODUCTS 
 
 method of fat estimation is most useful on account of its 
 rapidity, and, in practised hands, its exactness. 
 
 In summer, when milks curdle soon after they are 
 received, it is a great advantage to be able to deal with a 
 considerable number, and to ascertain within a short time 
 which of them are suspicious and will require further 
 work. 
 
 Fig. 2. — The Geeber Machine. 
 
 Gerber Process. — The Gerber apparatus consists of a 
 hollow disc about eighteen inches in diameter, in which 
 are clips for holding the glass tubes, which, when placed in 
 position, are covered by a plate that fits over the disc, form- 
 ing a hollow box. The disc and its contents are rotated 
 by giving fifteen to twenty sharp pulls with a string, 
 after the manner of a top, and the apparatus, which 
 runs in ball-bearings, then attains a velocity of 2,000 
 
MILK ' '29 
 
 revolutions per minute, and will run by itself for the three 
 minutes which suffice to separate out the fat completely. 
 The procedure, abstracted from the ' Working Instructions ' 
 furnished with the machine, is as follows. 10 c.c. of 
 sulphuric acid, from 1'820 to 1'825 specific gravity, are 
 placed in the bottle ; then 1 c.c. of amyl alcohol is added, 
 and finally 10 c.c. of the milk to be tested. The bottle is 
 then closed with the indiarubber stopper, and shaken 
 till all the ingredients are thoroughly mixed, and the 
 solution changes to a dark brown. TJiey are then rotated 
 in the machine for three minutes, and the fat read oil. If 
 the top of the fat column does not precisely correspond to 
 one of the graduations, it is readily made to do so by 
 moving the indiarubber stopper slightly. 
 
 When skimmed or separated milk is used, the milk must 
 be shaken for two to three minutes, and, after rotating, the 
 test-bottle is to be placed in hot water for a few minutes ; 
 this rotating and warming is to be repeated three separate 
 times. The same procedure applies to condensed milk, 
 working on a 10 per cent, solution. Other milk products 
 are tested thus : 
 
 1. Cream : Take from '5 to 1 gramme, add 6 c.c. hot water, 
 then 1 c.c. amyl alcohol, 6'5 c.c. of sulphuric acid, and 
 shake well ; add 6 c.c. hot water (60° to 70° C), and rotate 
 in the machine ; after a few minutes' rotating place in hot 
 water for a short time, and read off. 
 
 2. Butter ; Weigh out from '5 to 1 gramme, add 12 c.c. cold 
 water, 1 c.c. amyl alcohol, and 6"5 c.c. of sulphuric acid ; 
 shake well and rotate. Place in hot water for a few 
 minutes, and read off. 
 
 8. Cheese : If the cheese is soft, pound it in a mortar ; 
 if hard, pare it finely. Weigh out "5 to 1 gramme, add 6 c.c. 
 hot water, 6'5 c.c. of sulphuric acid, and shake well ; add 
 a further quantity of 7 c.c. of hot water, then run in about 
 
30 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 5 drops of amyl alcohol, close the bottle, and shake well. 
 After having rotated, add 1 c.c. amyl alcohol; then shake 
 gently, and place for a few minutes in hot water (60° to 
 70° C). Eotate a second time, place in hot water, and read 
 off. If skimmed cheese is being examined, rotate three 
 times, and make the second addition of hot water 8 c.c. 
 instead of seven. 
 
 (a) Bell's Method. — This method (used by the referees) is 
 particularly applicable to curdled milks. In such cases the. 
 entire contents of the sample bottle should be poured ou(j 
 into a large platinum dish and evaporated nearly to dryness, 
 When nearly dry, the milk solids are stirred with a glass 
 rod, so as to bring them into a state of fine division. The 
 solids are finally ground with ether and poured on to a 
 filter, and washed with the solvent till free from fat. The 
 ether having been distilled off, the fat is weighed. 
 
 This is without doubt the best way of treating curdled 
 Hiilks, in which much separation of the curd and serum hag 
 often taken place, thus rendering it impossible to . work 
 satisfactorily on a part of the sample. 
 
 (b) Adams' Method. — 5 c.c. of the milk are spotted on to 
 an Adams' paper, which is allowed to dry and then rolled 
 up. One and a half strips of the specially fat-free Adaijis' 
 paper will absorb ,10 c.c. of milk, but it is better not to put 
 more than, 5 c.c, on one paper. After the paper has dried 
 in the air, it is placed in the bath f pr a few minutes' final 
 dryingi and extracted in a Soxhlet extractor. A very'.han'dy 
 form.oj condenser is the hollow metal ball, which is more 
 efficient, than ,a three^foot tube condenser. The fat flask 
 should have a short wide neck, and weigh about 15,gi;anlmes ; 
 its. weight to two places of decimals should be marked on it 
 with a diamond. Sufficient ether (specific gravity . 720) 
 sho;uld be used to fill the Soxhlet. one and a half tinies, 
 and it should be, niade to siphon, ovoe twelve times at leasts 
 
.■',;.. . MIEK-- ..: -i . .. . 31 
 
 It is advisable to place a light screw of paper in the top of 
 the condenser tube to limit as far as possible the entrance 
 of air, which would deposit moisture inside the condenser 
 and wet the ether. Dry ether has no solvent action on 
 milk-sugar, so that nothing but fat will be extracted if the 
 ether is kept dry ; but if it contain water, milk-sugar will 
 come out with the fat, and if much moisture gets into the 
 ether it may' cause the coil to become damp, and then there 
 may be an error in either direction — i.e., an excess owing 
 to milk-sugar being weighed with the fat, or a loss from fat 
 remaining in the damp coil. 
 
 This is certainly the best method of fat extraction for 
 reference work where absolute figures are required. The 
 Adams' process is the of&cial method of the Society of 
 Public Analysts. 
 
 (c) Jffacfarlane's Method. — Macfarlane, chief analyst to -the 
 Canadian Government, dries 10 c.c. of milk on chrysotile 
 fibre, which is packed into a short wide glass funnel. After 
 drying and weighing, a number of these are placed together 
 in a tall glass cylinder and extracted at once with light 
 gasoline. On ire- weighing, the loss of weight' gives the 
 percentage of fat. 
 
 ' (d). Wemer-Sclimidt Method.— The Werner-Schmidt process 
 was first introduced into this country by. A. W. Stokes, a,nd 
 is briefly described in the Chemical News, vol. Iviii., p. 197. 
 In -the same paper (vol. Ix., p. 214) A. Wv Stokes published 
 in detail the exact procedure he recomrnends; together with 
 a table showing the .comparison of estimations made by it, 
 with the amount of fat calculated from the total solids and 
 gravity. We quote the description in hisiown words : 
 
 ' Intojspecjfil tubes (to be obtaineci of Messrs- To wnson 
 ^nd Mercer)£ which are partly graduated up to,' 50 cc, 
 pipette 10 c.e-. of. the milk, if fi;eBh,;ia,nd then pour direct 
 from its bottle, some. HCl rQtfghly.td.theraO c.t;. mark. . If 
 
32 THE ANALYSIS OF MILK AND MILK- PRODUCTS 
 
 the milk is a sour sample, -weigh out 10 grammes, and with 
 a small wash bottle containing strong HCl, wash the milk 
 into the tube till it is full to about the 20 c.c. mark. Now 
 boil the mixture, with frequent shaking, till it turns brown. 
 Merely heating, it to the boiling-point of water by immersion 
 in a water-bath is not sufficient. Leave for about three 
 minutes to stand ; the colour will darken considerably, 
 while thus standing, without further heat. Cool by 
 immersion in water ; fill up roughly to about the 60 c.c. 
 mark with ether. It is not necessary that the ether should 
 have been previously washed with water, unless it contains 
 more than 3 per cent, of alcohol. Cork the tube and shake 
 the mixture for half a minute ; let settle for five minutes. 
 Accurately pipette off 20 c.c. of the supernatant ethereal 
 solution into a tared beaker, evaporate off the ether, dry in 
 air-bath, and weigh the residual fat. It is advisable to 
 take at least 20 c.c. of the ethereal solution, so as to avoid 
 the errors of high multiplication. It is perfectly easy to 
 pipette off accurately 20 c.c. of the ethereal solution ; the 
 presence of the fat in the ether prevents the diflSculty that 
 is found in pipetting off ordinary ether. Now notice how 
 many c.c. of ethereal solution are left in the tube. Here 
 there is sometimes a slight difficulty, since above the sharp 
 line that separates the brown mixture of HCl and milk 
 from the colourless ethereal solution sometimes there floats 
 a fluffy narrow stratum of casein. If, however, three- 
 quarters of this stratum be assumed to be ether, a correct 
 reading will be made. From the whole quantity of ethereal 
 solution originally present the percentage of fat in the 
 milk is now calculated. 
 
 ' The whole process, doing at the same time a number of 
 samples, need not take more than twenty minutes. Its 
 accuracy is not excelled by any other process ; the process 
 is simple ; the reagents are those found in every laboratory ; 
 
MILK 33 
 
 almost all the ether can be recovered, if thought worth 
 
 while; the only special apparatus needed is a cheap 
 
 calibrated tube. 
 
 ' The HCl and the milk in the process should not be 
 
 boiled together more than two minutes, else the ether will 
 
 take up a caramel-like substance. Very highly watered 
 
 milks do not turn a dark brown because of the small 
 
 amount of milk-sugar present, while condensed and sugared 
 
 milks become almost black. 
 
 ' An example may show the calculation required : 
 
 ' 10 c.c. of a milk having the specific gravity 1031, 
 
 and giving 12 per cent, by weight of total solids, when 
 
 treated thus, gave, in 20 c.c. of ethereal solution, 
 
 0'277 grammes of fat. There were left in the tube 6"5 c.c. 
 
 of ethereal solution, making a total of 26'5 c.c. Then, 
 
 0-277 X 26-5 
 
 ^ = 3"67 per cent, m 100 c.c. of the milk. 
 
 Dividing this by the specific gravity 1081, we get 3-55 per 
 cent, by weight of fat in the sample. Calculating from 
 the specific gravity and total solids, the fat should be 
 3*54 per cent.' 
 
 Calculation Method. — Seeing that the presence of fat in 
 milk tends to lower its specific gravity, while the presence 
 of the solids-not-fat tends to raise it, it is evident that 
 there is a relation between the gravity, fat, and total 
 solids, which will enable us to calculate the third factor, 
 if the other two are known. 
 
 Formulae have been devised whereby, knowing two of the 
 factors, the third is easily calculated. The formula of 
 Hehner and Eichmond is the one now exclusively used, 
 this having been based on a very extensive series of 
 observations, in conjunction with perfect methods of 
 analysis. 
 
 The formula is as follows : 
 
 3 
 
34 THE ANALYSIS OF MILK AND MILK-PKODUCTS 
 
 P = -859T--2186G. 
 
 P represents the fat, T the total solids, and G the last 
 two units of the specific gravity, together with any 
 decimal. 
 
 The above formula suffices for ordinary milks, but for 
 skim milks it has been found necessary to slightly modify 
 it as under : 
 
 F = -8591 - •2186G - -05(1 - 2-5). 
 
 This correction need only be applied when G divided by 
 T exceeds 2"5. Extended tables are published (see the 
 Analyst, vol. xiii., p. 26) to facilitate calculation founded 
 on the above formulae, but a still more ready means of 
 using the f ormulse is by means of the invaluable instrument 
 described below. 
 
 Richmond's Slide Rule. — This consists of a wooden rule, 
 part of which is made to slide, and by its aid we can 
 calculate the fat from the total solids and gravity, or 
 the gravity from the total solids and fat, or the total solids 
 from the fat and gravity. The rule has three scales, two 
 of which are for total solids and fat respectively, which 
 are marked on the body of the rule, while that for specific 
 gravity is placed on the sliding part. The divisions are 
 as follows : Total solids, 1 inch divided into tenths ; fat, 
 1-164 inches divided into tenths; specific gravity, each 
 division = 0"254 inches. These numbers show the relation 
 according to the formula 
 
 T--254G = 1164F. 
 
 For example, supposing that a certain milk has a specific 
 gravity of 1032, and is found to contain 3'5 per cent, of 
 fat; the sliding portion of the rule is adjusted, until the 
 arrow points to 3'5 per cent, of fat. The figure represent- 
 ing total solids will now be exactly below the gravity figure. 
 
MILK 35 
 
 and in this particular case will show the total solids to be 
 12*2 per cent., and if we made an actual estimation by 
 evaporation, the figure obtained should vary but little from 
 this. Having obtained the total solids by the slide rule, we 
 subtract the fat, and if the solids-not-fat do not fall below 
 8'5 per cent., an actual estimation of the total solids is not 
 required. 
 
 Total Proteids. — The best method of determining these 
 is by calculation from the total nitrogen, which is deter- 
 mined by Kjeldahl's process. This method, which occupies 
 but a short time, and does not involve the use of com- 
 plicated apparatus, has almost entirely replaced the com- 
 bustion process. 
 
 It depends on the conversion of the nitrogenous matter 
 into ammonium sulphate, which is subsequently decom- 
 posed by an excess of alkali, the liberated ammonia being 
 distilled off and titrated. 
 
 About 5 grammes of the milk are weighed into a round- 
 bottomed, long-necked flask of about 300 c.c. capacity. 
 The milk is then evaporated to dryness over a water-bath, 
 the last traces of water being got rid of by drying in a 
 water-oven and by gently blowing warm air into the flask. 
 To the dry solids are then added 20 c.c. of concentrated 
 sulphuric acid, and 5 to 10 grammes of potassium sulphate, 
 and the whole heated for some time over a Bunsen flame. 
 At first frothing takes place and white fumes escape, con- 
 sisting chiefly of water vapour. 
 
 The flask should be placed in a slanting position, so as 
 to encourage condensation of the sulphuric acid vapours in 
 the neck as far as possible. When the liquid has become 
 clear and colourless, or nearly so, the flask is allowed to 
 cool ; 200 c.c. water is added, and the whole poured into 
 the funnel of the distilling apparatus. A further quantity 
 of about 200 c.c. of water is used to rinse out the flask, 
 
 3—2 
 
36 THE ANALYSIS OF MILK AND MILK- PRODUCTS 
 
 which is also poured mto the funnel, then 75 c.c. of 50 per 
 cent, soda hydrate solution. The stop-cock of the funnel 
 is closed and heat applied ; the ammoniacal steam is freed 
 from splashings by its passage with the glass ' anti-splasher.' 
 With this apparatus the most troublesome substances can 
 be dealt with. 
 
 Fig. 3. — Distillation Apparatus foe Kjeldahl. 
 
 The apparatus is constructed as follows : A copper flask, 
 capable of holding 500 c.c, is fitted with a rubber cork, 
 through which passes a Soxhlet tube, the other end of 
 which is closed by a rubber cork pierced by two holes ; 
 through one of these passes the stem of a tapped funnel, 
 and through the other the end of a block-tin tube, f inch 
 in diameter, which^is carried up about 18 inches, and then 
 down again, its other end passing through a rubber cork 
 
MILK 37 
 
 into a tapered glass connector, which dips into 50 c.c. of y'V 
 sulphuric acid contained in a 4-oz. flask, which is kept cool 
 by being placed in a vessel of cold water. 
 
 The ammoniacal steam condenses largely in the tin tube, 
 and is received in the acid. After about 250 c.c. of dis- 
 tillate have been collected, the stop-cock is opened and the 
 burner turned out. 
 
 The distillate is cooled by placing the flask under the 
 tap, and then titrated with -^-^ soda hydrate, till the excess 
 of acid is neutralized, using methyl-orange as indicator. 
 
 Each c.c. of ^ sulphuric acid neutralized by the ammo- 
 niacal distillate corresponds to "0014 gramme of nitrogen, 
 or "OOIV gramme of ammonia. 
 
 Working with ordinary reagents, it will be found that 
 a ' blank ' experiment usually requires '2 c.c. of ^rr sulphuric 
 acid, hence '2 c.c. should be subtracted from the number of 
 c.c. of ^ soda used to neutralize the 50 c.c. of acid taken. 
 
 The total proteids are found by multiplying the amount 
 of nitrogen found by the factor 6'33. This will give a 
 very close approximation to the total amount of proteids 
 present, as they contain on the average 15'8 per cent, of 
 nitrogen. 
 
 The Estimation of Sugar by Pavy's Method. — This method 
 is a modification of Pehling's process, where, instead of 
 weighing the copper in the form of cuprous oxide, as in the 
 ordinary method, the modification devised by Dr. Pavy 
 depends on the fact that cuprous oxide dissolves in 
 ammonia, forming a colourless liquid. When the sac- 
 charine liquid to be tested is run into the boiling Pavy 
 solution, instead of a bulky red precipitate falling and 
 obscuring the end reaction, the liquid changes from blue to 
 colourless. 
 
 To prepare the test solution, dissolve 20"4 grammes of 
 Eochelle salt and the same weight of caustic potash in 
 
38 THE ANALYSIS OF MILK AND MILK-PBODDCTS 
 
 distilled water ; dissolve separately 4-158 grammes of pure 
 cupric sulphate in more water with heat ; add the copper 
 solution to that first prepared, and when cold, add 300 c.c. 
 of strong ammonia, and distilled water to 1 litre. 
 
 If ordinary Fehling solution is to hand, the Pavy solu- 
 tion is most readily made as follows : To 120 c.c. of the 
 ordinary Fehling solution is added 300 c.c. of ammonia 
 (specific gravity 'SSO), and 400 c.c. 12 per cent, caustic 
 soda solution ; the mixed solutions are then made up to 
 1 litre. 100 c.c. of this solution has the same oxidizing 
 power on glucose as 10 c.c. of the ordinary Fehling solu- 
 tion. 
 
 Pavy's solution possesses a different oxidizing power on 
 maltose and lactose from that exerted by Fehling's solu- 
 tion. Its reaction on invert sugar under the above-named 
 conditions is only five-sixths of that exerted by Fehling's 
 test. Hence 120 c.c. of the latter are employed in making 
 the ammoniacal solution instead of 100, as would be the 
 case if they were strictly equivalent (A. H. Allen). 
 
 In working with Pavy's solution, either 50 or 100 c.c. 
 should be placed in a flask, and the milk, diluted with 
 fairly strong ammonia, run in from a burette, the nose of 
 which passes through a hole in the cork which closes the 
 mouth of the flask. Another tube, just passing through 
 the cork, is connected to a rubber tube leading the 
 ammoniacal vapour out of the window. The Pavy's solu- 
 tion is brought to a boil and maintained just boiling, and 
 then the diluted milk is, run in slowly till the blue colour is 
 just discharged. The reaction does not occur so rapidly as 
 when glucose is brought into contact with Pavy's solution. 
 
 A suitable dilution for the milk would be five parts of 
 milk made up to 100 c.c, with ammonia and water. 
 
 Polarimetric Method of Determination of Lactose. — Milk- 
 sugar can be readily determined by determining the 
 
MILK 
 
 39 
 
 rotation of the sample, after the removal of the fat and 
 proteids. The best form of instrument for the purpose is 
 one of the half-shadow type, for use with the sodium mono- 
 chromatic light. This, destroying all colour, causes a dark 
 shadow to appear when the instrument is used, and so 
 enabling colour-blind persons to use the instrument without 
 difficulty. 
 
 The specific rotatory power of a body, usually expressed 
 as [a]^, is the amount of angular rotation of the ray of 
 
 Fig. 4. — The Polarimeter. 
 
 polarized light in degrees which is produced when a solu- 
 tion of the substance, containing 1 granime in 1 c.c;, is 
 examined in a column 1 decimetre long. It is expressed by 
 the following formula : 
 
 Let o = the observed angle, c the strength in grammes per lOO c.c, 
 and I the length of the tube used in decimetres ; then — 
 
 100a 
 
 [»]" = 
 
 CX.I' 
 
40 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 The calculation of the amount of sugar corresponding to 
 the observed rotation may be determined by substitution in 
 the formula. 
 
 The specific rotatory power of lactose at 20° C. is 52*5° 
 when observed by the sodium flame. The rotatory power 
 is unaffected by the degree of concentration within the 
 limits met with in ordinary analytical work. It is also but 
 slightly affected by temperature, it being decreased by 
 about 0*042° for each successive rise of each degree of 
 temperature* 
 
 When milk-sugar is freshly dissolved in water it exhibits 
 the phenomenon known as ' birotation,' whereby it shows 
 a higher rotation than that given above. By standing, or 
 immediately on boiling and cooling, the rotatory power 
 falls to normal. This precaution is unnecessary when 
 operating upon milk. In preparing solutions of solid milk- 
 sugar care must therefore be taken to raise them to the 
 boiling-point before making up to a definite volume. 
 
 To determine the amount of milk-sugar by means of the 
 polarimeter, the proteids and fat must be removed. This 
 is best done by means of mercuric nitrate, prepared as 
 follows : Mercury is dissolved in an equal weight of nitric 
 acid of specific gravity 1'42; when dissolved, an equal 
 weight of water is added. 
 
 The process is as follows : 60 c.c. of the milk is placed 
 in a 100 c.c. flask, and 1 c.c. of the mercury solution added, 
 and water added up to the mark. The solution, after being 
 well shaken, is filtered through a wet filter. The liquid, 
 which should be quite clear, is now ready for examination 
 in the polarimeter tube. Several readings should be care- 
 fully made, and the average taken. Correction should be 
 made for the space occupied by the fat and proteids. This 
 is done by finding the volume of the fat by multiplying the 
 weight by 1'075, and that of the proteids by 0*8. 
 
MILK 41 
 
 For example : Specific gravity of sample of milk taken 
 1030*0. Pat, 4 per cent. Total proteids, 4 per cent. 
 
 Milk taken =60x l-03 = 61-8 grammes. 
 
 Weight of fat =4 per cent, of 61'8 = 2'47 grammes. 
 
 Volume of fat =2-47 x l-075 = 2-65 c.c. 
 
 Weight of proteids =4 per cent, of 61 '8 = 2 '47 grammes. 
 Volume of proteids=2-47x •8 = 1-97 c.c. 
 .•. The actual bulk of the sugar-containing liquid is : 
 100-(2-65+l-97) = 95-38 c.c. 
 
 In order to avoid the calculation involved in taking 
 60 c.c. of sample as above, an amount may be employed 
 which is a simple multiple of the standard amount of the 
 polarimeter to hand. In case of instruments adjusted so 
 that 16"19 grammes of sucrose ( = 20'56 grammes of milk- 
 sugar) in 100 c.c. of the solution produce 100 degrees on 
 the per cent, scale, therefore, 61"68 ( = 20*56 grammes 
 X 3) grammes of the sample are weighed out, treated with 
 mercury solution, and made up to 100 c.c. The joint 
 volume of the fat and proteids is calculated as above, and 
 the same volume so found is added to the 100 c.c. The 
 flask is well shaken and filtered as before. The bulk of the 
 sugar-containing liquid will now be exactly 100 c.c, and 
 the polarimeter readings divided by 3 (if a 200 mm. tube 
 be used) will give the percentages of hydrated lactose. 
 
 Detection and Estimation of Preservatives in Milk. — The 
 following are those which are most commonly employed : 
 (1) Borax and boric acid ; (2) Formaldehyde ; (3) Salicylic 
 acid ; (4) Potassium chromate. 
 
 This is the order in which we have most frequently met 
 with them, but there is no doubt that formaldehyde, being 
 the most effective, when it becomes more generally known, 
 will supersede boric acid as a preservative. Mixtures of 
 the above are sold to farmers and dairymen under pro- 
 tected names, with directions how, and in what quantity, 
 they are to be added. 
 
42 THE ANALYSIS OF MILK AND MILK-PEODtJCTS 
 
 The following table of the relative efSciency of various 
 preservatives, compiled by E. T. Thomson {Analyst^ 1896, 
 p. 65), is of value. The author kept measured quantities 
 of the same milk, to which the various preservatives were 
 added, in stoppered bottles, under identical conditions, and 
 examined them from time to time. 
 
 'Pl'PSdVWa +117(3 
 
 Grains of 
 Preserva- 
 
 After 
 
 After 
 
 After 
 
 After 
 
 After 
 8 days. 
 
 After 
 11 days. 
 
 used. 
 
 tive per 
 
 standing 
 
 standing 
 
 standing 
 
 standing 
 
 Lactic 
 
 Lacti 
 
 
 gal. o£ 
 milk. 
 
 2 days. 
 
 4 days. 
 
 6 days. 
 
 7 days. 
 
 Acid, 
 
 Acid, 
 
 
 
 
 
 
 per cent. 
 
 per cent. 
 
 (Pure milk) 
 
 — 
 
 Dis- 
 tinctly 
 turned 
 
 Slightly 
 sour 
 
 Sour 
 
 Sour and 
 curdled 
 
 •68 
 
 •71 
 
 Formic al- 
 
 8-75 
 
 Sweet 
 
 Sweet 
 
 Sweet 
 
 Sweet 
 
 Sweet 
 
 Sour and 
 
 dehyde 
 
 •0125% 
 
 
 
 
 
 •12 
 
 curdled 
 
 (40%) 
 
 
 
 
 
 
 
 •43 
 
 j; )) 
 
 17-5 
 •025% 
 
 Sweet 
 
 Sweet 
 
 Sweet 
 
 Sweet 
 
 Sweet 
 •10 
 
 Sweet 
 •14 
 
 5) )> 
 
 35 
 •05% 
 
 Sweet 
 
 Sweet 
 
 Sweet 
 
 Sweet 
 
 Sweet 
 •07 
 
 Sweet 
 
 •10: : 
 
 Boric, acid 
 
 35 
 •05% 
 
 Sweet 
 
 Sweet 
 
 Turned 
 
 Sour and 
 curdled 
 
 •42 
 
 •52 
 
 Boric acid 
 
 35 
 
 Sweet 
 
 Sweet 
 
 Sweet 
 
 Sweet 
 
 Sweet 
 
 . Sour 
 
 -f borax 
 
 (17^5 of 
 
 
 
 
 
 •10 
 
 •32 
 
 (calcu- 
 
 eacli) 
 
 
 
 
 
 
 
 lated to 
 
 
 
 
 
 
 
 
 boric acid) 
 
 
 
 
 
 
 
 
 Salicylic 
 acid 
 
 17^5 
 
 Sweet 
 
 Sweet 
 
 Sweet 
 
 Turned 
 
 Sour 
 
 •42 
 
 •025% 
 
 
 
 
 
 •26 
 
 
 )» 
 
 35 , 
 
 Sweet 
 
 Sweet 
 
 Sweet 
 
 Sweet 
 
 Sweet 
 
 Sonr 
 
 
 •05% 
 
 
 
 
 
 •10 
 
 •33 
 
 Benzoic 
 
 17^5 
 
 Sweet 
 
 Sweet 
 
 SHghtly 
 
 Sour 
 
 Sour 
 
 •52 
 
 acid 
 
 •025% 
 
 
 
 turned 
 
 
 •45 
 
 
 (1) Borax and Boric Acid, NajBAlOHsO; H3BO3.— 
 The above table shows that a mixture of boric acid and 
 borax in equal parts is more efficient than an equal weight 
 of boric acid. 
 
 The presence of either may be demonstrated by ashing 
 a few c.c.'s of the milk, and after moistening with strong 
 sulphuric acid and alcohol, applying a light, when a green 
 colour is imparted to the flame, which is more easily seen 
 by placing the dish in the dark, if either boric acid or 
 
MILK 43 
 
 borax be present. This reaction is fairly delicate, for as 
 small a quantity as 0"01 per cent, can be detected in this 
 manner. 
 
 Another and slightly more delicate mode of testing for 
 these is by treating the ash with dilute hydrochloric acid, 
 and on placing a small piece of freshly prepared turmeric 
 paper in the dish, it will turn reddish-brown, which is 
 turned to bluish-black when treated with a little sodium 
 carbonate solution. 
 
 Boric acid in milk may be estimated by Thomson's 
 method as follows : One to two grammes of sodium hydrate 
 are added to 100 c.c. of milk, and the whole evaporated to 
 dryness in a platinum dish. The residue is thoroughly 
 charred, heated with 20 c.c. of water, and hydrochloric acid 
 added drop by drop until all but the carbon is dissolved. 
 The whole is transferred to a 100 c.c. flask, the bulk not 
 being allowed to get above 50 to 60 c.c, and 0'5 gramme 
 dry calcium chloride added. To this mixture a few drops 
 of phenolphthalein solution are added, then a 10 per cent, 
 solution of caustic soda, till a permanent slight pink colour 
 is perceptible, and finally 25 c.c. of lime-water. In this 
 way all the phosphoric acid is precipitated as calcium 
 phosphate. The mixture is made up to 100 c.c, thoroughly 
 mixed, and filtered through a dry filter. To 50 c.c. of the 
 filtrate (equal to 50 grammes of the milk) normal sulphuric 
 acid is added till the pink colour is gone, then methyl 
 orange, and the addition of the acid continued until the 
 yellow is just changed to pink. Fifth-normal caustic soda 
 is now added till the liquid assumes the yellow tinge, excess 
 of soda being avoided. At this stage all acids likely to be 
 present exist as salts neutral to phenolphthalein, except 
 boric acid, which, being neutral to methyl orange, exists in 
 the free condition, and a little carbonic acid, which is 
 expelled by boiling for a few minutes. The solution is 
 
44 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 cooled, a little phenolphthalein added, and as much 
 glycerine as will give at least 30 per cent, of that substance 
 in the titrated solution, and titrated with fifth -normal 
 caustic soda till a distinct permanent pink colour is pro- 
 duced; each c.c. of fifth-normal soda is equal to 0-0124 
 gramme crystallized boric acid. A series of experiments 
 with this process showed that no boric acid was precipitated 
 along with the phosphate of calcium so long as the solution 
 operated upon did not contain more than 0"2 per cent, of 
 crystallized boric acid, but when stronger solutions were 
 tested, irregular results were obtained. The charring of 
 the milk is apt to drive off boric acid, but by carefully 
 carrying the incineration only so far as is necessary to 
 secure a residue which will yield a colourless solution, no 
 appreciable loss occurs. 
 
 A modification of Gooch's method is recommended by 
 C. E. Cassal {Analyst, 1890, p. 231), and is performed as 
 follows : In the case of cream about 50 grammes, and in 
 that of milk about 100 grammes of the sample are rendered 
 alkaline with caustic soda, evaporated to dryness and 
 incinerated. The ash, which need not of course be burnt 
 white, is ground up and transferred, washing in with a 
 little methyl alcohol and a few drops of water, to a conical 
 flask of from 200 c.c. to 300 c.c. capacity, provided with a 
 doubly-perforated caoutchouc cork, through which pass a 
 stopcock-funnel tube and a delivery tube. The flask is 
 attached to an ordinary condenser by means of a flexible 
 joint, in such a way as to admit of its being shaken round 
 occasionally, and is placed in an oil-bath. The mixture in 
 the flask is acidified with acetic acid, and 5 c.c. of methyl 
 alcohol is run in from the funnel tube and distilled. The 
 distillate is received in a weighed amount of pure lime 
 (ignited to constant weight). About 1 gramme is a con- 
 venient quantity. The lime is contained in a platinum 
 
MILK 45 
 
 dish capable of holding about 70 c.c, which is placed in a 
 glass receiver. 
 
 This receiver may be made from a ground-edged glass 
 vessel, covered with a perforated ground-glass plate. The 
 end of the condenser passes through a grooved or closely- 
 fitting cork in the perforation of the plate, and terminates 
 just above the lime in the platinum dish. 
 
 Distillation with 5 c.c. of methyl alcohol is repeated 
 several times, ten such treatments being ample for all 
 likely cases ; fewer are generally quite sufficient. A gela- 
 tinous mixture is obtained in the platinum dish ; it is well 
 stirred, and allowed to stand for some minutes ; it is then 
 evaporated in an oil-bath, and finally heated over the blow- 
 pipe flame to constant weight. The increase of weight is 
 boric acid, and the error does not in any case amount to 
 more than 1 milligramme plus, when dealing with from 0"1 
 to 0*25 or 0"3 gramme of anhydrous boric acid. To make 
 sure that the acid has been completely volatilized, the 
 residues iu the distillation-flasks are tested with turmeric- 
 paper. 
 
 The few drops of water added before the first distillation 
 greatly increases the rapidity of the removal of the boric 
 acid. 
 
 (2) Formaldehyde, CH2O. — Formaldehyde is, compara- 
 tively speaking, a new food preservative, and is by far the 
 most effective ; consequently a smaller amount being 
 required to preserve the milk, renders its detection and 
 estimation a somewhat difficult matter. 
 
 It is usually added in the form of a 40 per cent, solution, 
 which is called ' Formalin.' Two or three drops in a pint 
 of milk keeps it fresh for three or four days, and the 
 addition of 0"05 per cent, preserves milk for months. 
 
 In the trade a much more dilute solution of formaldehyde 
 is generally employed, namely, 1 part of formaldehyde to 
 
46 THE ANALYSIS OF MILK AND MILK-PBODUCTS 
 
 80 of water. S. Eideal states that a quarter of a pint of 
 such solution added to 17 or 18 gallons of milk keeps it 
 fresh for at least three days, and does not communicate 
 any smell or taste to the milk. 
 
 Formaldehyde may be detected by the following methods : 
 
 On tasting milk containing formaldehyde, a peculiar 
 sensation is noticed at the back of the throat, and when 
 strong hydrochloric acid is added to it (as in the Werner- 
 Schmidt process) the casein ■ turns yellow, and is less 
 soluble than that of pure milk. 
 
 The most reliable test is one pointed out by Otto Hehner. 
 It is based upon the fact that when milk, formaldehyde, 
 and sulphuric acid are mixed together a blue coloration 
 is formed. The best method of applying the test, according 
 to Eichmond and Boseley, is to dilute the milk with an 
 equal bulk of water and add sulphuric acid of 90 to 94 per 
 cent, strength. Under these conditions milk, in the absence 
 of formaldehyde, gives a slight greenish tinge at the junction 
 of the two liquids, while a violet ring is formed when 
 formaldehyde is present. This colour is permanent for 
 two or three days. In the absence of formaldehyde a 
 brownish-red colour is developed after some hours, not at 
 the junction of the two liquids, but lower down in the acid. 
 It cannot be mistaken by anyone who has had any ex- 
 perience with the test for the formaldehyde reaction. 
 
 It is stated that 1 part of formaldehyde in 200,000 parts 
 of milk can be easily detected by means of this test, but 
 the blue coloration is not obtained with milks containing 
 over 0"5 per cent. This method is represented to be much 
 more delicate than Schiff's test, which is as follows : 
 
 The reagent used is a solution of magenta decolorized 
 by sulphurous acid, care being taken not to have an excess 
 of sulphurous acid. This is added to the whey obtained by 
 coagulating the casein with a little dilute sulphuric acid 
 
MILK 47 
 
 andfiltering, the presence of formaldehyde being indicated 
 by a red coloration. 
 
 Otto Hehner prefers to conduct the operation by adding 
 five drops of the reagent to the distillate from 100 c.c. of 
 milk (amounting to about 25 c.c), placing the mixture in 
 a stoppered cylinder, and to observe the colour next morning, 
 and then to add a few drops of sulphurous acid solution. 
 After a short time any colour which may be due to oxidation 
 will have vanished, while that due to the presence of an 
 aldehyde remains. There is certainly a difference in the 
 tint produced by colour oxidation, which resembles that 
 of rosaniline, and that of the aldehyde compound, which 
 is violet ; and with those small traces with which we have 
 often to deal, only a comparison of the relative colours 
 would allow of anything like a safe conclusion being drawn. 
 
 Another test suggested by Otto Hehner is the following, 
 which is equally sensitive as the foregoing : 
 
 If to the distillate from a sample of milk, etc., one drop 
 of a dilute aqueous solution of 'phenol is added, and the 
 mixture poured upon strong sulphuric acid contained in a 
 test-tube, a bright crimson colour appears in the zone of 
 contact. This colour is still readily seen with one part of 
 formaldehyde in 200,000 of water. If there is more' than 
 one part in 100,000, there is seen above the red ring a 
 white, milky zone, while in stronger solutions a copious 
 white or slightly pink, curdy precipitate is obtained. This 
 reaction has the advantage of the one above referred to, 
 inasmuch as it is obtained with formaldehyde solutions 
 of all strengths, while the blue colour with milk is not 
 obtained with milk containing much formaldehyde. 
 
 Acetaldehyde also gives a coloration and a precipitate 
 with phenol and sulphuric acid, but it is orange-yellow, 
 not crimson. 
 
 Trillat, Kleeberg, and Plochl's tests are useful as con- 
 
48 THE ANALYSIS OF MILK AND MILK-PEODUCTS 
 
 firmatory ones, but are less sensitive than those mentioned 
 above. 
 
 E. T. Thompson has recently made experiments with the 
 object of proving the presence of this substance in milks, 
 and has found that a modification of the well-known re- 
 action with ammonia nitrate of silver gives a good indication 
 of its presence. To apply the test 100 c.c. of the milk are 
 carefully distilled until (say) 20 c.c. of distillate comes 
 over ; this is transferred to a stoppered tube, and about 5 
 drops of ammonia silver nitrate added. (This solution is 
 prepared by dissolving 1 gramme of silver nitrate crystals 
 in 30 c.c. of distilled water, adding dilute ammonia till the 
 precipitate at first formed is redissolved, and then making 
 up to 50 c.c. with water.) The mixture of the milk distillate 
 and the silver solution is now allowed to stand for several 
 hours in a dark place (as much as twelve to eighteen hours 
 may be necessary if very little formic aldehyde is present), 
 when, if formic aldehyde is present, a strong black colour 
 or deposit will be produced. A light brown colour should 
 be disregarded ; but, so far as his experience goes, the pro- 
 duction of a decided black under these circumstances is 
 only brought out by formic aldehyde, but possibly by other 
 aldehydes also. The usual method of heating with the 
 silver solution in order to obtain a silver mirror is of no 
 value with weak solutions of formic aldehyde. It was 
 found that genuine milks from various sources, when tested 
 by the method described, gave no reaction whatever, even 
 when the distillate was left mixed with the silver solution 
 for twenty-four hours ; or at most gave a slight brown 
 tinge. When as little as 2 grains of the 40 per cent, 
 formalin were added to 1 gallon of milk (which before 
 addition gave no reaction with this process), the distillate 
 from 100 c.c. gave a decided black colour, or deposit, 
 intense enough to render the mixture quite opaque. As 
 
MILK 49 
 
 2 grains per gallon is a quantity of formalin which would 
 be of little value in the preservation of milk, it is evident 
 that this method of testing is quite delicate enough for the 
 purpose. It ought to be noted that, if a milk contains 
 about 2 grains of formalin per gallon, the 20 c.c. distillate 
 from 100 c.c. of the milk appears to contain all the formic 
 aldehyde that will distil over, and distillates after that give 
 practically no reaction. A milk containing 7 or 8 grains 
 per gallon of the preservative may require the distillation 
 to be carried on till 30 or 40 c.c. are collected before it 
 ceases to show a reaction with the silver solution ; but in 
 all cases the reaction can be got by distilling over the 20 c.c, 
 or indeed 10 c.c. 
 
 (3) Salicylic Acid, C^HgOs. — The use of this acid as a 
 milk-preservative is gradually diminishing in this country, 
 though it is still largely employed on the Continent. 
 
 It is best detected by H. Pellet's method as given in 
 Allen's Commercial Organic Analysis, vol. iii., part i., p. 53. 
 200 c.c. of the milk are diluted with an equal measure of 
 water, then heated to 60° C, and treated with 1 c.c. of 
 acetic acid and an excess of mercuric nitrate free from 
 mercurouB salt. The salicylic acid is extracted from the 
 filtered solution by agitation with ether, and recognised by 
 evaporating a little of the ethereal solution to dryness, and 
 testing the residue with ferric chloride, which gives a violet 
 coloration with salicylic acid. 
 
 (4) Potassium Chromate, K2Cr20^. — We have no knowledge 
 of this substance occurring as a milk-preservative in this 
 country, but it is stated to be used on the Continent. It 
 is needless to point out the great danger to public health 
 which would accrue from the use of this highly dangerous 
 salt. 
 
 The following abstract of a paper by J. Froidevaux (Jour. 
 Pharm. Chim., 1896, pp. 155-158) appeared in the Analyst, 
 
 4 
 
50 THE ANALYSIS OF MILK AND MILK-PEODUCTS 
 
 1896, p. 285 : The author describes experiments made to 
 determine what amount of neutral potassium chromate is 
 necessary for preserving milk for an appreciable time, and 
 finds that at least 0*2 gramme per litre is required, an 
 amount which gives to the milk an intense and absolutely 
 abnormal colour. He considers that the proportion of 
 2 grammes to 50 litres of milk, which (according to Deniges) 
 the retailers of Bordeaux employ, is altogether insufficient 
 to retard coagulation. 
 
 With regard to the detection of chromates in milk, the 
 method of Deniges (addition of 1 c.c. of a 2 per cent, solution 
 of silver nitrate to 1 c.c. of milk) is satisfactory when the 
 amount of chromate exceeds O'Ol gramme per litre ; when 
 below this the colour is masked by the precipitated phos- 
 phates. The following method is preferred : The ash from 
 10 c.c. of milk is dissolved in a few drops of water acidified 
 with nitric acid, neutralized with magnesium carbonate, 
 and. the silver nitrate solution (preferably 20 per cent.) 
 added. 
 
 As a control test the ash from 10 c.c, of milk is taken up 
 in a few drops of water slightly acidulated with sulphuric 
 acid, and tincture of guaiacum added little by little. An 
 intense blue colour, which rapidly disappears, is produced 
 when chromates are present. The reaction will detect 0*02 
 to 0'03 gramme of chromate per litre. 
 
THE BACTERIOLOGT OF MILK 51 
 
 THE BACTERIOLOGY OF MILK. 
 
 Milk usually contains a large number of bacteria derived 
 for the most part from the external surroundings of the 
 cow. Where these are unclean, the number may reach 
 three or more millions per c.c. These can for experi> 
 mental purposes be completely separated by filtration 
 through Pasteur tubes, the tubes being cleaned at short 
 intervals. A thin watery serum constitutes the filtrate, the 
 whole of the fat being arrested with the organisms, so that 
 milk cannot therefore be freed from organisms for practical 
 purposes by any known system of filtration. 
 
 Milk is capable of being curdled by ferments, even in the 
 absence of an. acid reaction. The most notable of these 
 ferments is rennet, obtained from the stomach of a calf. 
 On slightly warming the milk, thick lumps of paracasein 
 are thrown down, and whey is separated. This action does 
 not occur when the milk has previously been boiled, but 
 it can then be induced by the addition of a little acid, 
 even COg. 
 
 Duclaux {Comptes Rendus, 1891) and Hueppe {Deutsche 
 Mediz. Wochensch., pp. 48, 49) first pointed out that such 
 ferments are conveyed by many different bacteria, which 
 precipitate the casein in the presence of a weakly acid, 
 amphioteric, or even neutral reaction. The numerous 
 tyrothrix bacilli isolated by Duclaux, the Bacillus pyo'- 
 cyaneus, yellow sarcina, and particularly the organisms 
 described by Pliigge (Zeitsch. f. Hygiene, xvii., p. 272), and 
 characterized by their capacity to peptonize milk, belong to 
 this class. Cohn (CentralU. f. Bacteriologie, ix., p. 653) pro- 
 duced the precipitation even by means of bacteria of which 
 
 4—2 
 
52 THE ANALYSIS OP MILK AND MILK-PRODUCTS 
 
 the vegetative capacity had been completely abolished with 
 chloroform, thus showing that the fermentative action was 
 due to a substance independent of the metabolic products 
 of the organism. These substances have been isolated by 
 Cohn and others ; they are destroyed in most cases at from 
 65° to 75° C. Some ferments, however, as, for example, 
 that described by Gorini {Hyg. Bundsch., 1893, p. 381), in 
 association with the Bacillus prodigiosus, resist as much as 
 an hour's exposure to 70° to 80°, and require at least half 
 an hour's exposure to 100° C. for their destruction. 
 
 The amount of the ferment varies with the species and 
 age of the culture and also with the temperature, much 
 more coagulating ferment being obtained at 20° than at 
 37°. The ferment works, however, as does rennet, much 
 more strongly at 37° than at lower temperatures. Its 
 action is impeded by the presence of alkalies. When 
 tryptic ferments are produced simultaneously, the coagu- 
 lating ferments, which are developed more slowly, may fail 
 to work, the casein being peptonized before the coagulat- 
 ing ferment has acted. 
 
 The chemistry of the action of the coagulating ferments 
 is mainly due to Einger {Journal of Physiology, xi., p. 464). 
 It probably occurs in two stages. In the first place, the 
 casein, or, as Einger considers, the caseinogen, is decom- 
 posed into one or more unknown intermediate products, 
 and in the second place these products are precipitated by 
 the lime-salts in the milk. Barium or magnesium salts 
 will have the same effect, but not the salts of the lighter 
 alkalies. This conception of the process explains the fact 
 that casein freed from sugar, fat and ash is not precipi- 
 tated by the ferment until some calcium salt is added. 
 The conditions of the two parts of the process are different. 
 The change in the casein is delayed by low temperatures, 
 stopped by alkalies, and hastened by dilute acids ; while 
 
THE BACTERIOLOGY OF MILK 55 
 
 the combination with calcium salts occurs readily, even at 
 freezing-point and in weakly alkaline solutions. After, 
 therefore, the ferment has acted on the casein, it may be 
 destroyed by alkalies without affecting the subsequent 
 coagulation ; and in this way it becomes clear that the 
 second part of the process is entirely independent of the 
 action of the ferment. 
 
 It is practically inevitable that milk as delivered from 
 the cow should contain a number, and usually a very large 
 number, of bacteria. The extent of their presence is, 
 however, affected by many circumstances, of which some 
 are also indications of unwholesomeness or danger. Many 
 of the organisms which are capable of causing disease do 
 so by producing toxic decomposition - products from the 
 mUk. Their vegetative capacity increases greatly with a 
 rise of temperature ; and it is therefore an essential con- 
 dition of sanitary milk-production, and especially of the 
 designation of a milk as suitable for children, that it should 
 be kept at a low temperature during the whole of • the in- 
 terval between being drawn and being consumed. 
 
 Milk has a great capacity for combining with aromatic 
 substances, and if kept in dirty or ill-ventilated surround- 
 ings will lose its characteristically pleasant flavour, a cir- 
 cumstance of which account should be taken in examina- 
 tion of samples. 
 
 There are very many so-called diseases of milk due to 
 the presence of various organisms. Thus, milk may be 
 ropy, bitter, slimy, blue, soapy, red, yellow, and even 
 putrefying ; and in each case the disease appears to be 
 caused by one, and possibly by more than one, organism. 
 Most of these organisms are probably derived from the 
 surroundings of the milk, and their presence can be abso- 
 lutely avoided by proper cleanliness in the biacteriological 
 sense. Thus, for instance, Adametz (Milch. Zeitung, 1889, 
 
54. THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 p. 48) found in two streams large numbers of a bacillus 
 which had the property of producing a high degree of 
 ropiness in sterilized milk ; so that under microscopical 
 examination no trace of the structure of the fat corpuscles 
 could be observed, although the fat had not been decom- 
 posed. This water may either have been used to wash out 
 the utensils, or. grass moistened with it may have been 
 conveyed into the stables as hay, and the organisms' have 
 found their way into the milk through the dust. This 
 explanation is not inconsistent with the fact that the 
 trouble of ropy milk has been known to disappear on the 
 substitution of a good for a bad fodder. The recognition 
 of ropy milk can be readily made by taking a few drops 
 between the fingers, when it will be found capable of being 
 drawn out into threads. 
 
 Bitter milk is recognisable by the taste, which is very 
 often also mouldy and accompanied by coagulation, though 
 the latter phenomenon may not occur until the milk is 
 warmed. Several organisms have been isolated which 
 have the capacity of producing bitterness in milk, notably 
 by Hueppe and Loffler. The bacillus of bitter milk proper 
 is that of Bleisch (Zeitschr. f. Hyg,, xiii., p. 81) — a facul- 
 tative anaerobic organism, consisting of a stout rod with 
 bundles of flagella, rapidly liquefying gelatine, producing a 
 thin, flat, grayish growth on agar and potato. 
 
 In milk it will after a week produce transparent yellow 
 streaks below the cream, the milk itself coagulating, and 
 the coagulum being subsequently, earlier or later, almost 
 completely dissolved. The bitter taste arises after the 
 second week ; there is no smell ; the reaction is acid. At 
 higher temperatures the milk becomes bitter, and gives the 
 biuret reaction after twenty-four hours, while spores are 
 produced which resist boiling for six hours. The micro- 
 coccus, of bitter milk is that of Qoh.n{Centralhl.f. Bakt., 
 
THE BACTERIOLOGY OF MILK 55 
 
 ix., p. 653), which coagulates milk, and then forms it into 
 a slimy solution, with a slightly sour and very bitter taste. 
 
 The peptonizing organisms were minutely investigated 
 and classified by Flugge {Zeitschr. f, Hyg., xvii., pp. 272, 
 342). In general, they require many hours' boiling for 
 their destruction ; boiling for an hour, for instance, being 
 sufficient to destroy the lactic acid and butyric acid bac- 
 teria, but not the peptonizing organism to which those of 
 bitter milk belong. Their development is greatly favoured 
 by warmth, so that they multiply with great rapidity at 
 room-temperature in summer time. It is not easy to 
 detect their action on milk with the eye, the commence- 
 ment of the peptonizing giving at the utmost a thin, clear 
 layer under the cream. Experimentally, several of these 
 organisms have been shown to produce serious intoxica- 
 tion, and it is some organisms of this group which are 
 responsible for infantile diarrhoea. Their significance is 
 probably even greater than this fact implies, because, if 
 not sufficient to produce acute specific disease in adults, 
 *hey or their decomposition-products in milk are liable to 
 set up digestive derangements which render the consumer 
 more accessible to other diseases of intestinal origin. 
 
 Slimy milk is attributable to various organisms. The 
 Micrococcus viscosus of Schmidt - Miihlheim {Archiv. f, 
 Physiol., xxvii., p. 490) is of 1 /a diameter, often occurs in 
 wreathed chains of fifteen or more cells, and gives a slime 
 analogous to that of plants, and derived from the milk- 
 sugar. The process seems to differ from that of slime- 
 production in wine in that it forms no mannite and no 
 carbonic acid. Hueppe also isolated a coccus, and numerous 
 bacilli have been discovered, all possessing the property of 
 making milk slimy. In the case of Guillebeau's Bacillus 
 Hessii (Ann. de Microg., iv., p. 225) the slimy character 
 disappears after two days' exposure to 35° C. The Bacillus 
 
56 THE ANALYSIS OF MILK AND MILK-PEODUCTS 
 
 lactis viscosus of Adametz mentioned above is notable for 
 the length of time daring which it operates, and the com- 
 pleteness with which it attacks the milk. Its effect is 
 apparent after four or five days, and is continued for four 
 weeks, by which time the milk-corpuscles have practically 
 disappeared and the milk is transparent. The casein is 
 not precipitated ; no acceleration of the process occurs on 
 a rise Of temperature, and there is no special smell. The 
 Bacillus lactis pituitosi of Loffler (Berlin Klin. Wochenschr., 
 1887, p. 631), on the other hand, gives a specific smell, 
 and renders the milk slimy and slightly acid, especially at 
 the lower part. Whether the viscous substance is derived 
 from the milk-sugar or the casein has not been determined. 
 
 The Bacillus cyanogenus, or Bacteriwm syncyaneum of 
 Ehrenberg, appears to cause blue milk. It is described at 
 length byHueppe {Mitth. a.d.k. Gesmidheitsavit, ii., p. 335), 
 and by Heim {Arbeiten. a.d.k. Gesundheitsamt, v., p. 518), 
 and figured in Lehmann and Neumann's Atlas. It gives 
 an alkaliae reaction, and produces neither coagulation nor 
 acidity. It usually yields two pigments, one of the ordinary 
 fluorescent type, and the other of a bluish to grayish colour, 
 which becomes more strongly blue up to azure in unsteri- 
 lized milk with an acid reaction. The addition, for instance, 
 of Bacillus acidi lactici a day or two after syncyaneus has 
 been introduced into the milk, shows this very clearly. 
 The addition of soda or potash produces a pink coloration. 
 Numerous other organisms are also capable of turning milk 
 blue. SchoU (Fortschr. d. Med., 1889) isolated six bacilli 
 which had this capacity. 
 
 Soapy milk may be produced by the Bacillus lactis 
 saponacei of Weigmann and Ziron {Gentralbl. f. Bakt., xv., 
 p. 464). It does not coagulate milk, but makes it slimy and 
 slightly ropy, with a faint soapy taste. It grows best at 10° C. 
 
 Bed milk may be produced by the Bacterium lactis 
 
THE BACTERIOLOGY OF MILK 57 
 
 erythrogenes of Hueppe, which is described in detail by 
 Grotenfelt [Fortschr. d. Med., 1889, ii., p. 41), It gives 
 milk a red coloration, which is developed best when the 
 medium is slightly alkaline and kept in the dark, and is 
 checked by acidity and light. On standing, the cream 
 rises as a yellowish layer, and the casein is precipitated, 
 though the reaction remains alkaline and the clear serum 
 is pink. Other organisms, as for instance the sarcina of 
 Menge {Centralbl. f. Bakt., vi., p. 22), and the red yeast or 
 Saccharomyces ruber of Demme (Festschrift, Hirschwald, 
 Berlin, 1890), may also colour milk red. The latter is 
 liable to cause infantile diarrhoea. Bed milk may also be 
 caused by the passage of blood corpuscles in the case of a 
 sick cow, or by the accidental presence of the Bacillus 
 prodigiosus. 
 
 Yellow .milk is stated to be produced by the Bacillus 
 synxanthus of Schroter, the colour being removed by acids 
 and restored by alkalies. 
 
 In the space at command, it has only been possible to 
 enumerate a small number of those organisms which have 
 been known to produce abnormal appearances in milk, and 
 it cannot be said that in the case of all of them a pathogenic 
 character has' been demonstrated ; but it is unquestionably 
 the case in regard to a large number, and none of the 
 appearances in question are either natural in milk or 
 proper to it. They cannot of course be classed as adultera- 
 tions; but on the same principle as that laid down by 
 Lehmann, which regards as adulteration anything which 
 does not normally belong to the food in question, it is 
 proper that milk having these appearances should be 
 rejected, and steps taken to prevent them. A rough-and- 
 ready but very useful and effective test is the rate at 
 which milk goes sour. For practical purposes, it may be 
 said that any milk which goes sour rapidly is a bad milk, 
 
58 THE ANALYSIS OF MILK AND MILK-PROBUCTS 
 
 although the converse is not neeessarily true. According 
 to Schatzmann, if a sample of milk be kept for twelve hours 
 at 40° C„ and within that time coagulates, it is to. some 
 extent defective; and in nine hours no change whatever 
 should appear to have occurred. The presence of colostrum 
 is a ground for the immediate condemnation of milk; it 
 can usually be detected by the presence of long elastic 
 yellowish threads. 
 
 The remedy for almost all of these diseases lies above 
 all in cleanliness, both as to the udders and body of the 
 cow, the stable, and the hands of those employed in milk- 
 ing. Clean milk-pans, pure water, and a cool, odourless 
 and clean store-room, are absolutely indispensable. The 
 milk-cans or other utensils are best made of earthenware 
 or well-tinned copper or iron, and they should be thoroughly 
 scalded and cleaned between each change of milk. 
 
 In addition to the faults which arise from specific 
 bacterial causes, milk may be watery in the case of ill-kept 
 and ill-fed cows. Salty milk may be watery (1027 to 1029 
 specific gravity). Salty milk is stated to occur only in 
 connection with inflammation of the udder. It is to be 
 detected not only by its taste and its high percentage of 
 ash, but by its low percentage of milk-sugar. According to 
 Klenze, 2*4 per cent, of small deposits of calcium carbonate 
 in the milk glands may give rise to sandy milk. 
 
 The faults mentioned so far are those which arise for the 
 most part through causes external to the cow, and are 
 diseases of the milk itself. Milk may, however, be no less 
 defective through disease in the cow. A large number of 
 diseases are known to be capable of being conveyed in this 
 way ; in many cases the milk itself is affected unfavourably, 
 ^0 that it would be rejected for its unpleasant smell or 
 taste ; but this is far from being invariable. Cows suffering 
 from pulmonary tuberculosis give,, for instance, an un- 
 
THE BACTERIOLOGY OF MILK 59 
 
 pleasant and unsavoury milk, which carries its own con- 
 demnation with it ; those, however, which suffer from 
 generalized tuberculosis may yield a milk which only in 
 advahced stages gives the slight yellow coloration, the 
 reduction of cream, fat and milk-sugar, and the increase of 
 albumen which has been attributed to such milk. Never- 
 theless, it has been shown that cows suffering from 
 generalized tuberculosis, whether it has or has not affected 
 the udders, are capable of conveying the bacterial infection 
 into the milk, which constitutes, therefore, a serious danger. 
 Scheurlen {Arheiten a.d.k. Ges.-Amt., vii., 1891) investi- 
 gated the extent to which milk can be freed from bacteria 
 suspended in it by the operation of a centrifugal machine ; 
 and his results, although not obtained with that object, are 
 of considerable assistance in arranging the examination of 
 milk. He found the curious result that, while of the large 
 majority of the bacteria contained in milk three-fourths 
 went into the cream on being centrifugalized, and the rest 
 stayed in the separated milk, and the same result was 
 obtained by merely leaving the milk to stand, and that 
 these results held good not only for the ordinary milk 
 bacteria, but also for anthrax, typhoid and cholera 
 organisms, the tubercle bacillus only remained to a small 
 extent in either the milk or the cream, and the large 
 majority was ejected under the centrifugal influence. 
 
 Scheurlen's method for demonstrating the tubercle 
 bacillus in milk was to steep it for twenty-four hours in 
 absolute alcohol, digest in ether for another twenty-four hours 
 in order to remove the fat, and stain according to Ziehl's 
 method. (For Ziehl's method, see 'Applied Bacteriology,' 
 Pearmain and Moor.) Ilkavitch (Munchener med. Wachen- 
 schr., 1892, p. 5) described a convenient method of applying 
 this result : 20 c.c. of milk are coagulated with citric acid, 
 filtered and dissolved in saturated aqueous sohition of 
 
60 THE ANALYSIS OE MILK AND MILK-PRODUCTS 
 
 Na3P04, The solution is treated with 6 c.c. of ether, the fat 
 which rises to the surface is decanted, and the remainder, 
 after the addition of one or two drops of acetic acid, is 
 centrifugalized in a copper tube with a screwed bottom. 
 The deposit is separated from the fluid by means of a 
 fairly well-fitting ball, which is let into the tube on a 
 stalk. The bottom of the tube is then scraped, the deposit 
 divided between two cover-glasses and stained for the 
 tubercle bacillus. 
 
 Drs. Woodhead and Macfadyen found the tubercle 
 bacillus in six samples of milk out of 600 samples ex- 
 amined. 
 
 The. question of the use of tuberculous milk has received 
 much more attention on the Continent than it has in this 
 country. In Denmark a most thorough and complete 
 system of inspection has been instituted with excellent 
 results ; cattle found to be tuberculous are at once isolated, 
 and, if, necessary, slaughtered and the body destroyed. 
 
 The great mortality amongst young children, due to 
 tubercular intestinal affections, is undoubtedly due to the 
 use of milk containing .the tubercle bacillus. Delicate 
 children are the most susceptible, as, owing to imperfect 
 nutrition and other causes, the system is unable to resist 
 the attack of the organism. Brouardel cites a case where 
 five out of fourteen young girls living together in a board- 
 ing-school became consumptive subsequent to the daily use 
 of milk from a tuberculous cow. 
 
 That the tubercle bacilli occurring in milk are virulent 
 has been proved by subjecting animals to subcutaneous 
 injection, and by feeding them with the infected milk. 
 
 Dr. Martin writes (Eoyal Commission on Tuberculosis, 
 1895) _: ' The milk of cows with tuberculosis of the udders 
 possesses a virulence which can only be described as extra- 
 ordinary. All animals inoculated showed tuberculosis in 
 
THE BACTEEI0L.06Y OF MILK 61 
 
 its most rapid form.' Dr. Woodhead, after investigating 
 the effects of unboiled tubercular milk, speaks in similar 
 terms of this virulence of milk derived from tuberculous 
 udders and inoculated into test animals. 
 
 These two observers had occasion to use milk from a cow 
 that had tuberculous disease in one quarter only of the 
 udder ; and they found the milk from the other three 
 quarters to be perfectly harmless on inoculation ; but the 
 mixed milk from the four teats was to all appearance 
 just as virulent as the milk from the diseased quarter. 
 Butter, skimmed milk, butter-milk, obtained from the 
 milk of a cow having tuberculous udders, all contained 
 tubercle bacilli. 
 
 To some extent, the chances of infection are reduced in 
 actual practice, as the milk as usually supplied to the 
 consumers is the mixed milk of a herd of cows, whereby a 
 tuberculous milk suffers considerable dilution with the milk 
 from healthy cows ; but this dilution, as shown by recent 
 experiments, only reduces the risk of infection, but does 
 not entirely do away with it. 
 
 Freudenreich examined twenty-eight samples of mixed 
 milk, and found out of this number four that proved to be 
 virulent when inoculated into guinea-pigs. Two of these 
 samples came from dairies where from twenty to thirty 
 cows were kept, and where in each case only one cow was 
 suspected to be affected with tuberculosis. The other two 
 samples, which were more virulent, came from dairies 
 where there was more than one suspected cow, and where 
 the udders of some of the animals were visibly tuber- 
 culous. 
 
 Affected milk may not, under ordinary circumstances, 
 induce tuberculosis, owing to its not containing a sufficient 
 number of organisms to constitute a ' toxic dose,' but, in 
 the case of persons rendered susceptible Owitig to disease 
 
62 THE ANALYSIS OP MILK AND MILK-PRODTJCTS 
 
 or weak health, or who have a constitutional predisposition 
 to consumption, the use of tuberculous milk constitutes a 
 very grave danger to health. 
 
 An admirable account of the animal tuberculoses by 
 Nocard has recently been translated by Scurfield, in which 
 the latest views as to the relation between the tuberculosis 
 of man and animals are set forth. In this work a detailed 
 account of Professor Bang's method for stamping out tuber- 
 culosis in cattle is to be found. 
 
 Professor Bang has shown that it is possible to produce 
 a healthy stock from one partially infected, by the simple 
 procedure of isolating those cattle that react to the tuber- 
 culin test. 
 
 Calves are rarely tuberculous, and if kept with healthy 
 animals, will probably not contract the disease, so that the 
 procedure adopted, of dividing the herd so that the healthy 
 and infected animals are kept entirely separate, and placing 
 all calves as they are born, whether from healthy or in- 
 fected cows, in the buildings occupied by healthy animals, 
 speedily resulted in a great diminution of the disease. The 
 testing with tuberculin was repeated every six months, 
 the healthy and the infected animals were provided with 
 entirely separate stables, although they were allowed to 
 mix when out in the fields. The plan appears to have been 
 attended with remarkable success, and deserves attention in 
 this country. 
 
 In addition to the diseases which may be caused in the 
 milk itself by external circumstances and to the organisms 
 of disease which may pass into it from an affected animal, 
 milk can also serve to transmit organisms of infectious 
 disease which neither cause abnormal modifications in the 
 milk nor arise from sickness in the cow. Such diseases are 
 notably scarlet fever, diphtheria, typhoid, and cholera. It 
 is p):obable that such transmission is due invariably either 
 
THE BACTERIOLOGY OF MILK 63 
 
 to the presence of the infection in the persons immediately 
 concerned in drawing or handling the milk, or to the 
 absence of cleanliness either in the water or in the other 
 surroundings. The meaning of cleanliness in the water in 
 the present sense is freedom from bacterial contamination ; 
 and it is unquestionable that unless an exceptionally pure 
 source of supply is at hand, no water should be allowed to 
 be used for dairy purposes unless it has been boiled, or, 
 preferably, passed through a Pasteur filter immediately 
 before use. It must be borne in mind that some organisms, 
 such as typhoid, find in milk a medium capable of favour- 
 ing their multiplication at an enormous rate, and the film 
 of damp which is inevitably left on a vessel after cleaning 
 may serve to contaminate the whole of the milk which 
 may be poured into it if it be kept for a few hours. The 
 standard of purity of water for dairy purposes is not to be 
 estimated chemically, nor in the presence of a medium in 
 which the number of bacteria in the water is liaJDle to in- 
 crease at so enormous a rate is the. criterion of a small 
 number of bacteria any substantial protection. The sole 
 criterion which can be admitted, and the only one which, 
 in the absence of direct chemical contamination such as 
 that which may arise from lead pipes, is necessary, is that 
 the water be absolutely sterilized before use. In the same 
 way the presence of dust must be avoided where milk is 
 about. There is no practicable method of purifying dust,- 
 and not only tuberculosis, but probably the peptonizing 
 organisins described above, both of which are liable to 
 assume forms of the highest resistance and permanence, 
 are largely conveyed by way of dust. 
 
 Finally, it must be remarked that under the influence of 
 various organisms, which are only now beginning- ' to be 
 identified, milk undergoes decompositions resulting in the 
 production of xiirect poisons— such, for infttaiice> as the 
 
64 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 tyrotoxicon of Vaughan. These are usually accompanied 
 in milk by, an odour which causes its rejection, and in 
 those cases where the milk is rapidly consumed after being 
 drawn it probably seldom happens that the organism has 
 the necessary time for development. It is accordingly in 
 cheese, where a somewhat pronounced bouquet is more 
 readily tolerated than in milk, and where much longer 
 periods are allowed for the development of the organisms, 
 that cases of such poisoning are most frequent. The 
 poisoning appears to be analogous to that which occurs 
 with decomposing meat, sausages, etc., and is liable to 
 assume acute forms, terminating fatally. 
 
 'The sterilization of milk for sanitary purposes is a 
 question allied to that of its preservation. The preserva- 
 tion of milk by freezing has been attempted, and is in some 
 measure successful. It is, however, only of temporary 
 value, and while it probably stops, the development of 
 pathogenic organisms for the time being, it does not 
 necessarily kill them. 
 
 In the Report of a special, analytical and biological 
 commission on milk -supply, held under the auspices of 
 the British Medical Journal (1895), the following reforms 
 were suggested for the better management of dairies, to 
 Secure a pure milk- supply : 
 
 1. That all milking be carried on in the open air, the 
 animals and operators standing on a material which is 
 capable of being thoroughly washed, such as a floor of, 
 concrete or cement. Such a floor could be easily laid down 
 in any convenient place which can be found. The site 
 chosen should be removed from inhabited parts as far as 
 possible, and should be provided with a plentiful water- 
 supply. Only in this way does it seem possible to avoid 
 the initial contamination with the colon bacillus. 
 
 2. That greater care should be expended on the personal 
 
THE BACTERIOLOGY OE MILK 65 
 
 cleanliness of the cows. The only too familiar picture of 
 the animal's hind- quarters, flanks, and side being thickly 
 plastered with mud and f seces is one that should be common 
 no longer. It would not be difi&cult to carry out this 
 change ; indeed, in the better managed of our large dairy 
 companies' farms such a condition no longer prevails, but 
 in the smaller farms it is but too frequently met with. 
 
 3. That the hands of the milker be thoroughly washed 
 before the operation of milking is commenced, and that 
 after once being washed, they be not again employed in 
 handling the cow otherwise than in the necessary operation 
 of milking. Any such handling should be succeeded by 
 another washiag in fresh water before again commencing 
 to milk. 
 
 4. That all milk-vendors' shops should be kept far 
 cleaner than is often the case at present. That all milk- 
 retailing shops should be compelled to provide proper 
 storage accommodation, and that the counters, etc., should 
 be tiled. 
 
 CREAST. 
 
 ' Devonshire ' or ' Cornish ' cream is prepared by warming 
 milk in pans for several hours, when the cream rises to the 
 top in a much more coherent layer and more rapidly than 
 if the milk is merely allowed to stand at room-temperature. 
 The cream obtained in this way contains 60 per cent, or 
 more of butter-fat. Owing to the partial sterilization that 
 it has thus undergone, cream prepared in this way keeps 
 sweet without preservatives being added much longer than 
 separated cream would. 
 
 Before the introduction of separators, cream used to be 
 prepared either by the above-mentioned method, or by 
 simply allowing milk to stand overnight and skimming off 
 
 5 
 
66 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 the cream in the morning. Such a rough method as this 
 would" of necessity result in a product containing very 
 varying amounts of fat, according to the temperature of the 
 milk and the shape of the vessel. Now, however, there are a 
 
 number of centrifugal machines 
 in the market, by the use of 
 which milk can be almost en- 
 tirely denuded of its cream. 
 The accompanying illustration 
 shows one of the well-known 
 ' Laval ' separators, by means 
 of which the cream can be so 
 completely separated as to leave 
 only 0"05 per cent, of fat in the 
 milk. Milk is slightly warmed 
 before being passed through the 
 separator, as it is found that 
 this causes a more complete 
 separation of the cream. 
 
 The percentage of fat in sepa- 
 rated cream is often as high as 
 65 per cent., but it is rarely 
 sent out for sale at this strength, 
 and the retailers dilute it down 
 considerably before it reaches 
 the consumer, in many cases 
 adding an equal volume of milk. 
 It is therefore evident that a standard for cream should 
 be fixed, and a reasonable standard would be that it should 
 contain not less than 45 per cent, of fat. 
 
 Cream is generally bought and sold on its taste and 
 appearance, not on the amount of fat that it contains. 
 There is hence a certain temptation to thicken it artificially 
 so as to enable a thin cream to sell at the best price, and 
 
 Fig. 5. — The Laval 
 Sepaeatoe. 
 
CREAM 67 
 
 we recently examined a sample containing only 40 per cent, 
 of fat which had been thickened with gelatine so as to 
 bring it to the same thickness as a sample containing 60 per 
 cent, of fat. The addition of any considerable quantity of 
 gelatine to cream causes it to assume a buttery consistency, 
 and it will no longer pull out into strings as genuine rich 
 separated cream will. 
 
 The difference is very apparent on comparing a sample 
 of cream adulterated in this way with a genume sample. 
 Gelatine in cream may be detected by carefully drying a 
 weighed portion, as in Bell's method for the analysis of 
 milk, removing the fat with ether, and taking up the 
 residue with the least possible quantity of boiling water. 
 On allowing to cool, if gelatine is present, the liquid will 
 set solid. 
 
 Another method, for which we are indebted to Mr. A. W. 
 Stokes, depends on the precipitation of gelatin by tannin. 
 Mix a weighed quantity of the suspected sample with warm 
 water, and add acetic acid to precipitate fat and albuminoids, 
 taking care to avoid excess ; filter, and to the clear liquid 
 add a few drops of strong solution of tannin. 
 
 A sample of genuine cream should be treated in the 
 same way for comparison. On addition of the tannin 
 solution, a slight precipitate is produced in the case of 
 genuine cream ; but in a sample adulterated with gelatine 
 a copious precipitate will be thrown down. 
 
 The amount of fat in cream is most readily estimated by 
 weighing out 2 grammes into a small dish, thoroughly 
 mixing with about 12 c.c. of water, pouring into a Leffmann- 
 Beam bottle, and treating in the same way as ordinary milk. 
 The result obtained is multiplied by the factor 7'77, as 
 explained in the analysis of cheese. The fat in cream 
 might also be estimated by the Werner- Schmidt method, 
 
 5—2 
 
68 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 or by the Adams' method ; but the process above described 
 is quite accurate and by far the quickest. 
 
 Separated cream almost invariably contains preservatives ; 
 one preservative, very commonly used, consists of borax 
 and boric acid, sweetened with saccharine. The use of a 
 preservative is practically a necessity in the cream trade, 
 as separated cream will begin to turn sour in as short a 
 time as four hours in the summer, if no preservative has 
 been added. 
 
CONDENSED MILK 
 
 69 
 
 CONDENSED MILE. 
 
 Attention has often been drawn to the serious conse- 
 quences of using inferior brands of condensed milk for the 
 feeding of children, a use to which they are largely applied, 
 and complaint has been made that sufficient disclosure of 
 the composition of these milks is not made to the purchaser. 
 It appears that some of the labels on some of the tins of 
 inferior condensed milk bear the words ' skimmed,' but in 
 such small type that the intimation might be easily over- 
 looked. To remedy this, it has been suggested that all 
 tins containing milk, such as that referred to, should be 
 required to bear labels on which the words ' condensed 
 skimmed milk ' are printed in large and legible type. It is 
 also suggested that it should be made compulsory that an 
 additional notification should be printed on the labels to 
 the effect that such milk is not suitable for the purpose of- 
 feeding young children. 
 
 The following number of condensed milk samples were 
 examined by the public analysts under the Food and Drugs 
 Act during the past six years : 
 
 
 Total Number Examined. 
 
 Adulterated. 
 
 Percentage of 
 Adulteration. 
 
 1890 
 
 . — 
 
 — 
 
 
 
 1891 
 
 — 
 
 — 
 
 — 
 
 1892 
 
 __ 
 
 — 
 
 — 
 
 189.<i 
 
 159 
 
 14 
 
 8-8 
 
 1894 
 
 — 
 
 — 
 
 — 
 
 1895 
 
 72 
 
 3 
 
 4-1 
 
70 THE ANALYSIS OP MILK AND MILK-PKODUCTS 
 
 Condensed milk seems to have been first prepared about 
 the year 1856, and is now an article of great importance, 
 particularly on account of the great quantity used in the 
 feeding of young children. 
 
 There are about fifty brands at present on sale in this 
 country, many of which are the same milk under different 
 labels. They may be roughly divided into four classes : 
 
 1. Unsweetened milks. 
 
 2. Sweetened milks. 
 
 3. Sweetened partly-skimmed milks. 
 
 4. Sweetened skimmed milks. 
 
 In most cases the degree of concentration is that obtained 
 by evaporating three volumes to one ; that is, the addition 
 of two volumes of water (to an unsweetened milk) will pro- 
 dace a strength equal to the original. 
 
 1. The unsweetened milks, of which there are at present 
 four different brands, are well prepared, and keep perfectly. 
 They contain the due proportion of fat. 
 
 2. This class forms by far the largest and most important 
 part of the whole • supply, and for the most part there is 
 nothing to complain of in them, except that the dilutions 
 recommended would in very nearly every case produce a 
 milk very much below standard. A sweetened condensed 
 milk generally contains rather more added cane-sugar than 
 milk solids. The following figures are typical of a good 
 specimen : 
 
 Fat ... 
 Proteids 
 Milk-sugar 
 Ash ... 
 Cane-sugar 
 
 11 "0 per cent. 
 10-0 „ 
 14-0 „ 
 2-2 „ 
 38-0 
 
 3. This class seems now to have almost disappeared. 
 
 4. Separated milks (generally miscalled skimmed milks) 
 are largely used by poor and ignorant people for infant- 
 
CONDENSED MILK 
 
 71 
 
 feeding, and there can be no doubt but that great harm 
 results from the practice. 
 
 In 1881 Dr. A. Voelcker {Analyst, vi., 221) published the 
 figures he obtained on three samples of unsweetened con- 
 densed milk, and five samples of sweetened condensed milk. 
 The unsweetened milks were rich in fat, but the sweetened 
 condensed milks were all, in his opinion, prepared from 
 partly skimmed milk. 
 
 In 1889 Harald Paber {Analyst, xiv., 141) examined two 
 samples of condensed milk, one the well-known ' Goat ' 
 brand, on which he obtained results that accord closely 
 with figures by several subsequent observers, while the 
 other was a sample of unsweetened condensed milk, the 
 brand of which he does not mention, but which was 
 evidently only condensed to one-half in process of manu- 
 facture, as he found fat 6-15, albuminoids 5-35, sugar 9-55, 
 and ash 1-44 per cent. The direction on the label recom- 
 mended the addition of seven parts of water, which would 
 give a liquid containing 2-8 per cent of solids ! The author 
 rightly points out that the use of such a mixture as the 
 chief or only food for infants might have fatal conse- 
 quences. 
 
 The following figures were obtained by Charles P. 
 Worcester (Massachusetts), on samples of so-called ' evapo- 
 rated cream ' (unsweetened condensed milk), and sweetened 
 condensed milks, in the year 1892 : 
 
 EVAPORATED CREAM. 
 
 Brand. 
 
 Total Solids. 
 
 Solids-not-Fat. 
 
 Fat. 
 
 Water. 
 
 Highland 
 
 Highland 
 
 St. Charles 
 St. Charles 
 
 41-44% 
 33-26 
 31-23 
 29-64 
 
 29-00% 
 26-14 
 27-40 
 23-87 
 
 12-44% 
 7-12 
 3-83 
 5-77 
 
 58-66% 
 66-74 
 68-77 
 70-36 
 
72 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 CONDENSED MILKS. 
 
 Brand. 
 
 Total SoUds. 
 
 SoUda-not-Pat. 
 
 Pat. 
 
 Water. 
 
 Newport 
 
 77-79% ■ 
 
 71-95% 
 
 6-84% 
 
 22-21% 
 
 Star 
 
 77-67 
 
 73-08 
 
 4-69 
 
 22-33 
 
 Red Cross 
 
 75-52 
 
 68-14 
 
 7-38 
 
 24-48 
 
 Superior 
 
 74-13 
 
 68-68 
 
 5-45 
 
 25-87 
 
 At antic and Pacific 
 
 
 
 
 
 Tea Company ... 
 
 73-10 
 
 67-25 
 
 5-85 
 
 26-90 
 
 Red Cross 
 
 71-66 
 
 63-95 
 
 7-71 
 
 28-34 
 
 In 1893 he obtained the following figures : 
 EVAPORATED CREAM. 
 
 Brand. 
 
 Total Solids. 
 
 Solids-not-Fat. 
 
 Fat. 
 
 Water. 
 
 St. Charles 
 Highland .. 
 St. Charles 
 
 31-87% 
 
 31-60 
 
 31-09 
 
 26-83% 
 
 25-70 
 
 26-74 
 
 5-04% 
 
 5-90 
 
 4-35 
 
 68-13% 
 
 68-40 
 
 68-91 
 
 CONDENSED MILK. 
 
 Brand. 
 
 Total Solids. 
 
 SoUds-not-Pat. 
 
 Pat. 
 
 Water. 
 
 Bahy 
 
 67-20% 
 
 59-70% 
 
 7-50% 
 
 32-80% 
 
 J. B. Smith 
 
 75-43 
 
 63-65 
 
 11-78 
 
 24-57 
 
 Newport 
 
 68-32 
 
 61-10 
 
 7-22 
 
 31-68 
 
 Puritan 
 
 78-36 
 
 68-00 
 
 10-35 
 
 21-65 
 
 Maine Jersey 
 
 69-02 
 
 62-75 
 
 6-27 
 
 30-98 
 
 Full Weight 
 
 74^45 
 
 67-16 
 
 7-29 
 
 25-56 
 
 Sovereign 
 
 71-75 
 
 65-00 
 
 6-75 
 
 28-25 
 
 Winner 
 
 68-00 
 
 62-00 
 
 6-00 
 
 32-00 
 
 Pifherman's 
 
 69-10 
 
 63-07 
 
 6-03 
 
 30-90 
 
 Champion... 
 
 74-20 
 
 65-50 
 
 8-70 
 
 25-80 
 
 Twitchel Champlin 
 
 
 
 
 
 Co 
 
 72-20 
 
 59-80 
 
 12-40 
 
 27-80 
 
 Dime 
 
 72-10 
 
 60-00 
 
 12-10 
 
 27-90 
 
CONDENSED MILK 
 
 73 
 
 In 1894 he obtained the following figures : 
 
 Brand. 
 
 Total SoUds. 
 
 SoUds-not-Pat. 
 
 Fat. 
 
 Water. 
 
 Watch 
 
 68-85% 
 
 64-85% 
 
 4-00% 
 
 31-15% 
 
 Tiptop 
 
 74-60 
 
 67-90 
 
 6-70 
 
 25-40 
 
 Highland 
 
 76-40 
 
 67-75 
 
 7-65 
 
 24-60 
 
 Maine Jersey 
 
 68-32 
 
 61-05 
 
 7-27 
 
 31-68 
 
 Dirigo 
 
 70-70 
 
 64-75 
 
 5-95 
 
 29-30 
 
 Full "Weight 
 
 76-68 
 
 68-85 
 
 7-73 
 
 23-32 
 
 Winner 
 
 71-66 
 
 66-19 
 
 4-47 
 
 28-34 
 
 Interstate 
 
 68-76 
 
 64-58 
 
 4-18 
 
 31-24 
 
 Gold Medal 
 
 74-33 
 
 66-50 
 
 7-83 
 
 25-67 
 
 J. B. Smith 
 
 69-60 
 
 64-75 
 
 4-85 
 
 30-40 
 
 Clover 
 
 74-23 
 
 68-25 
 
 5-98 
 
 25-77 
 
 Pine-tree 
 
 73-08 
 
 69-20 
 
 3-88 
 
 26-92 
 
 Leader 
 
 73-15 
 
 70-00 
 
 3-15 
 
 26-85 
 
 Out. of all the above samples only three, namely, ' evapo- 
 rated cream ' ' Highland ' brand (1892), and condensed 
 milks ' J. B. Smith ' and ' Puritan ' (1893), can be regarded 
 as prepared from whole milk, all the rest having evidently 
 been prepared from a mixture of whole milk and separated 
 milk. The report does not state what action, if any, was 
 taken. 
 
 In a paper on ' The Composition and Analysis of Con- 
 densed Milk,' printed in the Analyst for December, 1895, 
 we drew attention to the inadequate disclosure of the 
 contents made on most of the labels, and stated that it 
 ought not merely to be enacted that the words ' skimmed ' 
 or ' separated ' should be printed on the label in as con- 
 spicuous type as the words * condensed milk,' but that it 
 should be made compulsory to add, ' Skimmed milk is unfit 
 for the nourishment of children.' 
 
 In the case of the best condensed milks, prepared from 
 genuine milk with all its fat, the directions for dilution are 
 very open to criticism. The following table shows the 
 character of the liquid — it cannot be called milk— that is 
 produced by following out the directions on the labels of 
 half a dozen of the best brands of whole-cream milk .- 
 
74 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 Sweetened 
 
 Dilution recom- 
 mended for 
 
 Fat in such 
 
 Dilution recom- 
 mended for 
 Infants' Use. 
 
 Fat in such 
 
 Whole Milk. 
 
 Houseliold 
 
 Product. 
 
 Product. 
 
 
 Purposes. 
 
 
 
 
 A. 
 
 1 to 3 
 
 2-6% 
 
 1 to 5 
 
 1-8% 
 
 B. 
 
 1 „ 5 
 
 1-6 
 
 1 „ 14 
 
 0-7 
 
 C. 
 
 1 „ 5 
 
 1-6 
 
 1 „ 14 
 
 0-6 
 
 D. 
 
 1 „ 6 
 
 1-4 
 
 1 „15 
 
 0-7 
 
 E. 
 
 1 „ 5 
 
 2-1 
 
 1 „ 14 
 
 0-8 
 
 F. 
 
 1 „ 5 
 
 1-7 
 
 1 „ 14 
 
 0-7 
 
 G. 
 
 1 „ 5 
 
 1-7 
 
 1 „ 14 
 
 0-7 
 
 Human Milk. 
 
 
 — 
 
 
 3-6 
 
 If the dilution were carried out with only two parts 
 of water, so that the fat in the diluted milk would be equal 
 to that in the original milk before condensation, the fluid 
 so produced would be so rich in sugar as to be an un- 
 drinkable syrup in the case of the sweetened milks, and no 
 dilution can give them anything like the same composition 
 as the natural food for infants, human milk. This led us 
 to wonder whether it was possible to prepare a condensed 
 milk, at a price not exceeding that of the best brands of 
 condensed milk, which should, on dilution, yield a liquid 
 closely approximating to the composition of human milk. 
 An account of this preparation will be found at p. 83. 
 
 In the Analyst for December, 1895, Alfred H. Allen 
 published the analyses of a number of condensed milks 
 bought in the year 1894, and these figures are of great 
 interest as showing the change that has taken place, in 
 the disappearance from the market of the class of ' partially 
 skimmed milks.' His results are as follows : 
 
CONDENSED MILK 
 
 75 
 
 Brand. 
 
 Total 
 Solids. 
 
 Fat. 
 
 Protelds. 
 
 Description on Label. 
 
 Alderney 
 
 68-10% 
 
 11-05% 
 
 10-95% 
 
 Guaranteed to contain 60 
 per- cent, of original 
 cream. 
 
 Anglo-Swiss ... 
 
 73-70 
 
 9-70 
 
 9-87 
 
 Best unskimmed country 
 milk. 
 
 Arcadia 
 
 71-20 
 
 8-08 
 
 10-25 
 
 From best and purest 
 cow's milk. 
 
 Calf 
 
 66-30 
 
 — 
 
 10-19 
 
 Contains skimmed milk. 
 
 Cow 
 
 72-50 
 
 6-50 
 
 10-57 
 
 From partly - skimmed 
 milk. 
 
 Cowslip 
 
 72-10 
 
 018 
 
 10-20 
 
 Skimmed milk. 
 
 Daily 
 
 69-64 
 
 0-26 
 
 10-58 
 
 From skimmed milk. 
 
 Devon 
 
 70-60 
 
 8-50 
 
 10-63 
 
 
 Farm 
 
 66-60 
 
 0-12 
 
 10-14 
 
 i) J) }j 
 
 First Swiss ... 
 
 28-37 
 
 8-76 
 
 10-14 
 
 Unsweetened. 
 
 }) )) ■ * * 
 
 36-10 
 
 11-06 
 
 12-75 
 
 
 Foiirpenny ... 
 
 75-36 
 
 5-40 
 
 13-18 
 
 From pure fresh milk con- 
 taining all its cream. 
 
 Full weight ... 
 
 75-70 
 
 11-60 
 
 11-27 
 
 Warranted not skimmed. 
 
 Geranium 
 
 73-98 
 
 10-35 
 
 8-56 
 
 Full cream. 
 
 Goat ... 
 
 79-10 
 
 4-30 
 
 10-44 
 
 From skimmed milk. 
 
 Handy 
 
 67-38 
 
 0-17 
 
 10-20 
 
 J) »j )» 
 
 Home 
 
 69-44 
 
 0-91 
 
 9-13 
 
 tt )t )) 
 
 Milkman 
 
 73-66 
 
 11-80 
 
 11-40 
 
 Warranted to contain all 
 original cream. 
 
 Mother 
 
 70-60 
 
 5-57 
 
 8-32 
 
 From unskimmed milk. 
 
 Popular 
 
 71-75 
 
 2-47 
 
 10-14 
 
 From paitly - skimmed 
 milk. 
 
 Rose 
 
 72-40 
 
 10-30 
 
 9-75 
 
 Warranted not skimmed. 
 
 Shamrock 
 
 71-30 
 
 — 
 
 10-88 
 
 From skimmed milk. 
 
 St. Olaf 
 
 71-86 
 
 11-80 
 
 10-50 
 
 From pure unskimmed 
 milk. 
 
 Sunlight 
 
 71-76 
 
 5-60 
 
 8-63 
 
 From unskimmed milk. 
 
 Swiss (Nestle's) 
 
 75-00 
 
 13-50 
 
 10.-44 
 
 From pure milk, and only 
 small quantity of pure 
 cane-sugar. 
 
 >j » 
 
 70-41 
 
 11-03 
 
 10-51 
 
 From pure milk, and only 
 small quantity of pure 
 cane-sugar. 
 
 Threepenny ... 
 
 66-25 
 
 0-30 
 
 10-49 
 
 From partly - skimmed 
 milk. 
 
 Tip- top 
 
 74-25 
 
 8-12 
 
 8-82 
 
 Warranted not skimmed. 
 
 Tiking- 
 
 35-16 
 
 10-40 
 
 9-14 
 
 Unsweetened. 
 
 In the issue of the British Medical Journal of July 27, 
 1895, a report on seventeen samples of condensed milks 
 was published. The report forms part of a ' Eeport on the 
 Milk Supply of London.' The samples were obtained by 
 
76 THE ANALYSIS OP MILK AND MILK-PRODUCTS 
 
 the editor, and submitted to Dr. Bernard Dyer for. analysis. 
 AH the samples purchased were skimmed condensed milks, 
 the only sample of true condensed milk which was subse- 
 quently examined, as showing the amount of fat that con- 
 densed milk of good quality should contain, being a sample 
 of the '- Milkmaid ' brand. 
 
 In the report several of the labels were reproduced, which 
 gives it great additional interest, as it has caused some of 
 them to be withdrawn. The following table shows the 
 results obtained by Dr. Dyer, and the conclusions he draws 
 as to the proportions of whole and skimmed milk in each 
 sample. All the tins were sold as skimmed or separated 
 milk, so that no action could be taken against the vendors 
 under the Sale of Pood and Drugs Act, though some of 
 them are actually recommended for the food of infants. 
 
 Brand. 
 
 Shamrock ... 
 
 Swiss Dairy 
 
 Caif 
 
 Cup 
 
 Tea 
 
 Wheatsheaf 
 
 Marguerite 
 
 Daily 
 
 Gondola ... 
 
 As You Like It 
 
 Nutrient ... 
 
 Clipper 
 
 Groat 
 
 Handy 
 
 Cross 
 Cow 
 
 Home 
 Milkmaid ... 
 
 Fat. 
 
 0-79% 
 
 0-63 
 
 0-60 
 
 0-49 
 
 0-48 
 
 0-62 
 
 0-42 
 
 0-69 
 
 0-49 
 
 4-23 
 
 2-36 
 
 0-73 
 0-56 
 1-49 
 
 0-96 
 2-84 
 
 1-02 
 10-92 
 
 Conclusion. 
 
 All separated milk. 
 
 60 per cent, separated milk, 
 40 per cent, whole milk. 
 
 80 per oenti separated milk, 
 20 per cent, whple milk. 
 
 All separated milk. 
 
 )> )) 
 
 90 per cent, separated milk, 
 
 10 per cent, whole milk. 
 All separated milk. 
 70 per centi separated milk, 
 
 30 per cent, whole milk. 
 All separated milk. 
 Condensed milk of good 
 
 quality. 
 
 The following figures, which appeared in the Analyst for 
 December, 1895, represent the composition of all the con- 
 
CONDENSED MILK 
 
 77 
 
 densed milks then on the market that we were able to meet 
 with : 
 
 UNSWEETENED CONDENSED MILKS. 
 
 Brand. 
 
 Total SoUds. 
 
 Fat. 
 
 Milk Sugar. 
 
 , . Protelds. 
 
 First Swiss 
 
 Ideal 
 
 Hollandia 
 Viking ... 
 
 36-7% 
 38-0 
 43'0 
 34-2 
 
 10-5% 
 
 12-4 
 
 9-8 
 
 10-0 
 
 14-2% 
 16-0 
 18-5 
 13-3 
 
 9-7% 
 8-3 
 11-3 
 9-C 
 
 SWEETENED CONDENSED MILKS. 
 
 Brand. 
 
 Total 
 SoUds. 
 
 Fat. 
 
 Milk- 
 Sugar. 
 
 Proteids. 
 
 Cane- 
 Sugar. 
 
 Anglo-Swiss 
 
 74-4% 
 
 10-8% 
 
 16-0% 
 
 8-8% 
 
 37-1% 
 
 Cleeves 
 
 71-0 
 
 10-8 
 
 17-1 
 
 10-1 
 
 31-3 
 
 Clover Leaf 
 
 76-0 
 
 10-7 
 
 13-6 
 
 8-8 
 
 40-9 
 
 Darby and Joan ... 
 
 73-1 
 
 9-8 
 
 13-0 
 
 13-3 
 
 34-7 
 
 Fourpenny 
 
 76-5 
 
 10-4 
 
 13-0 
 
 9-8 
 
 41-3 
 
 Fern 
 
 67-7 
 
 10-7 
 
 15-0 
 
 10-6 
 
 29-8 
 
 Full Weight 
 
 76-5 
 
 11-0 
 
 13-5 
 
 12-3 
 
 37-2 
 
 Geranium 
 
 75-0 
 
 9-8 
 
 13-0 
 
 7-5 
 
 43-1 
 
 Gro-ahead 
 
 76-1 
 
 10-0 
 
 14-6 
 
 9-7 
 
 39-7 
 
 Gowan 
 
 72-0 
 
 10-8 
 
 13-4 
 
 10-5 
 
 35-6 
 
 Home and Colonial 
 
 72-6 
 
 13-5 
 
 17-0 
 
 9-7 
 
 30-5 
 
 Lipton's 
 
 71-0 
 
 9-3 
 
 14-5 
 
 7-9 
 
 37-1 
 
 Lucerne Lion 
 
 71-7 
 
 10-8 
 
 15-2 
 
 9-3 
 
 34-4 
 
 Mother 
 
 72-0 
 
 8-8 
 
 13-7 
 
 7-3 
 
 40-5 
 
 Milkmaid 
 
 76-3 
 
 11-0 
 
 14-6 
 
 9-7 
 
 38-7 
 
 Nestle's 
 
 77-2 
 
 13-7 
 
 15-0 
 
 9-7 
 
 37-2 
 
 Eose 
 
 76-6 
 
 12-4 
 
 17-6 
 
 8-3 
 
 36-1 
 
 Scandinavian 
 
 74-6 
 
 10-8 
 
 14-5 
 
 6-6 
 
 41-1 
 
 World's Tea Co. ... 
 
 75-0 
 
 9-3 
 
 — 
 
 — 
 
 — 
 
 SWEETENED SKIMMED MILKS. 
 
 Brand. 
 
 Total 
 Bolids. 
 
 Fat. 
 
 Milk- 
 Sugar. 
 
 Proteids. 
 
 Oane- 
 Sugar. 
 
 Beehive 
 
 77-7% 
 
 0-2% 
 
 — 
 
 — 
 
 — 
 
 Calf 
 
 
 
 58-0 
 
 1-0 
 
 16'0% 
 
 7-5% 
 
 31-9% 
 
 Cow 
 
 
 
 74-9 
 
 2-0 
 
 13-0 
 
 11-5 
 
 45-8 
 
 Cowslip 
 
 
 
 70-9 
 
 1-4 
 
 14-6 
 
 11-4 
 
 41-9 
 
 Cross 
 
 
 
 75-0 
 
 1-2 
 
 16-0 
 
 10-5 
 
 44-7 
 
 Cup 
 
 
 
 56-9 
 
 1-0 
 
 15-4 
 
 8-5 
 
 304 
 
 Daily 
 
 
 
 68-8 
 
 1-3 
 
 13-7 
 
 10-2 
 
 41-1 
 
 Daisy 
 
 
 
 64 
 
 0-5 
 
 — 
 
 — 
 
 — 
 
 Drummer '. 
 
 Boy 
 
 
 — 
 
 1-0 
 
 — 
 
 9-6 
 
 — 
 
78 THE ANALYSIS OF MILK AND MILK-PEODUCTS 
 
 SWEETENED SKIMMED M.ILKS— continued. 
 
 Brand. 
 
 Total 
 Solids. 
 
 Fat. 
 
 MUk- 
 Sugar. 
 
 Proteids. 
 
 Cane- 
 
 Sugar. 
 
 Farmhonae 
 
 77-0% 
 
 0-4% 
 
 
 
 
 
 
 
 Favourite 
 
 70-6 
 
 0-3 
 
 — 
 
 10-0% 
 
 — 
 
 Goat 
 
 71-0 
 
 1-2 
 
 12-0% 
 
 9-9 
 
 45-9% 
 
 Golden Eagle 
 
 
 
 1-0 
 
 
 
 6-8 
 
 
 Handy 
 
 75-5 
 
 0-3 
 
 17-0 
 
 12-3 
 
 44-3 
 
 Home 
 
 71-3 
 
 1-3 
 
 12-5 
 
 11-1 
 
 43-9 
 
 Household..,. 
 
 700 
 
 0-3 
 
 
 
 ■ 
 
 
 
 Imperial Dairy ... 
 
 70-4 
 
 3-7 
 
 12-6 
 
 11-3 
 
 41-2 
 
 Lancer 
 
 67-6 
 
 0-3 
 
 16-6 
 
 12-3 
 
 35-8 
 
 Lifeguard 
 
 65-8 
 
 0-3 
 
 
 
 
 
 
 Lovers 
 
 730 
 
 0-2 
 
 
 
 
 
 
 
 Alinstrel 
 
 75-3 
 
 0-2 
 
 16-4 
 
 9-7 
 
 48-3 
 
 Shamrock ... 
 
 71-6 
 
 0-5 
 
 18-4 
 
 11-8 
 
 .38-9 
 
 Springtime 
 
 74-0 
 
 0-3 
 
 
 — 
 
 — 
 
 Wasp 
 
 73-4 
 
 1-0 
 
 12-3 
 
 ll'O 
 
 47-2 
 
 The following figures were obtained by - one of us 
 (C. G. Moor) on five samples of condensed milk, examined 
 during January, 1897 : 
 
 Brand. 
 
 Sold as 
 
 Fat. 
 
 Ash. 
 
 Total 
 Solids. 
 
 Milkcan 
 Dolphin 
 Grass Country 
 Flamingo 
 Coopers 
 
 Partly skimmed 
 
 1) )) 
 Not skimmed 
 Skimmed milk 
 Unskimmed 
 
 0-2% 
 0-26 
 9-7 
 0-29 
 13-9 
 
 3-0% 
 
 2-2 
 
 2-0 
 
 2-1 
 
 2-0 
 
 74-4% 
 
 72-0 
 
 73-1 
 
 76-0 
 
 74-3 
 
 Attention has again been drawn to the question of 
 dilutions recently by A. H. Allen {Analyst, xxi., 281), who 
 has determined the percentage of fat contained in the fluids 
 obtained on following the directions on the tins of twelve 
 well-known brands of condensed milk ; of these, the dilu- 
 tions recommended for ordinary use would only in two 
 cases produce a milk containing as much as 3*0 per cent, 
 of fat, six other samples show ftbout 2"5 per cent., while 
 the remaining three samples showed only traces of fat. 
 
 Of the twelve samples, seven carry on their labels 
 directions for dilution for feeding infants, and, if we take 
 
CONDENSED MILK 79 
 
 the extreme dilution, the fat in the fluid produced averages 
 for the whole seven only 0"96 per cent. 
 
 The Analysis of Condensed Milk. — The analysis of con- 
 densed milk should comprise estimations of the total solids, 
 ash, proteids, milk-sugar and fat. The last four items added 
 together and subtracted from the total solids will give the 
 cane-sugar with fair accuracy. We have examined the ash 
 of a number of milks for tin and lead, but they have been 
 absent in every case. In examining any tinned article for 
 tin or lead, great care must be taken to use a sharp tin- 
 opener, or fragments of metal may be torn off and vitiate 
 the result. The tin should be cut open and the contents 
 should be thoroughly mixed, and 10 grammes weighed out, 
 and made up to 100 c.c. We have now a 10 per cent, 
 solution, which serves conveniently for the following 
 estimations : 
 
 Total Solids. — 20 c.c. of the solution are evaporated in a 
 platinum dish till constant in weight ; this will take five or 
 six hours. 
 
 Ash. — The same quantity will serve for the determination 
 of ash, and should be ignited at as low a temperature as 
 possible. The ash in condensed milk varies, but on the 
 average is somewhat over 2 per cent. 
 
 Proteids. — 10 c.c. of the solution are evaporated to 
 dryness in an 8-ounce oxygen flask, and the nitrogen 
 determined by the Kjeldhal process ; the proteids are then 
 calculated by the 6*3 factor. 
 
 In May, 1893, Eichmond and Boseley read a paper 
 before the Society of Public Analysts, entitled ' Points in 
 the Analysis of Condensed Milks,' in which they give 
 methods for the estimation of the casein and the albumin, 
 and point out some of the errors liable to occur in the use 
 of Eitthausen's process for determining proteids. They 
 recommend the Adams process for the extraction of the fat. 
 
80 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 Milk-Sugar. — 10 c.c. of the solution are made up tct 
 100 c.c. by the addition of 40 c.c. of water and 50 c.c. of 
 ammonia. This gives us a 1 per cent, solution, which is 
 of a convenient strength for the estimation of the sugar by 
 Pavy's method. In applying Pavy's test to milk-sugar, 
 the reagent must be kept boiling briskly while the milk- 
 sugar solution is run in slowly. The reaction takes longer 
 in the case of milk-sugar than in the case of glucose, so 
 that unless the titration is conducted slowly, the milk-sugar 
 solution may be added in excess of the quantity required to 
 complete the reaction. Using ordinary Pavy's solution, it 
 will be found in the case of most condensed milks that 
 about 35 c.c. of the 1 per cent, solution of condensed milk 
 will be equivalent to 50 c.c. of Pavy's solution, and the 
 calculation will be as follows : 
 
 If 35 c.c. dilute milk = 50 c.c, Pavy's Solution, 
 
 Or 70 c.c. dilute milk =100 c.c. Pavy's Solution, 
 
 Or •? grammes sample = 100 c.c. Pavy's Solution, 
 
 „ „ = '05 grammes of glucose, 
 
 ■f nrv j: , "05 X 100 , 
 
 .■. 100 grammes of sample = = grammes glucose. 
 
 •05x100x100 4. - .,, 
 
 „ „ = -= — ^ =percent. of milk-sugar 
 
 in the original sample. 
 
 Fat. — Two quantities of 5 c.c. of the solution are placed 
 on two Adams papers, which are well dried and then 
 extracted with dried ether. 
 
 The method of drying with sand or calcium sulphate, 
 and then extracting with ether, is unsatisfactory, as the fat 
 is extracted with great difficulty. 
 
 We have attempted to apply the Leffmann-Beam machine 
 to the estimation of fat in condensed milk, and it succeeds 
 fairly well if the following details are adhered to: Place 
 10 c.c. of the 10 per cent, solution in a Leffinann-Beam 
 bottle with 3 c.c. of the hydrochloric acid and fusel-oil 
 
CONDENSED; MILK 81 
 
 mixture, shake well, and add 15 c.c. of sulphuric acid, 85 
 per cent., and fill up to the mark to the hot mixture of 
 sulphuric acid and water. The bottle is then whirled for 
 three minutes. The fat will not all come out at once ; and 
 after whirling, the bottle should be placed in the water-bath 
 and then whirled again, when the entire amount of fat will 
 be separated. The chief objection to this method is the 
 large factor that has to be employed, namely, 15"5, showing 
 that we are working on 1 gramme of the original sample. 
 
 It is rarely possible to get the figures representing the 
 separately determined constituents when added up to agree 
 with the total solids in an ordinary milk, but the error is 
 usually small ; hence, in the case of sweetened condensed 
 milks it is usual to subtract the sum of the ash, fat, 
 proteids, and milk-sugar from the total solids, and to 
 consider the difference added sugar. 
 
 In the case just referred to respecting the difficulty of 
 getting agreements between the sum of the separately 
 determined constituents and the total solids, some of the 
 anomalies observed may be due to the fact that milk-sugar 
 may or may not be dehydrated, according to the manner in 
 which the evaporation has been conducted ; or, again, the 
 proteids may have become altered during the process of 
 condensation, causing them to be incorrectly represented 
 by the ordinary factor. 
 
 Condensed sweetened milk used to be used as an 
 adulterant of cow's milk, but it is not now employed for 
 that purpose. Unsweetened condensed milk is sold in 
 large tins, which are made use of by the retailer to 
 supplement his daily supply on occasions when inilk is 
 short, and a special milk is sold for this purpose, which is 
 prepared in Italy from milk containing about two-thirds 
 the proper amount of fat. It has the following composi- 
 tion: 
 
 6 
 
82 THE ANALYSIS OF MILK AND MILK-PKODUCTS 
 
 Fat 
 
 9*5 per cent. 
 
 Milk-sugar 
 
 ... 16-5 „ 
 
 Proteids 
 
 ... 14-7 „ 
 
 Ash 
 
 _ _T • i 1 
 
 ... 3-5 „ 
 
 1 ■ _ _ ^ _' Jl 
 
 It is heavily preserved with boric acid. 
 
 Attempts have frequently been made to replace the fat in 
 separated condensed milk by working in some foreign oil or 
 fat, just as margarine cheese is prepared by adding mar- 
 garine fat to skim milk and then curdling in the ordinary 
 way. 
 
 The following is an abstract of English Patent 2,081, 
 1896: 
 
 ' Milk Products, Condensed, Improvements in the Manu- 
 facture of- W. P. Maclaren and A. Smith Fleming, of 
 Glasgow. 
 
 ' The object of this invention is to prepare a wholesome 
 condensed milk from separated milk. Forty gallons of 
 milk are mixed with 28 lb, of sugar and condensed to about 
 100 lb. To this is then added an emulsive compound con- 
 sisting of 12 lb. of refined beef-oil (oleo-oil), 8 lb. of lime- 
 water, 1 lb. of cornflour, and 12 lb. of sugar. The 
 emulsifying is preferably carried out in a steam- jacketed 
 pan provided with an agitator, the lime-water being gradu- 
 ally added during the mixing operation.' 
 
 In a newspaper dated 1895, Warnampool (Victoria), is 
 an account of the exhibition by the inventor, a Mr. Munro, 
 of a condensed milk prepared by the addition to condensed 
 skimmed milk of a ' compound ' (presumably an emulsified 
 fat). The report states that the inventor in noway claimed 
 that the condensed milk so prepared was fit for feeding young 
 children — in fact, he proposed to state that it was not 
 suitable for this purpose on the labels. 
 
 Foreign fat in condensed milk would be readily detected 
 by extracting 200 grammes of the sample with ether, after 
 
CONDENSED MILK 83 
 
 grinding with enough anhydrous sulphate to form a dry 
 powder, and examining the fat when quite free from the 
 solvent by the Valenta or Eeichert tests. 
 
 A sample of condensed milk may be regarded as genuine 
 ■which fulfils the following requirements : 
 
 1. The fat must not be less than 10 per cent., and must, 
 be true butter-fat. 
 
 2. The albuminoids, estimated by multiplying the 
 nitrogen figure by the 6'3 factor, must not exceed the fat 
 figure. 
 
 3. The sample must be free from preservatives (except 
 sugar), starch, and all other foreign matters. 
 
 Humanized Condensed Milk. — It has long been recognised 
 that, however pure cow's milk may be, the proportion of 
 albuminoids which it contains is necessarily higher than 
 that in human milk, and must therefore be more difficult 
 of digestion for infants. Various methods have been used 
 for removing a portion of the excess of albuminoids, and 
 the preparations so obtained are called ' humanized milks,' 
 because they closely approximate to human milk in 
 chemical constitution. 
 
 After considering these methods, one of us (C. G. M.) 
 found it practicable to produce a milk of similar composi- 
 tion in a condensed form, having the following composition : 
 
 Fat 
 
 .. 13-5 
 
 Albuminoids 
 
 .. 7-0 
 
 Milk-sugar 
 
 . 21-2 
 
 Mineral matter 
 
 2-0 
 
 Water 
 
 .. 56-0 
 
 100-0 
 
 When diluted with three parts of water its composition 
 will closely approximate to that of human milk. It con- 
 tains no cane-sugar, colouring matter, or preservative. 
 
 It would be improper, in a work of this nature, to make 
 
 6—2 
 
84 THE ANALYSIS OF MILK AND MILK-PRODUCTS. 
 
 any remarks as to the results that have been obtained on 
 the use of this preparation, but we trust our readers will 
 take an interest in our attempt to provide an infants' food 
 of a scientifically correct composition, and the price of 
 which is low enough to bring it within the reach even of 
 the poor. 
 
BUTTER 85 
 
 BUTTER. 
 
 By continuously shaking or beating milk at a high 
 temperature, the fat corpuscles can be divided, and their 
 number increased. If, however, the milk is cooled, and 
 is then shaken or beaten, the fat corpuscles adhere 
 together and form butter, a thin bluish butter-milk re- 
 maining behind. 
 
 The amount of butter -fat in butter prepared from cow's 
 milk is about 85 per cent., the remainder being water, 
 casein, or curd, and generally added salt. 
 
 Butter varies in colour from white to deep yellow, and 
 is more or less granular in character. 
 
 The butter-fat is very complicated in composition, con- 
 sisting as it does of fatty acids in combination with glycerol, 
 forming triglycerides. 
 
 The fatty acids that enter into the composition of butter- 
 fat are : Butyric, caproic, caprylic, capric, myristic, palmitic, 
 stearic, and oleic acids. The first four are soluble in water, 
 and are therefore known as ' soluble fatty acids ' ; the 
 latter, being insoluble, are known as ' insoluble fatty acids.' 
 
 Dr. J. Bell has published the following analysis of a 
 sample of butter-fat : 
 
 Butyric acid ... 6'1% 
 
 Caproic, caprylic, and capric acids 2'1 
 
 Myristic, palmitic, and stearic acids ... 49-4 
 
 Oleic acid 36°1 
 
 Glycerol (calculated) 12 '5 
 
 The proportion of butyric acid and its immediate homo- 
 
86 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 logues produced by the saponification of butter-fat ranges 
 between 5 and 8 per cent. 
 
 The amount of glycerol in butter-fat was first deter- 
 mined by Chevreul, who obtained 11 '85 per cent, by 
 direct weighing of the isolated glycerol. Benedict and 
 Zsigmondy, by oxidizing the glycerol with permanganate 
 of potash, and determining the oxalic acid so formed, 
 have found from 10;2 to 11"6 per cent, of glycerol to be 
 formed by the saponification of butter-fat. A. H. Allen 
 has confirmed these experiments. 
 
 These analytical results show that butter-fat is essentially 
 a tnixture of various triglycerides, those of butyric, palmitic, 
 and oleic acids being the leading constituents : 
 
 Tributyrin Tripalmitin Triolein 
 
 CsHsCO.CH^Oe 03H5(O.Oi6H8iO)3 CBHsCO.OiaHssO)^. 
 
 Some experiments of Dr. J. Bell indicate that the 
 glycerides contain several acid radicles in the same 
 molecule, and therefore the butyrin cannot be separated 
 by any process of fractional solution from the less soluble 
 glycerides of palmitic and oleic acids. Hence butter-fat 
 probably contains complex glycerides of the following 
 characters : 
 
 rCO^HjO. 
 
 ,iC 
 
 isO. 
 
 fO.OiHjO. 
 . tO.CisHssO. 
 
 Such a complex glyceride would yield on saponification 
 fatty acids and glycerol in the same proportion as would 
 be obtained from a mixture of butyrin, palmitin, and olein 
 in the ratio of their molecular weights (A. H. Allen, 
 ' Commercial Organic Analysis,' vol. ii.). 
 
 Margarine. -;- Margarine, oleo-margarine, butterine, or 
 Dutch butter, .as it used to be termed, is prepared by 
 churning meltied and clarified animal fats, usually beef or 
 mutton fat, occasionally lard (vegetable oils are now but 
 
BUTTER 
 
 87 
 
 rarely employed), with skim milk, milk, or cream ; in this 
 way the curd or casein found in the margarine contracts 
 more or less the flavour of genuine cow's butter. When 
 margarine is carefully prepared and duly coloured, it is not 
 easy to tell the same from pure butter by taste or smell. 
 
 Any fictitious butter can now only be legally sold in 
 this country under the term ' margarine,' and must be so 
 marked by a label bearing this name in letters not less 
 than 1| inches high. If unlabelled, an inspector may 
 require the shopkeeper to supply him with the article as 
 butter, and convictions are often obtained in this manner 
 (see the Margarine Act). 
 
 Margarine differs from butter in yielding only traces of 
 ' soluble ' fatty acids. It consists mainly of the glycerides 
 of oleic, stearic and palmitic acids. The absence of glyceryl 
 butyrate, which is the chief characteristic of butter-fat, is 
 the most valuable means of distinguishing between butter 
 and margarine. On this difference depend the Eeichert, 
 Valenta, and Hehner tests, the three most trustworthy tests 
 we have for butter. 
 
 The following number of samples of butter were examined 
 by Public Analysts during the past few years : 
 
 
 Total Number 
 Examined. 
 
 Adulterated. 
 
 Adulteration. 
 
 1890 
 
 2,743 
 
 316 
 
 11-6% 
 
 1891 
 
 3,558 
 
 551 
 
 15-5 
 
 1892 
 
 4,743 
 
 725 
 
 15-3 
 
 1893 
 
 5,784 
 
 794 
 
 13-7 
 
 1894 
 
 6,419 
 
 667 
 
 10-4 
 
 1895 
 
 7,186 
 
 590 
 
 82 
 
 Eight hundred and ninety butters were sampled by the 
 Customs and analyzed in the Inland Eevenue Laboratory 
 between May, 1895, and February, 1896, with the view of 
 ascertaining whether imported butter was much adulterated. 
 
88 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 Of these 106 were regarded as adulterated, but the methods 
 of analysis employed are not stated. 
 
 The Analysis of Butter and the Detection of Foreign Fat. — 
 Professor Tichborne has made an elaborate series of ex- 
 periments on the quantitative estimation of margarine in 
 mixtures of butter and margarine. He worked on mixtures 
 of margarine (made up without the addition of butter) and 
 pure butter of known origin. 
 
 He made mixtures containing 50 per cent., 10 per cent., 
 and 5 per cent, respectively of margarine, and examined 
 their melting-points and specific gravity, and applied the 
 Eeichert and the soluble and insoluble acid tests. 
 
 In each case he gives the inference to be drawn from the 
 result of the experiment both when the iigures yielded by the 
 butter and margarine are known, and when (as in ordinary 
 working) they are unknown. It would have added greatly 
 to the value of these observations if determinations had 
 also been made by the Valenta test and by the oleo- 
 refractometer. 
 
 Professor Tichborne' s figures are printed in the appendix 
 of the Eeport from the Select Committee on Food Products 
 Adulteration (price 2s. 5^d. ; Eyre and Spottiswoode). 
 
 Water. — Ordinary good butter should contain about 
 
 12 per cent, of water; anything over 16 per cent, should be 
 held to be adulteration. Out of 1,500 samples of English 
 and foreign butter examined by Vieth, Eichmond, Bell, 
 and others, only 0*6 per cent, contained over 16 per cent, 
 of water; the larger number contained between 11 and 
 
 13 per cent, of water. Where the amount exceeds 16 per 
 cent, it has either been left in by careless manufacture 
 or fraudulently incorporated. Professor Tichborne is ap- 
 parently in favour of allowing more than 16 per cent, of water 
 in Irish salt butters, but we cannot believe that they could 
 not be made so that the water should not exceed 16 per cent. 
 
BUTTER 89 
 
 The amount of water can be best estimated by drying 
 10 grammes in a platinum dish at 105° C. until practically 
 constant in weight ; this will generally be when no 
 crackling noise can be heard when the ear is brought near 
 the dish. 
 
 Salt. — The residue, after burning off the fat from the 
 above, can be taken as salt for all practical purposes. 
 The salt in butter may amount to, but does not often 
 exceed, 10 per cent. Dr. Bell found as much as 15 per 
 cent, in one sample of English butter. 
 
 Casein or Curd can be estimated by drying the butter as 
 above, transferring to a filter, and washed with ether until 
 fat-free, and the residue, which consists of casein, weighed. 
 The casein varies from 0'3 to 4 per cent. 
 
 Examination of the Fat. — The sample is put into a beaker 
 and placed in the water-bath for a short time, when the 
 water and curd will settle to the bottom. The fat is then 
 decanted and filtered through a dry filter-paper. If this is 
 carefully done the fat will be quite clear and bright. 
 
 The fat is now examined by the Eeichert-Meissl and 
 the Valenta acetic acid test. The indications given by 
 these two tests are generally all that is necessary for 
 all ordinary purposes, but in the case of suspicious or 
 adulterated samples it may be desirable to determine the 
 soluble and insoluble fatty acid by the Hehner method, 
 and also to take the specific gravity of the fat. 
 
 The Eeichert Method of estimating foreign fat in butter 
 is as follows : 5 grammes of the fat at as low a temperature 
 as it will keep fluid are weighed into a flask, and 2 c.c. 
 50 per cent. NaHO and 30 c.c. rectified spirit are then 
 added and the flask attached to a reflux condenser. The 
 flask is then heated over a water-bath, and the contents 
 allowed to boil briskly for twenty minutes. The flask is 
 then detached from the condenser and the alcohol boiled 
 
90 THE ANALYSIS OP MILK AND MILK-PRODUCTS 
 
 off ; the last traces are removed by gentle blowing with the 
 bellows ; 100 o.c. of hot water' are then added to the flask 
 and shaken until the soap is entirely dissolved. 40 c.c. 
 dilute sulphuric acid (40 grammes to the litre) is then run 
 in together with a small piece of pumice-stone to prevent 
 explosive boiling. The flask is quickly attached to an 
 ordinary condenser, and heated with a naked Bunsen flame 
 until 110 c.c. have distilled over ; the distillation should 
 last thirty minutes. The distillate is mixed and filtered. 
 100 c.c. is then titrated with xrr soda or baryta, using 
 phenolphthalein as indicator. As 110 c.c. was distilled 
 and only 100 c.c. titrated, we have to add ^ to the number 
 of c.c. of XTT alkali required. 
 
 Leffmann and B6am have published a process wherein 
 glycerine is used instead of alcohol to perform the 
 saponification. This process is much quicker than the 
 alcohol method, with which it gives concordant results. 
 Special flasks with a bulb blown in the neck, which can be 
 placed on the fine balance, can be obtained, which are very 
 much more convenient than using a flask into which the 
 fat has to be weighed by subtraction. Genuine butter-fat 
 requires from 24 c.c. to 32 c.c. ^u alkali for neutralization. 
 
 Margarine-fat and the vegetable oils when tested by 
 this process give distillates which only require from 
 0-2 to 1-0 c.c. /t5 alkaU. 
 
 Prom the above data we can calculate approximately 
 the amount of butter-fat in a mixture. Many chemists 
 use half the above quantities ; that is, they work on 
 2*5 grammes of fat, and obtain a distillate of 50 c.c, which 
 is titrated direct without filtering. Hence, in speaking 
 of the Eeichert figure it is important to know which 
 procedure has been employed. Of course the amount of 
 alkali required by the distillates in this case is only about 
 half that in which 5 grammes of fat were saponified. 
 
BUTTER 
 
 91 
 
 Professor Tichborne obtained the following resultsj 
 employing Eeichert's test : 
 
 Sample. 
 
 Result of Test 
 required 
 
 of Decinormal 
 Baryta. 
 
 Inference (1) 
 
 knowing figures 
 
 given by the Butter 
 
 and Margarine. 
 
 Inference (2) 
 
 not knowing figures 
 
 given by Butter 
 
 and Margarine. 
 
 Pure butter 
 
 Margarine 
 
 Mixture, 50 per cent. 
 
 „ 10 „ 
 
 27-0 CC. 
 2-2 cc. 
 14-3 
 
 24-3 
 
 48-8% marga- 
 rine 
 
 9-0% marga- 
 rine 
 
 Pure. 
 
 Margarine. 
 46-9% marga- 
 rine. 
 
 7-7%' 
 
 The pure butter used in the first series of experiments 
 was specially made by the Educational Dairy Company. 
 
 II. 
 
 Pure butter 
 
 27-5 cc. 
 
 _ 
 
 Pure. 
 
 Margarine 
 
 3-4 
 
 — 
 
 Margarine. 
 
 Mixture, 50 percent. 
 
 15-3 
 
 50"6% marga- 
 
 48-6% marga- 
 
 
 
 rine 
 
 rine. 
 
 » 10 „ 
 
 25-0 
 
 10-3% marga- 
 rine 
 
 4-5% 
 
 » 5 „ 
 
 26-3 
 
 4-9% marga- 
 rine 
 
 Pure. 
 
 In the above experiments 5 grammes of fat was saponi- 
 fied by heating with a solution of caustic soda in glycerine, 
 and the resulting soap dissolved in 90 cc. of water and 
 acidified with 60 cc. of sulphuric acid (40 grammes of 
 sulphuric in a litre of water), and 100 cc. were distilled 
 and titrated with decinormal alkali. These experiments 
 confirm the experience of other observers, who agree in 
 regarding the Eeichert test as the most valuable we have 
 at present. 
 
 The Valenta Acetic Acid Test. — This test depends on the 
 interiniscibility of butter-fat and strong acetic acid at a 
 low temperature, whereas animal and vegetable fats do 
 not form a clear mixture, except at a much higher tem- 
 
92 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 perature. The acetic acid used must be about 99 per cent. 
 This will give a turbidity with genuine butter-fat at from 
 32° to 36° C. It is best to set the acid against a sample 
 of butter-fat of known purity rather than by titration. 
 
 This test is carried out as follows : A test-tube is graduated 
 by running two quantities of 8 c.c. of water from a burette 
 or pipette ; file scratches are made exactly opposite the 
 dark line forming the meniscus. The tube is then dried, 
 and 8 c.c. each of the fat and acetic acid poured into the 
 tube. The mixture is warmed to 40° C, when, if the 
 sample is butter-fat, the contents of the tube will become 
 clear, whereas margarine will not dissolve in the acid at a 
 lower temperature than 75° C. Mixtures of butter and 
 margarine will of course require intermediate temperatures. 
 In practice we do not note the temperature at which solution 
 takes place, but the point at which turbidity (the reverse 
 of solution) begins to appear on removing. the source of 
 heat. This is done by stirring the mixture, previously 
 heated till clear, with a thermometer, and noting the 
 temperature as soon as the point of turbidity is reached. 
 It makes its first appearance as a tail following the thermo- 
 meter bulb. (See a paper on ' The Acetic Acid Test,' by 
 the authors, Analyst, July, 1894.) 
 
 The Determination of the ' Soluble ' and ' Insoluble ' Fatty 
 Acids (Hehner's Method). — This method was first devised by 
 Angell and Hehner, but the original process has been 
 modified and improved by Allen, Muter, and others. 
 
 The following procedure is recommended as the most 
 satisfactory by A. H. Allen in his * Commercial Organic 
 Analysis,' vol. ii. 
 
 Before commencing the operation, the following standard 
 solutions must be prepared : 
 
 (a) Dissolve 14 grammes of good stick-potash in 500 c.c. 
 of rectified spirit, or methylated spirit which has been 
 
BUTTER 93 
 
 redistilled with caustic alkali, and allow the liquid to 
 stand till clear. This solution will be approximately 
 seminormal. 
 
 (b) A standard hydrochloric or sulphuric acid of approxi- 
 mately seminormal strength. 
 
 (c) Accurately prepared decinormal causticsoda. Each 
 I'O c.c. contains "0040 gramme of NaHO, and neutralizes 
 •0088 gramme of butyric acid, C4H802- 
 
 A quantity of the butter-fat is melted in a small beaker, 
 a small glass rod introduced, and the whole allowed to 
 cool and then weighed. It is remelted, stirred thoroughly, 
 and about 5 grammes poured into a strong 6-oz. bottle. 
 The exact weight of fat taken is ascertained by reweighing 
 the beaker containing the residual fat. 
 
 By means of a fast-delivering pipette, 50 c.c. measure 
 of the alcoholic potash (solution a) is run into the bottle, 
 and the pipette drained exactly thirty seconds. At the 
 same time another quantity of 50 c.c. is measured off in 
 an exactly similar manner into an empty flask. 
 
 The bottle is fitted with an indiarubber stopper, which 
 is tightly wired down, and is placed in the water-oven, 
 and from time to time removed and agitated, avoiding 
 contact between the liquid and the stopper. In about 
 half ah hour the liquid will appear perfectly homogeneous, 
 and when this is the case the saponification is complete, 
 and the bottle may be removed. When sufficiently cool 
 the stopper is removed, and the contents of the bottle 
 rinsed with boiling water into a flask of about 250 c.c. 
 capacity, whic|i is placed over a steam-bath, together with 
 the flask containing merely alcoholic potash, until the 
 alcohol has evaporated. 
 
 Into each of the two flasks is now run about 1 c.c. more 
 seminormal acid (solution b) than is required to neutralize 
 the potash, and the quantity used accurately noted. The 
 
94 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 flask containing the decomposed butter-fat is nearly filled 
 with boiling water, a cork with a long upright tube fitted 
 to it, and the whole allowed to stand on the water-bath 
 until the separated fatty acids form a clear stratum on the 
 surface of the liquid. When this occurs, the flask and 
 contents are allowed to become perfectly cold. 
 
 Meanwhile, the blank experiment is completed by care- 
 fully titrating the contents of the flask with the decinormal 
 soda, a few drops of an alcoholic solution of phenolphthalein 
 being added to indicate the point of neutrality. 
 
 The fatty acids having quite solidified, the resultant cake 
 is detached by gently agitating the flask, so as to allow the 
 liquid to be poured out, but avoiding fracture of the cake. 
 The liquid is passed through a filter to catch any flakes of 
 fatty acid, and is collected in a capacious flask. If any 
 genuine butter be contained in the sample, the filtrate 
 will have a marked odour of butyric acid, especially on 
 warming. 
 
 Boiling water is next poured into the flask containing 
 the fatty acids, a cork and long glass tube attached, and 
 the liquid cautiously heated till it begins to boil, when the 
 flask is removed and strongly agitated till the melted fatty 
 acids form a sort of emulsion with the water. When the 
 fatty acids have again separated ais an oily layer, the 
 contents of the flask should be thoroughly cooled, the cake 
 of fatty acids detached, and the liquid filtered as before. 
 This process of alternate washing in the flask by agitation 
 with boiling water, followed by cooling, and filtration of 
 the wash-water, is repeated three times, the washings 
 being added to the first filtrate. It is often difficult or 
 impossible to obtain the wash- water wholly free from acid 
 reaction, but when the operation is judged to be complete, 
 the washings may be collected separately and titrated with 
 decinormal soda. If the measure of this solution required 
 
BITTTBR 95 
 
 for neutralization does not exceed 0*2 c.c, further washing 
 of the fatty acids is unnecessary. 
 
 The mixed washings and filtrate are next made up to 
 1,000 c.c, or some other definite measure, and an aliquot 
 part carefully titrated with decinormal soda (solution c). 
 The volume required is calculated to the whole liquid. 
 The number so obtained represents the measure of deci- 
 normal soda neutralized by the soluble fatty acids of the 
 butter-fat taken, plus that corresponding to the excess of 
 standard acid used. This last will have been previously 
 ascertained by the blank experiment. The amount of 
 soda employed in this is deducted from the total amount 
 required by the butter-fat quantity, when the difference 
 is the number of cubic centimetres of standard soda corre- 
 sponding to the soluble fatty acids. This volume multiplied 
 by the factor 0"0088 gives the butyric acid in the weight of 
 butter -fat employed. Thus, suppose an experiment to 
 have given the following figures : Weight of butter-fat takenj 
 5'120 grammes ; decinormal soda required in the blank 
 experiment, 3'90 c.c. ; decinormal soda required to 
 neutralize one-fifth of the solution of the soluble fatty 
 acids, 6'25 c.c. Then, 
 
 •0088C31-2y-9)xl00^^.^0 p^^ ^^^, 
 
 The flask containing the cake of insoluble fatty acids is 
 thoroughly drained, and then placed on the water- bath to 
 melt the contents, which are poured as completely as 
 possible into the (wet) filter through which the aqueous 
 liquid was previously passed. The fatty acids are then 
 washed on the filter with boiling water, to remove the last 
 traces of sparingly soluble acids. The funnel, with the 
 filter containing the fatty acids, is then placed in a small 
 beaker and put into the water-bath until all the fatty acid 
 that will has run through. 
 
96 THE ANALYSIS OF MILK AND MILK-PRODTJCTS 
 
 The flask, funnel, and filter-paper are well washed with 
 ether, and the washings evaporated in a flask to obtain 
 the last traces of fatty acids. The beaker containing the 
 bulk of the insoluble fatty acids, together with the smaller 
 quantity recovered from the ether washings, are dried to 
 constant weight at 100° C. and the two weights added ; the 
 sum gives the weight of the insoluble fatty acids contained 
 in the weight of butter-fat taken. 
 
 The soluble fatty acids, calculated as butyric acid, should 
 amount to at least 5 per cent., any notably smaller pro- 
 portion being due to adulteration. The insoluble fatty 
 acids from genuine butter-fat rarely exceed 88| per cent., 
 occasionally reaching 89 per cent., but a sample ought 
 scarcely to be regarded as certainly adulterated unless the 
 insoluble acids exceed 89^ per cent. As a standard for 
 calculation, 88 per cent, of insoluble acids may be regarded 
 as a fair average, the soluble acids being taken at 5| per 
 cent. 
 
 According to J. Bell, the proportion of soluble acids, 
 calculated as butyric acid, not unfrequently falls as low 
 as 4"5, and the percentage of insoluble acids sometimes 
 slightly exceeds 89 "0 per cent. 
 
 The percentage of adulterant in a butter-fat may be 
 calculated from the following formula, in which F is the 
 percentage of foreign fat, and I that of the insoluble fatty 
 acids : 
 
 F = (I_88)xl3-3. 
 
 Or each 0"1 per cent, of soluble acids above 0*5 may be 
 regarded as showing the presence of 2 per cent, of butter- 
 fat. 
 
 Determining the soluble fatty acids. Professor Tichborne 
 found as follows : 
 
BUTTER 
 
 97 
 
 
 I. 
 
 
 
 Sample. 
 
 Result of Test. 
 
 Inference (1) 
 
 knowiug figures 
 
 given by the Butter 
 
 and Margarine. 
 
 Inference (2) 
 
 not knowing figures 
 
 -given by Butter 
 
 and Margarine. 
 
 Pure butter 
 
 Margarine 
 
 Mixture, 50 per cent. 
 ,1 10 „ 
 
 6-07% 
 0-74 
 3-40 
 5-40 
 
 50% 
 12-5 
 
 Pure. 
 
 Margarine. 
 29% 
 Pure. 
 
 
 II 
 
 
 
 Pure butter 
 
 6-3% 
 
 _ 
 
 Pure. 
 
 Margarine 
 
 0-82 
 
 — 
 
 Margarine. 
 
 Mixture, 50 per cent. 
 
 3-77 
 
 47-5% 
 
 20-8% 
 
 I, 10 „ 
 
 5-61 
 
 12-4 
 
 Pure. 
 
 II 5 „ 
 
 6-0 
 
 5-4 
 
 Pure. 
 
 The soluble fatty acids were estimated at the same time 
 as the insoluble acids, by estimating the acidity in the 
 filtrate and washings from the latter due to dissolved fatty 
 acids. 
 
 Determining the insoluble fatty acids, Professor Tich- 
 borne found as follows : 
 
 I. 
 
 Sample. 
 
 Result of Test. 
 
 Inference (1) 
 
 knowing figures 
 
 given by the pure 
 
 Butter & Margarine. 
 
 Inference (2) 
 
 not knowing figures 
 
 given by Butter 
 
 and Margarine. 
 
 Pure butter 
 
 Margarine 
 
 Mixture, 50 per cent. 
 
 ♦ 
 „ 10 ■ „ 
 
 90-07% 
 
 94-54 
 92-6% 92-3% 
 shows ex- 
 actly 50% 
 90-5% 
 
 57-7% 
 11-1 
 
 Slightly adul- 
 terated. 
 Margarine. 
 67-8% 
 
 28-3 
 
 Pure butter 
 
 Margarine 
 
 Mixture, 50 per cent. 
 
 II. 
 
 10 
 5 
 
 88-8% 
 
 
 
 Pure. 
 
 95-1 
 
 
 
 Margarine. 
 
 91-8 
 
 47-6% 
 
 52% marga- 
 rine. 
 
 89-4 
 
 9-2 
 
 7-5% marga- 
 rine. 
 
 89-1 
 
 4-7 
 
 Pure. 
 
98 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 The insoluble fatty acids were determined by saponifying 
 the butter-fat with an alcoholic solution of soda, by heating 
 in a closed vessel in a water-oven. Water was added, and 
 the alcohol evaporated, and the fatty acids precipitated by 
 sulphuric acid, and after repeated washings with water, 
 dried and weighed. Professor Tichborne remarks that ' this 
 test requires great delicacy of manipulation, even the error 
 of experiment causing a great difference. Thus (in the case 
 91'8 per cent.), 91"95 per cent, would show 50 per cent.' 
 
 The Specific Gravity of Butter-Fat. — The method of taking 
 the specific gravity at the ordinary temperature is now but 
 seldom employed, it being more convenient to take the 
 density at temperatures at which the fat is in a molten 
 condition. The temperatures that give the best results, 
 and those that are generally employed, are 37"8° C. 
 ( = 100° F.) and 99° to 100° C. 
 
 Dr. James Bell prefers the former temperature. The 
 fat at about 110° F. is poured into an ordinary specific 
 gravity bottle, which is then allowed to stand in water at 
 exactly 100° F. for a few minutes. The stopper is then 
 pushed well home, the bottle wiped dry, cooled, and 
 weighed. The weight is then compared with water as 
 1,000 at the same temperature. As the result of the 
 examination of a great number of samples, pure butter-fat 
 is found to range from 910'7 to 913'5, the greatier number 
 falling between 911 and 913. Margarine examined under 
 the same conditions ranges from 901*5 to 906"0. 
 
 Allen, Estcourt, and others, take the specific gravity 
 of butter-fat at the temperature of boiling water (98° to 
 100° C), comparing with water at 15*5° C. as unity. 
 
 This is best done with the Sprengel tube. The tube is 
 filled with the melted fat by sucking the contracted end of 
 the tube, the wider end being immersed in the fat. The 
 tube is then placed in water in a state of rapid ebullition, 
 
BUTTER 
 
 99 
 
 contained in a beaker of such a size that the capillary ends 
 are only just free of the boiling water. When the expan- 
 sion ceases the tube is set to the mark by the application 
 of filter-paper to the capillary orifice. The tube is then 
 withdrawn, dried, cooled, and weighed. The weight of the 
 Sprengel tube and the weight of the water contained in it 
 at 15'5° C. being known, the weight of fat contained at 
 99° C. divided by the weight of water at 15-5° C. will give 
 the density of the fat at the temperature of boiling water. 
 
 The following are the limits for butter-fat and margarine 
 at a temperature of about 99'5° C. compared with water at 
 15-5° C: 
 
 Butter-fat -8653 to -8668 
 
 Margarine -8560 „ -8600 
 
 The above method may be replaced by the Westphal 
 balance, as recommended by Estcourt. The oil or fat is 
 contained in a wide test-tube, which is immersed in boiling 
 water ; the tube is arranged with a collar, or some other 
 device, to protect the balance from the steam. The plummet 
 is then dipped into the fat, and the specific gravity found 
 as soon as the maximum temperature is reached. 
 
 The best form of apparatus with which to carry out this 
 method is that devised by Charles Estcourt, a full account 
 of which will be found in Allen's ' Commercial Organic 
 AnalysiSj' vol. ii., p. 16. 
 
 Determining the specific gravity. Professor Tichborne 
 found as follows : 
 
 I. 
 
 
 
 Inference (1) 
 
 Inference (2) 
 
 Sample. 
 
 Besult of Test. 
 
 knowing figures 
 given by the Butter 
 
 not knowing figures 
 given by the Butter 
 
 
 
 and Margarine. 
 
 and Margarine. 
 
 Pure butter 
 
 911-2% 
 
 
 
 Pure. 
 
 Margarine 
 
 905-7 
 
 — 
 
 Margarine, with 
 a little butter. 
 
 Mixture, 50 per cent. 
 
 908-7 
 
 45-5% 
 
 23^4% 
 
 „ 10 „ 
 
 910-6 
 
 10-9 
 
 Pure. 
 
 7—2 
 
100 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 II. 
 
 Sample. 
 
 Result of Test. 
 
 Inference (1) 
 
 knowing figures 
 
 given by the Butter 
 
 and Margarine. 
 
 Inference (2) not 
 
 knowing figured 
 
 given by the Butter 
 
 and Margarine. 
 
 Pure butter 
 
 911-6% 
 
 
 
 Pure. 
 
 Margarine 
 
 905-3 
 
 
 Margarine, with 
 a little butter. 
 
 Mixture, 50 per cent. 
 
 908-7 
 
 46% marga- 
 
 23% marga- 
 
 
 
 rine 
 
 rine.. 
 
 „ 10 „ 
 
 910-8 
 
 11-1% marga- 
 rine 
 
 Pure. 
 
 1! 5 „ 
 
 911-28 
 
 500% marga- 
 rine 
 
 Pure. 
 
 The specific gravity test consists in taking the weight of 
 a certain volume of the fat measured at 100° F. and com- 
 paring it with the weight of the- same volume of water 
 measured at 100° P. 
 
 The Koettstorfer Saponification Equivalent of butter-fat 
 varies from 242 to 253, the mean figure for margarine 
 being about 284. 
 
 The Iodine Absorption of butter-fat ranges from 23 to 
 38 per cent., margarine giving from 40 to 55 per cent, of 
 iffdine absorbed. These results, though interesting in 
 themselves, are not of much value when determining the 
 amount of foreign fat in butter. 
 
 The Oleorefractometer, an instrument devised by Messrs. 
 Jean and Amagat, has recently come into use for testing 
 butter and other fats. This instrument is so arranged that 
 a ray of Ught from a lamp is passed through a chamber 
 containing lard-oil, which is the 'type' or standard oil. 
 In the centre of this chamber is placed a hollow prism, 
 which is filled with the oil or fat under examination. The 
 light, in passing through thp prism, is deflected to a more 
 or less degree, which casts la sharp shadow on a scale, 
 which is placed in the focus (j)f the eye-piece of a telescope 
 attached to the body of the | oil-chamber. A collimator is 
 
- BUTTER 101 
 
 placed in front of the telescope, and the oil-chamber is 
 surrounded by a case to contain warm water, so that ob- 
 servations may be made at any required temperature. The 
 scale is divided into divisions, both right and left of the 
 zero mark. Vegetable oils deflect light to the right and 
 animal oils to the left of the zero. These deflections are 
 noted as -f- or — ; that is, right or left of the zero mark, 
 as the case may be. The liquid oils are examined at 22° C. 
 Butter, margarine, lard, etc., that are not liquefied at this 
 temperature are examined at 45° C. 
 
 As the result of the examination of several hundred of 
 samples examined in this instrument, pure butter-fat gives 
 a deflection of from —35° to —23°, margarine deflecting 
 from -18° to -10°. 
 
 On keeping, butter may turn rancid, especially if the 
 butter-milk has not been well washed out ; but the changes 
 it undergoes would not under ordinary circumstances be 
 sufficient to very seriously invalidate the Eeichert, Valenta, 
 or ' Soluble ' and ' Insoluble ' fatty acid determinations. 
 
 Some experiments to ascertain the change that butter 
 undergoes on keeping are published in the Analyst by 
 Alfred H. Allen and one of us, giving the analysis of butter 
 made when the butter was fresh, and again after keeping 
 for six years in tin boxes in the laboratory. Two samples 
 were examined; and in the case of one of these very little 
 6hange was observable, while the other had decomposed 
 more, and attacked the iron of the box. Such decomposi- 
 tion as would occur in the course of two or three months 
 would probably never be sufficient to invalidate the results 
 of analysis. 
 
 Mr. A. W. Stokes uses the microscope for sorting the 
 samples of butter. He cuts a piece off the sample to obtain 
 a cleau surface ; this surface is lightly scraped all along so 
 as to get as representative a portion as possible. The 
 
102 THE ANALYSIS OV MILK AND MILK-PHODUCTS 
 
 scrapings should not measure more than about one-tenth 
 of the size of an ordinary pea. This is transferred to a 
 small microscopical cover-glass, which is then pressed down 
 on to the ordinary microscopical slip, so that the butter lies 
 in the form of a wedge between the two glasses, one side 
 being pressed very thin, while the opposite side is left very 
 thick. The slide thus prepared is placed under the 1-inch 
 power of the microscope ; the microscope must be provided 
 with a good polarizing apparatus. The thinnest edge of 
 the slide should be so arranged that it cuts across half of 
 the field of view, the rest of the field being left blank. 
 Gas-light should be used. 
 
 Now, on rotating the Nicol's prisms, so that the two 
 prisms are at right angles, the whole of the field will 
 remain equally dark if the sample be pure butter. If, 
 however, as much as 20 per cent, of margarine be present 
 in the sample, it will be impossible, however the prisms 
 may be placed, to darken the side of the field occupied by 
 the sample. Margarine gives, when thus viewed, a similar 
 appearance to that of a cloud in a dark sky. 
 
 No selenite plate should be used. The butter should not 
 be melted," and since most of the adulterated samples of 
 butter contain over 20 per cent, of margarine, this micro- 
 scopical method will rarely pass an adulterated sample. 
 After an experience on many thousands of samples during 
 the past twenty years, Mr. Stokes still believes in the 
 efiiciency of this simple and rapid method in the hands of 
 a competent microscopist. 
 
 Butter has occasionally been adulterated with sesame-oil, 
 which is readily detected by the use of Baudouin's test, 
 which is as follows : Dissolve 0*1 gramme of cane-sugar in 
 10 c.c. of hydrochloric acid of specific gravity 1*2 ; add to 
 this 20 c.c. of the oil to be tested ; shake thoroughly, and 
 allow to stand. In the presence of even 2 per cent, of 
 
BUTTER 103 
 
 sesame-oil the aqueous liquid will become of a crimson 
 colour. 
 
 The addition of hydrochloric acid alone to butter-fat 
 sometimes produces a pink colour owing to the presence of 
 a tar -dye, so that care must be taken not to put down to 
 the presence of sesame-oil a colour due to a dye. 
 
 It occasionally happens that a butter is found which 
 gives a positive reaction with Conroy's test for cotton-seed- 
 oil ; but if this reaction is obtained, it must not be taken 
 for granted that cotton-oil has been added to the butter, as 
 we found by experiment that when a cow, whose butter 
 normally gave no reaction with Conroy's test, had half a 
 pound of cotton-seed cake added to her daily food, the 
 butter at once gave an unmistakable darkening with 
 Conroy's reagent. 
 
 Artificial colouring matters are very frequently added 
 to butter, among which may be mentioned the following : 
 annatto, turmeric, saffron, saffronette, marigold, and the 
 azo-dyes, the last being of somewhat recent introduction 
 for this purpose. The introduction of artificial colouring 
 into butter is not regarded as an adulteration in this 
 country. It is worth noting that the butter from some 
 Jersey cows often has a very deep yellow colour. 
 
 If the colouring matter of a butter can be extracted with 
 alcohol, foreign colouring is undoubtedly present, as the 
 natural colpuring matter is not soluble in alcohol. 
 
 Butter-fat on exposure to light and air loses its yellow 
 colour, and acquires the smell and colour of tallow. 
 
 Preservatives in the form of boric and salicylic acids 
 are sometimes added to butter. We have found boric acid 
 recently in a large number of samples, especially in those 
 coming from abroad. 
 
 Boric Acid is detected in the ash, as given under milk. . 
 
 Salicylic Acid may be estimated by the method a& used 
 
104 THE ANALYSIS OP MILK AND MILK -PRODUCTS 
 
 in the Paris Municipal Laboratory: 20 grammes of the 
 sample is repeatedly exhausted by a solution of bicarbonate 
 of soda, which converts the salicylic acid in the soluble 
 sodium salicylate. The aqueous liquid is acidulated with 
 dilute Ha SO4, extracted with ether, and a little mercurous 
 nitrate added -to the residue after evaporating off the ether, 
 when a precipitate nearly insoluble in water is obtained. 
 This is filtered off, washed, and decomposed by dilute 
 H-2S04, free salicylic acid again resulting. It is redissolved 
 in ether, the solvent evaporated off, and the residue warmed 
 from 80° to 100° C. until nearly dry. In order to remove 
 any other acid present, the residue is extracted with 
 neutral petroleum ether, the ethereal solution diluted 
 with an equal volume of 95 per cent, alcohol, and 
 titrated with xV alkali, using phenolphthalein as indicator 
 (1 c.c. of t\ NaHO =0-0138 of salicylic acid). 
 
 For further identification, the salicylic acid may be 
 liberated again with a corresponding amount of xet HCl, 
 and the liquid tested with a drop of FeaCle solution, when a 
 violet coloration should be obtained. 
 
 Bacteria in Butter. — Butter can serve as a vehicle for 
 many infectious diseases. Gasperini, for instance {Oiornali 
 delta Societa dei Igiene, 1890), made butter out of milk 
 artificially contaminated with tubercle organisms, and found 
 that it retained its virulence even up to 120 days. That 
 the risk of tuberculous infection so conveyed is not small 
 was shown by Bucarferro (Giornali dei Med. prat. Medico, 
 1890, pp. 2-3, 201). He found the organism in nine samples 
 of butter purchased on the market. His inoculations of 
 this butter into guinea-pigs showed the interesting circum- 
 stance that a portion of the infection was a form of tuber- 
 culosis which it is stated does not occur in cows, and 
 must have therefore been due to external contamination of 
 the milk, while another part was true bovine tuberculosis. 
 
BUTTER 105 
 
 Laser (Zeit. f. Hyg., x., p. 513) found that the cholera and 
 typhoid organisms could survive for about a week in butter. 
 
 Butter becomes rancid through chemical processes which 
 are at present but imperfectly understood. It is certain 
 that the presence of light and oxygen, as well as a high 
 temperature, favour the decomposition. Eancid butter is 
 extremely liable to cause intestinal disturbances, and is 
 unfit for food. Where the process has not gone far it is 
 possible, by re-churning it with milk or boiling it in water 
 containing a little soda, to get rid of the fatty acids which 
 cause rancidity. 
 
 Considerable effort has been made, particularly abroad, 
 to prepare butter from sterilized cream with pure cultures. 
 In many parts of Germany pure cultures are obtainable 
 commercially for the preparation of butter with particular 
 aromas ; and butter so prepared is stated in many cases to 
 be much superior, both in capacity for keeping and in 
 aroma, to that produced in the ordinary way. 
 
106 THE ANALYSIS OF MILK AND MILK-PRODXTCTS 
 
 CHEESE. 
 
 The manufacture of cheese by empirical methods has 
 been carried on from the earliest times, and was a means 
 of storing and preserving milk many centuries before any 
 other method was devised. 
 
 Cheese is made from the milk of the cow, goat, and 
 sheep, though cows' milk alone is used for cheese-making 
 in this country. Cheese is made from whole milk, 
 skimmed, or separated, milk, and from milk enriched with 
 cream. The first step in cheese-making is to produce a 
 curd of perfectly uniform consistency, which, though it is 
 not sterile, is yet sufficiently free from bacteria to enable 
 the subsequent means that can be applied (variations of 
 temperature, moisture, etc.) to be effectual in encouraging 
 the development of those ferments, moulds, or bacteria, 
 whose presence is desired, while avoiding the growth of 
 others which might injure the flavour of the finished 
 cheese. Seeing how very empirical the process of cheese- 
 making has been until quite recently, one cannot help 
 feeling surprised at the success which is uniformly attained, 
 failures being only of comparatively accidental occurrence. 
 
 If it were admissible to sterilize the milk before curdling 
 and to keep it sterile throughout, and to add pure cultures 
 of the desired bacteria to the exclusion of all others, one 
 might expect that there would be no failures at all in so far 
 as they might be produced by the accidental predominance 
 of undesirable organisms. The cheese-maker, however, 
 cannot start with his medium (milk) entirely sterile, and 
 
CHEESE 107 
 
 therefore his efforts have to be directed towards the 
 restraint of the growth of injurious organisms, and towards 
 the development of those that are desired, so that it is 
 obvious that his efforts may, through the skill and know- 
 ledge attained by experience, be attended with uniform 
 success,' but cannot be expected to be invariably successful. 
 As above stated, the application of suf&cient heat to sterilize 
 milk produces alterations in the casein which interfere with 
 the action of rennet, and causes the curd to lose its power 
 of coherence. ' There is no doubt that the different kinds 
 of cheese owe their particular property or characteristics to 
 the action of definite bacteria, or classes of bacteria. Since 
 it is possible to prepare any kind of cheese from a given 
 quantity of milk in a given place, or at any time, and 
 according to its nature to obtain it from this milk, it 
 follows that all the kinds of bacteria which are necessary 
 for the manufacture of the cheese in question must be 
 present universally and invariably in the milk. These 
 bacteria must have an extraordinarily wide occurrence. The 
 art of cheese-making consists in the preparation of the 
 fresh cheese mass of each different kind in such a way that 
 those kinds of bacteria which are active in the ripening of 
 that particular cheese must be developed to a predominant 
 extent ' (Fleisohmann). 
 
 The following general process of cheese-making applies 
 with modifications to the manufacture of every kind of 
 cheese. 
 
 The mUk, whether in its natural condition or with added 
 or abstracted fat, is warmed up to 80° F. or above, and 
 curdled by the addition of rennet. Sometimes rennet is 
 not employed, and curdling is effected by the acidity pro- 
 duced by certain of the bacteria commonly found in milk. 
 
 After curdling has taken place the curd is then cut up 
 by means of an apparatus specially designed to divide it 
 
108 THE ANALYSIS OF MILK AND MILK-PEO DUCTS 
 
 into small fragments, without Squeezing out the fat more 
 than is possible. 
 
 In this country rennet is commonly employed, but, 
 whichever method is adopted, the curd is allowed to stand 
 for some time after the whey is drawn off, and this 
 increases the amount of acidity. The time required for 
 coagulation by rennet is about an hour. The amount 
 of acidity in the milk both before the addition of the rennet 
 and after the formation of the curd, is of considerable 
 practical importance, and the amount should not vary 
 greatly from '20 per cent., while after the whey is run 
 off the acidity in the curd increases somewhat. 
 
 After the curd has been cut up, it is allowed to remain 
 with or without the whey till it has ' ripened,' which 
 condition is judged of by testing whether a fragment on 
 being touched by a hot iron bar pulls out to a thread of 
 a certain length ; after this it is placed in the press, which 
 is made of wood or metal, and pressure applied gradually. 
 The pressure must not be too suddenly applied, or the 
 moisture in the centre will not escape, so that the result- 
 ing cheese will be too damp, which is also the case if 
 the pressure is insuiScient. 
 
 In using rennet too rapid coagulation must be avoided, 
 which might produce a curd so firm that it would refuse 
 to break up finely enough for subsequent working, while 
 if the period of coagulation is too extended, it is more 
 difficult to keep the temperature equable throughout, and 
 if this is not attended to the coagulum will not be of 
 a uniform nature. 
 
 Some cheeses are eaten fresh, while others are kept 
 in order that ' ripening,' on which the development of the 
 peculiar flavour for which the cheese is valued depends, 
 may take place. The normal ripening of cheese is due 
 to the action of fairly well-defined organisms, and does 
 
CHEESE 109 
 
 not occur in their absence. Their effect is to first pepto- 
 nize the albumen, and in later stages to cause still further 
 decomposition with evolution (Zeit. f. Hyg., xiv. 267) of 
 ammonia, leucine, and tyrosine. If the organisms are 
 excluded, as, for instance, by using sterilized milk, and 
 according to Schaffer and Bodzynski (analyzed in Koch's 
 Annual for 1890, p. 92) and Freudenreich (analyzed in Koch's 
 Annual for 1891, p. 135), of sterilized rennet, or in the 
 presence of anti-septics, no ripening of cheese occurs. 
 At the same time considerable difference of opinion prevails 
 as to the actual organisms to which the ripening pro- 
 cesses are due. Considerable work is being done on the 
 subject, and no doubt ultimately trustworthy data will 
 exist corresponding to those which are available in the 
 ease of beer. 
 
 It may be said that at present we have only an imperfect 
 knowledge of the organisms which are responsible for the 
 development of cheese. Moulds as well as bacteria un- 
 questionably take part in it, and many have been definitely 
 related to particular processes of ripening. At the same 
 time a description would occupy more space than we have 
 at our disposal. The taints of cheese are similarly refer- 
 able to organisms of which some have been isolated. 
 
 The capacity of milk to nourish organisms pathogenic 
 to man is possessed in some measure by cheese, and in 
 these cases where the organisms do flourish in that 
 medium the far greater length of time for which they 
 are kept promotes the production of poisons of extreme 
 virulence, as, for instance, the tyro-toxicon of Vaughan. 
 It is probable that some at least of these organisms are 
 related to, or are identical with, those which produce cases 
 of poisoning in meat, and in particular in sausages, salt 
 fish, 'and ham, where the same length of preservation occurs 
 to promote the formation of _an extremely vii^ulent poison. 
 
110 THE ANALYSIS OP MILK AND MILK-PRODUCTS 
 
 The Bacillus enteritidis of Gartner, and the Bacillus 
 enteritidis sporagines of Klein are of this order. 
 
 The mineral acids cannot be used to produce coagulation 
 because they could not be sufficiently washed out subse- 
 quently to allow the desired ripening to take place, and for 
 the same reason it is not possible to utilize (except in a few 
 cases) the natural tendency of milk to become acid, for the 
 large quantity of lactic acid and the large number of 
 rapidly-growing lactic acid bacilli would prevent or greatly 
 retard the growth of other organisms whose presence is 
 desired. 
 
 The following characteristics are representative of the 
 curd produced by rennet : it is elastic, not greasy or sticky, 
 and though, of course, not free from micro-organisms, 
 contains none that are of specially luxuriant growth, so 
 that it serves as an available culture medium for bacteria 
 that may either be purposely introduced at the time or 
 have fallen in from the air, being disseminated throughout 
 the building by previous operations, or for such bacteria 
 or moulds that may subsequently develop inwards from 
 the exterior of the finished cheese. 
 
 Fleischmann considers that the ripening of rennet cheeses 
 may be regarded as largely due to ferments which operate 
 throughout the whole of the cheese, while in the case of 
 cheeses in which curdling was brought about by natural 
 acidity, changes take place chiefly from the exterior to the 
 interior. Cheeses made with rennet contain a considerable 
 proportion of calcium phosphate enclosed in the casein. 
 When cheese is made by curdling due to natural acidity, 
 the curd is of quite a different character to t]^e rennet curd, 
 and is not elastic, but sticky and greasy. As it is highly 
 acid, only certain organisms can develop, and hence a 
 lesser number of cheeses can be prepared from curd 
 obtained in this way. There is less calcium phosphate 
 
CHEESE 111 
 
 in this kind of cheese than in rennet cheese,' as it is 
 dissolved in the lactic acid, and runs off in the whey. 
 
 Alfred Smetham, " Eeport on Cheese-making Experi- 
 ments at Worleston in 1893," says : 
 
 'AH those conriected with the Stilton cheese industry 
 know that it is difficult to obtain in a new dairy the 
 green " fade " for which this class of cheese is so justly 
 noted, but that in a dairy in which good Stilton cheese 
 has been manufactured for years it is next to impossible 
 to obtain it without it. This is because in those dairies 
 in which Stilton cheese has been produced for many years 
 the spores of the mould necessary for the production of 
 the green " fade " are floating about in the air, and find 
 a ready entrance into the curd in the processes of 
 manufacture.' 
 
 Eegarding the accidents of cheese-making, Mr. Smetham 
 quotes Dr. H. W. Conn (Connecticut Agricultural Eeport, 
 1892) respecting the possibility of our having increased 
 control over the quality of cheese by a more exact know- 
 ledge of the bacteria of cheese-making : 
 
 ' The cheese-maker has not yet learnt to cultivate 
 bacteria as the brewer has learned to cultivate his yeast. 
 Some day, I think we may say in the not far distant 
 future, after our experiment stations have had time to 
 work upon this matter a little longer, the cheese-maker 
 is going to be told of some way in which he can cultivate 
 bacteria as the brewer does his yeast, and then he will 
 know what kinds of bacteria produce a badly-ripened 
 cheese, and what kinds will produce an exceedingly good 
 cheese.' 
 
 Rennet. — The fourth stomach of the calf has long been 
 known for its milk-curdling properties, which are due to a 
 ferment termed chymosin. This ferment is invariably 
 present in the healthy human stomach, aild also in that of 
 
112 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 most animals. The chymosin may be extracted from the 
 calf's stomach with water ; but a better way is to digest 
 with weak acid for twenty-four hours, and then carefully 
 neutralize the resulting liquid. Such solutions of rennet 
 cannot be sterilized, as the ferment is injured by a tempeira- 
 ture of 60° C, and are therefore preserved by the addition 
 of a preservative such as salicylic acid. Alcohol pre- 
 cipitates the ferment in an impure form, and dry prepara- 
 tions are made of this precipitate, which keep better, and 
 are some thirty to a hundred and fifty times as strong as 
 the liquid extracts. Their strength is determined by making 
 a small scale experiment with a standard amount of milk 
 of a certain degree of acidity and rennet under given con- 
 ditions of temperature, and noting how long coagulation 
 takes to occur. From the time observed to be necessary 
 for the formation of curd, a calculation is made as to how 
 many volumes of milk would be coagulated in forty minutes. 
 This test applies to the determination of the value of the 
 rennet, and must not be confused with the testing of the 
 milk with rennet, as mentioned in the account of Cheddar- 
 cheese making. There are various brands of rennet on the 
 market — Hansen's, Blumenthal's, Nicholl's, etc. They 
 range from a strength of from 2,000 to 10,000 in the liquid 
 form, and up to 300,000 in powder. The liquid forms are 
 most commonly employed. 
 
 Eennet will act in neutral or slightly alkaline milk, but 
 the coagulation takes longer, and unless there is a certain 
 amount of acidity in the milk, the curd will not contract so 
 as to expel the whey sufficiently. If too much whey 
 remains behind, the milk-sugar is liable to cause fermenta- 
 tion and swelHng of the cheese during curing, while too 
 much acidity, on the other hand, has the opposite effect, and 
 produces a cheese that is too hard and dry, and conse- 
 quently wanting in flavour. 
 
CHEESE 113 
 
 'Starter' for cheese -making consists of a culture of 
 Bacillus acidi lactici in milk, and is obtained by adding a 
 pure culture of the organism to sterile milk. It is not in 
 practice considered necessary that an absolutely pure 
 culture should be used, and therefore some soured milk 
 (in which this organism predominates) is obtained from 
 any dairy well known for its good cheeses, and fresh 
 ' starter ' is made as required for each day by adding a 
 little of the last batch of ' starter ' to sweet milk direct from 
 the cow, just as leaven for bread-making used to be pre- 
 pared in olden times. 
 
 CLASSIFICATION OF CHEESES. 
 
 Cheeses may be classified in various ways, as, for example, 
 
 1. Whether hard or soft. 
 
 2. According to the amount of fat in the raw material — 
 i.e., whether made from 
 
 (a) Cream ; 
 
 (6) Cream and milk ; 
 
 (c) Whole milk ; 
 
 (d) Partly-skimmed milk ; 
 
 (e) Skim milk. 
 
 3. Whether made by curdling with rennet or by natural 
 souring. 
 
 4. Whether eaten fresh or after being kept to ripen. 
 For the sake of convenience, we have dealt with the 
 
 various kinds of cheese used in this country, whether of 
 foreign or home manufacture, in alphabetical oi^der; and 
 we have given figures obtained by ourselves, with others 
 published by Cameron and Aikman, and Leffmann and 
 Beam. 
 American Cheese. — This is a Cheddar cheese, and usually 
 
 8 
 
114 THE ANALYSIS OF MILK AND MILK-PBODUCTS 
 
 contains an excellent amount of fat, averaging about 
 33*0 per cent. 
 
 According to W. Pleischmann (translated by Aikman and 
 Wright, 1896), much cheese has lately been made in 
 America by the American Cheddar process from milk partly 
 deprived of its fat. This may be the case ; but among 
 about sixty samples of American cheese bought during the 
 past two years, we have not found any that has shown 
 much below 30 per cent, of fat. 
 
 In the preparation of American Cheddar, the procedure 
 in use in making Cheddar cheese in England is followed 
 with but slight variations. 
 
 AMEEICAN CHEESE. 
 
 Authority. 
 
 Water. 
 
 Froteids. 
 
 Fat. 
 
 Ash. 
 
 p. and M. 
 
 29-8 % 
 
 30-3% 
 
 33-9% 
 
 3-7% 
 
 
 29-1 
 
 28-1 
 
 35-3 
 
 3-7 
 
 
 24-1 
 
 — 
 
 32-0 
 
 3-9 
 
 
 27-0 
 
 — 
 
 30-1 
 
 4-5 
 
 
 25-0 
 
 
 
 20-1 
 
 7-9 
 
 
 27-2 
 
 — 
 
 30-9 
 
 4-4 
 
 
 28-1 
 
 — 
 
 33-0 
 
 ,4-6 
 
 0. and A. 
 
 22-59 
 
 37-20 
 
 35-41 
 
 4-80 
 
 ^» 
 
 31-80 
 
 36-00 
 
 28-70 
 
 3-50 
 
 Camembert. — Camembert is made in France, and im- 
 ported into this country in considerable quantities. It is 
 usually kept two months before being eaten. 
 
 Authority. 
 C. and A. 
 L. and B. 
 P. and M. 
 
 CAMEMBERT. 
 
 Water. 
 
 51-30 % 
 51-9 
 47-9 
 43-4 
 
 Froteids. 
 19-00 % 
 
 18-9 
 21-8 
 24-4 
 
 Pat. 
 
 21-50 % 
 21-0 
 21-9 
 22-6 
 
 Ash. 
 
 4-70% 
 4-7 
 4-7 
 3-8 
 
 Cheddar. — Cheddar is the typical cheese of English manu- 
 facture, and its composition, and the conditions under 
 which the greatest success is attainable, have formed the 
 subject of many researches. Of re'Cent work on this point. 
 
CHEESE 115 
 
 ■we may mention that of F. J. Lloyd, particularly his 
 report for 1895, entitled ' Observations on Cheddar-Cheese 
 Making," in the Bath and West of England Society's 
 Journal. 
 
 We are indebted to Mr. John Benson, who supplied us 
 ■with samples of typical cheeses, for the following descrip- 
 tion of the manufacturing details employed in Cheddar- 
 cheese-making : 
 
 'The Cheddar cheeses ■were' prepared by taking 40 
 gallons of " night's milk," cooling to 60° P., and allowing 
 to stand. Next morning the cream that had risen was 
 skimmed off and the milk heated to 90° F., and returned to 
 the vat, together with 40 gallons of " morning's milk." The 
 whole (80 gallons in all) was then raised to a tempe;rature 
 of 84° F., and 2 quarts of " starter" were added to bring 
 the milk to the desired ripeness for adding rennet. At the 
 expiration of twenty minutes the milk was ripe — i.e., con- 
 tained about '22 per cent, of acid ; then rennet was added 
 in the proportion of 1 drachm of rennet to every 4 gallons 
 of milk. After fifty minutes, coagulation had taken place 
 sufficiently to render the curd fit to be cut with knives. It 
 was then stirred for ten minutes with the hand, and then 
 ■with a rake, and the temperature gradually raised to 
 100° F. in fifty minutes. A fragment was then tested with 
 a hot iron, and drew out into strings a quarter of an inch 
 long. The whey was then drained off and the curd scooped 
 out into the cooler, allowed to mat together for ten minutes 
 and then cut up in square blocks, turned over, and allowed 
 iio lie another twenty minutes, when it was in a fit condition 
 ior milling. It was then tested with the hot iron, and 
 drew out into threads two inches long. The curd was then 
 ground up, weighed, and returned to the cooler, where it 
 was stirred with the hands for ten minutes, then spread 
 ■out in the cooler, covered with a cloth, and allowed to 
 
 8—2 
 
lis THE ANALYSIS OF MILK AND MILK-PEODDCTS 
 
 stand for twenty minutes. It was then broken up again 
 and well stirred and ealted, 1 oz. of salt being added to 
 3 lbs. of curd, and then vatted, and then put to press with 
 very light pressure, the time taken from renneting to 
 vatting being five hours. Next day the cheese was then 
 taken out and a fresh cloth put on, and a pressure of 
 30 cwt. was applied during the night, and the cheese was 
 kept in the press for three days, after which it was bandaged 
 and taken to the curing-room and kept at a temperature of 
 62° F. till ripe, which required four months. 
 
 
 
 CHEDDAR. 
 
 
 
 Authority. 
 
 Water. 
 
 Protelde. 
 
 Fat. 
 
 Ash. 
 
 C. and A. 
 
 27-83 
 
 % 44-47 % 
 
 24-04 % 
 
 3-66% 
 
 
 28 34 
 
 47-03 
 
 21-01 
 
 3-62 
 
 P. and M. 
 
 33 
 
 27-4 
 
 29-6 
 
 4-3 
 
 
 35-5 
 
 27-8 
 
 26-6 
 
 4-2 
 
 
 33-8 
 
 26-7 
 
 30-5 
 
 4-1 
 
 
 33-3 
 
 27-6 
 
 30-6 
 
 3-6 
 
 In two samples of Cheddar cheese supplied us by Mr, 
 Benson (March, 1897) we found 32-3 and 33-8 per cent, of 
 fat. 
 
 Cheshire Cheese. — We are indebted to Mr. Alfred Smetham 
 for the following figures on three Cheshire cheeses, which 
 were made on three different farms, in July, 1892, by a 
 skilled cheese-maker under his supervision. The analyses 
 were made when the cheese were three months old : 
 
 
 Water. 
 
 Fat. 
 
 Proteids. 
 
 Milk Sugar. 
 
 Ash. 
 
 A 
 
 39-5 % 
 
 29-5 % 
 
 24-8 % 
 
 2-1% 
 
 ■ 3-9,% • 
 
 B 
 
 37-7 
 
 34-1 
 
 23-7 
 
 1-3 
 
 3-2 
 
 C 
 
 40-8 
 
 28-8 
 
 22-6 
 
 3-9 
 
 3-7 
 
 The procedure adopted in the manufacture of these, 
 cheeses may be regarded as typical, and was as follows : 
 
CHEESE 117 
 
 ' A mixture of evening and morning milk was curdled 
 with Van Hasselt's rennet (1 o.z. being required for every 
 22 gallons of milk) at a temperature of 86" F. Breaking 
 down of the curd commenced in fifty-five minutes, and 
 lasted fifty minutes. The temperature was then raised to 
 88°P., and after one hour and fifty minutes the whey was 
 drawn off, and twenty minutes later the curd was broken 
 up by hand. The curd was then weighed, and 7 oz. of 
 salt were added to every 20 lbs. of curd and mixed in 
 through the curd-mill. After pressing, the cheese was 
 kept for jthree months, and at the end of this time was fit 
 for sale.' 
 
 In the case of. sample A we give Mr. Smetham's figures, 
 ; not merely for the cheese, but also for the milk, whey and 
 curd compared with the finished cheese : 
 
 
 Milk, 
 
 Whey, 
 
 Curd, 
 
 Cheese, ' 
 
 June, 28,, 1892. 
 
 June 28, 1892, 
 
 June 28, 1892. 
 
 Sept. 16, 1892. 
 
 Water 
 
 «8'17 % 
 
 93-33 % 
 
 47-90 % 
 
 39-55 % 
 
 Fat 
 
 3-23 
 
 ■24 
 
 26-00 
 
 29-66 
 
 Casein and albumen 
 
 
 
 , 
 
 
 (Nx'6i) 
 
 3-30 
 
 ■88 
 
 20-37 . 
 
 24-83 
 
 Milk-sugar, etc. ... 
 
 4-60 
 
 5-06 
 
 3-4V 
 
 2-11 
 
 Ash 
 
 ■70 
 
 •49 
 
 2^26 
 
 3-95 
 
 100-CO 100-00 100-00 100-00 
 
 CHESHIRE CHEESE. 
 
 Authority. 
 
 Water. 
 
 Protelds. 
 
 Fat. 
 
 Ash. 
 
 P. and M. 
 
 ' 37-8 % 
 
 25-7 % 
 
 31-3% 
 
 4-2% 
 
 ;j 
 
 31-6 
 
 26-5 
 
 35-3 
 
 4-4 
 
 L. and B. 
 
 30-4 
 
 36-1 
 
 25-5 
 
 4-8 
 
 Cream Cheese. — Cream . cheese may be made either by 
 coagulating with rennet, or by allowing the cream to sour 
 naturally; in either case it should contain about 75 per 
 cent, of butter-fat. We are indebted to Mr. Alfred 
 Smetham for two analyses of cream cheese. No. 1 had 
 . been prepared by natural souring ; the mode of preparation 
 of the other was not known. ; 
 
11-8 THE ANALYSIS OF MILK AND MILK-PEODUCTS 
 
 
 No. 1. 
 
 No. 2. 
 
 Water 
 
 .. 11-44 % 
 
 19-71 % 
 
 Butter-fat 
 
 .. 86-03 
 
 74-03 
 
 Casein 
 
 -87 
 
 5-12 
 
 Milk-sugar, etc. 
 
 -61 
 
 -54 
 
 Mineral matters 
 
 .. 1-05 
 
 •60 
 
 100-00 
 
 100-00 
 
 In the July number of the Analyst, 1894, we published 
 the following figures on two cream cheeses : 
 
 
 Water. 
 
 Ash. 
 
 Fat. 
 
 Proteids. 
 
 Double cream 
 Bondon 
 
 57-6% 
 39-5 
 
 3-4% 
 
 •7 
 
 39-3% 
 24-4 
 
 19-0% 
 9-4 
 
 Derbyshire Cheese. — The evening's milk is cooled and set 
 to stand as in Cheddar-cheese-making, and in the morning 
 the cream is skimmed off, the milk heated to 90° E., and 
 the cream mixed with morning's milk and returned to the 
 vat. The whole is heated to 84° F., and rennet is added 
 while the milk is only slightly acid ; if it is not acid enough 
 1 per cent, of ' starter ' is added. The coagulation must 
 be rather firmer than is required for Cheddar-cheese- 
 making, and the curd is cut in pieces about the size of 
 broad beans. It is then stirred gently with the hands 
 for ten minutes, and the temperature gradually raised 
 to 95° P. After sufficient acid has developed the whey 
 is drawn off, while the amount of acid is equal to that 
 in the same stage of Cheddar-cheese-making, but the 
 curd is in a much moister condition. The curd is now 
 drawn up to the end of the vat, covered with a cloth 
 and pressed with rack and weights, left for twenty minutes, 
 and then broken up lightly by hand to allow the whey 
 to escape, and then pressed again as before. In about 
 fifty [to sixty minutes after drawing off the whey the 
 curd should be ready for milling, and should draw in 
 threads 1 inch long when tested with a hot iron, and 
 
CHEESE 119 
 
 have a tough, leafy appearance. The curd is then ground 
 roughly so that as much moisture as possible is retained 
 and stirred well for fifteen minutes, and salt added in 
 the proportion of an ounce to every 4 lbs. of curd. The 
 curd is then placed in hoops holding about 35 lbs. each, 
 they are slightly pressed during the night, the cloths 
 changed in the morning and pressure increased. They 
 are pressed for two days and bandaged, and removed 
 to a curing-room kept at 62° F. The cheeses are ready 
 to cut in two months, and should have a soft, flaky 
 texture. 
 
 We are indebted to Mr. Benson for the above description 
 and for a sample of Derby cheese which contained 26"9 per 
 cent, of fat. 
 
 DUTCH CHEESE. 
 
 Authority. Water. Casein. Fat. Ash. 
 
 P. and M. 37-6 % 32-6 % 10-6 % 6-3 % 
 
 28-2 29-1 22-5 6-5 
 
 In consequence of the variations exhibited in the amount 
 of fat, a further number -of samples of Dutch cheese were 
 examined by one of us, which gave the following amounts 
 of fat (February, 1897. C. G. Moor) : 
 
 mple. 
 
 Fat, per cent. 
 
 Valeiita test. 
 
 Price per lb. 
 
 1 
 
 15-5 
 
 23-0° 
 
 C. 
 
 6d. 
 
 2 
 
 10-7 
 
 24-0 
 
 
 6d. 
 
 3 
 
 15-5 
 
 24-0 
 
 
 6d. 
 
 4 
 
 16-2 
 
 26-0 
 
 
 6d. 
 
 5 
 
 28-4 
 
 27-0 
 
 
 6d. 
 
 6 
 
 26-1 
 
 30-0 
 
 
 8d. 
 
 7 
 
 14-6 
 
 30-0 
 
 
 7Ad. 
 
 8 
 
 21-0 
 
 30-0 
 
 
 7|d. 
 
 9 
 
 15-0 
 
 30-0 
 
 
 6|d. 
 
 Gloucester Cheese. — Two varieties of this cheese are 
 made: Single Gloucester, and Double Gloucester. The 
 distinction between these cheeses is one of size only, and 
 has no reference whatever to quality. 
 
120 THE ANALYSIS OV MILK AND MILK-PRODUCTS 
 GLOUCESTER. 
 
 Authority. Water. Proteids. Fat. Ash. 
 
 C. andA. 21 -41% 49-12% 25-38% 4-09% 
 
 „ 35-82 37-96 21-97 4-25 . 
 
 P. andM. 33-1 31-8 23-5 5-0 
 
 37-4 28-3 28-1 4-6 
 
 Gorgonzola. — Gorgonzola when ripe is permeated by 
 moulds. The spores of the moulds are introduced' by 
 Adding in the process of manufacture powdered bread 
 crusts on which moulds have been allowed to grow. 
 GORGONZOLA. 
 
 Authority. Water. Proteids. Fat. Aeh. 
 
 C. and A. 43-56 % 24-17 % 27-95 % 4-32 % 
 
 P. andM. 40-3 27-7 26-1 5-3 
 
 „ 33-9 25-8 26-7 4-6 
 
 Gruyere. — The term Gruyere is applied to two kinds of 
 
 hard Swiss cheese, which are prepared from cow's milk. 
 
 Much of the cheese sold in England as Gruyere is known 
 
 as Emmenthaler in Switzerland. 
 
 GRtTYl&RE. 
 
 Authority. 
 
 Water. 
 
 Proteids. 
 
 Fat. 
 
 Aah. 
 
 P. and M. 
 
 28-2% 
 
 31-3 % 
 
 28-6% 
 
 4-7% 
 
 )} 
 
 35-7 
 
 28-7 
 
 31-8 
 
 3-7 
 
 C. and A. 
 
 40-00 
 
 31-25 
 
 24-00 
 
 3-00 
 
 )) 
 
 34-68 
 
 31-41 
 
 28-93 
 
 3-85 
 
 L. and B. 
 
 32-0 
 
 35-1 
 
 28-0 
 
 4-8 
 
 Leicester Cheese.— ^The manufacture of this cheese is 
 as follows : 
 
 The evening's milk is cooled to 64° P., and in the 
 morning the cream is skimmed off and the milk heated 
 to 90° F., and the cream mixed with a volume of morning's 
 milk equal to the evening's and returned, the whole being 
 brought to 84° F. If ' starter ' is used, it should be in 
 less proportion than for either Cheddar or Derby cheeses 
 as the development of the acid in the curd must be slower 
 than in the above-named cheeses to obtain good results. 
 The proper quantity, of rennet is then added, together 
 with the colouring .(annate), and about fifty minutes 
 
CHEESE 121 
 
 should elapse before tlie curd- is ready for cutting, when 
 it is cut up rather smaller than in Cheddar-cheese-inakingj 
 and then scalded lip to 98° F. in about sixty minutes 
 after cutting up. The curd should be well cooked and 
 then rolled to the end of the vat, and before the whey 
 is drawn off the curd is pressed in the vat with rack 
 and weights, the acidity at this, stage being rather less 
 than in the corresponding stage of Cheddar or Derby 
 cheeses. When the whey has been drawn off, the curd 
 is cut into pieces about 4 inches square, and spread out 
 on the bottom of the vat to let the whey drain off* The 
 cutting-up and pressing is repeated three times j and in 
 about an hour after drawing off the whey the curd should 
 be ready for grinding, and at this stage should draw into 
 threads f inch in length when tested with a hot iron. 
 After grinding finely the curd is salted, adding about 
 an ounce of salt to 3J lbs. of curd, which is then vatted 
 into hoops holding about 50 lbs. each, and very light' 
 pressure is applied till the cheese is turned at night, 
 pressure being increased next day, and the cheese turned 
 daily for three days. The cheese is then rubbed all 
 over with hot lard, bandaged, and taken to a curing-room 
 kfept at 62° F. The cheese ripens in about five months, 
 and shows a slightly granular texture, midway between 
 Cheddar and Derby for firmness, and should keep perfect 
 in flavour for several months. 
 
 The important points in the manufacture of this cheese 
 are to start with milk in proper condition, and to keep 
 control over the development of the acid, which must not 
 be too rapid, for if the acidity develops too rapidly before 
 the curd is properly 'cooked,' the resulting cheese will 
 be soft and soapy, and will not keep. The object, there- 
 fore, is to get a dry, granular curd with a moderate 
 amount of acidity ; this ensures slow curing , and the 
 
122 THE ANALYSIS OF MILK AND MILK-PRODUCTS 
 
 production of a cheese of good flavour and keeping 
 properties. 
 
 A sample received from Mr. Benson in April, 1897, gavQ 
 29'2 per cent, of fat. 
 
 Parmesan Cheese. — Parmesan cheese is prepared fronj: 
 partly skimmed goat's milk. 
 
 PARMESAN. 
 
 Authority. Water. Protelds. Fat. Ash. 
 
 P.andM. 32-5% 43-6% 17-1% 6-2% ■ 
 
 C. and A. 27-56 44-08 15-95 5-72 
 
 Roquefort Cheese. — Eoquefort cheese is prepared from 
 partly-skimmed ewe's milk, and is highly prized. 
 
 ROQUEFORT. 
 
 Authority. Water. Protelds. Fat. Ash. 
 
 C. and A. 19-30 % 43-28 % 32-30 % — 
 
 P. audM. 29-6 28-3 30-0 6-7 
 
 L. and B. 26-5 .32-9 32-3 4-4 
 
 Skim-Milk Cheese. — Skim-milk cheese may be prepared 
 either by natural souring, or by the use of rennet. 
 
 Authority. 
 
 Water. 
 
 Casein. 
 
 Fat. 
 
 Ash. 
 
 p. and M. 
 
 63-1 % 
 
 17-9% 
 
 6-5% 
 
 1-4% 
 
 C. and A. 
 
 43-14 
 
 49-79 
 
 0-86 
 
 6-^1 
 
 L. and B. 
 
 43-8 
 
 45-0 
 
 5-9 
 
 5-2 
 
 Stilton. — Stilton cheese is prepared in this country, in 
 America, and on the Continent. It is prepared from 
 a mixture of whole milk and cream, or whole milk 
 alone. 
 
 
 
 STILTON. 
 
 
 
 Authority. 
 
 Water. 
 
 Casein. 
 
 Fat. 
 
 Ash. 
 
 P. and M. 
 
 19-4 % 
 
 21-1 % 
 
 42-2% 
 
 2-6 % 
 
 )I 
 
 21-2 
 
 26-3 
 
 46-8 
 
 2-l» 
 
 C. and A. 
 
 32-11 
 
 24-31 
 
 37-36 
 
 3-93 
 
 » 
 
 38-28 
 
 23-93 
 
 .30-59 
 
 3-20 
 
 We found, March, 1897, 42-3 and 41-5 per cent, of fat 
 in two Stilton cheeses prepared at the Midland Dairy 
 Institute. They were made under the direction of Mr. -John 
 
CHEESE 123 
 
 Benson, to whom we are indebted for the following 
 description of the precise details of their manufacture, 
 which may be regarded as typical : 
 
 'The Stilton cheeses were prepared from fresh milk 
 containing 3'7 per cent, of fat. After the milk had been 
 heated up to 84° P., it was curdled by the addition of 
 a sufficient quantity of Hansen's rennet, in the proportion 
 of 1 drachm of rennet to 4 gallons of milk. This addition 
 was sufficient to produce a firm coagulum in one and a 
 quarter hours. The curd was then ladled out into cloths 
 and tightened during the day, the whey having been 
 allowed to remain with the curd for one hour after ladling 
 out, and seven hours later the coagulum was turned out 
 on to a draining sink in a firm condition; the curd was 
 then cut up into pieces about 4 inches square, and turned 
 occasionally. At the expiration of twenty-four hours, 
 counting from when the milk was first set, the curd was 
 judged to be sufficiently ripe for salting and hooping. At 
 this stage the whey draining from the curd would contain 
 some I'O per cent, of lactic acid ; salt was then added 
 in the proportion of 1 oz. to 4 lbs. of curd. The moulds 
 or hoops were then filled and turned the same evening, 
 and kept in the "making-room " (temperature 65° F.) for 
 nine days ; at the end of this period the sides of the 
 cheeses became creamy, and they were then taken to 
 the " coating-room " (temperature 60° F.) where they 
 were scraped and bandaged. This scraping and bandaging 
 was repeated on three successive days, after which the 
 bandages were removed, and the cheeses kept in the 
 coating-room till the outside became dry and crinkled, 
 when they were taken to the "curing-room," which is 
 kept damp and at a temperature not exceeding 65° F. 
 In this manner the cheeses were kept to ripen for three 
 months, and were turned daily. As soon as blue moulds 
 
124 THE ANALYSIS OF MILK AND MILK- PRODUCTS 
 
 ; began to appear in the interior oi the cheeses they were 
 Removed to a cellar, where they are kept until sold. The 
 temperature of the cellar is about 56° F., and that of 
 the ripening-room 62° F. In the case of these particular 
 cheeses, about 12 lbs. of milk were required, to produce 
 1 lb. of ripened cheese, which was fit for sale when four 
 and a half months old. The curing-room and cellar are 
 both kept damp to encourage the growth of the blue 
 mould to penetrate to the interior of the cheese.' 
 
 Wensleydale Cheese. — This is one of the blue-mould 
 cheeses, and is made chiefly in the north of Yorkshire. 
 The method of manufacture is somewhat like that for 
 .Stilton. Fresh milk only is used, and is coagulated at 
 80° to 82° F., the curd is cut into rather large pieces 
 at the end of an hour, and the contents of the vat 
 ■heated up to 86° F. The curd is stirred by, hand for an 
 hour and then allowed to sink in the whey and remain for 
 an hour and a half or thereabouts ; it is then ladled 
 out on to open cloths, after drawing off the whey. A 
 board with loose weights is placed on the cloths to assist 
 in expelling the whey. The curd is ground some five 
 or six hours after renneting, and is then only slightly 
 sour. Salt is added in the proportion of 1 oz. to every 
 4 lbs. of curd, and the curd is filled loosely into the 
 hoops. The cheese remains without pressure over-night, 
 and next morning is turned into a dry cloth, and put 
 into the press with a very slight pressure for one day. 
 The curd when put into the press is only slightly acid, 
 and contains a good deal of whey, but as it is left without 
 .pressure all night, sufficient acidity is developed to 
 obviate any danger of fermentation taking place, and 
 ■the open texture, produced by the absence of much 
 pressure, allows the free admission of air, which is 
 necessary for the formation of the blue mould, the 
 
CHEESE 125 
 
 spores of this mould being in the air of the room. A 
 further quantity of whey drains from the cheese in the 
 curing-room, and this increases the openness of the 
 interior of the cheese. The cheeses are cured at 60° P., 
 and the curing-room is kept damp to render the cheese 
 soft, and favour the growth of the mould. The cheeses 
 are made Stilton shaped, and require about the same 
 time to ripen. 
 
 A sample received from Mr. Benson gave 31'5 per cent, 
 of fat (April, 1897). 
 
 • Adulteration of Cheese. — The only adulteration of cheese 
 at the present day is the substitution of foreign fats for the 
 true butterrfat, such cheeses being known as margarine, or 
 ' filled ' cheeses. They are prepared as follows : Skim-milk 
 is churned up with an emulsion of clarified animal fats 
 (margarine), and the mixture curdled, pressed, and treated 
 in the ordinary way. The sale of this product as cheese 
 without any qualifying description is an offence under the 
 Pood and Drugs Act, and convictions have been from time 
 to time obtained ; so that very little margarine cheese is to 
 be found in this country. 
 
 Margarine cheese is prepared at Dunragit, N.B., at Ham- 
 burg, in Holland, and in America. Its sale and its pre- 
 paration are both prohibited in Canada, New York State, 
 Pennsylvania, Massachusetts, and Wisconsin. 
 
 The smell of margarine cheese is peculiar, and its de- 
 tection analytically is so exceedingly easy and certain that 
 there is little chance of its ever being sold in England, 
 except on rare occasions. The methods for its detection 
 are given under the ' Analysis of Cheese.' 
 
 In 1894, we examined a sample of margarine cheese that 
 was purchased as American Cheddar, the figures obtained 
 being as follows : 
 
 Water, Protelds. Fat. Ash. 
 
 30-6 % 30-8 % 27-7 % 8-6 % 
 
126 THE ANALYSIS OF MILK AND MILK-PKODUCTS 
 
 The fat, when tested with an equal volume of strong 
 glacial acetic acid, gave a turbidity at 82'0° C. In July and 
 September, 1895, E. Bodmer examined two samples of 
 chfeese purchased under the Pood and Drugs Act, which 
 proved to be margarine cheeses. They contained 42*4 and 
 81-7 per cent, of fat respectively, and the Eeichert figure ob- 
 tained on the fat showed them to be almost entirely devoid 
 of butter-fat. The samples were sold as American cheeses, 
 and in each case the vendors were convicted and fined. 
 
 The following number of samples of cheese have been 
 examined by the Public Analysts during the past six years : 
 
 
 Total Number 
 
 Adulterated. 
 
 Percentage of 
 
 
 examined. 
 
 Adulteration. 
 
 1890 
 
 130 
 
 8 
 
 6-15% 
 
 1891 
 
 143 
 
 5, 
 
 36 
 
 1892 
 
 
 
 
 
 
 
 1893 
 
 330 
 
 12 
 
 36 
 
 1894 
 
 421 
 
 18 
 
 4-2 
 
 1895 
 
 697 
 
 19 
 
 3-2 
 
 In 1884, Macfarlane, in Canada, examined fifty -six 
 samples of cheese, all of which proved to be genuine. 
 
 In Massachusetts there is no legal standard for the fat 
 in cheese, but of a number of samples examined by Charles 
 P. Worcester during the past three years the fat averaged 
 about 35 per cent. He used the Babcock centrifugal 
 machine for determining the percentage of fat. 
 
 In 1885, a special New York State brand was adopted for 
 ■' pure cream cheese,' which is said to have accomplished 
 much in restricting the sale of the spurious article. 
 
 Wynter Blyth (' Foods and their Adulterations ') says, 
 ■' A cheese which shows under 10 per cent, of fat may with 
 j)ropriety be called skim,' and with this we heartily agree. 
 
 At present we may purchase as ' cream cheese ' what is 
 
CHEESE 127 
 
 not even a milk cheese, and if cream cheese is really 
 desired it must be asked as ' double cream,' or under some 
 fancy name. 
 
 There is no standard for cheese in this country, and con- 
 sequently a skimmed'milk cheese ■ containing only traces of 
 fat may be sold as cheese without any qualification, while 
 we are confronted with the anomaly that if margarine-fat is 
 added, it is an offence to sell the resulting product as cheese 
 unless its character is declared. 
 
 It is therefore evident that a standard for cheese should 
 be adopted, and the following would be a fair one, namely : 
 not less than 30 per cent, of true butter-fat, while no starch 
 or other extraneous matter should be present. Any cheese 
 that does not come up to this standard should be required 
 to be plainly marked and sold as 'prepared from milk, 
 from which a portion of the fat has been removed.' 
 
 The Analysis of Cheese. — ^Water is estimated by drying 
 5 grammes of the sample in thin slices at a temperature of 
 105° C. till constant in weight. 
 
 The ash is estimated on the above by igniting at as low 
 as possible a temperature. It will vary very largely 
 (without taking into consideration the added salt) according 
 to the amount of acidity that was permitted in the process 
 of manufacture, the higher the acidity the more calcium 
 salts will have been dissolved and run off with the whey. 
 
 The proteids are obtained by multiplying the nitrogen 
 figure by the factor 6"3.^ 
 
 Fat. — Many methods may be erhployed to estimate the 
 fat in cheese. After having given some considerable atten- 
 tion to the matter, we prefer to use one of the two following 
 methods : 
 
 . Ether Extraction Process. — Fifty grammes of the cheese is 
 ground up in a mortar with a fairly large quantity of sand. 
 The powder so obtained is placed in a tall stoppered 
 
128 THE ANALYSIS OF' MILK AND MILK-PRODUCTS 
 
 cylinder, and extracted by means of four successive portions 
 of ether, using in all about 500 c.c. The ether- washings, 
 are then made up to a definite volume and an aliquot 
 portion taken, the ether evaporated, and the residual fat 
 ■weighed in the usual way. 
 
 Mechanical Process. — When it is merely necessary to 
 estimate the fat, it can be quickly and accurately deter-, 
 mined by means of the following modification of the. 
 Leffmann-Beam process for milk : 
 
 Two grammes of the cheese are taken and reduced to as 
 fine a state of division as possible ; this is then transferred 
 to a small dish and treated on the water-bath with 30 c.c. 
 concentrated hydrochloric acid until solution is effected and 
 the solution is of a dark purplish colour, The mixture is 
 then poured into a Leffmann-Beam bottle, and the dish 
 rinsed with the HCl-fusel-oil mixture into the bottle, 
 and, finally, enough of the hot strong acid added to fill 
 the bottle to the mark. It is then centrifugated for one 
 minute. 
 
 The Leffmann-Beam bottles are graduated so that ten 
 divisions equal I'O per cent, by weight on the 15'55 
 grammes ( = 15 c.c.) of milk taken. 
 
 It follows therefore that the factor, in order to make use 
 of the bottle, will be : 
 
 15-55 „ „„ 
 
 ^-=7-77 
 
 With very little practice concordant readings are easily 
 obtained, which agree with the ether extraction process 
 already explained. 
 
 To obtain fat on which to do further work, it is generally 
 sufficient to chop up about 50 grammes of the sample, 
 which is hung up in a muslin bag in the water-bath ; the 
 fat will generally run out clear. The fat can also be 
 
CHEESE, 129 
 
 obtained from the remainder of the ether by evaporation in 
 the process already given. 
 
 The fat should be examined by means of the Eeichert 
 process, as described under butter, to prove that it is true 
 milk-fat. Starch should be tested for by taking a portion 
 of the fat-free residue, boiling with water, filtering, and 
 testing the filtrate for starch, by means of a dilute solution 
 of iodine in potassium iodide. No blue coloration should 
 be obtained. 
 
INDEX 
 
 Abnobmal milks, 19 
 Accidents of cheese-making, 111 
 Acetic acid test for butter, 92 
 Acidity in cheese-making, 108 
 
 ,, in milk, 3 
 Action of heat on milk, 5 
 Adams' method, 30 
 Adulteration of butter, 88 
 
 „ of condensed milk, 82 
 
 ,, of cheese, 127 
 
 „ of cream, 65 
 
 ,, of mUk, 19 
 
 Age of cows, influence of, 10 
 Albuminoids, estimation of, 35 
 American Cheddar, 113 
 Amphoteric reaction, 3 
 Aniline dyes in butter, 103 
 Artificial souring of milk, 113 
 Ash of butter, 89 
 „ of cheese, 127 
 „ of condensed milk, 79 
 ,, of milk, 25 
 Ass's milk, 9 
 
 Bacteria in butter, 103 
 
 „ in cheese, 109 
 
 „ in mUk, 51-65 
 Bang, on the tuberculosis of cattle, 
 
 62 
 Bell's method, 30 
 Blue milk, 56 
 Bondon cheese, 118 
 Borax, 41 
 Boric acid, estimation of, 41 
 
 „ tests for, 41 
 
 Brewers' grains, 11 
 Butter, adulteration of, 88 
 
 „ analysis of, 85 
 
 „ bacteria in, 103 
 
 Butter colourings, 103 
 
 „ composition of, 85 
 
 ,, curd in, 89 
 
 ,, foreign fat in, 88 
 
 „ -milk, 20 
 
 ,, preservatives in, 103 
 
 ,, salt in, 89 
 
 „ water in, 88 
 
 Calculation method, 33 
 Camembert cheese, 114 
 Centrifugal machines, 27, 28 
 Cheddar cheese, 114 
 Cheese, adulteration of, 127 
 „ American, 113 
 ,, analysis of, 127 
 
 ash of, 127 
 „ Bondon, 118 
 ,, Camembert, 114 
 „ composition of, 106 
 ,, Cheddar, 114 
 ,, Cheshire, 116 
 ,, cream, 117 
 ,, Derbyshire, 118 
 
 Dutch, 119 
 ,, Gloucester, 119 
 ,, Gorgonzola, 120 
 „ Gruyfere, 120 
 ,, Leicester, 120 
 „ margarine, 125 
 „ Parmesan, 122 
 „ Eoquefort, 122 
 „ ripening of, 108 
 ,, skim nulk, 122 
 „ Stilton, 122 
 „ Wensleydale, 124 
 Citric acid in Milk, 7 
 Colostrum, 8 
 Composition of milk, 4 
 
INDEX 
 
 131 
 
 Condensed milk, adulteration of, 
 82 
 ,, ,j analysis of, 79 
 
 „ ,, ash of, 79 
 
 ,, ,, composition of, 
 
 71 
 ,, ,, dilution of, 74, 
 
 78 
 ,, ,, fat in, 83 
 
 „ ,, humanised, 83 
 
 ,, „ milk-sugar in, 
 
 80 
 „ „ varieties, 70 
 
 Cream, adulteration of, 65 
 ,, analysis of, 67 
 ,, composition of, 65 
 „ cheese, 117 
 „ Devonshire, 64 
 ,, fat in, 66 
 ,, gelatine in, 67- 
 ,, separated, 66 
 
 Dairy reforms, 64 
 Danger of tuberculous milk, 60 
 Derby cheese, 118 
 Diseases of milk, 53 
 
 ,, propagated by milk, 59 
 Dutch cheese, 119 
 
 Effect of keeping on butter, 101 
 Efi&ciency of preservatives, 42 
 Estimation of fat in butter, 88 
 „ ,, in cheese, 127 
 
 ,, „ in condensed 
 
 mUk, 83 
 ,, ,, in cream, 66 
 
 „ ,, in milk, 25 
 
 Ewe milk, 9 
 
 Fat globules, 2 
 Pat of milk, 25 
 Fehling's te&t, 37 
 ruled cheese, 125 
 Foreign fat in butter, 88 
 Formalin, tests for, 45 
 Frozen milk, 20 
 
 Gelatine in cream, 67 
 Gerber machine, 28 
 Gloucester cheese, 119 
 Goat-mUk, 9 
 Gooch's method, 44 
 Gorgonzola cheese, 120 
 Gravity of butter-fat, 98 
 
 Gravity of margarine-fat, 98 
 
 of milk, 21 
 Gruyere cheese, 120 
 
 Hehner's test for formalin, 47 
 ,, process for soL and insol. 
 
 acids, 92 
 Human mUk, 8 
 Humanised condensed milk, 83 
 
 Influence of breed on milk, 10 
 lodme absorption of butter-fat, 
 100 
 
 Keeping of butter, effect of, 101 
 Kephir, 21 
 
 Kjeldahl's process, 35 
 Koumiss, 21 
 
 Lactalbumen, 5 
 Lactic acid fermentation, 3 
 Lactose, 6, 87 
 Lactoscope, Feser, 21 
 Leflmann-Beam machine, 27 
 Leicester cheese, 120 
 
 Macfarlane's method, 31 
 Margarine, 87 
 Margarine cheese, 125 
 
 ,, detection of, in butter, 
 84 
 Mechanical methods of fat estima- 
 tion, 26 
 Microscopical examination of 
 
 butter, 101 
 MUk, adulteration of, 19 
 analysis of, 21 
 .ash, 7 
 ass's, 9 
 cow's, 1 
 epidemics, 62 
 fat, 4 
 proteids, 4 
 standards, 17 
 sugar, 6 
 
 Nitrogen, estimation of, 35 
 
 Nuclein, 5 
 
 Number of butter samples ex- 
 amined by public analysts, 87 
 
 Number of cheese samples 
 examined by public analysts, 
 126 
 
 Number of samples of condensed 
 9—2 
 
132 
 
 INDEX 
 
 by public 
 
 milk examined 
 analysts, 69 
 Number of milk samples examined 
 by public analysts, 12 
 
 Oleo-refractometer, 100 
 ,, margarine, 87 
 
 Pavy's process, 37 
 
 Peptonising organisms in milk, 
 
 55 
 Polarimetric method for sugar, 
 
 38 
 Potassium chromate in milk, 49 
 Preservatives, 41 
 Proteids in milk, 35 
 Public analysts' standard for mUk, 
 
 18 
 Pure cultures used in dairy work, 
 
 105 
 
 Eapid souring of milk, 113 
 Bechnagel's phenomenon, 22 
 Bed mUk, 56 
 Beichert process, 89 
 Bennet, 111 
 
 „ tests for strength of, 112 
 Biohmond's slide-rule, 34 
 Bipening of cheese, 108 
 Bopy milk, 53 
 
 Salicylic acid in butter, 103 
 
 ,, „ milk, 49 
 
 Sampling of milk, 22 
 Schiiff's test for formalin, 46 
 Separated milk, 19 
 Separator, 65 
 Sesame-oil, test for, 102 
 Skimmed milk, 19 
 Soapy milk, 53 
 Sow's milk, 9 
 Specific gravity of butter-fat, 98 
 
 Specific gravity of margarine fat, 
 98 
 
 „ ,, of milk, 21 
 
 Standard for cheese, 127 
 
 ,, „ condensed milk, 83 
 
 ,, „ cream, 66 
 
 „ „ milk, 17 
 
 Starter in cheese-making, 113 
 Sterilization of milk, 64 
 StUton cheese, 122 
 Stringy mUk, 53 
 
 Sugar, estimation of, in condensed 
 milk, 80 
 ,, ,, in mOk, 37 , 
 
 Table of corrected gravities for 
 
 temperature, 23 
 Testing of rennet, 112 
 Tichborne's experiments on butter 
 
 analysis, 97 
 Total solids of milk, 24 
 Tubercle bacillus in butter, 104 
 
 ,, „ in milk, 59 
 
 Tyrotoxicon, 64, 109 
 
 Unsweetened condensed milk, 77 
 
 Valenta test, 91 
 
 Variations in composition of milk, 
 11 
 
 Water, importance of purity of, 63 
 
 ,, in butter, 88 
 
 ,, in cheese, 127 
 Werner-Schmidt, 31 
 Wensleydale cheese, 124 
 Westphal balance, 22 
 
 Yellow milk, 57 
 
 Ziehl's method for tubercle bacil- 
 lus, 59 
 
Microscopes Suitable for every Class of Research. 
 
 FIJffEST QUALITY WOUKMANSIUP ONLY. 
 
 PERFECTION IN EVERY DETAIL. 
 
 WATSON a SONS' 
 
 Edinburgh Student's Microscope. 
 
 This well-known instrument is made in four different 
 forms, and can be had with either the Tripod or Horse- 
 shoe pattern of foot. It is miequalled for stability 
 and precision of working parts by any other Student's 
 or medium-sized Microscope. 
 
 Prices -From S3 lOs. 
 
 ' H ' Tripod Edinburgh Student's Microscope, as 
 figured. Price £9 lOs. 
 
 Dr. Henri ^an Heurck's Microscope, 
 
 FOR HIGH POWER WORK AND PHOTO- 
 MICROGRAPHY. 
 
 Made exduswel/y ly W. WATSON & SONS, to tl,e 
 
 specification of t)r. HENMI VAN HEVliCK, 
 
 of Antwerp. 
 
 Fitted with Fine Adjustments of utmost sensitiveness 
 
 and precision, not liable to derangement by wear. 
 Has Backwork Draw-tube to correct objectiTes for thick- 
 ness of cover-glass. Can be used with either Continental 
 or English Objective^. 
 Has Fine Adjustment to Subatage. 
 The Stand is specially designed to afford the greatest 
 possible convenience for manipulation. 
 
 PRICE.— As figured (but without centring screws or 
 divisions to Stage), with one eye-piece, £18 lOs. 
 
 PHOTO-MICROGRAPHIC APPARATUS of the most 
 approved designs. Vertical, Horizontal, and for 
 Instantaneous work. 
 MICROSCOPES, OBJECTIVES AND ACCESSORIES 
 Of every description. 
 Full niuatrated Catalogue of Microaeopes and Jppamtus (No. 2), containing full particulars of the 
 above, forwarded post free on application. 
 MICROSCOPIC OBJECTS— W. WATSON & SONS issue a CLASHIFIED LIST (No. 3), reweserding 
 a stock of over 0-0,000 first-dass preparations, forwarded free on application to 
 
 OPTICIANS TO H.M. GOVERNMENT, 
 
 313. Hieli Holborn, london, W.C, ENGLAND ; and 78, Swanston Street, 
 Melbourne, AUSTEALIA. 
 
 Awarded the GRAND PRIX for Microscopes at the Antwerp International Exhibition, 1894 ; Five 
 
 H™t Awards at the World's Fair, Chicago 1893 ; Two Gold Medals, Pans Universal B^itation, 
 
 1889 ; & Thirty-two, Gold & other Medals at the principal International Exlubitions of the Woria. 
 
 £:sa7.A>.EIX<IS^X:X3 1837. 
 
HEARSON'S PATENT 
 
 BIOLOGICAL INCUBATORS, 
 
 For the Cultivation of Micro-Organisms, 
 
 For researches in Embryology and Histology, for 
 
 preparing and embedding objects for Section 
 
 Cutting, and for regulating Chronometers. 
 
 CHAS. HEARSON & CO., LIMITED, 
 
 235, RE6EST STREET, LOOON, W. 
 
 Descriptive Pamphlet and Prices on application.